INSTALL?=install
RM ?= rm -f
DOC_REF = origin/man
+HTML_REF = origin/html
infodir?=$(prefix)/share/info
MAKEINFO=makeinfo
quick-install:
sh ./install-doc-quick.sh $(DOC_REF) $(DESTDIR)$(mandir)
+quick-install-html:
+ sh ./install-doc-quick.sh $(HTML_REF) $(DESTDIR)$(htmldir)
+
.PHONY: .FORCE-GIT-VERSION-FILE
* Many commands did not use the correct working tree location when used
with GIT_WORK_TREE environment settings.
+* Some systems needs to use compatibility fnmach and regex libraries
+ independent from each other; the compat/ area has been reorganized to
+ allow this.
* "git apply --unidiff-zero" incorrectly applied a -U0 patch that inserts
a new line before the second line.
+* "git blame -c" did not exactly work like "git annotate" when range
+ boundaries are involved.
+
+* "git checkout file" when file is still unmerged checked out contents from
+ a random high order stage, which was confusing.
+
* "git clone $there $here/" with extra trailing slashes after explicit
local directory name $here did not work as expected.
+* "git diff" on tracked contents with CRLF line endings did not drive "less"
+ intelligently when showing added or removed lines.
+
* "git diff --dirstat -M" did not add changes in subdirectories up
correctly for renamed paths.
+* "git diff --cumulative" did not imply "--dirstat".
+
* "git for-each-ref refs/heads/" did not work as expected.
-* "git log --grep=pattern -i" did not ignore case.
+* "git gui" allowed users to feed patch without any context to be applied.
+
+* "git gui" botched parsing "diff" output when a line that begins with two
+ dashes and a space gets removed or a line that begins with two pluses
+ and a space gets added.
+
+* "git gui" translation updates and i18n fixes.
+
+* "git index-pack" is more careful against disk corruption while completing
+ a thin pack.
+
+* "git log -i --grep=pattern" did not ignore case; neither "git log -E
+ --grep=pattern" triggered extended regexp.
* "git log --pretty="%ad" --date=short" did not use short format when
showing the timestamp.
+* "git log --author=author" match incorrectly matched with the
+ timestamp part of "author " line in commit objects.
+
+* "git log -F --author=author" did not work at all.
+
* Build procedure for "git shell" that used stub versions of some
functions and globals was not understood by linkers on some platforms.
+* "git stash" was fooled by a stat-dirty but otherwise unmodified paths
+ and refused to work until the user refreshed the index.
+
+* "git svn" was broken on Perl before 5.8 with recent fixes to reduce
+ use of temporary files.
+
* "git verify-pack -v" did not work correctly when given more than one
packfile.
--
exec >/var/tmp/1
-O=v1.6.0.1-49-g6a42cfe
+O=v1.6.0.1-78-g3632cfc
echo O=$(git describe maint)
git shortlog --no-merges $O..maint
(subsystems)
-* ...
+* gitk can call out to git-gui to view "git blame" output; git-gui in turn
+ can run gitk from its blame view.
(portability)
* "git push --tags --all $there" failed with generic usage message without
telling saying these two options are incompatible.
+* "git log --author/--committer" match used to potentially match the
+ timestamp part, exposing internal implementation detail. Also these did
+ not work with --fixed-strings match at all.
+
--
exec >/var/tmp/1
-O=v1.6.0.1-215-g9b8ae93
+O=v1.6.0.1-266-gaf9552f
echo O=$(git describe master)
git shortlog --no-merges $O..master ^maint
Values 'man', 'info', 'web' and 'html' are supported. 'man' is
the default. 'web' and 'html' are the same.
+help.autocorrect::
+ Automatically correct and execute mistyped commands after
+ waiting for the given number of deciseconds (0.1 sec). If more
+ than one command can be deduced from the entered text, nothing
+ will be executed. If the value of this option is negative,
+ the corrected command will be executed immediately. If the
+ value is 0 - the command will be just shown but not executed.
+ This is the default.
+
http.proxy::
Override the HTTP proxy, normally configured using the 'http_proxy'
environment variable (see linkgit:curl[1]). This can be overridden
Annotates each line in the given file with information from the commit
which introduced the line. Optionally annotate from a given revision.
+The only difference between this command and linkgit:git-blame[1] is that
+they use slightly different output formats, and this command exists only
+for backward compatibility to support existing scripts, and provide more
+familiar command name for people coming from other SCM systems.
+
OPTIONS
-------
include::blame-options.txt[]
0 nothing nothing nothing (does not happen)
1 nothing nothing exists use M
2 nothing exists nothing remove path from index
- 3 nothing exists exists use M
+ 3 nothing exists exists, use M if "initial checkout"
+ H == M keep index otherwise
+ exists fail
+ H != M
clean I==H I==M
------------------
merge, but it would not show in `git diff-index --cached $M`
output after two-tree merge.
+Case #3 is slightly tricky and needs explanation. The result from this
+rule logically should be to remove the path if the user staged the removal
+of the path and then swiching to a new branch. That however will prevent
+the initial checkout from happening, so the rule is modified to use M (new
+tree) only when the contents of the index is empty. Otherwise the removal
+of the path is kept as long as $H and $M are the same.
3-Way Merge
~~~~~~~~~~~
Note about konqueror
--------------------
-When 'konqueror' is specified by the a command line option or a
+When 'konqueror' is specified by a command line option or a
configuration variable, we launch 'kfmclient' to try to open the HTML
man page on an already opened konqueror in a new tab if possible.
branch of the `git.git` repository.
Documentation for older releases are available here:
-* link:v1.6.1/git.html[documentation for release 1.6.1]
+* link:v1.6.0.2/git.html[documentation for release 1.6.0.2]
* release notes for
- link:RelNotes-1.6.1.txt[1.6.1],
+ link:RelNotes-1.6.0.2.txt[1.6.0.2],
+ link:RelNotes-1.6.0.1.txt[1.6.0.1],
link:RelNotes-1.6.0.txt[1.6.0].
* link:v1.5.6.5/git.html[documentation for release 1.5.6.5]
- `bibtex` suitable for files with BibTeX coded references.
-- `java` suitable for source code in the Java lanugage.
+- `html` suitable for HTML/XHTML documents.
+
+- `java` suitable for source code in the Java language.
- `pascal` suitable for source code in the Pascal/Delphi language.
+- `php` suitable for source code in the PHP language.
+
- `python` suitable for source code in the Python language.
- `ruby` suitable for source code in the Ruby language.
- `tex` suitable for source code for LaTeX documents.
-- `html` suitable for HTML/XHTML documents.
-
Performing a three-way merge
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
merge.stat::
- Whether to print the diffstat between ORIG_HEAD and merge result
+ Whether to print the diffstat between ORIG_HEAD and the merge result
at the end of the merge. True by default.
merge.log::
#!/bin/sh
GVF=GIT-VERSION-FILE
-DEF_VER=v1.6.0.GIT
+DEF_VER=v1.6.0.2.GIT
LF='
'
to do a global install, you can do
$ make prefix=/usr all doc info ;# as yourself
- # make prefix=/usr install install-doc install-info ;# as root
+ # make prefix=/usr install install-doc install-html install-info ;# as root
(or prefix=/usr/local, of course). Just like any program suite
that uses $prefix, the built results have some paths encoded,
$ make configure ;# as yourself
$ ./configure --prefix=/usr ;# as yourself
$ make all doc ;# as yourself
- # make install install-doc ;# as root
+ # make install install-doc install-html;# as root
Issues of note:
inclined to install the tools, the default build target
("make all") does _not_ build them.
+ "make doc" builds documentation in man and html formats; there are
+ also "make man", "make html" and "make info". Note that "make html"
+ requires asciidoc, but not xmlto. "make man" (and thus make doc)
+ requires both.
+
+ "make install-doc" installs documentation in man format only; there
+ are also "make install-man", "make install-html" and "make
+ install-info".
+
Building and installing the info file additionally requires
makeinfo and docbook2X. Version 0.8.3 is known to work.
The documentation is written for AsciiDoc 7, but "make
ASCIIDOC8=YesPlease doc" will let you format with AsciiDoc 8.
- Alternatively, pre-formatted documentation are available in
+ Alternatively, pre-formatted documentation is available in
"html" and "man" branches of the git repository itself. For
example, you could:
http://www.kernel.org/pub/software/scm/git/docs/
+ There are also "make quick-install-doc" and "make quick-install-html"
+ which install preformatted man pages and html documentation.
+ This does not require asciidoc/xmlto, but it only works from within
+ a cloned checkout of git.git with these two extra branches, and will
+ not work for the maintainer for obvious chicken-and-egg reasons.
+
It has been reported that docbook-xsl version 1.72 and 1.73 are
buggy; 1.72 misformats manual pages for callouts, and 1.73 needs
the patch in contrib/patches/docbook-xsl-manpages-charmap.patch
PROGRAMS += git-patch-id$X
PROGRAMS += git-receive-pack$X
PROGRAMS += git-send-pack$X
+PROGRAMS += git-shell$X
PROGRAMS += git-show-index$X
PROGRAMS += git-unpack-file$X
PROGRAMS += git-update-server-info$X
LIB_H += grep.h
LIB_H += hash.h
LIB_H += help.h
+LIB_H += levenshtein.h
LIB_H += list-objects.h
LIB_H += ll-merge.h
LIB_H += log-tree.h
LIB_OBJS += help.o
LIB_OBJS += ident.o
LIB_OBJS += interpolate.o
+LIB_OBJS += levenshtein.o
LIB_OBJS += list-objects.o
LIB_OBJS += ll-merge.o
LIB_OBJS += lockfile.o
endif
NO_STRLCPY = YesPlease
NO_MEMMEM = YesPlease
+ COMPAT_CFLAGS += -Icompat/regex
+ COMPAT_OBJS += compat/regex/regex.o
endif
ifeq ($(uname_S),SunOS)
NEEDS_SOCKET = YesPlease
BASIC_LDFLAGS += -L/usr/local/lib
DIR_HAS_BSD_GROUP_SEMANTICS = YesPlease
THREADED_DELTA_SEARCH = YesPlease
+ COMPAT_CFLAGS += -Icompat/regex
+ COMPAT_OBJS += compat/regex/regex.o
endif
ifeq ($(uname_S),OpenBSD)
NO_STRCASESTR = YesPlease
NEEDS_LIBICONV = YesPlease
endif
BASIC_CFLAGS += -I/usr/pkg/include
- BASIC_LDFLAGS += -L/usr/pkg/lib
- ALL_LDFLAGS += -Wl,-rpath,/usr/pkg/lib
+ BASIC_LDFLAGS += -L/usr/pkg/lib $(CC_LD_DYNPATH)/usr/pkg/lib
THREADED_DELTA_SEARCH = YesPlease
endif
ifeq ($(uname_S),AIX)
INTERNAL_QSORT = UnfortunatelyYes
NEEDS_LIBICONV=YesPlease
BASIC_CFLAGS += -D_LARGE_FILES
+ COMPAT_CFLAGS += -Icompat/regex
+ COMPAT_OBJS += compat/regex/regex.o
endif
ifeq ($(uname_S),GNU)
# GNU/Hurd
NO_PERL_MAKEMAKER = YesPlease
NO_POSIX_ONLY_PROGRAMS = YesPlease
NO_ST_BLOCKS_IN_STRUCT_STAT = YesPlease
- COMPAT_CFLAGS += -D__USE_MINGW_ACCESS -DNOGDI -Icompat
+ COMPAT_CFLAGS += -D__USE_MINGW_ACCESS -DNOGDI -Icompat -Icompat/regex -Icompat/fnmatch
COMPAT_CFLAGS += -DSNPRINTF_SIZE_CORR=1
COMPAT_CFLAGS += -DSTRIP_EXTENSION=\".exe\"
- COMPAT_OBJS += compat/mingw.o compat/fnmatch.o compat/regex.o compat/winansi.o
+ COMPAT_OBJS += compat/mingw.o compat/fnmatch/fnmatch.o compat/regex/regex.o compat/winansi.o
EXTLIBS += -lws2_32
X = .exe
gitexecdir = ../libexec/git-core
endif
endif
-ifdef NO_R_TO_GCC_LINKER
- # Some gcc does not accept and pass -R to the linker to specify
- # the runtime dynamic library path.
- CC_LD_DYNPATH = -Wl,-rpath=
-else
- CC_LD_DYNPATH = -R
+ifndef CC_LD_DYNPATH
+ ifdef NO_R_TO_GCC_LINKER
+ # Some gcc does not accept and pass -R to the linker to specify
+ # the runtime dynamic library path.
+ CC_LD_DYNPATH = -Wl,-rpath,
+ else
+ CC_LD_DYNPATH = -R
+ endif
endif
ifdef NO_CURL
ifndef NO_POSIX_ONLY_PROGRAMS
PROGRAMS += git-daemon$X
PROGRAMS += git-imap-send$X
- PROGRAMS += git-shell$X
endif
ifndef NO_OPENSSL
OPENSSL_LIBSSL = -lssl
doc:
$(MAKE) -C Documentation all
+man:
+ $(MAKE) -C Documentation man
+
+html:
+ $(MAKE) -C Documentation html
+
info:
$(MAKE) -C Documentation info
$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(bindir_SQ)'
$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(gitexec_instdir_SQ)'
$(INSTALL) $(ALL_PROGRAMS) '$(DESTDIR_SQ)$(gitexec_instdir_SQ)'
- $(INSTALL) git$X git-upload-pack$X git-receive-pack$X git-upload-archive$X git-shell$X '$(DESTDIR_SQ)$(bindir_SQ)'
+ $(INSTALL) git$X git-upload-pack$X git-receive-pack$X git-upload-archive$X git-shell$X git-cvsserver '$(DESTDIR_SQ)$(bindir_SQ)'
$(MAKE) -C templates DESTDIR='$(DESTDIR_SQ)' install
$(MAKE) -C perl prefix='$(prefix_SQ)' DESTDIR='$(DESTDIR_SQ)' install
ifndef NO_TCLTK
quick-install-doc:
$(MAKE) -C Documentation quick-install
+quick-install-html:
+ $(MAKE) -C Documentation quick-install-html
+
### Maintainer's dist rules
static int reverse;
static int blank_boundary;
static int incremental;
-static int cmd_is_annotate;
static int xdl_opts = XDF_NEED_MINIMAL;
static struct string_list mailmap;
if (suspect->commit->object.flags & UNINTERESTING) {
if (blank_boundary)
memset(hex, ' ', length);
- else if (!cmd_is_annotate) {
+ else if (!(opt & OUTPUT_ANNOTATE_COMPAT)) {
length--;
putchar('^');
}
};
struct parse_opt_ctx_t ctx;
-
- cmd_is_annotate = !strcmp(argv[0], "annotate");
+ int cmd_is_annotate = !strcmp(argv[0], "annotate");
git_config(git_blame_config, NULL);
init_revisions(&revs, NULL);
parse_done:
argc = parse_options_end(&ctx);
+ if (cmd_is_annotate)
+ output_option |= OUTPUT_ANNOTATE_COMPAT;
+
if (DIFF_OPT_TST(&revs.diffopt, FIND_COPIES_HARDER))
opt |= (PICKAXE_BLAME_COPY | PICKAXE_BLAME_MOVE |
PICKAXE_BLAME_COPY_HARDER);
*
* Careful: order of argument flags does matter. For example,
*
- * git-checkout-index -a -f file.c
+ * git checkout-index -a -f file.c
*
* Will first check out all files listed in the cache (but not
* overwrite any old ones), and then force-checkout "file.c" a
* second time (ie that one _will_ overwrite any old contents
* with the same filename).
*
- * Also, just doing "git-checkout-index" does nothing. You probably
- * meant "git-checkout-index -a". And if you want to force it, you
- * want "git-checkout-index -f -a".
+ * Also, just doing "git checkout-index" does nothing. You probably
+ * meant "git checkout-index -a". And if you want to force it, you
+ * want "git checkout-index -f -a".
*
* Intuitiveness is not the goal here. Repeatability is. The
* reason for the "no arguments means no work" thing is that
* from scripts you are supposed to be able to do things like
*
- * find . -name '*.h' -print0 | xargs -0 git-checkout-index -f --
+ * find . -name '*.h' -print0 | xargs -0 git checkout-index -f --
*
* or:
*
- * find . -name '*.h' -print0 | git-checkout-index -f -z --stdin
+ * find . -name '*.h' -print0 | git checkout-index -f -z --stdin
*
* which will force all existing *.h files to be replaced with
* their cached copies. If an empty command line implied "all",
}
if (!state.quiet) {
- fprintf(stderr, "git-checkout-index: %s ", name);
+ fprintf(stderr, "git checkout-index: %s ", name);
if (!has_same_name)
fprintf(stderr, "is not in the cache");
else if (checkout_stage)
return 0;
}
+static int skip_same_name(struct cache_entry *ce, int pos)
+{
+ while (++pos < active_nr &&
+ !strcmp(active_cache[pos]->name, ce->name))
+ ; /* skip */
+ return pos;
+}
+
+
static int checkout_paths(struct tree *source_tree, const char **pathspec)
{
int pos;
if (report_path_error(ps_matched, pathspec, 0))
return 1;
+ /* Any unmerged paths? */
+ for (pos = 0; pos < active_nr; pos++) {
+ struct cache_entry *ce = active_cache[pos];
+ if (pathspec_match(pathspec, NULL, ce->name, 0)) {
+ if (!ce_stage(ce))
+ continue;
+ errs = 1;
+ error("path '%s' is unmerged", ce->name);
+ pos = skip_same_name(ce, pos) - 1;
+ }
+ }
+ if (errs)
+ return 1;
+
/* Now we are committed to check them out */
memset(&state, 0, sizeof(state));
state.force = 1;
for (pos = 0; pos < active_nr; pos++) {
struct cache_entry *ce = active_cache[pos];
if (pathspec_match(pathspec, NULL, ce->name, 0)) {
- errs |= checkout_entry(ce, &state, NULL);
+ if (!ce_stage(ce)) {
+ errs |= checkout_entry(ce, &state, NULL);
+ continue;
+ }
+ pos = skip_same_name(ce, pos) - 1;
}
}
}
/* 2-way merge to the new branch */
+ topts.initial_checkout = (!active_nr &&
+ (old->commit == new->commit));
topts.update = 1;
topts.merge = 1;
topts.gently = opts->merge;
typename(expect));
}
-static const char commit_tree_usage[] = "git-commit-tree <sha1> [-p <sha1>]* < changelog";
+static const char commit_tree_usage[] = "git commit-tree <sha1> [-p <sha1>]* < changelog";
static void new_parent(struct commit *parent, struct commit_list **parents_p)
{
}
else
printf("%lu objects, %lu kilobytes\n",
- loose, loose_size / 2);
+ loose, loose_size / 1024);
return 0;
}
line[len-1] = 0;
if (get_sha1_hex(line, sha1))
return -1;
- obj = lookup_object(sha1);
- obj = obj ? obj : parse_object(sha1);
+ obj = lookup_unknown_object(sha1);
+ if (!obj || !obj->parsed)
+ obj = parse_object(sha1);
if (!obj)
return -1;
if (obj->type == OBJ_COMMIT)
if (!ret && nr_heads) {
/* If the heads to pull were given, we should have
* consumed all of them by matching the remote.
- * Otherwise, 'git-fetch remote no-such-ref' would
+ * Otherwise, 'git fetch remote no-such-ref' would
* silently succeed without issuing an error.
*/
for (i = 0; i < nr_heads; i++)
/*
* Not fetched to a tracking branch? We need to fetch
* it anyway to allow this branch's "branch.$name.merge"
- * to be honored by git-pull, but we do not have to
+ * to be honored by 'git pull', but we do not have to
* fail if branch.$name.merge is misconfigured to point
* at a nonexisting branch. If we were indeed called by
- * git-pull, it will notice the misconfiguration because
+ * 'git pull', it will notice the misconfiguration because
* there is no entry in the resulting FETCH_HEAD marked
* for merging.
*/
* The refs we are going to fetch are in to_fetch (nr_heads in
* total). If running
*
- * $ git-rev-list --objects to_fetch[0] to_fetch[1] ... --not --all
+ * $ git rev-list --objects to_fetch[0] to_fetch[1] ... --not --all
*
* does not error out, that means everything reachable from the
* refs we are going to fetch exists and is connected to some of
arg++;
}
if (argc < arg + 2 - commits_on_stdin) {
- usage("git-http-fetch [-c] [-t] [-a] [-v] [--recover] [-w ref] [--stdin] commit-id url");
+ usage("git http-fetch [-c] [-t] [-a] [-v] [--recover] [-w ref] [--stdin] commit-id url");
return 1;
}
if (commits_on_stdin) {
}
url = argv[arg];
if (url && url[strlen(url)-1] != '/') {
- rewritten_url = malloc(strlen(url)+2);
+ rewritten_url = xmalloc(strlen(url)+2);
strcpy(rewritten_url, url);
strcat(rewritten_url, "/");
url = rewritten_url;
fprintf(stderr,
"Some loose object were found to be corrupt, but they might be just\n"
"a false '404 Not Found' error message sent with incorrect HTTP\n"
-"status code. Suggest running git-fsck.\n");
+"status code. Suggest running 'git fsck'.\n");
}
walker_free(walker);
/* Note: if ".git/hooks" file exists in the repository being
* re-initialized, /etc/core-git/templates/hooks/update would
- * cause git-init to fail here. I think this is sane but
+ * cause "git init" to fail here. I think this is sane but
* it means that the set of templates we ship by default, along
* with the way the namespace under .git/ is organized, should
* be really carefully chosen.
#endif
static const char pack_usage[] = "\
-git-pack-objects [{ -q | --progress | --all-progress }] \n\
+git pack-objects [{ -q | --progress | --all-progress }] \n\
[--max-pack-size=N] [--local] [--incremental] \n\
[--window=N] [--window-memory=N] [--depth=N] \n\
[--no-reuse-delta] [--no-reuse-object] [--delta-base-offset] \n\
/*
* Compare the objects in the offset order, in order to emulate the
- * "git-rev-list --objects" output that produced the pack originally.
+ * "git rev-list --objects" output that produced the pack originally.
*/
static int ofscmp(const void *a_, const void *b_)
{
}
-static const char read_tree_usage[] = "git
-read-tree (<sha> | [[-m [--trivial] [--aggressive] | --reset | --prefix=<prefix>] [-u | -i]] [--exclude-per-directory=<gitignore>] [--index-output=<file>] <sha1> [<sha2> [<sha3>]])";
+static const char read_tree_usage[] = "git
read-tree (<sha> | [[-m [--trivial] [--aggressive] | --reset | --prefix=<prefix>] [-u | -i]] [--exclude-per-directory=<gitignore>] [--index-output=<file>] <sha1> [<sha2> [<sha3>]])";
static struct lock_file lock_file;
break;
case 2:
opts.fn = twoway_merge;
+ opts.initial_checkout = !active_nr;
break;
case 3:
default:
static void show_object(struct object_array_entry *p)
{
/* An object with name "foo\n0000000..." can be used to
- * confuse downstream git-pack-objects very badly.
+ * confuse downstream "git pack-objects" very badly.
*/
const char *ep = strchr(p->name, '\n');
"from both the file and the HEAD\n"
"(use -f to force removal)", name);
else if (!index_only) {
- /* It's not dangerous to git-rm --cached a
+ /* It's not dangerous to "git rm --cached" a
* file if the index matches the file or the
* HEAD, since it means the deleted content is
* still available somewhere.
po.out = fd;
po.git_cmd = 1;
if (start_command(&po))
- die("git-pack-objects failed (%s)", strerror(errno));
+ die("git pack-objects failed (%s)", strerror(errno));
/*
* We feed the pack-objects we just spawned with revision
static const char tar_tree_usage[] =
"git tar-tree [--remote=<repo>] <tree-ish> [basedir]\n"
-"*** Note that this command is now deprecated; use git-archive instead.";
+"*** Note that this command is now deprecated; use \"git archive\" instead.";
int cmd_tar_tree(int argc, const char **argv, const char *prefix)
{
/*
- * git-tar-tree is now a wrapper around git-archive --format=tar
+ * "git tar-tree" is now a wrapper around "git archive --format=tar"
*
* $0 --remote=<repo> arg... ==>
- * git-archive --format=tar --remote=<repo> arg...
+ * git archive --format=tar --remote=<repo> arg...
* $0 tree-ish ==>
- * git-archive --format=tar tree-ish
+ * git archive --format=tar tree-ish
* $0 tree-ish basedir ==>
- * git-archive --format-tar --prefix=basedir tree-ish
+ * git archive --format-tar --prefix=basedir tree-ish
*/
int i;
const char **nargv = xcalloc(sizeof(*nargv), argc + 2);
char *basedir_arg;
int nargc = 0;
- nargv[nargc++] = "git-archive";
+ nargv[nargc++] = "archive";
nargv[nargc++] = "--format=tar";
if (2 <= argc && !prefixcmp(argv[1], "--remote=")) {
nargv[nargc] = NULL;
fprintf(stderr,
- "*** git-tar-tree is now deprecated.\n"
- "*** Running git-archive instead.\n***");
+ "*** \"git tar-tree\" is now deprecated.\n"
+ "*** Running \"git archive\" instead.\n***");
for (i = 0; i < nargc; i++) {
fputc(' ', stderr);
sq_quote_print(stderr, nargv[i]);
#include "fsck.h"
static int dry_run, quiet, recover, has_errors, strict;
-static const char unpack_usage[] = "git-unpack-objects [-n] [-q] [-r] [--strict] < pack-file";
+static const char unpack_usage[] = "git unpack-objects [-n] [-q] [-r] [--strict] < pack-file";
/* We always read in 4kB chunks. */
static unsigned char buffer[4096];
* Default to not allowing changes to the list of files. The
* tool doesn't actually care, but this makes it harder to add
* files to the revision control by mistake by doing something
- * like "git-update-index *" and suddenly having all the object
+ * like "git update-index *" and suddenly having all the object
* files be revision controlled.
*/
static int allow_add;
/* This reads lines formatted in one of three formats:
*
* (1) mode SP sha1 TAB path
- * The first format is what "git-apply --index-info"
+ * The first format is what "git apply --index-info"
* reports, and used to reconstruct a partial tree
* that is used for phony merge base tree when falling
* back on 3-way merge.
*
* (2) mode SP type SP sha1 TAB path
- * The second format is to stuff git-ls-tree output
+ * The second format is to stuff "git ls-tree" output
* into the index file.
*
* (3) mode SP sha1 SP stage TAB path
* This format is to put higher order stages into the
- * index file and matches git-ls-files --stage output.
+ * index file and matches "git ls-files --stage" output.
*/
errno = 0;
ul = strtoul(buf.buf, &ptr, 8);
extern const char git_more_info_string[];
extern void list_common_cmds_help(void);
-extern void help_unknown_cmd(const char *cmd);
+extern const char *help_unknown_cmd(const char *cmd);
extern void prune_packed_objects(int);
extern int read_line_with_nul(char *buf, int size, FILE *file);
extern int fmt_merge_msg(int merge_summary, struct strbuf *in,
return (isalpha(ch) || ch == '_' || ch == '$');
}
+static void show_line_to_eol(const char *line, int len, const char *reset)
+{
+ int saw_cr_at_eol = 0;
+ if (len < 0)
+ len = strlen(line);
+ saw_cr_at_eol = (len && line[len-1] == '\r');
+
+ printf("%.*s%s%s\n", len - saw_cr_at_eol, line,
+ reset,
+ saw_cr_at_eol ? "\r" : "");
+}
+
static void dump_sline(struct sline *sline, unsigned long cnt, int num_parent,
int use_color)
{
else
putchar(' ');
}
- printf("%s%s\n", ll->line, c_reset);
+ show_line_to_eol(ll->line, -1, c_reset);
ll = ll->next;
}
if (cnt < lno)
putchar(' ');
p_mask <<= 1;
}
- printf("%.*s%s\n", sl->len, sl->bol, c_reset);
+ show_line_to_eol(sl->bol, sl->len, c_reset);
}
}
}
+++ /dev/null
-/* Copyright (C) 1991, 92, 93, 96, 97, 98, 99 Free Software Foundation, Inc.
- This file is part of the GNU C Library.
-
- This library is free software; you can redistribute it and/or
- modify it under the terms of the GNU Library General Public License as
- published by the Free Software Foundation; either version 2 of the
- License, or (at your option) any later version.
-
- This library is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- Library General Public License for more details.
-
- You should have received a copy of the GNU Library General Public
- License along with this library; see the file COPYING.LIB. If not,
- write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
- Boston, MA 02111-1307, USA. */
-
-#if HAVE_CONFIG_H
-# include <config.h>
-#endif
-
-/* Enable GNU extensions in fnmatch.h. */
-#ifndef _GNU_SOURCE
-# define _GNU_SOURCE 1
-#endif
-
-#include <errno.h>
-#include <fnmatch.h>
-#include <ctype.h>
-
-#if HAVE_STRING_H || defined _LIBC
-# include <string.h>
-#else
-# include <strings.h>
-#endif
-
-#if defined STDC_HEADERS || defined _LIBC
-# include <stdlib.h>
-#endif
-
-/* For platform which support the ISO C amendement 1 functionality we
- support user defined character classes. */
-#if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H)
-/* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. */
-# include <wchar.h>
-# include <wctype.h>
-#endif
-
-/* Comment out all this code if we are using the GNU C Library, and are not
- actually compiling the library itself. This code is part of the GNU C
- Library, but also included in many other GNU distributions. Compiling
- and linking in this code is a waste when using the GNU C library
- (especially if it is a shared library). Rather than having every GNU
- program understand `configure --with-gnu-libc' and omit the object files,
- it is simpler to just do this in the source for each such file. */
-
-#if defined _LIBC || !defined __GNU_LIBRARY__
-
-
-# if defined STDC_HEADERS || !defined isascii
-# define ISASCII(c) 1
-# else
-# define ISASCII(c) isascii(c)
-# endif
-
-# ifdef isblank
-# define ISBLANK(c) (ISASCII (c) && isblank (c))
-# else
-# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
-# endif
-# ifdef isgraph
-# define ISGRAPH(c) (ISASCII (c) && isgraph (c))
-# else
-# define ISGRAPH(c) (ISASCII (c) && isprint (c) && !isspace (c))
-# endif
-
-# define ISPRINT(c) (ISASCII (c) && isprint (c))
-# define ISDIGIT(c) (ISASCII (c) && isdigit (c))
-# define ISALNUM(c) (ISASCII (c) && isalnum (c))
-# define ISALPHA(c) (ISASCII (c) && isalpha (c))
-# define ISCNTRL(c) (ISASCII (c) && iscntrl (c))
-# define ISLOWER(c) (ISASCII (c) && islower (c))
-# define ISPUNCT(c) (ISASCII (c) && ispunct (c))
-# define ISSPACE(c) (ISASCII (c) && isspace (c))
-# define ISUPPER(c) (ISASCII (c) && isupper (c))
-# define ISXDIGIT(c) (ISASCII (c) && isxdigit (c))
-
-# define STREQ(s1, s2) ((strcmp (s1, s2) == 0))
-
-# if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H)
-/* The GNU C library provides support for user-defined character classes
- and the functions from ISO C amendement 1. */
-# ifdef CHARCLASS_NAME_MAX
-# define CHAR_CLASS_MAX_LENGTH CHARCLASS_NAME_MAX
-# else
-/* This shouldn't happen but some implementation might still have this
- problem. Use a reasonable default value. */
-# define CHAR_CLASS_MAX_LENGTH 256
-# endif
-
-# ifdef _LIBC
-# define IS_CHAR_CLASS(string) __wctype (string)
-# else
-# define IS_CHAR_CLASS(string) wctype (string)
-# endif
-# else
-# define CHAR_CLASS_MAX_LENGTH 6 /* Namely, `xdigit'. */
-
-# define IS_CHAR_CLASS(string) \
- (STREQ (string, "alpha") || STREQ (string, "upper") \
- || STREQ (string, "lower") || STREQ (string, "digit") \
- || STREQ (string, "alnum") || STREQ (string, "xdigit") \
- || STREQ (string, "space") || STREQ (string, "print") \
- || STREQ (string, "punct") || STREQ (string, "graph") \
- || STREQ (string, "cntrl") || STREQ (string, "blank"))
-# endif
-
-/* Avoid depending on library functions or files
- whose names are inconsistent. */
-
-# if !defined _LIBC && !defined getenv
-extern char *getenv ();
-# endif
-
-# ifndef errno
-extern int errno;
-# endif
-
-/* This function doesn't exist on most systems. */
-
-# if !defined HAVE___STRCHRNUL && !defined _LIBC
-static char *
-__strchrnul (s, c)
- const char *s;
- int c;
-{
- char *result = strchr (s, c);
- if (result == NULL)
- result = strchr (s, '\0');
- return result;
-}
-# endif
-
-# ifndef internal_function
-/* Inside GNU libc we mark some function in a special way. In other
- environments simply ignore the marking. */
-# define internal_function
-# endif
-
-/* Match STRING against the filename pattern PATTERN, returning zero if
- it matches, nonzero if not. */
-static int internal_fnmatch __P ((const char *pattern, const char *string,
- int no_leading_period, int flags))
- internal_function;
-static int
-internal_function
-internal_fnmatch (pattern, string, no_leading_period, flags)
- const char *pattern;
- const char *string;
- int no_leading_period;
- int flags;
-{
- register const char *p = pattern, *n = string;
- register unsigned char c;
-
-/* Note that this evaluates C many times. */
-# ifdef _LIBC
-# define FOLD(c) ((flags & FNM_CASEFOLD) ? tolower (c) : (c))
-# else
-# define FOLD(c) ((flags & FNM_CASEFOLD) && ISUPPER (c) ? tolower (c) : (c))
-# endif
-
- while ((c = *p++) != '\0')
- {
- c = FOLD (c);
-
- switch (c)
- {
- case '?':
- if (*n == '\0')
- return FNM_NOMATCH;
- else if (*n == '/' && (flags & FNM_FILE_NAME))
- return FNM_NOMATCH;
- else if (*n == '.' && no_leading_period
- && (n == string
- || (n[-1] == '/' && (flags & FNM_FILE_NAME))))
- return FNM_NOMATCH;
- break;
-
- case '\\':
- if (!(flags & FNM_NOESCAPE))
- {
- c = *p++;
- if (c == '\0')
- /* Trailing \ loses. */
- return FNM_NOMATCH;
- c = FOLD (c);
- }
- if (FOLD ((unsigned char) *n) != c)
- return FNM_NOMATCH;
- break;
-
- case '*':
- if (*n == '.' && no_leading_period
- && (n == string
- || (n[-1] == '/' && (flags & FNM_FILE_NAME))))
- return FNM_NOMATCH;
-
- for (c = *p++; c == '?' || c == '*'; c = *p++)
- {
- if (*n == '/' && (flags & FNM_FILE_NAME))
- /* A slash does not match a wildcard under FNM_FILE_NAME. */
- return FNM_NOMATCH;
- else if (c == '?')
- {
- /* A ? needs to match one character. */
- if (*n == '\0')
- /* There isn't another character; no match. */
- return FNM_NOMATCH;
- else
- /* One character of the string is consumed in matching
- this ? wildcard, so *??? won't match if there are
- less than three characters. */
- ++n;
- }
- }
-
- if (c == '\0')
- /* The wildcard(s) is/are the last element of the pattern.
- If the name is a file name and contains another slash
- this does mean it cannot match. */
- return ((flags & FNM_FILE_NAME) && strchr (n, '/') != NULL
- ? FNM_NOMATCH : 0);
- else
- {
- const char *endp;
-
- endp = __strchrnul (n, (flags & FNM_FILE_NAME) ? '/' : '\0');
-
- if (c == '[')
- {
- int flags2 = ((flags & FNM_FILE_NAME)
- ? flags : (flags & ~FNM_PERIOD));
-
- for (--p; n < endp; ++n)
- if (internal_fnmatch (p, n,
- (no_leading_period
- && (n == string
- || (n[-1] == '/'
- && (flags
- & FNM_FILE_NAME)))),
- flags2)
- == 0)
- return 0;
- }
- else if (c == '/' && (flags & FNM_FILE_NAME))
- {
- while (*n != '\0' && *n != '/')
- ++n;
- if (*n == '/'
- && (internal_fnmatch (p, n + 1, flags & FNM_PERIOD,
- flags) == 0))
- return 0;
- }
- else
- {
- int flags2 = ((flags & FNM_FILE_NAME)
- ? flags : (flags & ~FNM_PERIOD));
-
- if (c == '\\' && !(flags & FNM_NOESCAPE))
- c = *p;
- c = FOLD (c);
- for (--p; n < endp; ++n)
- if (FOLD ((unsigned char) *n) == c
- && (internal_fnmatch (p, n,
- (no_leading_period
- && (n == string
- || (n[-1] == '/'
- && (flags
- & FNM_FILE_NAME)))),
- flags2) == 0))
- return 0;
- }
- }
-
- /* If we come here no match is possible with the wildcard. */
- return FNM_NOMATCH;
-
- case '[':
- {
- /* Nonzero if the sense of the character class is inverted. */
- static int posixly_correct;
- register int not;
- char cold;
-
- if (posixly_correct == 0)
- posixly_correct = getenv ("POSIXLY_CORRECT") != NULL ? 1 : -1;
-
- if (*n == '\0')
- return FNM_NOMATCH;
-
- if (*n == '.' && no_leading_period && (n == string
- || (n[-1] == '/'
- && (flags
- & FNM_FILE_NAME))))
- return FNM_NOMATCH;
-
- if (*n == '/' && (flags & FNM_FILE_NAME))
- /* `/' cannot be matched. */
- return FNM_NOMATCH;
-
- not = (*p == '!' || (posixly_correct < 0 && *p == '^'));
- if (not)
- ++p;
-
- c = *p++;
- for (;;)
- {
- unsigned char fn = FOLD ((unsigned char) *n);
-
- if (!(flags & FNM_NOESCAPE) && c == '\\')
- {
- if (*p == '\0')
- return FNM_NOMATCH;
- c = FOLD ((unsigned char) *p);
- ++p;
-
- if (c == fn)
- goto matched;
- }
- else if (c == '[' && *p == ':')
- {
- /* Leave room for the null. */
- char str[CHAR_CLASS_MAX_LENGTH + 1];
- size_t c1 = 0;
-# if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H)
- wctype_t wt;
-# endif
- const char *startp = p;
-
- for (;;)
- {
- if (c1 == CHAR_CLASS_MAX_LENGTH)
- /* The name is too long and therefore the pattern
- is ill-formed. */
- return FNM_NOMATCH;
-
- c = *++p;
- if (c == ':' && p[1] == ']')
- {
- p += 2;
- break;
- }
- if (c < 'a' || c >= 'z')
- {
- /* This cannot possibly be a character class name.
- Match it as a normal range. */
- p = startp;
- c = '[';
- goto normal_bracket;
- }
- str[c1++] = c;
- }
- str[c1] = '\0';
-
-# if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H)
- wt = IS_CHAR_CLASS (str);
- if (wt == 0)
- /* Invalid character class name. */
- return FNM_NOMATCH;
-
- if (__iswctype (__btowc ((unsigned char) *n), wt))
- goto matched;
-# else
- if ((STREQ (str, "alnum") && ISALNUM ((unsigned char) *n))
- || (STREQ (str, "alpha") && ISALPHA ((unsigned char) *n))
- || (STREQ (str, "blank") && ISBLANK ((unsigned char) *n))
- || (STREQ (str, "cntrl") && ISCNTRL ((unsigned char) *n))
- || (STREQ (str, "digit") && ISDIGIT ((unsigned char) *n))
- || (STREQ (str, "graph") && ISGRAPH ((unsigned char) *n))
- || (STREQ (str, "lower") && ISLOWER ((unsigned char) *n))
- || (STREQ (str, "print") && ISPRINT ((unsigned char) *n))
- || (STREQ (str, "punct") && ISPUNCT ((unsigned char) *n))
- || (STREQ (str, "space") && ISSPACE ((unsigned char) *n))
- || (STREQ (str, "upper") && ISUPPER ((unsigned char) *n))
- || (STREQ (str, "xdigit") && ISXDIGIT ((unsigned char) *n)))
- goto matched;
-# endif
- }
- else if (c == '\0')
- /* [ (unterminated) loses. */
- return FNM_NOMATCH;
- else
- {
- normal_bracket:
- if (FOLD (c) == fn)
- goto matched;
-
- cold = c;
- c = *p++;
-
- if (c == '-' && *p != ']')
- {
- /* It is a range. */
- unsigned char cend = *p++;
- if (!(flags & FNM_NOESCAPE) && cend == '\\')
- cend = *p++;
- if (cend == '\0')
- return FNM_NOMATCH;
-
- if (cold <= fn && fn <= FOLD (cend))
- goto matched;
-
- c = *p++;
- }
- }
-
- if (c == ']')
- break;
- }
-
- if (!not)
- return FNM_NOMATCH;
- break;
-
- matched:
- /* Skip the rest of the [...] that already matched. */
- while (c != ']')
- {
- if (c == '\0')
- /* [... (unterminated) loses. */
- return FNM_NOMATCH;
-
- c = *p++;
- if (!(flags & FNM_NOESCAPE) && c == '\\')
- {
- if (*p == '\0')
- return FNM_NOMATCH;
- /* XXX 1003.2d11 is unclear if this is right. */
- ++p;
- }
- else if (c == '[' && *p == ':')
- {
- do
- if (*++p == '\0')
- return FNM_NOMATCH;
- while (*p != ':' || p[1] == ']');
- p += 2;
- c = *p;
- }
- }
- if (not)
- return FNM_NOMATCH;
- }
- break;
-
- default:
- if (c != FOLD ((unsigned char) *n))
- return FNM_NOMATCH;
- }
-
- ++n;
- }
-
- if (*n == '\0')
- return 0;
-
- if ((flags & FNM_LEADING_DIR) && *n == '/')
- /* The FNM_LEADING_DIR flag says that "foo*" matches "foobar/frobozz". */
- return 0;
-
- return FNM_NOMATCH;
-
-# undef FOLD
-}
-
-
-int
-fnmatch (pattern, string, flags)
- const char *pattern;
- const char *string;
- int flags;
-{
- return internal_fnmatch (pattern, string, flags & FNM_PERIOD, flags);
-}
-
-#endif /* _LIBC or not __GNU_LIBRARY__. */
+++ /dev/null
-/* Copyright (C) 1991, 92, 93, 96, 97, 98, 99 Free Software Foundation, Inc.
- This file is part of the GNU C Library.
-
- The GNU C Library is free software; you can redistribute it and/or
- modify it under the terms of the GNU Library General Public License as
- published by the Free Software Foundation; either version 2 of the
- License, or (at your option) any later version.
-
- The GNU C Library is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- Library General Public License for more details.
-
- You should have received a copy of the GNU Library General Public
- License along with the GNU C Library; see the file COPYING.LIB. If not,
- write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
- Boston, MA 02111-1307, USA. */
-
-#ifndef _FNMATCH_H
-#define _FNMATCH_H 1
-
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-#if defined __cplusplus || (defined __STDC__ && __STDC__) || defined WINDOWS32
-# if !defined __GLIBC__ || !defined __P
-# undef __P
-# define __P(protos) protos
-# endif
-#else /* Not C++ or ANSI C. */
-# undef __P
-# define __P(protos) ()
-/* We can get away without defining `const' here only because in this file
- it is used only inside the prototype for `fnmatch', which is elided in
- non-ANSI C where `const' is problematical. */
-#endif /* C++ or ANSI C. */
-
-#ifndef const
-# if (defined __STDC__ && __STDC__) || defined __cplusplus
-# define __const const
-# else
-# define __const
-# endif
-#endif
-
-/* We #undef these before defining them because some losing systems
- (HP-UX A.08.07 for example) define these in <unistd.h>. */
-#undef FNM_PATHNAME
-#undef FNM_NOESCAPE
-#undef FNM_PERIOD
-
-/* Bits set in the FLAGS argument to `fnmatch'. */
-#define FNM_PATHNAME (1 << 0) /* No wildcard can ever match `/'. */
-#define FNM_NOESCAPE (1 << 1) /* Backslashes don't quote special chars. */
-#define FNM_PERIOD (1 << 2) /* Leading `.' is matched only explicitly. */
-
-#if !defined _POSIX_C_SOURCE || _POSIX_C_SOURCE < 2 || defined _GNU_SOURCE
-# define FNM_FILE_NAME FNM_PATHNAME /* Preferred GNU name. */
-# define FNM_LEADING_DIR (1 << 3) /* Ignore `/...' after a match. */
-# define FNM_CASEFOLD (1 << 4) /* Compare without regard to case. */
-#endif
-
-/* Value returned by `fnmatch' if STRING does not match PATTERN. */
-#define FNM_NOMATCH 1
-
-/* This value is returned if the implementation does not support
- `fnmatch'. Since this is not the case here it will never be
- returned but the conformance test suites still require the symbol
- to be defined. */
-#ifdef _XOPEN_SOURCE
-# define FNM_NOSYS (-1)
-#endif
-
-/* Match NAME against the filename pattern PATTERN,
- returning zero if it matches, FNM_NOMATCH if not. */
-extern int fnmatch __P ((__const char *__pattern, __const char *__name,
- int __flags));
-
-#ifdef __cplusplus
-}
-#endif
-
-#endif /* fnmatch.h */
--- /dev/null
+/* Copyright (C) 1991, 92, 93, 96, 97, 98, 99 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ This library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Library General Public License as
+ published by the Free Software Foundation; either version 2 of the
+ License, or (at your option) any later version.
+
+ This library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Library General Public License for more details.
+
+ You should have received a copy of the GNU Library General Public
+ License along with this library; see the file COPYING.LIB. If not,
+ write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ Boston, MA 02111-1307, USA. */
+
+#if HAVE_CONFIG_H
+# include <config.h>
+#endif
+
+/* Enable GNU extensions in fnmatch.h. */
+#ifndef _GNU_SOURCE
+# define _GNU_SOURCE 1
+#endif
+
+#include <errno.h>
+#include <fnmatch.h>
+#include <ctype.h>
+
+#if HAVE_STRING_H || defined _LIBC
+# include <string.h>
+#else
+# include <strings.h>
+#endif
+
+#if defined STDC_HEADERS || defined _LIBC
+# include <stdlib.h>
+#endif
+
+/* For platform which support the ISO C amendement 1 functionality we
+ support user defined character classes. */
+#if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H)
+/* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. */
+# include <wchar.h>
+# include <wctype.h>
+#endif
+
+/* Comment out all this code if we are using the GNU C Library, and are not
+ actually compiling the library itself. This code is part of the GNU C
+ Library, but also included in many other GNU distributions. Compiling
+ and linking in this code is a waste when using the GNU C library
+ (especially if it is a shared library). Rather than having every GNU
+ program understand `configure --with-gnu-libc' and omit the object files,
+ it is simpler to just do this in the source for each such file. */
+
+#if defined _LIBC || !defined __GNU_LIBRARY__
+
+
+# if defined STDC_HEADERS || !defined isascii
+# define ISASCII(c) 1
+# else
+# define ISASCII(c) isascii(c)
+# endif
+
+# ifdef isblank
+# define ISBLANK(c) (ISASCII (c) && isblank (c))
+# else
+# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
+# endif
+# ifdef isgraph
+# define ISGRAPH(c) (ISASCII (c) && isgraph (c))
+# else
+# define ISGRAPH(c) (ISASCII (c) && isprint (c) && !isspace (c))
+# endif
+
+# define ISPRINT(c) (ISASCII (c) && isprint (c))
+# define ISDIGIT(c) (ISASCII (c) && isdigit (c))
+# define ISALNUM(c) (ISASCII (c) && isalnum (c))
+# define ISALPHA(c) (ISASCII (c) && isalpha (c))
+# define ISCNTRL(c) (ISASCII (c) && iscntrl (c))
+# define ISLOWER(c) (ISASCII (c) && islower (c))
+# define ISPUNCT(c) (ISASCII (c) && ispunct (c))
+# define ISSPACE(c) (ISASCII (c) && isspace (c))
+# define ISUPPER(c) (ISASCII (c) && isupper (c))
+# define ISXDIGIT(c) (ISASCII (c) && isxdigit (c))
+
+# define STREQ(s1, s2) ((strcmp (s1, s2) == 0))
+
+# if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H)
+/* The GNU C library provides support for user-defined character classes
+ and the functions from ISO C amendement 1. */
+# ifdef CHARCLASS_NAME_MAX
+# define CHAR_CLASS_MAX_LENGTH CHARCLASS_NAME_MAX
+# else
+/* This shouldn't happen but some implementation might still have this
+ problem. Use a reasonable default value. */
+# define CHAR_CLASS_MAX_LENGTH 256
+# endif
+
+# ifdef _LIBC
+# define IS_CHAR_CLASS(string) __wctype (string)
+# else
+# define IS_CHAR_CLASS(string) wctype (string)
+# endif
+# else
+# define CHAR_CLASS_MAX_LENGTH 6 /* Namely, `xdigit'. */
+
+# define IS_CHAR_CLASS(string) \
+ (STREQ (string, "alpha") || STREQ (string, "upper") \
+ || STREQ (string, "lower") || STREQ (string, "digit") \
+ || STREQ (string, "alnum") || STREQ (string, "xdigit") \
+ || STREQ (string, "space") || STREQ (string, "print") \
+ || STREQ (string, "punct") || STREQ (string, "graph") \
+ || STREQ (string, "cntrl") || STREQ (string, "blank"))
+# endif
+
+/* Avoid depending on library functions or files
+ whose names are inconsistent. */
+
+# if !defined _LIBC && !defined getenv
+extern char *getenv ();
+# endif
+
+# ifndef errno
+extern int errno;
+# endif
+
+/* This function doesn't exist on most systems. */
+
+# if !defined HAVE___STRCHRNUL && !defined _LIBC
+static char *
+__strchrnul (s, c)
+ const char *s;
+ int c;
+{
+ char *result = strchr (s, c);
+ if (result == NULL)
+ result = strchr (s, '\0');
+ return result;
+}
+# endif
+
+# ifndef internal_function
+/* Inside GNU libc we mark some function in a special way. In other
+ environments simply ignore the marking. */
+# define internal_function
+# endif
+
+/* Match STRING against the filename pattern PATTERN, returning zero if
+ it matches, nonzero if not. */
+static int internal_fnmatch __P ((const char *pattern, const char *string,
+ int no_leading_period, int flags))
+ internal_function;
+static int
+internal_function
+internal_fnmatch (pattern, string, no_leading_period, flags)
+ const char *pattern;
+ const char *string;
+ int no_leading_period;
+ int flags;
+{
+ register const char *p = pattern, *n = string;
+ register unsigned char c;
+
+/* Note that this evaluates C many times. */
+# ifdef _LIBC
+# define FOLD(c) ((flags & FNM_CASEFOLD) ? tolower (c) : (c))
+# else
+# define FOLD(c) ((flags & FNM_CASEFOLD) && ISUPPER (c) ? tolower (c) : (c))
+# endif
+
+ while ((c = *p++) != '\0')
+ {
+ c = FOLD (c);
+
+ switch (c)
+ {
+ case '?':
+ if (*n == '\0')
+ return FNM_NOMATCH;
+ else if (*n == '/' && (flags & FNM_FILE_NAME))
+ return FNM_NOMATCH;
+ else if (*n == '.' && no_leading_period
+ && (n == string
+ || (n[-1] == '/' && (flags & FNM_FILE_NAME))))
+ return FNM_NOMATCH;
+ break;
+
+ case '\\':
+ if (!(flags & FNM_NOESCAPE))
+ {
+ c = *p++;
+ if (c == '\0')
+ /* Trailing \ loses. */
+ return FNM_NOMATCH;
+ c = FOLD (c);
+ }
+ if (FOLD ((unsigned char) *n) != c)
+ return FNM_NOMATCH;
+ break;
+
+ case '*':
+ if (*n == '.' && no_leading_period
+ && (n == string
+ || (n[-1] == '/' && (flags & FNM_FILE_NAME))))
+ return FNM_NOMATCH;
+
+ for (c = *p++; c == '?' || c == '*'; c = *p++)
+ {
+ if (*n == '/' && (flags & FNM_FILE_NAME))
+ /* A slash does not match a wildcard under FNM_FILE_NAME. */
+ return FNM_NOMATCH;
+ else if (c == '?')
+ {
+ /* A ? needs to match one character. */
+ if (*n == '\0')
+ /* There isn't another character; no match. */
+ return FNM_NOMATCH;
+ else
+ /* One character of the string is consumed in matching
+ this ? wildcard, so *??? won't match if there are
+ less than three characters. */
+ ++n;
+ }
+ }
+
+ if (c == '\0')
+ /* The wildcard(s) is/are the last element of the pattern.
+ If the name is a file name and contains another slash
+ this does mean it cannot match. */
+ return ((flags & FNM_FILE_NAME) && strchr (n, '/') != NULL
+ ? FNM_NOMATCH : 0);
+ else
+ {
+ const char *endp;
+
+ endp = __strchrnul (n, (flags & FNM_FILE_NAME) ? '/' : '\0');
+
+ if (c == '[')
+ {
+ int flags2 = ((flags & FNM_FILE_NAME)
+ ? flags : (flags & ~FNM_PERIOD));
+
+ for (--p; n < endp; ++n)
+ if (internal_fnmatch (p, n,
+ (no_leading_period
+ && (n == string
+ || (n[-1] == '/'
+ && (flags
+ & FNM_FILE_NAME)))),
+ flags2)
+ == 0)
+ return 0;
+ }
+ else if (c == '/' && (flags & FNM_FILE_NAME))
+ {
+ while (*n != '\0' && *n != '/')
+ ++n;
+ if (*n == '/'
+ && (internal_fnmatch (p, n + 1, flags & FNM_PERIOD,
+ flags) == 0))
+ return 0;
+ }
+ else
+ {
+ int flags2 = ((flags & FNM_FILE_NAME)
+ ? flags : (flags & ~FNM_PERIOD));
+
+ if (c == '\\' && !(flags & FNM_NOESCAPE))
+ c = *p;
+ c = FOLD (c);
+ for (--p; n < endp; ++n)
+ if (FOLD ((unsigned char) *n) == c
+ && (internal_fnmatch (p, n,
+ (no_leading_period
+ && (n == string
+ || (n[-1] == '/'
+ && (flags
+ & FNM_FILE_NAME)))),
+ flags2) == 0))
+ return 0;
+ }
+ }
+
+ /* If we come here no match is possible with the wildcard. */
+ return FNM_NOMATCH;
+
+ case '[':
+ {
+ /* Nonzero if the sense of the character class is inverted. */
+ static int posixly_correct;
+ register int not;
+ char cold;
+
+ if (posixly_correct == 0)
+ posixly_correct = getenv ("POSIXLY_CORRECT") != NULL ? 1 : -1;
+
+ if (*n == '\0')
+ return FNM_NOMATCH;
+
+ if (*n == '.' && no_leading_period && (n == string
+ || (n[-1] == '/'
+ && (flags
+ & FNM_FILE_NAME))))
+ return FNM_NOMATCH;
+
+ if (*n == '/' && (flags & FNM_FILE_NAME))
+ /* `/' cannot be matched. */
+ return FNM_NOMATCH;
+
+ not = (*p == '!' || (posixly_correct < 0 && *p == '^'));
+ if (not)
+ ++p;
+
+ c = *p++;
+ for (;;)
+ {
+ unsigned char fn = FOLD ((unsigned char) *n);
+
+ if (!(flags & FNM_NOESCAPE) && c == '\\')
+ {
+ if (*p == '\0')
+ return FNM_NOMATCH;
+ c = FOLD ((unsigned char) *p);
+ ++p;
+
+ if (c == fn)
+ goto matched;
+ }
+ else if (c == '[' && *p == ':')
+ {
+ /* Leave room for the null. */
+ char str[CHAR_CLASS_MAX_LENGTH + 1];
+ size_t c1 = 0;
+# if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H)
+ wctype_t wt;
+# endif
+ const char *startp = p;
+
+ for (;;)
+ {
+ if (c1 == CHAR_CLASS_MAX_LENGTH)
+ /* The name is too long and therefore the pattern
+ is ill-formed. */
+ return FNM_NOMATCH;
+
+ c = *++p;
+ if (c == ':' && p[1] == ']')
+ {
+ p += 2;
+ break;
+ }
+ if (c < 'a' || c >= 'z')
+ {
+ /* This cannot possibly be a character class name.
+ Match it as a normal range. */
+ p = startp;
+ c = '[';
+ goto normal_bracket;
+ }
+ str[c1++] = c;
+ }
+ str[c1] = '\0';
+
+# if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H)
+ wt = IS_CHAR_CLASS (str);
+ if (wt == 0)
+ /* Invalid character class name. */
+ return FNM_NOMATCH;
+
+ if (__iswctype (__btowc ((unsigned char) *n), wt))
+ goto matched;
+# else
+ if ((STREQ (str, "alnum") && ISALNUM ((unsigned char) *n))
+ || (STREQ (str, "alpha") && ISALPHA ((unsigned char) *n))
+ || (STREQ (str, "blank") && ISBLANK ((unsigned char) *n))
+ || (STREQ (str, "cntrl") && ISCNTRL ((unsigned char) *n))
+ || (STREQ (str, "digit") && ISDIGIT ((unsigned char) *n))
+ || (STREQ (str, "graph") && ISGRAPH ((unsigned char) *n))
+ || (STREQ (str, "lower") && ISLOWER ((unsigned char) *n))
+ || (STREQ (str, "print") && ISPRINT ((unsigned char) *n))
+ || (STREQ (str, "punct") && ISPUNCT ((unsigned char) *n))
+ || (STREQ (str, "space") && ISSPACE ((unsigned char) *n))
+ || (STREQ (str, "upper") && ISUPPER ((unsigned char) *n))
+ || (STREQ (str, "xdigit") && ISXDIGIT ((unsigned char) *n)))
+ goto matched;
+# endif
+ }
+ else if (c == '\0')
+ /* [ (unterminated) loses. */
+ return FNM_NOMATCH;
+ else
+ {
+ normal_bracket:
+ if (FOLD (c) == fn)
+ goto matched;
+
+ cold = c;
+ c = *p++;
+
+ if (c == '-' && *p != ']')
+ {
+ /* It is a range. */
+ unsigned char cend = *p++;
+ if (!(flags & FNM_NOESCAPE) && cend == '\\')
+ cend = *p++;
+ if (cend == '\0')
+ return FNM_NOMATCH;
+
+ if (cold <= fn && fn <= FOLD (cend))
+ goto matched;
+
+ c = *p++;
+ }
+ }
+
+ if (c == ']')
+ break;
+ }
+
+ if (!not)
+ return FNM_NOMATCH;
+ break;
+
+ matched:
+ /* Skip the rest of the [...] that already matched. */
+ while (c != ']')
+ {
+ if (c == '\0')
+ /* [... (unterminated) loses. */
+ return FNM_NOMATCH;
+
+ c = *p++;
+ if (!(flags & FNM_NOESCAPE) && c == '\\')
+ {
+ if (*p == '\0')
+ return FNM_NOMATCH;
+ /* XXX 1003.2d11 is unclear if this is right. */
+ ++p;
+ }
+ else if (c == '[' && *p == ':')
+ {
+ do
+ if (*++p == '\0')
+ return FNM_NOMATCH;
+ while (*p != ':' || p[1] == ']');
+ p += 2;
+ c = *p;
+ }
+ }
+ if (not)
+ return FNM_NOMATCH;
+ }
+ break;
+
+ default:
+ if (c != FOLD ((unsigned char) *n))
+ return FNM_NOMATCH;
+ }
+
+ ++n;
+ }
+
+ if (*n == '\0')
+ return 0;
+
+ if ((flags & FNM_LEADING_DIR) && *n == '/')
+ /* The FNM_LEADING_DIR flag says that "foo*" matches "foobar/frobozz". */
+ return 0;
+
+ return FNM_NOMATCH;
+
+# undef FOLD
+}
+
+
+int
+fnmatch (pattern, string, flags)
+ const char *pattern;
+ const char *string;
+ int flags;
+{
+ return internal_fnmatch (pattern, string, flags & FNM_PERIOD, flags);
+}
+
+#endif /* _LIBC or not __GNU_LIBRARY__. */
--- /dev/null
+/* Copyright (C) 1991, 92, 93, 96, 97, 98, 99 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Library General Public License as
+ published by the Free Software Foundation; either version 2 of the
+ License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Library General Public License for more details.
+
+ You should have received a copy of the GNU Library General Public
+ License along with the GNU C Library; see the file COPYING.LIB. If not,
+ write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ Boston, MA 02111-1307, USA. */
+
+#ifndef _FNMATCH_H
+#define _FNMATCH_H 1
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#if defined __cplusplus || (defined __STDC__ && __STDC__) || defined WINDOWS32
+# if !defined __GLIBC__ || !defined __P
+# undef __P
+# define __P(protos) protos
+# endif
+#else /* Not C++ or ANSI C. */
+# undef __P
+# define __P(protos) ()
+/* We can get away without defining `const' here only because in this file
+ it is used only inside the prototype for `fnmatch', which is elided in
+ non-ANSI C where `const' is problematical. */
+#endif /* C++ or ANSI C. */
+
+#ifndef const
+# if (defined __STDC__ && __STDC__) || defined __cplusplus
+# define __const const
+# else
+# define __const
+# endif
+#endif
+
+/* We #undef these before defining them because some losing systems
+ (HP-UX A.08.07 for example) define these in <unistd.h>. */
+#undef FNM_PATHNAME
+#undef FNM_NOESCAPE
+#undef FNM_PERIOD
+
+/* Bits set in the FLAGS argument to `fnmatch'. */
+#define FNM_PATHNAME (1 << 0) /* No wildcard can ever match `/'. */
+#define FNM_NOESCAPE (1 << 1) /* Backslashes don't quote special chars. */
+#define FNM_PERIOD (1 << 2) /* Leading `.' is matched only explicitly. */
+
+#if !defined _POSIX_C_SOURCE || _POSIX_C_SOURCE < 2 || defined _GNU_SOURCE
+# define FNM_FILE_NAME FNM_PATHNAME /* Preferred GNU name. */
+# define FNM_LEADING_DIR (1 << 3) /* Ignore `/...' after a match. */
+# define FNM_CASEFOLD (1 << 4) /* Compare without regard to case. */
+#endif
+
+/* Value returned by `fnmatch' if STRING does not match PATTERN. */
+#define FNM_NOMATCH 1
+
+/* This value is returned if the implementation does not support
+ `fnmatch'. Since this is not the case here it will never be
+ returned but the conformance test suites still require the symbol
+ to be defined. */
+#ifdef _XOPEN_SOURCE
+# define FNM_NOSYS (-1)
+#endif
+
+/* Match NAME against the filename pattern PATTERN,
+ returning zero if it matches, FNM_NOMATCH if not. */
+extern int fnmatch __P ((__const char *__pattern, __const char *__name,
+ int __flags));
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* fnmatch.h */
+++ /dev/null
-/* Extended regular expression matching and search library,
- version 0.12.
- (Implements POSIX draft P10003.2/D11.2, except for
- internationalization features.)
-
- Copyright (C) 1993 Free Software Foundation, Inc.
-
- This program is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 2, or (at your option)
- any later version.
-
- This program is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- GNU General Public License for more details.
-
- You should have received a copy of the GNU General Public License
- along with this program; if not, write to the Free Software
- Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */
-
-/* AIX requires this to be the first thing in the file. */
-#if defined (_AIX) && !defined (REGEX_MALLOC)
- #pragma alloca
-#endif
-
-#define _GNU_SOURCE
-
-/* We need this for `regex.h', and perhaps for the Emacs include files. */
-#include <sys/types.h>
-
-/* We used to test for `BSTRING' here, but only GCC and Emacs define
- `BSTRING', as far as I know, and neither of them use this code. */
-#include <string.h>
-#ifndef bcmp
-#define bcmp(s1, s2, n) memcmp ((s1), (s2), (n))
-#endif
-#ifndef bcopy
-#define bcopy(s, d, n) memcpy ((d), (s), (n))
-#endif
-#ifndef bzero
-#define bzero(s, n) memset ((s), 0, (n))
-#endif
-
-#include <stdlib.h>
-
-
-/* Define the syntax stuff for \<, \>, etc. */
-
-/* This must be nonzero for the wordchar and notwordchar pattern
- commands in re_match_2. */
-#ifndef Sword
-#define Sword 1
-#endif
-
-#ifdef SYNTAX_TABLE
-
-extern char *re_syntax_table;
-
-#else /* not SYNTAX_TABLE */
-
-/* How many characters in the character set. */
-#define CHAR_SET_SIZE 256
-
-static char re_syntax_table[CHAR_SET_SIZE];
-
-static void
-init_syntax_once ()
-{
- register int c;
- static int done = 0;
-
- if (done)
- return;
-
- bzero (re_syntax_table, sizeof re_syntax_table);
-
- for (c = 'a'; c <= 'z'; c++)
- re_syntax_table[c] = Sword;
-
- for (c = 'A'; c <= 'Z'; c++)
- re_syntax_table[c] = Sword;
-
- for (c = '0'; c <= '9'; c++)
- re_syntax_table[c] = Sword;
-
- re_syntax_table['_'] = Sword;
-
- done = 1;
-}
-
-#endif /* not SYNTAX_TABLE */
-
-#define SYNTAX(c) re_syntax_table[c]
-
-\f
-/* Get the interface, including the syntax bits. */
-#include "regex.h"
-
-/* isalpha etc. are used for the character classes. */
-#include <ctype.h>
-
-#ifndef isascii
-#define isascii(c) 1
-#endif
-
-#ifdef isblank
-#define ISBLANK(c) (isascii (c) && isblank (c))
-#else
-#define ISBLANK(c) ((c) == ' ' || (c) == '\t')
-#endif
-#ifdef isgraph
-#define ISGRAPH(c) (isascii (c) && isgraph (c))
-#else
-#define ISGRAPH(c) (isascii (c) && isprint (c) && !isspace (c))
-#endif
-
-#define ISPRINT(c) (isascii (c) && isprint (c))
-#define ISDIGIT(c) (isascii (c) && isdigit (c))
-#define ISALNUM(c) (isascii (c) && isalnum (c))
-#define ISALPHA(c) (isascii (c) && isalpha (c))
-#define ISCNTRL(c) (isascii (c) && iscntrl (c))
-#define ISLOWER(c) (isascii (c) && islower (c))
-#define ISPUNCT(c) (isascii (c) && ispunct (c))
-#define ISSPACE(c) (isascii (c) && isspace (c))
-#define ISUPPER(c) (isascii (c) && isupper (c))
-#define ISXDIGIT(c) (isascii (c) && isxdigit (c))
-
-#ifndef NULL
-#define NULL 0
-#endif
-
-/* We remove any previous definition of `SIGN_EXTEND_CHAR',
- since ours (we hope) works properly with all combinations of
- machines, compilers, `char' and `unsigned char' argument types.
- (Per Bothner suggested the basic approach.) */
-#undef SIGN_EXTEND_CHAR
-#if __STDC__
-#define SIGN_EXTEND_CHAR(c) ((signed char) (c))
-#else /* not __STDC__ */
-/* As in Harbison and Steele. */
-#define SIGN_EXTEND_CHAR(c) ((((unsigned char) (c)) ^ 128) - 128)
-#endif
-\f
-/* Should we use malloc or alloca? If REGEX_MALLOC is not defined, we
- use `alloca' instead of `malloc'. This is because using malloc in
- re_search* or re_match* could cause memory leaks when C-g is used in
- Emacs; also, malloc is slower and causes storage fragmentation. On
- the other hand, malloc is more portable, and easier to debug.
-
- Because we sometimes use alloca, some routines have to be macros,
- not functions -- `alloca'-allocated space disappears at the end of the
- function it is called in. */
-
-#ifdef REGEX_MALLOC
-
-#define REGEX_ALLOCATE malloc
-#define REGEX_REALLOCATE(source, osize, nsize) realloc (source, nsize)
-
-#else /* not REGEX_MALLOC */
-
-/* Emacs already defines alloca, sometimes. */
-#ifndef alloca
-
-/* Make alloca work the best possible way. */
-#ifdef __GNUC__
-#define alloca __builtin_alloca
-#else /* not __GNUC__ */
-#if HAVE_ALLOCA_H
-#include <alloca.h>
-#else /* not __GNUC__ or HAVE_ALLOCA_H */
-#ifndef _AIX /* Already did AIX, up at the top. */
-char *alloca ();
-#endif /* not _AIX */
-#endif /* not HAVE_ALLOCA_H */
-#endif /* not __GNUC__ */
-
-#endif /* not alloca */
-
-#define REGEX_ALLOCATE alloca
-
-/* Assumes a `char *destination' variable. */
-#define REGEX_REALLOCATE(source, osize, nsize) \
- (destination = (char *) alloca (nsize), \
- bcopy (source, destination, osize), \
- destination)
-
-#endif /* not REGEX_MALLOC */
-
-
-/* True if `size1' is non-NULL and PTR is pointing anywhere inside
- `string1' or just past its end. This works if PTR is NULL, which is
- a good thing. */
-#define FIRST_STRING_P(ptr) \
- (size1 && string1 <= (ptr) && (ptr) <= string1 + size1)
-
-/* (Re)Allocate N items of type T using malloc, or fail. */
-#define TALLOC(n, t) ((t *) malloc ((n) * sizeof (t)))
-#define RETALLOC(addr, n, t) ((addr) = (t *) realloc (addr, (n) * sizeof (t)))
-#define REGEX_TALLOC(n, t) ((t *) REGEX_ALLOCATE ((n) * sizeof (t)))
-
-#define BYTEWIDTH 8 /* In bits. */
-
-#define STREQ(s1, s2) ((strcmp (s1, s2) == 0))
-
-#define MAX(a, b) ((a) > (b) ? (a) : (b))
-#define MIN(a, b) ((a) < (b) ? (a) : (b))
-
-typedef char boolean;
-#define false 0
-#define true 1
-\f
-/* These are the command codes that appear in compiled regular
- expressions. Some opcodes are followed by argument bytes. A
- command code can specify any interpretation whatsoever for its
- arguments. Zero bytes may appear in the compiled regular expression.
-
- The value of `exactn' is needed in search.c (search_buffer) in Emacs.
- So regex.h defines a symbol `RE_EXACTN_VALUE' to be 1; the value of
- `exactn' we use here must also be 1. */
-
-typedef enum
-{
- no_op = 0,
-
- /* Followed by one byte giving n, then by n literal bytes. */
- exactn = 1,
-
- /* Matches any (more or less) character. */
- anychar,
-
- /* Matches any one char belonging to specified set. First
- following byte is number of bitmap bytes. Then come bytes
- for a bitmap saying which chars are in. Bits in each byte
- are ordered low-bit-first. A character is in the set if its
- bit is 1. A character too large to have a bit in the map is
- automatically not in the set. */
- charset,
-
- /* Same parameters as charset, but match any character that is
- not one of those specified. */
- charset_not,
-
- /* Start remembering the text that is matched, for storing in a
- register. Followed by one byte with the register number, in
- the range 0 to one less than the pattern buffer's re_nsub
- field. Then followed by one byte with the number of groups
- inner to this one. (This last has to be part of the
- start_memory only because we need it in the on_failure_jump
- of re_match_2.) */
- start_memory,
-
- /* Stop remembering the text that is matched and store it in a
- memory register. Followed by one byte with the register
- number, in the range 0 to one less than `re_nsub' in the
- pattern buffer, and one byte with the number of inner groups,
- just like `start_memory'. (We need the number of inner
- groups here because we don't have any easy way of finding the
- corresponding start_memory when we're at a stop_memory.) */
- stop_memory,
-
- /* Match a duplicate of something remembered. Followed by one
- byte containing the register number. */
- duplicate,
-
- /* Fail unless at beginning of line. */
- begline,
-
- /* Fail unless at end of line. */
- endline,
-
- /* Succeeds if at beginning of buffer (if emacs) or at beginning
- of string to be matched (if not). */
- begbuf,
-
- /* Analogously, for end of buffer/string. */
- endbuf,
-
- /* Followed by two byte relative address to which to jump. */
- jump,
-
- /* Same as jump, but marks the end of an alternative. */
- jump_past_alt,
-
- /* Followed by two-byte relative address of place to resume at
- in case of failure. */
- on_failure_jump,
-
- /* Like on_failure_jump, but pushes a placeholder instead of the
- current string position when executed. */
- on_failure_keep_string_jump,
-
- /* Throw away latest failure point and then jump to following
- two-byte relative address. */
- pop_failure_jump,
-
- /* Change to pop_failure_jump if know won't have to backtrack to
- match; otherwise change to jump. This is used to jump
- back to the beginning of a repeat. If what follows this jump
- clearly won't match what the repeat does, such that we can be
- sure that there is no use backtracking out of repetitions
- already matched, then we change it to a pop_failure_jump.
- Followed by two-byte address. */
- maybe_pop_jump,
-
- /* Jump to following two-byte address, and push a dummy failure
- point. This failure point will be thrown away if an attempt
- is made to use it for a failure. A `+' construct makes this
- before the first repeat. Also used as an intermediary kind
- of jump when compiling an alternative. */
- dummy_failure_jump,
-
- /* Push a dummy failure point and continue. Used at the end of
- alternatives. */
- push_dummy_failure,
-
- /* Followed by two-byte relative address and two-byte number n.
- After matching N times, jump to the address upon failure. */
- succeed_n,
-
- /* Followed by two-byte relative address, and two-byte number n.
- Jump to the address N times, then fail. */
- jump_n,
-
- /* Set the following two-byte relative address to the
- subsequent two-byte number. The address *includes* the two
- bytes of number. */
- set_number_at,
-
- wordchar, /* Matches any word-constituent character. */
- notwordchar, /* Matches any char that is not a word-constituent. */
-
- wordbeg, /* Succeeds if at word beginning. */
- wordend, /* Succeeds if at word end. */
-
- wordbound, /* Succeeds if at a word boundary. */
- notwordbound /* Succeeds if not at a word boundary. */
-
-#ifdef emacs
- ,before_dot, /* Succeeds if before point. */
- at_dot, /* Succeeds if at point. */
- after_dot, /* Succeeds if after point. */
-
- /* Matches any character whose syntax is specified. Followed by
- a byte which contains a syntax code, e.g., Sword. */
- syntaxspec,
-
- /* Matches any character whose syntax is not that specified. */
- notsyntaxspec
-#endif /* emacs */
-} re_opcode_t;
-\f
-/* Common operations on the compiled pattern. */
-
-/* Store NUMBER in two contiguous bytes starting at DESTINATION. */
-
-#define STORE_NUMBER(destination, number) \
- do { \
- (destination)[0] = (number) & 0377; \
- (destination)[1] = (number) >> 8; \
- } while (0)
-
-/* Same as STORE_NUMBER, except increment DESTINATION to
- the byte after where the number is stored. Therefore, DESTINATION
- must be an lvalue. */
-
-#define STORE_NUMBER_AND_INCR(destination, number) \
- do { \
- STORE_NUMBER (destination, number); \
- (destination) += 2; \
- } while (0)
-
-/* Put into DESTINATION a number stored in two contiguous bytes starting
- at SOURCE. */
-
-#define EXTRACT_NUMBER(destination, source) \
- do { \
- (destination) = *(source) & 0377; \
- (destination) += SIGN_EXTEND_CHAR (*((source) + 1)) << 8; \
- } while (0)
-
-#ifdef DEBUG
-static void
-extract_number (dest, source)
- int *dest;
- unsigned char *source;
-{
- int temp = SIGN_EXTEND_CHAR (*(source + 1));
- *dest = *source & 0377;
- *dest += temp << 8;
-}
-
-#ifndef EXTRACT_MACROS /* To debug the macros. */
-#undef EXTRACT_NUMBER
-#define EXTRACT_NUMBER(dest, src) extract_number (&dest, src)
-#endif /* not EXTRACT_MACROS */
-
-#endif /* DEBUG */
-
-/* Same as EXTRACT_NUMBER, except increment SOURCE to after the number.
- SOURCE must be an lvalue. */
-
-#define EXTRACT_NUMBER_AND_INCR(destination, source) \
- do { \
- EXTRACT_NUMBER (destination, source); \
- (source) += 2; \
- } while (0)
-
-#ifdef DEBUG
-static void
-extract_number_and_incr (destination, source)
- int *destination;
- unsigned char **source;
-{
- extract_number (destination, *source);
- *source += 2;
-}
-
-#ifndef EXTRACT_MACROS
-#undef EXTRACT_NUMBER_AND_INCR
-#define EXTRACT_NUMBER_AND_INCR(dest, src) \
- extract_number_and_incr (&dest, &src)
-#endif /* not EXTRACT_MACROS */
-
-#endif /* DEBUG */
-\f
-/* If DEBUG is defined, Regex prints many voluminous messages about what
- it is doing (if the variable `debug' is nonzero). If linked with the
- main program in `iregex.c', you can enter patterns and strings
- interactively. And if linked with the main program in `main.c' and
- the other test files, you can run the already-written tests. */
-
-#ifdef DEBUG
-
-/* We use standard I/O for debugging. */
-#include <stdio.h>
-
-/* It is useful to test things that ``must'' be true when debugging. */
-#include <assert.h>
-
-static int debug = 0;
-
-#define DEBUG_STATEMENT(e) e
-#define DEBUG_PRINT1(x) if (debug) printf (x)
-#define DEBUG_PRINT2(x1, x2) if (debug) printf (x1, x2)
-#define DEBUG_PRINT3(x1, x2, x3) if (debug) printf (x1, x2, x3)
-#define DEBUG_PRINT4(x1, x2, x3, x4) if (debug) printf (x1, x2, x3, x4)
-#define DEBUG_PRINT_COMPILED_PATTERN(p, s, e) \
- if (debug) print_partial_compiled_pattern (s, e)
-#define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2) \
- if (debug) print_double_string (w, s1, sz1, s2, sz2)
-
-
-extern void printchar ();
-
-/* Print the fastmap in human-readable form. */
-
-void
-print_fastmap (fastmap)
- char *fastmap;
-{
- unsigned was_a_range = 0;
- unsigned i = 0;
-
- while (i < (1 << BYTEWIDTH))
- {
- if (fastmap[i++])
- {
- was_a_range = 0;
- printchar (i - 1);
- while (i < (1 << BYTEWIDTH) && fastmap[i])
- {
- was_a_range = 1;
- i++;
- }
- if (was_a_range)
- {
- printf ("-");
- printchar (i - 1);
- }
- }
- }
- putchar ('\n');
-}
-
-
-/* Print a compiled pattern string in human-readable form, starting at
- the START pointer into it and ending just before the pointer END. */
-
-void
-print_partial_compiled_pattern (start, end)
- unsigned char *start;
- unsigned char *end;
-{
- int mcnt, mcnt2;
- unsigned char *p = start;
- unsigned char *pend = end;
-
- if (start == NULL)
- {
- printf ("(null)\n");
- return;
- }
-
- /* Loop over pattern commands. */
- while (p < pend)
- {
- switch ((re_opcode_t) *p++)
- {
- case no_op:
- printf ("/no_op");
- break;
-
- case exactn:
- mcnt = *p++;
- printf ("/exactn/%d", mcnt);
- do
- {
- putchar ('/');
- printchar (*p++);
- }
- while (--mcnt);
- break;
-
- case start_memory:
- mcnt = *p++;
- printf ("/start_memory/%d/%d", mcnt, *p++);
- break;
-
- case stop_memory:
- mcnt = *p++;
- printf ("/stop_memory/%d/%d", mcnt, *p++);
- break;
-
- case duplicate:
- printf ("/duplicate/%d", *p++);
- break;
-
- case anychar:
- printf ("/anychar");
- break;
-
- case charset:
- case charset_not:
- {
- register int c;
-
- printf ("/charset%s",
- (re_opcode_t) *(p - 1) == charset_not ? "_not" : "");
-
- assert (p + *p < pend);
-
- for (c = 0; c < *p; c++)
- {
- unsigned bit;
- unsigned char map_byte = p[1 + c];
-
- putchar ('/');
-
- for (bit = 0; bit < BYTEWIDTH; bit++)
- if (map_byte & (1 << bit))
- printchar (c * BYTEWIDTH + bit);
- }
- p += 1 + *p;
- break;
- }
-
- case begline:
- printf ("/begline");
- break;
-
- case endline:
- printf ("/endline");
- break;
-
- case on_failure_jump:
- extract_number_and_incr (&mcnt, &p);
- printf ("/on_failure_jump/0/%d", mcnt);
- break;
-
- case on_failure_keep_string_jump:
- extract_number_and_incr (&mcnt, &p);
- printf ("/on_failure_keep_string_jump/0/%d", mcnt);
- break;
-
- case dummy_failure_jump:
- extract_number_and_incr (&mcnt, &p);
- printf ("/dummy_failure_jump/0/%d", mcnt);
- break;
-
- case push_dummy_failure:
- printf ("/push_dummy_failure");
- break;
-
- case maybe_pop_jump:
- extract_number_and_incr (&mcnt, &p);
- printf ("/maybe_pop_jump/0/%d", mcnt);
- break;
-
- case pop_failure_jump:
- extract_number_and_incr (&mcnt, &p);
- printf ("/pop_failure_jump/0/%d", mcnt);
- break;
-
- case jump_past_alt:
- extract_number_and_incr (&mcnt, &p);
- printf ("/jump_past_alt/0/%d", mcnt);
- break;
-
- case jump:
- extract_number_and_incr (&mcnt, &p);
- printf ("/jump/0/%d", mcnt);
- break;
-
- case succeed_n:
- extract_number_and_incr (&mcnt, &p);
- extract_number_and_incr (&mcnt2, &p);
- printf ("/succeed_n/0/%d/0/%d", mcnt, mcnt2);
- break;
-
- case jump_n:
- extract_number_and_incr (&mcnt, &p);
- extract_number_and_incr (&mcnt2, &p);
- printf ("/jump_n/0/%d/0/%d", mcnt, mcnt2);
- break;
-
- case set_number_at:
- extract_number_and_incr (&mcnt, &p);
- extract_number_and_incr (&mcnt2, &p);
- printf ("/set_number_at/0/%d/0/%d", mcnt, mcnt2);
- break;
-
- case wordbound:
- printf ("/wordbound");
- break;
-
- case notwordbound:
- printf ("/notwordbound");
- break;
-
- case wordbeg:
- printf ("/wordbeg");
- break;
-
- case wordend:
- printf ("/wordend");
-
-#ifdef emacs
- case before_dot:
- printf ("/before_dot");
- break;
-
- case at_dot:
- printf ("/at_dot");
- break;
-
- case after_dot:
- printf ("/after_dot");
- break;
-
- case syntaxspec:
- printf ("/syntaxspec");
- mcnt = *p++;
- printf ("/%d", mcnt);
- break;
-
- case notsyntaxspec:
- printf ("/notsyntaxspec");
- mcnt = *p++;
- printf ("/%d", mcnt);
- break;
-#endif /* emacs */
-
- case wordchar:
- printf ("/wordchar");
- break;
-
- case notwordchar:
- printf ("/notwordchar");
- break;
-
- case begbuf:
- printf ("/begbuf");
- break;
-
- case endbuf:
- printf ("/endbuf");
- break;
-
- default:
- printf ("?%d", *(p-1));
- }
- }
- printf ("/\n");
-}
-
-
-void
-print_compiled_pattern (bufp)
- struct re_pattern_buffer *bufp;
-{
- unsigned char *buffer = bufp->buffer;
-
- print_partial_compiled_pattern (buffer, buffer + bufp->used);
- printf ("%d bytes used/%d bytes allocated.\n", bufp->used, bufp->allocated);
-
- if (bufp->fastmap_accurate && bufp->fastmap)
- {
- printf ("fastmap: ");
- print_fastmap (bufp->fastmap);
- }
-
- printf ("re_nsub: %d\t", bufp->re_nsub);
- printf ("regs_alloc: %d\t", bufp->regs_allocated);
- printf ("can_be_null: %d\t", bufp->can_be_null);
- printf ("newline_anchor: %d\n", bufp->newline_anchor);
- printf ("no_sub: %d\t", bufp->no_sub);
- printf ("not_bol: %d\t", bufp->not_bol);
- printf ("not_eol: %d\t", bufp->not_eol);
- printf ("syntax: %d\n", bufp->syntax);
- /* Perhaps we should print the translate table? */
-}
-
-
-void
-print_double_string (where, string1, size1, string2, size2)
- const char *where;
- const char *string1;
- const char *string2;
- int size1;
- int size2;
-{
- unsigned this_char;
-
- if (where == NULL)
- printf ("(null)");
- else
- {
- if (FIRST_STRING_P (where))
- {
- for (this_char = where - string1; this_char < size1; this_char++)
- printchar (string1[this_char]);
-
- where = string2;
- }
-
- for (this_char = where - string2; this_char < size2; this_char++)
- printchar (string2[this_char]);
- }
-}
-
-#else /* not DEBUG */
-
-#undef assert
-#define assert(e)
-
-#define DEBUG_STATEMENT(e)
-#define DEBUG_PRINT1(x)
-#define DEBUG_PRINT2(x1, x2)
-#define DEBUG_PRINT3(x1, x2, x3)
-#define DEBUG_PRINT4(x1, x2, x3, x4)
-#define DEBUG_PRINT_COMPILED_PATTERN(p, s, e)
-#define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2)
-
-#endif /* not DEBUG */
-\f
-/* Set by `re_set_syntax' to the current regexp syntax to recognize. Can
- also be assigned to arbitrarily: each pattern buffer stores its own
- syntax, so it can be changed between regex compilations. */
-reg_syntax_t re_syntax_options = RE_SYNTAX_EMACS;
-
-
-/* Specify the precise syntax of regexps for compilation. This provides
- for compatibility for various utilities which historically have
- different, incompatible syntaxes.
-
- The argument SYNTAX is a bit mask comprised of the various bits
- defined in regex.h. We return the old syntax. */
-
-reg_syntax_t
-re_set_syntax (syntax)
- reg_syntax_t syntax;
-{
- reg_syntax_t ret = re_syntax_options;
-
- re_syntax_options = syntax;
- return ret;
-}
-\f
-/* This table gives an error message for each of the error codes listed
- in regex.h. Obviously the order here has to be same as there. */
-
-static const char *re_error_msg[] =
- { NULL, /* REG_NOERROR */
- "No match", /* REG_NOMATCH */
- "Invalid regular expression", /* REG_BADPAT */
- "Invalid collation character", /* REG_ECOLLATE */
- "Invalid character class name", /* REG_ECTYPE */
- "Trailing backslash", /* REG_EESCAPE */
- "Invalid back reference", /* REG_ESUBREG */
- "Unmatched [ or [^", /* REG_EBRACK */
- "Unmatched ( or \\(", /* REG_EPAREN */
- "Unmatched \\{", /* REG_EBRACE */
- "Invalid content of \\{\\}", /* REG_BADBR */
- "Invalid range end", /* REG_ERANGE */
- "Memory exhausted", /* REG_ESPACE */
- "Invalid preceding regular expression", /* REG_BADRPT */
- "Premature end of regular expression", /* REG_EEND */
- "Regular expression too big", /* REG_ESIZE */
- "Unmatched ) or \\)", /* REG_ERPAREN */
- };
-\f
-/* Subroutine declarations and macros for regex_compile. */
-
-static void store_op1 (), store_op2 ();
-static void insert_op1 (), insert_op2 ();
-static boolean at_begline_loc_p (), at_endline_loc_p ();
-static boolean group_in_compile_stack ();
-static reg_errcode_t compile_range ();
-
-/* Fetch the next character in the uncompiled pattern---translating it
- if necessary. Also cast from a signed character in the constant
- string passed to us by the user to an unsigned char that we can use
- as an array index (in, e.g., `translate'). */
-#define PATFETCH(c) \
- do {if (p == pend) return REG_EEND; \
- c = (unsigned char) *p++; \
- if (translate) c = translate[c]; \
- } while (0)
-
-/* Fetch the next character in the uncompiled pattern, with no
- translation. */
-#define PATFETCH_RAW(c) \
- do {if (p == pend) return REG_EEND; \
- c = (unsigned char) *p++; \
- } while (0)
-
-/* Go backwards one character in the pattern. */
-#define PATUNFETCH p--
-
-
-/* If `translate' is non-null, return translate[D], else just D. We
- cast the subscript to translate because some data is declared as
- `char *', to avoid warnings when a string constant is passed. But
- when we use a character as a subscript we must make it unsigned. */
-#define TRANSLATE(d) (translate ? translate[(unsigned char) (d)] : (d))
-
-
-/* Macros for outputting the compiled pattern into `buffer'. */
-
-/* If the buffer isn't allocated when it comes in, use this. */
-#define INIT_BUF_SIZE 32
-
-/* Make sure we have at least N more bytes of space in buffer. */
-#define GET_BUFFER_SPACE(n) \
- while (b - bufp->buffer + (n) > bufp->allocated) \
- EXTEND_BUFFER ()
-
-/* Make sure we have one more byte of buffer space and then add C to it. */
-#define BUF_PUSH(c) \
- do { \
- GET_BUFFER_SPACE (1); \
- *b++ = (unsigned char) (c); \
- } while (0)
-
-
-/* Ensure we have two more bytes of buffer space and then append C1 and C2. */
-#define BUF_PUSH_2(c1, c2) \
- do { \
- GET_BUFFER_SPACE (2); \
- *b++ = (unsigned char) (c1); \
- *b++ = (unsigned char) (c2); \
- } while (0)
-
-
-/* As with BUF_PUSH_2, except for three bytes. */
-#define BUF_PUSH_3(c1, c2, c3) \
- do { \
- GET_BUFFER_SPACE (3); \
- *b++ = (unsigned char) (c1); \
- *b++ = (unsigned char) (c2); \
- *b++ = (unsigned char) (c3); \
- } while (0)
-
-
-/* Store a jump with opcode OP at LOC to location TO. We store a
- relative address offset by the three bytes the jump itself occupies. */
-#define STORE_JUMP(op, loc, to) \
- store_op1 (op, loc, (to) - (loc) - 3)
-
-/* Likewise, for a two-argument jump. */
-#define STORE_JUMP2(op, loc, to, arg) \
- store_op2 (op, loc, (to) - (loc) - 3, arg)
-
-/* Like `STORE_JUMP', but for inserting. Assume `b' is the buffer end. */
-#define INSERT_JUMP(op, loc, to) \
- insert_op1 (op, loc, (to) - (loc) - 3, b)
-
-/* Like `STORE_JUMP2', but for inserting. Assume `b' is the buffer end. */
-#define INSERT_JUMP2(op, loc, to, arg) \
- insert_op2 (op, loc, (to) - (loc) - 3, arg, b)
-
-
-/* This is not an arbitrary limit: the arguments which represent offsets
- into the pattern are two bytes long. So if 2^16 bytes turns out to
- be too small, many things would have to change. */
-#define MAX_BUF_SIZE (1L << 16)
-
-
-/* Extend the buffer by twice its current size via realloc and
- reset the pointers that pointed into the old block to point to the
- correct places in the new one. If extending the buffer results in it
- being larger than MAX_BUF_SIZE, then flag memory exhausted. */
-#define EXTEND_BUFFER() \
- do { \
- unsigned char *old_buffer = bufp->buffer; \
- if (bufp->allocated == MAX_BUF_SIZE) \
- return REG_ESIZE; \
- bufp->allocated <<= 1; \
- if (bufp->allocated > MAX_BUF_SIZE) \
- bufp->allocated = MAX_BUF_SIZE; \
- bufp->buffer = (unsigned char *) realloc (bufp->buffer, bufp->allocated);\
- if (bufp->buffer == NULL) \
- return REG_ESPACE; \
- /* If the buffer moved, move all the pointers into it. */ \
- if (old_buffer != bufp->buffer) \
- { \
- b = (b - old_buffer) + bufp->buffer; \
- begalt = (begalt - old_buffer) + bufp->buffer; \
- if (fixup_alt_jump) \
- fixup_alt_jump = (fixup_alt_jump - old_buffer) + bufp->buffer;\
- if (laststart) \
- laststart = (laststart - old_buffer) + bufp->buffer; \
- if (pending_exact) \
- pending_exact = (pending_exact - old_buffer) + bufp->buffer; \
- } \
- } while (0)
-
-
-/* Since we have one byte reserved for the register number argument to
- {start,stop}_memory, the maximum number of groups we can report
- things about is what fits in that byte. */
-#define MAX_REGNUM 255
-
-/* But patterns can have more than `MAX_REGNUM' registers. We just
- ignore the excess. */
-typedef unsigned regnum_t;
-
-
-/* Macros for the compile stack. */
-
-/* Since offsets can go either forwards or backwards, this type needs to
- be able to hold values from -(MAX_BUF_SIZE - 1) to MAX_BUF_SIZE - 1. */
-typedef int pattern_offset_t;
-
-typedef struct
-{
- pattern_offset_t begalt_offset;
- pattern_offset_t fixup_alt_jump;
- pattern_offset_t inner_group_offset;
- pattern_offset_t laststart_offset;
- regnum_t regnum;
-} compile_stack_elt_t;
-
-
-typedef struct
-{
- compile_stack_elt_t *stack;
- unsigned size;
- unsigned avail; /* Offset of next open position. */
-} compile_stack_type;
-
-
-#define INIT_COMPILE_STACK_SIZE 32
-
-#define COMPILE_STACK_EMPTY (compile_stack.avail == 0)
-#define COMPILE_STACK_FULL (compile_stack.avail == compile_stack.size)
-
-/* The next available element. */
-#define COMPILE_STACK_TOP (compile_stack.stack[compile_stack.avail])
-
-
-/* Set the bit for character C in a list. */
-#define SET_LIST_BIT(c) \
- (b[((unsigned char) (c)) / BYTEWIDTH] \
- |= 1 << (((unsigned char) c) % BYTEWIDTH))
-
-
-/* Get the next unsigned number in the uncompiled pattern. */
-#define GET_UNSIGNED_NUMBER(num) \
- { if (p != pend) \
- { \
- PATFETCH (c); \
- while (ISDIGIT (c)) \
- { \
- if (num < 0) \
- num = 0; \
- num = num * 10 + c - '0'; \
- if (p == pend) \
- break; \
- PATFETCH (c); \
- } \
- } \
- }
-
-#define CHAR_CLASS_MAX_LENGTH 6 /* Namely, `xdigit'. */
-
-#define IS_CHAR_CLASS(string) \
- (STREQ (string, "alpha") || STREQ (string, "upper") \
- || STREQ (string, "lower") || STREQ (string, "digit") \
- || STREQ (string, "alnum") || STREQ (string, "xdigit") \
- || STREQ (string, "space") || STREQ (string, "print") \
- || STREQ (string, "punct") || STREQ (string, "graph") \
- || STREQ (string, "cntrl") || STREQ (string, "blank"))
-\f
-/* `regex_compile' compiles PATTERN (of length SIZE) according to SYNTAX.
- Returns one of error codes defined in `regex.h', or zero for success.
-
- Assumes the `allocated' (and perhaps `buffer') and `translate'
- fields are set in BUFP on entry.
-
- If it succeeds, results are put in BUFP (if it returns an error, the
- contents of BUFP are undefined):
- `buffer' is the compiled pattern;
- `syntax' is set to SYNTAX;
- `used' is set to the length of the compiled pattern;
- `fastmap_accurate' is zero;
- `re_nsub' is the number of subexpressions in PATTERN;
- `not_bol' and `not_eol' are zero;
-
- The `fastmap' and `newline_anchor' fields are neither
- examined nor set. */
-
-static reg_errcode_t
-regex_compile (pattern, size, syntax, bufp)
- const char *pattern;
- int size;
- reg_syntax_t syntax;
- struct re_pattern_buffer *bufp;
-{
- /* We fetch characters from PATTERN here. Even though PATTERN is
- `char *' (i.e., signed), we declare these variables as unsigned, so
- they can be reliably used as array indices. */
- register unsigned char c, c1;
-
- /* A random tempory spot in PATTERN. */
- const char *p1;
-
- /* Points to the end of the buffer, where we should append. */
- register unsigned char *b;
-
- /* Keeps track of unclosed groups. */
- compile_stack_type compile_stack;
-
- /* Points to the current (ending) position in the pattern. */
- const char *p = pattern;
- const char *pend = pattern + size;
-
- /* How to translate the characters in the pattern. */
- char *translate = bufp->translate;
-
- /* Address of the count-byte of the most recently inserted `exactn'
- command. This makes it possible to tell if a new exact-match
- character can be added to that command or if the character requires
- a new `exactn' command. */
- unsigned char *pending_exact = 0;
-
- /* Address of start of the most recently finished expression.
- This tells, e.g., postfix * where to find the start of its
- operand. Reset at the beginning of groups and alternatives. */
- unsigned char *laststart = 0;
-
- /* Address of beginning of regexp, or inside of last group. */
- unsigned char *begalt;
-
- /* Place in the uncompiled pattern (i.e., the {) to
- which to go back if the interval is invalid. */
- const char *beg_interval;
-
- /* Address of the place where a forward jump should go to the end of
- the containing expression. Each alternative of an `or' -- except the
- last -- ends with a forward jump of this sort. */
- unsigned char *fixup_alt_jump = 0;
-
- /* Counts open-groups as they are encountered. Remembered for the
- matching close-group on the compile stack, so the same register
- number is put in the stop_memory as the start_memory. */
- regnum_t regnum = 0;
-
-#ifdef DEBUG
- DEBUG_PRINT1 ("\nCompiling pattern: ");
- if (debug)
- {
- unsigned debug_count;
-
- for (debug_count = 0; debug_count < size; debug_count++)
- printchar (pattern[debug_count]);
- putchar ('\n');
- }
-#endif /* DEBUG */
-
- /* Initialize the compile stack. */
- compile_stack.stack = TALLOC (INIT_COMPILE_STACK_SIZE, compile_stack_elt_t);
- if (compile_stack.stack == NULL)
- return REG_ESPACE;
-
- compile_stack.size = INIT_COMPILE_STACK_SIZE;
- compile_stack.avail = 0;
-
- /* Initialize the pattern buffer. */
- bufp->syntax = syntax;
- bufp->fastmap_accurate = 0;
- bufp->not_bol = bufp->not_eol = 0;
-
- /* Set `used' to zero, so that if we return an error, the pattern
- printer (for debugging) will think there's no pattern. We reset it
- at the end. */
- bufp->used = 0;
-
- /* Always count groups, whether or not bufp->no_sub is set. */
- bufp->re_nsub = 0;
-
-#if !defined (emacs) && !defined (SYNTAX_TABLE)
- /* Initialize the syntax table. */
- init_syntax_once ();
-#endif
-
- if (bufp->allocated == 0)
- {
- if (bufp->buffer)
- { /* If zero allocated, but buffer is non-null, try to realloc
- enough space. This loses if buffer's address is bogus, but
- that is the user's responsibility. */
- RETALLOC (bufp->buffer, INIT_BUF_SIZE, unsigned char);
- }
- else
- { /* Caller did not allocate a buffer. Do it for them. */
- bufp->buffer = TALLOC (INIT_BUF_SIZE, unsigned char);
- }
- if (!bufp->buffer) return REG_ESPACE;
-
- bufp->allocated = INIT_BUF_SIZE;
- }
-
- begalt = b = bufp->buffer;
-
- /* Loop through the uncompiled pattern until we're at the end. */
- while (p != pend)
- {
- PATFETCH (c);
-
- switch (c)
- {
- case '^':
- {
- if ( /* If at start of pattern, it's an operator. */
- p == pattern + 1
- /* If context independent, it's an operator. */
- || syntax & RE_CONTEXT_INDEP_ANCHORS
- /* Otherwise, depends on what's come before. */
- || at_begline_loc_p (pattern, p, syntax))
- BUF_PUSH (begline);
- else
- goto normal_char;
- }
- break;
-
-
- case '$':
- {
- if ( /* If at end of pattern, it's an operator. */
- p == pend
- /* If context independent, it's an operator. */
- || syntax & RE_CONTEXT_INDEP_ANCHORS
- /* Otherwise, depends on what's next. */
- || at_endline_loc_p (p, pend, syntax))
- BUF_PUSH (endline);
- else
- goto normal_char;
- }
- break;
-
-
- case '+':
- case '?':
- if ((syntax & RE_BK_PLUS_QM)
- || (syntax & RE_LIMITED_OPS))
- goto normal_char;
- handle_plus:
- case '*':
- /* If there is no previous pattern... */
- if (!laststart)
- {
- if (syntax & RE_CONTEXT_INVALID_OPS)
- return REG_BADRPT;
- else if (!(syntax & RE_CONTEXT_INDEP_OPS))
- goto normal_char;
- }
-
- {
- /* Are we optimizing this jump? */
- boolean keep_string_p = false;
-
- /* 1 means zero (many) matches is allowed. */
- char zero_times_ok = 0, many_times_ok = 0;
-
- /* If there is a sequence of repetition chars, collapse it
- down to just one (the right one). We can't combine
- interval operators with these because of, e.g., `a{2}*',
- which should only match an even number of `a's. */
-
- for (;;)
- {
- zero_times_ok |= c != '+';
- many_times_ok |= c != '?';
-
- if (p == pend)
- break;
-
- PATFETCH (c);
-
- if (c == '*'
- || (!(syntax & RE_BK_PLUS_QM) && (c == '+' || c == '?')))
- ;
-
- else if (syntax & RE_BK_PLUS_QM && c == '\\')
- {
- if (p == pend) return REG_EESCAPE;
-
- PATFETCH (c1);
- if (!(c1 == '+' || c1 == '?'))
- {
- PATUNFETCH;
- PATUNFETCH;
- break;
- }
-
- c = c1;
- }
- else
- {
- PATUNFETCH;
- break;
- }
-
- /* If we get here, we found another repeat character. */
- }
-
- /* Star, etc. applied to an empty pattern is equivalent
- to an empty pattern. */
- if (!laststart)
- break;
-
- /* Now we know whether or not zero matches is allowed
- and also whether or not two or more matches is allowed. */
- if (many_times_ok)
- { /* More than one repetition is allowed, so put in at the
- end a backward relative jump from `b' to before the next
- jump we're going to put in below (which jumps from
- laststart to after this jump).
-
- But if we are at the `*' in the exact sequence `.*\n',
- insert an unconditional jump backwards to the .,
- instead of the beginning of the loop. This way we only
- push a failure point once, instead of every time
- through the loop. */
- assert (p - 1 > pattern);
-
- /* Allocate the space for the jump. */
- GET_BUFFER_SPACE (3);
-
- /* We know we are not at the first character of the pattern,
- because laststart was nonzero. And we've already
- incremented `p', by the way, to be the character after
- the `*'. Do we have to do something analogous here
- for null bytes, because of RE_DOT_NOT_NULL? */
- if (TRANSLATE (*(p - 2)) == TRANSLATE ('.')
- && zero_times_ok
- && p < pend && TRANSLATE (*p) == TRANSLATE ('\n')
- && !(syntax & RE_DOT_NEWLINE))
- { /* We have .*\n. */
- STORE_JUMP (jump, b, laststart);
- keep_string_p = true;
- }
- else
- /* Anything else. */
- STORE_JUMP (maybe_pop_jump, b, laststart - 3);
-
- /* We've added more stuff to the buffer. */
- b += 3;
- }
-
- /* On failure, jump from laststart to b + 3, which will be the
- end of the buffer after this jump is inserted. */
- GET_BUFFER_SPACE (3);
- INSERT_JUMP (keep_string_p ? on_failure_keep_string_jump
- : on_failure_jump,
- laststart, b + 3);
- pending_exact = 0;
- b += 3;
-
- if (!zero_times_ok)
- {
- /* At least one repetition is required, so insert a
- `dummy_failure_jump' before the initial
- `on_failure_jump' instruction of the loop. This
- effects a skip over that instruction the first time
- we hit that loop. */
- GET_BUFFER_SPACE (3);
- INSERT_JUMP (dummy_failure_jump, laststart, laststart + 6);
- b += 3;
- }
- }
- break;
-
-
- case '.':
- laststart = b;
- BUF_PUSH (anychar);
- break;
-
-
- case '[':
- {
- boolean had_char_class = false;
-
- if (p == pend) return REG_EBRACK;
-
- /* Ensure that we have enough space to push a charset: the
- opcode, the length count, and the bitset; 34 bytes in all. */
- GET_BUFFER_SPACE (34);
-
- laststart = b;
-
- /* We test `*p == '^' twice, instead of using an if
- statement, so we only need one BUF_PUSH. */
- BUF_PUSH (*p == '^' ? charset_not : charset);
- if (*p == '^')
- p++;
-
- /* Remember the first position in the bracket expression. */
- p1 = p;
-
- /* Push the number of bytes in the bitmap. */
- BUF_PUSH ((1 << BYTEWIDTH) / BYTEWIDTH);
-
- /* Clear the whole map. */
- bzero (b, (1 << BYTEWIDTH) / BYTEWIDTH);
-
- /* charset_not matches newline according to a syntax bit. */
- if ((re_opcode_t) b[-2] == charset_not
- && (syntax & RE_HAT_LISTS_NOT_NEWLINE))
- SET_LIST_BIT ('\n');
-
- /* Read in characters and ranges, setting map bits. */
- for (;;)
- {
- if (p == pend) return REG_EBRACK;
-
- PATFETCH (c);
-
- /* \ might escape characters inside [...] and [^...]. */
- if ((syntax & RE_BACKSLASH_ESCAPE_IN_LISTS) && c == '\\')
- {
- if (p == pend) return REG_EESCAPE;
-
- PATFETCH (c1);
- SET_LIST_BIT (c1);
- continue;
- }
-
- /* Could be the end of the bracket expression. If it's
- not (i.e., when the bracket expression is `[]' so
- far), the ']' character bit gets set way below. */
- if (c == ']' && p != p1 + 1)
- break;
-
- /* Look ahead to see if it's a range when the last thing
- was a character class. */
- if (had_char_class && c == '-' && *p != ']')
- return REG_ERANGE;
-
- /* Look ahead to see if it's a range when the last thing
- was a character: if this is a hyphen not at the
- beginning or the end of a list, then it's the range
- operator. */
- if (c == '-'
- && !(p - 2 >= pattern && p[-2] == '[')
- && !(p - 3 >= pattern && p[-3] == '[' && p[-2] == '^')
- && *p != ']')
- {
- reg_errcode_t ret
- = compile_range (&p, pend, translate, syntax, b);
- if (ret != REG_NOERROR) return ret;
- }
-
- else if (p[0] == '-' && p[1] != ']')
- { /* This handles ranges made up of characters only. */
- reg_errcode_t ret;
-
- /* Move past the `-'. */
- PATFETCH (c1);
-
- ret = compile_range (&p, pend, translate, syntax, b);
- if (ret != REG_NOERROR) return ret;
- }
-
- /* See if we're at the beginning of a possible character
- class. */
-
- else if (syntax & RE_CHAR_CLASSES && c == '[' && *p == ':')
- { /* Leave room for the null. */
- char str[CHAR_CLASS_MAX_LENGTH + 1];
-
- PATFETCH (c);
- c1 = 0;
-
- /* If pattern is `[[:'. */
- if (p == pend) return REG_EBRACK;
-
- for (;;)
- {
- PATFETCH (c);
- if (c == ':' || c == ']' || p == pend
- || c1 == CHAR_CLASS_MAX_LENGTH)
- break;
- str[c1++] = c;
- }
- str[c1] = '\0';
-
- /* If isn't a word bracketed by `[:' and:`]':
- undo the ending character, the letters, and leave
- the leading `:' and `[' (but set bits for them). */
- if (c == ':' && *p == ']')
- {
- int ch;
- boolean is_alnum = STREQ (str, "alnum");
- boolean is_alpha = STREQ (str, "alpha");
- boolean is_blank = STREQ (str, "blank");
- boolean is_cntrl = STREQ (str, "cntrl");
- boolean is_digit = STREQ (str, "digit");
- boolean is_graph = STREQ (str, "graph");
- boolean is_lower = STREQ (str, "lower");
- boolean is_print = STREQ (str, "print");
- boolean is_punct = STREQ (str, "punct");
- boolean is_space = STREQ (str, "space");
- boolean is_upper = STREQ (str, "upper");
- boolean is_xdigit = STREQ (str, "xdigit");
-
- if (!IS_CHAR_CLASS (str)) return REG_ECTYPE;
-
- /* Throw away the ] at the end of the character
- class. */
- PATFETCH (c);
-
- if (p == pend) return REG_EBRACK;
-
- for (ch = 0; ch < 1 << BYTEWIDTH; ch++)
- {
- if ( (is_alnum && ISALNUM (ch))
- || (is_alpha && ISALPHA (ch))
- || (is_blank && ISBLANK (ch))
- || (is_cntrl && ISCNTRL (ch))
- || (is_digit && ISDIGIT (ch))
- || (is_graph && ISGRAPH (ch))
- || (is_lower && ISLOWER (ch))
- || (is_print && ISPRINT (ch))
- || (is_punct && ISPUNCT (ch))
- || (is_space && ISSPACE (ch))
- || (is_upper && ISUPPER (ch))
- || (is_xdigit && ISXDIGIT (ch)))
- SET_LIST_BIT (ch);
- }
- had_char_class = true;
- }
- else
- {
- c1++;
- while (c1--)
- PATUNFETCH;
- SET_LIST_BIT ('[');
- SET_LIST_BIT (':');
- had_char_class = false;
- }
- }
- else
- {
- had_char_class = false;
- SET_LIST_BIT (c);
- }
- }
-
- /* Discard any (non)matching list bytes that are all 0 at the
- end of the map. Decrease the map-length byte too. */
- while ((int) b[-1] > 0 && b[b[-1] - 1] == 0)
- b[-1]--;
- b += b[-1];
- }
- break;
-
-
- case '(':
- if (syntax & RE_NO_BK_PARENS)
- goto handle_open;
- else
- goto normal_char;
-
-
- case ')':
- if (syntax & RE_NO_BK_PARENS)
- goto handle_close;
- else
- goto normal_char;
-
-
- case '\n':
- if (syntax & RE_NEWLINE_ALT)
- goto handle_alt;
- else
- goto normal_char;
-
-
- case '|':
- if (syntax & RE_NO_BK_VBAR)
- goto handle_alt;
- else
- goto normal_char;
-
-
- case '{':
- if (syntax & RE_INTERVALS && syntax & RE_NO_BK_BRACES)
- goto handle_interval;
- else
- goto normal_char;
-
-
- case '\\':
- if (p == pend) return REG_EESCAPE;
-
- /* Do not translate the character after the \, so that we can
- distinguish, e.g., \B from \b, even if we normally would
- translate, e.g., B to b. */
- PATFETCH_RAW (c);
-
- switch (c)
- {
- case '(':
- if (syntax & RE_NO_BK_PARENS)
- goto normal_backslash;
-
- handle_open:
- bufp->re_nsub++;
- regnum++;
-
- if (COMPILE_STACK_FULL)
- {
- RETALLOC (compile_stack.stack, compile_stack.size << 1,
- compile_stack_elt_t);
- if (compile_stack.stack == NULL) return REG_ESPACE;
-
- compile_stack.size <<= 1;
- }
-
- /* These are the values to restore when we hit end of this
- group. They are all relative offsets, so that if the
- whole pattern moves because of realloc, they will still
- be valid. */
- COMPILE_STACK_TOP.begalt_offset = begalt - bufp->buffer;
- COMPILE_STACK_TOP.fixup_alt_jump
- = fixup_alt_jump ? fixup_alt_jump - bufp->buffer + 1 : 0;
- COMPILE_STACK_TOP.laststart_offset = b - bufp->buffer;
- COMPILE_STACK_TOP.regnum = regnum;
-
- /* We will eventually replace the 0 with the number of
- groups inner to this one. But do not push a
- start_memory for groups beyond the last one we can
- represent in the compiled pattern. */
- if (regnum <= MAX_REGNUM)
- {
- COMPILE_STACK_TOP.inner_group_offset = b - bufp->buffer + 2;
- BUF_PUSH_3 (start_memory, regnum, 0);
- }
-
- compile_stack.avail++;
-
- fixup_alt_jump = 0;
- laststart = 0;
- begalt = b;
- /* If we've reached MAX_REGNUM groups, then this open
- won't actually generate any code, so we'll have to
- clear pending_exact explicitly. */
- pending_exact = 0;
- break;
-
-
- case ')':
- if (syntax & RE_NO_BK_PARENS) goto normal_backslash;
-
- if (COMPILE_STACK_EMPTY)
- {
- if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)
- goto normal_backslash;
- else
- return REG_ERPAREN;
- }
-
- handle_close:
- if (fixup_alt_jump)
- { /* Push a dummy failure point at the end of the
- alternative for a possible future
- `pop_failure_jump' to pop. See comments at
- `push_dummy_failure' in `re_match_2'. */
- BUF_PUSH (push_dummy_failure);
-
- /* We allocated space for this jump when we assigned
- to `fixup_alt_jump', in the `handle_alt' case below. */
- STORE_JUMP (jump_past_alt, fixup_alt_jump, b - 1);
- }
-
- /* See similar code for backslashed left paren above. */
- if (COMPILE_STACK_EMPTY)
- {
- if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)
- goto normal_char;
- else
- return REG_ERPAREN;
- }
-
- /* Since we just checked for an empty stack above, this
- ``can't happen''. */
- assert (compile_stack.avail != 0);
- {
- /* We don't just want to restore into `regnum', because
- later groups should continue to be numbered higher,
- as in `(ab)c(de)' -- the second group is #2. */
- regnum_t this_group_regnum;
-
- compile_stack.avail--;
- begalt = bufp->buffer + COMPILE_STACK_TOP.begalt_offset;
- fixup_alt_jump
- = COMPILE_STACK_TOP.fixup_alt_jump
- ? bufp->buffer + COMPILE_STACK_TOP.fixup_alt_jump - 1
- : 0;
- laststart = bufp->buffer + COMPILE_STACK_TOP.laststart_offset;
- this_group_regnum = COMPILE_STACK_TOP.regnum;
- /* If we've reached MAX_REGNUM groups, then this open
- won't actually generate any code, so we'll have to
- clear pending_exact explicitly. */
- pending_exact = 0;
-
- /* We're at the end of the group, so now we know how many
- groups were inside this one. */
- if (this_group_regnum <= MAX_REGNUM)
- {
- unsigned char *inner_group_loc
- = bufp->buffer + COMPILE_STACK_TOP.inner_group_offset;
-
- *inner_group_loc = regnum - this_group_regnum;
- BUF_PUSH_3 (stop_memory, this_group_regnum,
- regnum - this_group_regnum);
- }
- }
- break;
-
-
- case '|': /* `\|'. */
- if (syntax & RE_LIMITED_OPS || syntax & RE_NO_BK_VBAR)
- goto normal_backslash;
- handle_alt:
- if (syntax & RE_LIMITED_OPS)
- goto normal_char;
-
- /* Insert before the previous alternative a jump which
- jumps to this alternative if the former fails. */
- GET_BUFFER_SPACE (3);
- INSERT_JUMP (on_failure_jump, begalt, b + 6);
- pending_exact = 0;
- b += 3;
-
- /* The alternative before this one has a jump after it
- which gets executed if it gets matched. Adjust that
- jump so it will jump to this alternative's analogous
- jump (put in below, which in turn will jump to the next
- (if any) alternative's such jump, etc.). The last such
- jump jumps to the correct final destination. A picture:
- _____ _____
- | | | |
- | v | v
- a | b | c
-
- If we are at `b', then fixup_alt_jump right now points to a
- three-byte space after `a'. We'll put in the jump, set
- fixup_alt_jump to right after `b', and leave behind three
- bytes which we'll fill in when we get to after `c'. */
-
- if (fixup_alt_jump)
- STORE_JUMP (jump_past_alt, fixup_alt_jump, b);
-
- /* Mark and leave space for a jump after this alternative,
- to be filled in later either by next alternative or
- when know we're at the end of a series of alternatives. */
- fixup_alt_jump = b;
- GET_BUFFER_SPACE (3);
- b += 3;
-
- laststart = 0;
- begalt = b;
- break;
-
-
- case '{':
- /* If \{ is a literal. */
- if (!(syntax & RE_INTERVALS)
- /* If we're at `\{' and it's not the open-interval
- operator. */
- || ((syntax & RE_INTERVALS) && (syntax & RE_NO_BK_BRACES))
- || (p - 2 == pattern && p == pend))
- goto normal_backslash;
-
- handle_interval:
- {
- /* If got here, then the syntax allows intervals. */
-
- /* At least (most) this many matches must be made. */
- int lower_bound = -1, upper_bound = -1;
-
- beg_interval = p - 1;
-
- if (p == pend)
- {
- if (syntax & RE_NO_BK_BRACES)
- goto unfetch_interval;
- else
- return REG_EBRACE;
- }
-
- GET_UNSIGNED_NUMBER (lower_bound);
-
- if (c == ',')
- {
- GET_UNSIGNED_NUMBER (upper_bound);
- if (upper_bound < 0) upper_bound = RE_DUP_MAX;
- }
- else
- /* Interval such as `{1}' => match exactly once. */
- upper_bound = lower_bound;
-
- if (lower_bound < 0 || upper_bound > RE_DUP_MAX
- || lower_bound > upper_bound)
- {
- if (syntax & RE_NO_BK_BRACES)
- goto unfetch_interval;
- else
- return REG_BADBR;
- }
-
- if (!(syntax & RE_NO_BK_BRACES))
- {
- if (c != '\\') return REG_EBRACE;
-
- PATFETCH (c);
- }
-
- if (c != '}')
- {
- if (syntax & RE_NO_BK_BRACES)
- goto unfetch_interval;
- else
- return REG_BADBR;
- }
-
- /* We just parsed a valid interval. */
-
- /* If it's invalid to have no preceding re. */
- if (!laststart)
- {
- if (syntax & RE_CONTEXT_INVALID_OPS)
- return REG_BADRPT;
- else if (syntax & RE_CONTEXT_INDEP_OPS)
- laststart = b;
- else
- goto unfetch_interval;
- }
-
- /* If the upper bound is zero, don't want to succeed at
- all; jump from `laststart' to `b + 3', which will be
- the end of the buffer after we insert the jump. */
- if (upper_bound == 0)
- {
- GET_BUFFER_SPACE (3);
- INSERT_JUMP (jump, laststart, b + 3);
- b += 3;
- }
-
- /* Otherwise, we have a nontrivial interval. When
- we're all done, the pattern will look like:
- set_number_at <jump count> <upper bound>
- set_number_at <succeed_n count> <lower bound>
- succeed_n <after jump addr> <succed_n count>
- <body of loop>
- jump_n <succeed_n addr> <jump count>
- (The upper bound and `jump_n' are omitted if
- `upper_bound' is 1, though.) */
- else
- { /* If the upper bound is > 1, we need to insert
- more at the end of the loop. */
- unsigned nbytes = 10 + (upper_bound > 1) * 10;
-
- GET_BUFFER_SPACE (nbytes);
-
- /* Initialize lower bound of the `succeed_n', even
- though it will be set during matching by its
- attendant `set_number_at' (inserted next),
- because `re_compile_fastmap' needs to know.
- Jump to the `jump_n' we might insert below. */
- INSERT_JUMP2 (succeed_n, laststart,
- b + 5 + (upper_bound > 1) * 5,
- lower_bound);
- b += 5;
-
- /* Code to initialize the lower bound. Insert
- before the `succeed_n'. The `5' is the last two
- bytes of this `set_number_at', plus 3 bytes of
- the following `succeed_n'. */
- insert_op2 (set_number_at, laststart, 5, lower_bound, b);
- b += 5;
-
- if (upper_bound > 1)
- { /* More than one repetition is allowed, so
- append a backward jump to the `succeed_n'
- that starts this interval.
-
- When we've reached this during matching,
- we'll have matched the interval once, so
- jump back only `upper_bound - 1' times. */
- STORE_JUMP2 (jump_n, b, laststart + 5,
- upper_bound - 1);
- b += 5;
-
- /* The location we want to set is the second
- parameter of the `jump_n'; that is `b-2' as
- an absolute address. `laststart' will be
- the `set_number_at' we're about to insert;
- `laststart+3' the number to set, the source
- for the relative address. But we are
- inserting into the middle of the pattern --
- so everything is getting moved up by 5.
- Conclusion: (b - 2) - (laststart + 3) + 5,
- i.e., b - laststart.
-
- We insert this at the beginning of the loop
- so that if we fail during matching, we'll
- reinitialize the bounds. */
- insert_op2 (set_number_at, laststart, b - laststart,
- upper_bound - 1, b);
- b += 5;
- }
- }
- pending_exact = 0;
- beg_interval = NULL;
- }
- break;
-
- unfetch_interval:
- /* If an invalid interval, match the characters as literals. */
- assert (beg_interval);
- p = beg_interval;
- beg_interval = NULL;
-
- /* normal_char and normal_backslash need `c'. */
- PATFETCH (c);
-
- if (!(syntax & RE_NO_BK_BRACES))
- {
- if (p > pattern && p[-1] == '\\')
- goto normal_backslash;
- }
- goto normal_char;
-
-#ifdef emacs
- /* There is no way to specify the before_dot and after_dot
- operators. rms says this is ok. --karl */
- case '=':
- BUF_PUSH (at_dot);
- break;
-
- case 's':
- laststart = b;
- PATFETCH (c);
- BUF_PUSH_2 (syntaxspec, syntax_spec_code[c]);
- break;
-
- case 'S':
- laststart = b;
- PATFETCH (c);
- BUF_PUSH_2 (notsyntaxspec, syntax_spec_code[c]);
- break;
-#endif /* emacs */
-
-
- case 'w':
- laststart = b;
- BUF_PUSH (wordchar);
- break;
-
-
- case 'W':
- laststart = b;
- BUF_PUSH (notwordchar);
- break;
-
-
- case '<':
- BUF_PUSH (wordbeg);
- break;
-
- case '>':
- BUF_PUSH (wordend);
- break;
-
- case 'b':
- BUF_PUSH (wordbound);
- break;
-
- case 'B':
- BUF_PUSH (notwordbound);
- break;
-
- case '`':
- BUF_PUSH (begbuf);
- break;
-
- case '\'':
- BUF_PUSH (endbuf);
- break;
-
- case '1': case '2': case '3': case '4': case '5':
- case '6': case '7': case '8': case '9':
- if (syntax & RE_NO_BK_REFS)
- goto normal_char;
-
- c1 = c - '0';
-
- if (c1 > regnum)
- return REG_ESUBREG;
-
- /* Can't back reference to a subexpression if inside of it. */
- if (group_in_compile_stack (compile_stack, c1))
- goto normal_char;
-
- laststart = b;
- BUF_PUSH_2 (duplicate, c1);
- break;
-
-
- case '+':
- case '?':
- if (syntax & RE_BK_PLUS_QM)
- goto handle_plus;
- else
- goto normal_backslash;
-
- default:
- normal_backslash:
- /* You might think it would be useful for \ to mean
- not to translate; but if we don't translate it
- it will never match anything. */
- c = TRANSLATE (c);
- goto normal_char;
- }
- break;
-
-
- default:
- /* Expects the character in `c'. */
- normal_char:
- /* If no exactn currently being built. */
- if (!pending_exact
-
- /* If last exactn not at current position. */
- || pending_exact + *pending_exact + 1 != b
-
- /* We have only one byte following the exactn for the count. */
- || *pending_exact == (1 << BYTEWIDTH) - 1
-
- /* If followed by a repetition operator. */
- || *p == '*' || *p == '^'
- || ((syntax & RE_BK_PLUS_QM)
- ? *p == '\\' && (p[1] == '+' || p[1] == '?')
- : (*p == '+' || *p == '?'))
- || ((syntax & RE_INTERVALS)
- && ((syntax & RE_NO_BK_BRACES)
- ? *p == '{'
- : (p[0] == '\\' && p[1] == '{'))))
- {
- /* Start building a new exactn. */
-
- laststart = b;
-
- BUF_PUSH_2 (exactn, 0);
- pending_exact = b - 1;
- }
-
- BUF_PUSH (c);
- (*pending_exact)++;
- break;
- } /* switch (c) */
- } /* while p != pend */
-
-
- /* Through the pattern now. */
-
- if (fixup_alt_jump)
- STORE_JUMP (jump_past_alt, fixup_alt_jump, b);
-
- if (!COMPILE_STACK_EMPTY)
- return REG_EPAREN;
-
- free (compile_stack.stack);
-
- /* We have succeeded; set the length of the buffer. */
- bufp->used = b - bufp->buffer;
-
-#ifdef DEBUG
- if (debug)
- {
- DEBUG_PRINT1 ("\nCompiled pattern: ");
- print_compiled_pattern (bufp);
- }
-#endif /* DEBUG */
-
- return REG_NOERROR;
-} /* regex_compile */
-\f
-/* Subroutines for `regex_compile'. */
-
-/* Store OP at LOC followed by two-byte integer parameter ARG. */
-
-static void
-store_op1 (op, loc, arg)
- re_opcode_t op;
- unsigned char *loc;
- int arg;
-{
- *loc = (unsigned char) op;
- STORE_NUMBER (loc + 1, arg);
-}
-
-
-/* Like `store_op1', but for two two-byte parameters ARG1 and ARG2. */
-
-static void
-store_op2 (op, loc, arg1, arg2)
- re_opcode_t op;
- unsigned char *loc;
- int arg1, arg2;
-{
- *loc = (unsigned char) op;
- STORE_NUMBER (loc + 1, arg1);
- STORE_NUMBER (loc + 3, arg2);
-}
-
-
-/* Copy the bytes from LOC to END to open up three bytes of space at LOC
- for OP followed by two-byte integer parameter ARG. */
-
-static void
-insert_op1 (op, loc, arg, end)
- re_opcode_t op;
- unsigned char *loc;
- int arg;
- unsigned char *end;
-{
- register unsigned char *pfrom = end;
- register unsigned char *pto = end + 3;
-
- while (pfrom != loc)
- *--pto = *--pfrom;
-
- store_op1 (op, loc, arg);
-}
-
-
-/* Like `insert_op1', but for two two-byte parameters ARG1 and ARG2. */
-
-static void
-insert_op2 (op, loc, arg1, arg2, end)
- re_opcode_t op;
- unsigned char *loc;
- int arg1, arg2;
- unsigned char *end;
-{
- register unsigned char *pfrom = end;
- register unsigned char *pto = end + 5;
-
- while (pfrom != loc)
- *--pto = *--pfrom;
-
- store_op2 (op, loc, arg1, arg2);
-}
-
-
-/* P points to just after a ^ in PATTERN. Return true if that ^ comes
- after an alternative or a begin-subexpression. We assume there is at
- least one character before the ^. */
-
-static boolean
-at_begline_loc_p (pattern, p, syntax)
- const char *pattern, *p;
- reg_syntax_t syntax;
-{
- const char *prev = p - 2;
- boolean prev_prev_backslash = prev > pattern && prev[-1] == '\\';
-
- return
- /* After a subexpression? */
- (*prev == '(' && (syntax & RE_NO_BK_PARENS || prev_prev_backslash))
- /* After an alternative? */
- || (*prev == '|' && (syntax & RE_NO_BK_VBAR || prev_prev_backslash));
-}
-
-
-/* The dual of at_begline_loc_p. This one is for $. We assume there is
- at least one character after the $, i.e., `P < PEND'. */
-
-static boolean
-at_endline_loc_p (p, pend, syntax)
- const char *p, *pend;
- int syntax;
-{
- const char *next = p;
- boolean next_backslash = *next == '\\';
- const char *next_next = p + 1 < pend ? p + 1 : NULL;
-
- return
- /* Before a subexpression? */
- (syntax & RE_NO_BK_PARENS ? *next == ')'
- : next_backslash && next_next && *next_next == ')')
- /* Before an alternative? */
- || (syntax & RE_NO_BK_VBAR ? *next == '|'
- : next_backslash && next_next && *next_next == '|');
-}
-
-
-/* Returns true if REGNUM is in one of COMPILE_STACK's elements and
- false if it's not. */
-
-static boolean
-group_in_compile_stack (compile_stack, regnum)
- compile_stack_type compile_stack;
- regnum_t regnum;
-{
- int this_element;
-
- for (this_element = compile_stack.avail - 1;
- this_element >= 0;
- this_element--)
- if (compile_stack.stack[this_element].regnum == regnum)
- return true;
-
- return false;
-}
-
-
-/* Read the ending character of a range (in a bracket expression) from the
- uncompiled pattern *P_PTR (which ends at PEND). We assume the
- starting character is in `P[-2]'. (`P[-1]' is the character `-'.)
- Then we set the translation of all bits between the starting and
- ending characters (inclusive) in the compiled pattern B.
-
- Return an error code.
-
- We use these short variable names so we can use the same macros as
- `regex_compile' itself. */
-
-static reg_errcode_t
-compile_range (p_ptr, pend, translate, syntax, b)
- const char **p_ptr, *pend;
- char *translate;
- reg_syntax_t syntax;
- unsigned char *b;
-{
- unsigned this_char;
-
- const char *p = *p_ptr;
- int range_start, range_end;
-
- if (p == pend)
- return REG_ERANGE;
-
- /* Even though the pattern is a signed `char *', we need to fetch
- with unsigned char *'s; if the high bit of the pattern character
- is set, the range endpoints will be negative if we fetch using a
- signed char *.
-
- We also want to fetch the endpoints without translating them; the
- appropriate translation is done in the bit-setting loop below. */
- range_start = ((unsigned char *) p)[-2];
- range_end = ((unsigned char *) p)[0];
-
- /* Have to increment the pointer into the pattern string, so the
- caller isn't still at the ending character. */
- (*p_ptr)++;
-
- /* If the start is after the end, the range is empty. */
- if (range_start > range_end)
- return syntax & RE_NO_EMPTY_RANGES ? REG_ERANGE : REG_NOERROR;
-
- /* Here we see why `this_char' has to be larger than an `unsigned
- char' -- the range is inclusive, so if `range_end' == 0xff
- (assuming 8-bit characters), we would otherwise go into an infinite
- loop, since all characters <= 0xff. */
- for (this_char = range_start; this_char <= range_end; this_char++)
- {
- SET_LIST_BIT (TRANSLATE (this_char));
- }
-
- return REG_NOERROR;
-}
-\f
-/* Failure stack declarations and macros; both re_compile_fastmap and
- re_match_2 use a failure stack. These have to be macros because of
- REGEX_ALLOCATE. */
-
-
-/* Number of failure points for which to initially allocate space
- when matching. If this number is exceeded, we allocate more
- space, so it is not a hard limit. */
-#ifndef INIT_FAILURE_ALLOC
-#define INIT_FAILURE_ALLOC 5
-#endif
-
-/* Roughly the maximum number of failure points on the stack. Would be
- exactly that if always used MAX_FAILURE_SPACE each time we failed.
- This is a variable only so users of regex can assign to it; we never
- change it ourselves. */
-int re_max_failures = 2000;
-
-typedef const unsigned char *fail_stack_elt_t;
-
-typedef struct
-{
- fail_stack_elt_t *stack;
- unsigned size;
- unsigned avail; /* Offset of next open position. */
-} fail_stack_type;
-
-#define FAIL_STACK_EMPTY() (fail_stack.avail == 0)
-#define FAIL_STACK_PTR_EMPTY() (fail_stack_ptr->avail == 0)
-#define FAIL_STACK_FULL() (fail_stack.avail == fail_stack.size)
-#define FAIL_STACK_TOP() (fail_stack.stack[fail_stack.avail])
-
-
-/* Initialize `fail_stack'. Do `return -2' if the alloc fails. */
-
-#define INIT_FAIL_STACK() \
- do { \
- fail_stack.stack = (fail_stack_elt_t *) \
- REGEX_ALLOCATE (INIT_FAILURE_ALLOC * sizeof (fail_stack_elt_t)); \
- \
- if (fail_stack.stack == NULL) \
- return -2; \
- \
- fail_stack.size = INIT_FAILURE_ALLOC; \
- fail_stack.avail = 0; \
- } while (0)
-
-
-/* Double the size of FAIL_STACK, up to approximately `re_max_failures' items.
-
- Return 1 if succeeds, and 0 if either ran out of memory
- allocating space for it or it was already too large.
-
- REGEX_REALLOCATE requires `destination' be declared. */
-
-#define DOUBLE_FAIL_STACK(fail_stack) \
- ((fail_stack).size > re_max_failures * MAX_FAILURE_ITEMS \
- ? 0 \
- : ((fail_stack).stack = (fail_stack_elt_t *) \
- REGEX_REALLOCATE ((fail_stack).stack, \
- (fail_stack).size * sizeof (fail_stack_elt_t), \
- ((fail_stack).size << 1) * sizeof (fail_stack_elt_t)), \
- \
- (fail_stack).stack == NULL \
- ? 0 \
- : ((fail_stack).size <<= 1, \
- 1)))
-
-
-/* Push PATTERN_OP on FAIL_STACK.
-
- Return 1 if was able to do so and 0 if ran out of memory allocating
- space to do so. */
-#define PUSH_PATTERN_OP(pattern_op, fail_stack) \
- ((FAIL_STACK_FULL () \
- && !DOUBLE_FAIL_STACK (fail_stack)) \
- ? 0 \
- : ((fail_stack).stack[(fail_stack).avail++] = pattern_op, \
- 1))
-
-/* This pushes an item onto the failure stack. Must be a four-byte
- value. Assumes the variable `fail_stack'. Probably should only
- be called from within `PUSH_FAILURE_POINT'. */
-#define PUSH_FAILURE_ITEM(item) \
- fail_stack.stack[fail_stack.avail++] = (fail_stack_elt_t) item
-
-/* The complement operation. Assumes `fail_stack' is nonempty. */
-#define POP_FAILURE_ITEM() fail_stack.stack[--fail_stack.avail]
-
-/* Used to omit pushing failure point id's when we're not debugging. */
-#ifdef DEBUG
-#define DEBUG_PUSH PUSH_FAILURE_ITEM
-#define DEBUG_POP(item_addr) *(item_addr) = POP_FAILURE_ITEM ()
-#else
-#define DEBUG_PUSH(item)
-#define DEBUG_POP(item_addr)
-#endif
-
-
-/* Push the information about the state we will need
- if we ever fail back to it.
-
- Requires variables fail_stack, regstart, regend, reg_info, and
- num_regs be declared. DOUBLE_FAIL_STACK requires `destination' be
- declared.
-
- Does `return FAILURE_CODE' if runs out of memory. */
-
-#define PUSH_FAILURE_POINT(pattern_place, string_place, failure_code) \
- do { \
- char *destination; \
- /* Must be int, so when we don't save any registers, the arithmetic \
- of 0 + -1 isn't done as unsigned. */ \
- int this_reg; \
- \
- DEBUG_STATEMENT (failure_id++); \
- DEBUG_STATEMENT (nfailure_points_pushed++); \
- DEBUG_PRINT2 ("\nPUSH_FAILURE_POINT #%u:\n", failure_id); \
- DEBUG_PRINT2 (" Before push, next avail: %d\n", (fail_stack).avail);\
- DEBUG_PRINT2 (" size: %d\n", (fail_stack).size);\
- \
- DEBUG_PRINT2 (" slots needed: %d\n", NUM_FAILURE_ITEMS); \
- DEBUG_PRINT2 (" available: %d\n", REMAINING_AVAIL_SLOTS); \
- \
- /* Ensure we have enough space allocated for what we will push. */ \
- while (REMAINING_AVAIL_SLOTS < NUM_FAILURE_ITEMS) \
- { \
- if (!DOUBLE_FAIL_STACK (fail_stack)) \
- return failure_code; \
- \
- DEBUG_PRINT2 ("\n Doubled stack; size now: %d\n", \
- (fail_stack).size); \
- DEBUG_PRINT2 (" slots available: %d\n", REMAINING_AVAIL_SLOTS);\
- } \
- \
- /* Push the info, starting with the registers. */ \
- DEBUG_PRINT1 ("\n"); \
- \
- for (this_reg = lowest_active_reg; this_reg <= highest_active_reg; \
- this_reg++) \
- { \
- DEBUG_PRINT2 (" Pushing reg: %d\n", this_reg); \
- DEBUG_STATEMENT (num_regs_pushed++); \
- \
- DEBUG_PRINT2 (" start: 0x%x\n", regstart[this_reg]); \
- PUSH_FAILURE_ITEM (regstart[this_reg]); \
- \
- DEBUG_PRINT2 (" end: 0x%x\n", regend[this_reg]); \
- PUSH_FAILURE_ITEM (regend[this_reg]); \
- \
- DEBUG_PRINT2 (" info: 0x%x\n ", reg_info[this_reg]); \
- DEBUG_PRINT2 (" match_null=%d", \
- REG_MATCH_NULL_STRING_P (reg_info[this_reg])); \
- DEBUG_PRINT2 (" active=%d", IS_ACTIVE (reg_info[this_reg])); \
- DEBUG_PRINT2 (" matched_something=%d", \
- MATCHED_SOMETHING (reg_info[this_reg])); \
- DEBUG_PRINT2 (" ever_matched=%d", \
- EVER_MATCHED_SOMETHING (reg_info[this_reg])); \
- DEBUG_PRINT1 ("\n"); \
- PUSH_FAILURE_ITEM (reg_info[this_reg].word); \
- } \
- \
- DEBUG_PRINT2 (" Pushing low active reg: %d\n", lowest_active_reg);\
- PUSH_FAILURE_ITEM (lowest_active_reg); \
- \
- DEBUG_PRINT2 (" Pushing high active reg: %d\n", highest_active_reg);\
- PUSH_FAILURE_ITEM (highest_active_reg); \
- \
- DEBUG_PRINT2 (" Pushing pattern 0x%x: ", pattern_place); \
- DEBUG_PRINT_COMPILED_PATTERN (bufp, pattern_place, pend); \
- PUSH_FAILURE_ITEM (pattern_place); \
- \
- DEBUG_PRINT2 (" Pushing string 0x%x: `", string_place); \
- DEBUG_PRINT_DOUBLE_STRING (string_place, string1, size1, string2, \
- size2); \
- DEBUG_PRINT1 ("'\n"); \
- PUSH_FAILURE_ITEM (string_place); \
- \
- DEBUG_PRINT2 (" Pushing failure id: %u\n", failure_id); \
- DEBUG_PUSH (failure_id); \
- } while (0)
-
-/* This is the number of items that are pushed and popped on the stack
- for each register. */
-#define NUM_REG_ITEMS 3
-
-/* Individual items aside from the registers. */
-#ifdef DEBUG
-#define NUM_NONREG_ITEMS 5 /* Includes failure point id. */
-#else
-#define NUM_NONREG_ITEMS 4
-#endif
-
-/* We push at most this many items on the stack. */
-#define MAX_FAILURE_ITEMS ((num_regs - 1) * NUM_REG_ITEMS + NUM_NONREG_ITEMS)
-
-/* We actually push this many items. */
-#define NUM_FAILURE_ITEMS \
- ((highest_active_reg - lowest_active_reg + 1) * NUM_REG_ITEMS \
- + NUM_NONREG_ITEMS)
-
-/* How many items can still be added to the stack without overflowing it. */
-#define REMAINING_AVAIL_SLOTS ((fail_stack).size - (fail_stack).avail)
-
-
-/* Pops what PUSH_FAIL_STACK pushes.
-
- We restore into the parameters, all of which should be lvalues:
- STR -- the saved data position.
- PAT -- the saved pattern position.
- LOW_REG, HIGH_REG -- the highest and lowest active registers.
- REGSTART, REGEND -- arrays of string positions.
- REG_INFO -- array of information about each subexpression.
-
- Also assumes the variables `fail_stack' and (if debugging), `bufp',
- `pend', `string1', `size1', `string2', and `size2'. */
-
-#define POP_FAILURE_POINT(str, pat, low_reg, high_reg, regstart, regend, reg_info)\
-{ \
- DEBUG_STATEMENT (fail_stack_elt_t failure_id;) \
- int this_reg; \
- const unsigned char *string_temp; \
- \
- assert (!FAIL_STACK_EMPTY ()); \
- \
- /* Remove failure points and point to how many regs pushed. */ \
- DEBUG_PRINT1 ("POP_FAILURE_POINT:\n"); \
- DEBUG_PRINT2 (" Before pop, next avail: %d\n", fail_stack.avail); \
- DEBUG_PRINT2 (" size: %d\n", fail_stack.size); \
- \
- assert (fail_stack.avail >= NUM_NONREG_ITEMS); \
- \
- DEBUG_POP (&failure_id); \
- DEBUG_PRINT2 (" Popping failure id: %u\n", failure_id); \
- \
- /* If the saved string location is NULL, it came from an \
- on_failure_keep_string_jump opcode, and we want to throw away the \
- saved NULL, thus retaining our current position in the string. */ \
- string_temp = POP_FAILURE_ITEM (); \
- if (string_temp != NULL) \
- str = (const char *) string_temp; \
- \
- DEBUG_PRINT2 (" Popping string 0x%x: `", str); \
- DEBUG_PRINT_DOUBLE_STRING (str, string1, size1, string2, size2); \
- DEBUG_PRINT1 ("'\n"); \
- \
- pat = (unsigned char *) POP_FAILURE_ITEM (); \
- DEBUG_PRINT2 (" Popping pattern 0x%x: ", pat); \
- DEBUG_PRINT_COMPILED_PATTERN (bufp, pat, pend); \
- \
- /* Restore register info. */ \
- high_reg = (unsigned) POP_FAILURE_ITEM (); \
- DEBUG_PRINT2 (" Popping high active reg: %d\n", high_reg); \
- \
- low_reg = (unsigned) POP_FAILURE_ITEM (); \
- DEBUG_PRINT2 (" Popping low active reg: %d\n", low_reg); \
- \
- for (this_reg = high_reg; this_reg >= low_reg; this_reg--) \
- { \
- DEBUG_PRINT2 (" Popping reg: %d\n", this_reg); \
- \
- reg_info[this_reg].word = POP_FAILURE_ITEM (); \
- DEBUG_PRINT2 (" info: 0x%x\n", reg_info[this_reg]); \
- \
- regend[this_reg] = (const char *) POP_FAILURE_ITEM (); \
- DEBUG_PRINT2 (" end: 0x%x\n", regend[this_reg]); \
- \
- regstart[this_reg] = (const char *) POP_FAILURE_ITEM (); \
- DEBUG_PRINT2 (" start: 0x%x\n", regstart[this_reg]); \
- } \
- \
- DEBUG_STATEMENT (nfailure_points_popped++); \
-} /* POP_FAILURE_POINT */
-\f
-/* re_compile_fastmap computes a ``fastmap'' for the compiled pattern in
- BUFP. A fastmap records which of the (1 << BYTEWIDTH) possible
- characters can start a string that matches the pattern. This fastmap
- is used by re_search to skip quickly over impossible starting points.
-
- The caller must supply the address of a (1 << BYTEWIDTH)-byte data
- area as BUFP->fastmap.
-
- We set the `fastmap', `fastmap_accurate', and `can_be_null' fields in
- the pattern buffer.
-
- Returns 0 if we succeed, -2 if an internal error. */
-
-int
-re_compile_fastmap (bufp)
- struct re_pattern_buffer *bufp;
-{
- int j, k;
- fail_stack_type fail_stack;
-#ifndef REGEX_MALLOC
- char *destination;
-#endif
- /* We don't push any register information onto the failure stack. */
- unsigned num_regs = 0;
-
- register char *fastmap = bufp->fastmap;
- unsigned char *pattern = bufp->buffer;
- unsigned long size = bufp->used;
- const unsigned char *p = pattern;
- register unsigned char *pend = pattern + size;
-
- /* Assume that each path through the pattern can be null until
- proven otherwise. We set this false at the bottom of switch
- statement, to which we get only if a particular path doesn't
- match the empty string. */
- boolean path_can_be_null = true;
-
- /* We aren't doing a `succeed_n' to begin with. */
- boolean succeed_n_p = false;
-
- assert (fastmap != NULL && p != NULL);
-
- INIT_FAIL_STACK ();
- bzero (fastmap, 1 << BYTEWIDTH); /* Assume nothing's valid. */
- bufp->fastmap_accurate = 1; /* It will be when we're done. */
- bufp->can_be_null = 0;
-
- while (p != pend || !FAIL_STACK_EMPTY ())
- {
- if (p == pend)
- {
- bufp->can_be_null |= path_can_be_null;
-
- /* Reset for next path. */
- path_can_be_null = true;
-
- p = fail_stack.stack[--fail_stack.avail];
- }
-
- /* We should never be about to go beyond the end of the pattern. */
- assert (p < pend);
-
-#ifdef SWITCH_ENUM_BUG
- switch ((int) ((re_opcode_t) *p++))
-#else
- switch ((re_opcode_t) *p++)
-#endif
- {
-
- /* I guess the idea here is to simply not bother with a fastmap
- if a backreference is used, since it's too hard to figure out
- the fastmap for the corresponding group. Setting
- `can_be_null' stops `re_search_2' from using the fastmap, so
- that is all we do. */
- case duplicate:
- bufp->can_be_null = 1;
- return 0;
-
-
- /* Following are the cases which match a character. These end
- with `break'. */
-
- case exactn:
- fastmap[p[1]] = 1;
- break;
-
-
- case charset:
- for (j = *p++ * BYTEWIDTH - 1; j >= 0; j--)
- if (p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH)))
- fastmap[j] = 1;
- break;
-
-
- case charset_not:
- /* Chars beyond end of map must be allowed. */
- for (j = *p * BYTEWIDTH; j < (1 << BYTEWIDTH); j++)
- fastmap[j] = 1;
-
- for (j = *p++ * BYTEWIDTH - 1; j >= 0; j--)
- if (!(p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH))))
- fastmap[j] = 1;
- break;
-
-
- case wordchar:
- for (j = 0; j < (1 << BYTEWIDTH); j++)
- if (SYNTAX (j) == Sword)
- fastmap[j] = 1;
- break;
-
-
- case notwordchar:
- for (j = 0; j < (1 << BYTEWIDTH); j++)
- if (SYNTAX (j) != Sword)
- fastmap[j] = 1;
- break;
-
-
- case anychar:
- /* `.' matches anything ... */
- for (j = 0; j < (1 << BYTEWIDTH); j++)
- fastmap[j] = 1;
-
- /* ... except perhaps newline. */
- if (!(bufp->syntax & RE_DOT_NEWLINE))
- fastmap['\n'] = 0;
-
- /* Return if we have already set `can_be_null'; if we have,
- then the fastmap is irrelevant. Something's wrong here. */
- else if (bufp->can_be_null)
- return 0;
-
- /* Otherwise, have to check alternative paths. */
- break;
-
-
-#ifdef emacs
- case syntaxspec:
- k = *p++;
- for (j = 0; j < (1 << BYTEWIDTH); j++)
- if (SYNTAX (j) == (enum syntaxcode) k)
- fastmap[j] = 1;
- break;
-
-
- case notsyntaxspec:
- k = *p++;
- for (j = 0; j < (1 << BYTEWIDTH); j++)
- if (SYNTAX (j) != (enum syntaxcode) k)
- fastmap[j] = 1;
- break;
-
-
- /* All cases after this match the empty string. These end with
- `continue'. */
-
-
- case before_dot:
- case at_dot:
- case after_dot:
- continue;
-#endif /* not emacs */
-
-
- case no_op:
- case begline:
- case endline:
- case begbuf:
- case endbuf:
- case wordbound:
- case notwordbound:
- case wordbeg:
- case wordend:
- case push_dummy_failure:
- continue;
-
-
- case jump_n:
- case pop_failure_jump:
- case maybe_pop_jump:
- case jump:
- case jump_past_alt:
- case dummy_failure_jump:
- EXTRACT_NUMBER_AND_INCR (j, p);
- p += j;
- if (j > 0)
- continue;
-
- /* Jump backward implies we just went through the body of a
- loop and matched nothing. Opcode jumped to should be
- `on_failure_jump' or `succeed_n'. Just treat it like an
- ordinary jump. For a * loop, it has pushed its failure
- point already; if so, discard that as redundant. */
- if ((re_opcode_t) *p != on_failure_jump
- && (re_opcode_t) *p != succeed_n)
- continue;
-
- p++;
- EXTRACT_NUMBER_AND_INCR (j, p);
- p += j;
-
- /* If what's on the stack is where we are now, pop it. */
- if (!FAIL_STACK_EMPTY ()
- && fail_stack.stack[fail_stack.avail - 1] == p)
- fail_stack.avail--;
-
- continue;
-
-
- case on_failure_jump:
- case on_failure_keep_string_jump:
- handle_on_failure_jump:
- EXTRACT_NUMBER_AND_INCR (j, p);
-
- /* For some patterns, e.g., `(a?)?', `p+j' here points to the
- end of the pattern. We don't want to push such a point,
- since when we restore it above, entering the switch will
- increment `p' past the end of the pattern. We don't need
- to push such a point since we obviously won't find any more
- fastmap entries beyond `pend'. Such a pattern can match
- the null string, though. */
- if (p + j < pend)
- {
- if (!PUSH_PATTERN_OP (p + j, fail_stack))
- return -2;
- }
- else
- bufp->can_be_null = 1;
-
- if (succeed_n_p)
- {
- EXTRACT_NUMBER_AND_INCR (k, p); /* Skip the n. */
- succeed_n_p = false;
- }
-
- continue;
-
-
- case succeed_n:
- /* Get to the number of times to succeed. */
- p += 2;
-
- /* Increment p past the n for when k != 0. */
- EXTRACT_NUMBER_AND_INCR (k, p);
- if (k == 0)
- {
- p -= 4;
- succeed_n_p = true; /* Spaghetti code alert. */
- goto handle_on_failure_jump;
- }
- continue;
-
-
- case set_number_at:
- p += 4;
- continue;
-
-
- case start_memory:
- case stop_memory:
- p += 2;
- continue;
-
-
- default:
- abort (); /* We have listed all the cases. */
- } /* switch *p++ */
-
- /* Getting here means we have found the possible starting
- characters for one path of the pattern -- and that the empty
- string does not match. We need not follow this path further.
- Instead, look at the next alternative (remembered on the
- stack), or quit if no more. The test at the top of the loop
- does these things. */
- path_can_be_null = false;
- p = pend;
- } /* while p */
-
- /* Set `can_be_null' for the last path (also the first path, if the
- pattern is empty). */
- bufp->can_be_null |= path_can_be_null;
- return 0;
-} /* re_compile_fastmap */
-\f
-/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
- ENDS. Subsequent matches using PATTERN_BUFFER and REGS will use
- this memory for recording register information. STARTS and ENDS
- must be allocated using the malloc library routine, and must each
- be at least NUM_REGS * sizeof (regoff_t) bytes long.
-
- If NUM_REGS == 0, then subsequent matches should allocate their own
- register data.
-
- Unless this function is called, the first search or match using
- PATTERN_BUFFER will allocate its own register data, without
- freeing the old data. */
-
-void
-re_set_registers (bufp, regs, num_regs, starts, ends)
- struct re_pattern_buffer *bufp;
- struct re_registers *regs;
- unsigned num_regs;
- regoff_t *starts, *ends;
-{
- if (num_regs)
- {
- bufp->regs_allocated = REGS_REALLOCATE;
- regs->num_regs = num_regs;
- regs->start = starts;
- regs->end = ends;
- }
- else
- {
- bufp->regs_allocated = REGS_UNALLOCATED;
- regs->num_regs = 0;
- regs->start = regs->end = (regoff_t) 0;
- }
-}
-\f
-/* Searching routines. */
-
-/* Like re_search_2, below, but only one string is specified, and
- doesn't let you say where to stop matching. */
-
-int
-re_search (bufp, string, size, startpos, range, regs)
- struct re_pattern_buffer *bufp;
- const char *string;
- int size, startpos, range;
- struct re_registers *regs;
-{
- return re_search_2 (bufp, NULL, 0, string, size, startpos, range,
- regs, size);
-}
-
-
-/* Using the compiled pattern in BUFP->buffer, first tries to match the
- virtual concatenation of STRING1 and STRING2, starting first at index
- STARTPOS, then at STARTPOS + 1, and so on.
-
- STRING1 and STRING2 have length SIZE1 and SIZE2, respectively.
-
- RANGE is how far to scan while trying to match. RANGE = 0 means try
- only at STARTPOS; in general, the last start tried is STARTPOS +
- RANGE.
-
- In REGS, return the indices of the virtual concatenation of STRING1
- and STRING2 that matched the entire BUFP->buffer and its contained
- subexpressions.
-
- Do not consider matching one past the index STOP in the virtual
- concatenation of STRING1 and STRING2.
-
- We return either the position in the strings at which the match was
- found, -1 if no match, or -2 if error (such as failure
- stack overflow). */
-
-int
-re_search_2 (bufp, string1, size1, string2, size2, startpos, range, regs, stop)
- struct re_pattern_buffer *bufp;
- const char *string1, *string2;
- int size1, size2;
- int startpos;
- int range;
- struct re_registers *regs;
- int stop;
-{
- int val;
- register char *fastmap = bufp->fastmap;
- register char *translate = bufp->translate;
- int total_size = size1 + size2;
- int endpos = startpos + range;
-
- /* Check for out-of-range STARTPOS. */
- if (startpos < 0 || startpos > total_size)
- return -1;
-
- /* Fix up RANGE if it might eventually take us outside
- the virtual concatenation of STRING1 and STRING2. */
- if (endpos < -1)
- range = -1 - startpos;
- else if (endpos > total_size)
- range = total_size - startpos;
-
- /* If the search isn't to be a backwards one, don't waste time in a
- search for a pattern that must be anchored. */
- if (bufp->used > 0 && (re_opcode_t) bufp->buffer[0] == begbuf && range > 0)
- {
- if (startpos > 0)
- return -1;
- else
- range = 1;
- }
-
- /* Update the fastmap now if not correct already. */
- if (fastmap && !bufp->fastmap_accurate)
- if (re_compile_fastmap (bufp) == -2)
- return -2;
-
- /* Loop through the string, looking for a place to start matching. */
- for (;;)
- {
- /* If a fastmap is supplied, skip quickly over characters that
- cannot be the start of a match. If the pattern can match the
- null string, however, we don't need to skip characters; we want
- the first null string. */
- if (fastmap && startpos < total_size && !bufp->can_be_null)
- {
- if (range > 0) /* Searching forwards. */
- {
- register const char *d;
- register int lim = 0;
- int irange = range;
-
- if (startpos < size1 && startpos + range >= size1)
- lim = range - (size1 - startpos);
-
- d = (startpos >= size1 ? string2 - size1 : string1) + startpos;
-
- /* Written out as an if-else to avoid testing `translate'
- inside the loop. */
- if (translate)
- while (range > lim
- && !fastmap[(unsigned char)
- translate[(unsigned char) *d++]])
- range--;
- else
- while (range > lim && !fastmap[(unsigned char) *d++])
- range--;
-
- startpos += irange - range;
- }
- else /* Searching backwards. */
- {
- register char c = (size1 == 0 || startpos >= size1
- ? string2[startpos - size1]
- : string1[startpos]);
-
- if (!fastmap[(unsigned char) TRANSLATE (c)])
- goto advance;
- }
- }
-
- /* If can't match the null string, and that's all we have left, fail. */
- if (range >= 0 && startpos == total_size && fastmap
- && !bufp->can_be_null)
- return -1;
-
- val = re_match_2 (bufp, string1, size1, string2, size2,
- startpos, regs, stop);
- if (val >= 0)
- return startpos;
-
- if (val == -2)
- return -2;
-
- advance:
- if (!range)
- break;
- else if (range > 0)
- {
- range--;
- startpos++;
- }
- else
- {
- range++;
- startpos--;
- }
- }
- return -1;
-} /* re_search_2 */
-\f
-/* Declarations and macros for re_match_2. */
-
-static int bcmp_translate ();
-static boolean alt_match_null_string_p (),
- common_op_match_null_string_p (),
- group_match_null_string_p ();
-
-/* Structure for per-register (a.k.a. per-group) information.
- This must not be longer than one word, because we push this value
- onto the failure stack. Other register information, such as the
- starting and ending positions (which are addresses), and the list of
- inner groups (which is a bits list) are maintained in separate
- variables.
-
- We are making a (strictly speaking) nonportable assumption here: that
- the compiler will pack our bit fields into something that fits into
- the type of `word', i.e., is something that fits into one item on the
- failure stack. */
-typedef union
-{
- fail_stack_elt_t word;
- struct
- {
- /* This field is one if this group can match the empty string,
- zero if not. If not yet determined, `MATCH_NULL_UNSET_VALUE'. */
-#define MATCH_NULL_UNSET_VALUE 3
- unsigned match_null_string_p : 2;
- unsigned is_active : 1;
- unsigned matched_something : 1;
- unsigned ever_matched_something : 1;
- } bits;
-} register_info_type;
-
-#define REG_MATCH_NULL_STRING_P(R) ((R).bits.match_null_string_p)
-#define IS_ACTIVE(R) ((R).bits.is_active)
-#define MATCHED_SOMETHING(R) ((R).bits.matched_something)
-#define EVER_MATCHED_SOMETHING(R) ((R).bits.ever_matched_something)
-
-
-/* Call this when have matched a real character; it sets `matched' flags
- for the subexpressions which we are currently inside. Also records
- that those subexprs have matched. */
-#define SET_REGS_MATCHED() \
- do \
- { \
- unsigned r; \
- for (r = lowest_active_reg; r <= highest_active_reg; r++) \
- { \
- MATCHED_SOMETHING (reg_info[r]) \
- = EVER_MATCHED_SOMETHING (reg_info[r]) \
- = 1; \
- } \
- } \
- while (0)
-
-
-/* This converts PTR, a pointer into one of the search strings `string1'
- and `string2' into an offset from the beginning of that string. */
-#define POINTER_TO_OFFSET(ptr) \
- (FIRST_STRING_P (ptr) ? (ptr) - string1 : (ptr) - string2 + size1)
-
-/* Registers are set to a sentinel when they haven't yet matched. */
-#define REG_UNSET_VALUE ((char *) -1)
-#define REG_UNSET(e) ((e) == REG_UNSET_VALUE)
-
-
-/* Macros for dealing with the split strings in re_match_2. */
-
-#define MATCHING_IN_FIRST_STRING (dend == end_match_1)
-
-/* Call before fetching a character with *d. This switches over to
- string2 if necessary. */
-#define PREFETCH() \
- while (d == dend) \
- { \
- /* End of string2 => fail. */ \
- if (dend == end_match_2) \
- goto fail; \
- /* End of string1 => advance to string2. */ \
- d = string2; \
- dend = end_match_2; \
- }
-
-
-/* Test if at very beginning or at very end of the virtual concatenation
- of `string1' and `string2'. If only one string, it's `string2'. */
-#define AT_STRINGS_BEG(d) ((d) == (size1 ? string1 : string2) || !size2)
-#define AT_STRINGS_END(d) ((d) == end2)
-
-
-/* Test if D points to a character which is word-constituent. We have
- two special cases to check for: if past the end of string1, look at
- the first character in string2; and if before the beginning of
- string2, look at the last character in string1. */
-#define WORDCHAR_P(d) \
- (SYNTAX ((d) == end1 ? *string2 \
- : (d) == string2 - 1 ? *(end1 - 1) : *(d)) \
- == Sword)
-
-/* Test if the character before D and the one at D differ with respect
- to being word-constituent. */
-#define AT_WORD_BOUNDARY(d) \
- (AT_STRINGS_BEG (d) || AT_STRINGS_END (d) \
- || WORDCHAR_P (d - 1) != WORDCHAR_P (d))
-
-
-/* Free everything we malloc. */
-#ifdef REGEX_MALLOC
-#define FREE_VAR(var) if (var) free (var); var = NULL
-#define FREE_VARIABLES() \
- do { \
- FREE_VAR (fail_stack.stack); \
- FREE_VAR (regstart); \
- FREE_VAR (regend); \
- FREE_VAR (old_regstart); \
- FREE_VAR (old_regend); \
- FREE_VAR (best_regstart); \
- FREE_VAR (best_regend); \
- FREE_VAR (reg_info); \
- FREE_VAR (reg_dummy); \
- FREE_VAR (reg_info_dummy); \
- } while (0)
-#else /* not REGEX_MALLOC */
-/* Some MIPS systems (at least) want this to free alloca'd storage. */
-#define FREE_VARIABLES() alloca (0)
-#endif /* not REGEX_MALLOC */
-
-
-/* These values must meet several constraints. They must not be valid
- register values; since we have a limit of 255 registers (because
- we use only one byte in the pattern for the register number), we can
- use numbers larger than 255. They must differ by 1, because of
- NUM_FAILURE_ITEMS above. And the value for the lowest register must
- be larger than the value for the highest register, so we do not try
- to actually save any registers when none are active. */
-#define NO_HIGHEST_ACTIVE_REG (1 << BYTEWIDTH)
-#define NO_LOWEST_ACTIVE_REG (NO_HIGHEST_ACTIVE_REG + 1)
-\f
-/* Matching routines. */
-
-#ifndef emacs /* Emacs never uses this. */
-/* re_match is like re_match_2 except it takes only a single string. */
-
-int
-re_match (bufp, string, size, pos, regs)
- struct re_pattern_buffer *bufp;
- const char *string;
- int size, pos;
- struct re_registers *regs;
- {
- return re_match_2 (bufp, NULL, 0, string, size, pos, regs, size);
-}
-#endif /* not emacs */
-
-
-/* re_match_2 matches the compiled pattern in BUFP against the
- the (virtual) concatenation of STRING1 and STRING2 (of length SIZE1
- and SIZE2, respectively). We start matching at POS, and stop
- matching at STOP.
-
- If REGS is non-null and the `no_sub' field of BUFP is nonzero, we
- store offsets for the substring each group matched in REGS. See the
- documentation for exactly how many groups we fill.
-
- We return -1 if no match, -2 if an internal error (such as the
- failure stack overflowing). Otherwise, we return the length of the
- matched substring. */
-
-int
-re_match_2 (bufp, string1, size1, string2, size2, pos, regs, stop)
- struct re_pattern_buffer *bufp;
- const char *string1, *string2;
- int size1, size2;
- int pos;
- struct re_registers *regs;
- int stop;
-{
- /* General temporaries. */
- int mcnt;
- unsigned char *p1;
-
- /* Just past the end of the corresponding string. */
- const char *end1, *end2;
-
- /* Pointers into string1 and string2, just past the last characters in
- each to consider matching. */
- const char *end_match_1, *end_match_2;
-
- /* Where we are in the data, and the end of the current string. */
- const char *d, *dend;
-
- /* Where we are in the pattern, and the end of the pattern. */
- unsigned char *p = bufp->buffer;
- register unsigned char *pend = p + bufp->used;
-
- /* We use this to map every character in the string. */
- char *translate = bufp->translate;
-
- /* Failure point stack. Each place that can handle a failure further
- down the line pushes a failure point on this stack. It consists of
- restart, regend, and reg_info for all registers corresponding to
- the subexpressions we're currently inside, plus the number of such
- registers, and, finally, two char *'s. The first char * is where
- to resume scanning the pattern; the second one is where to resume
- scanning the strings. If the latter is zero, the failure point is
- a ``dummy''; if a failure happens and the failure point is a dummy,
- it gets discarded and the next next one is tried. */
- fail_stack_type fail_stack;
-#ifdef DEBUG
- static unsigned failure_id = 0;
- unsigned nfailure_points_pushed = 0, nfailure_points_popped = 0;
-#endif
-
- /* We fill all the registers internally, independent of what we
- return, for use in backreferences. The number here includes
- an element for register zero. */
- unsigned num_regs = bufp->re_nsub + 1;
-
- /* The currently active registers. */
- unsigned lowest_active_reg = NO_LOWEST_ACTIVE_REG;
- unsigned highest_active_reg = NO_HIGHEST_ACTIVE_REG;
-
- /* Information on the contents of registers. These are pointers into
- the input strings; they record just what was matched (on this
- attempt) by a subexpression part of the pattern, that is, the
- regnum-th regstart pointer points to where in the pattern we began
- matching and the regnum-th regend points to right after where we
- stopped matching the regnum-th subexpression. (The zeroth register
- keeps track of what the whole pattern matches.) */
- const char **regstart = NULL, **regend = NULL;
-
- /* If a group that's operated upon by a repetition operator fails to
- match anything, then the register for its start will need to be
- restored because it will have been set to wherever in the string we
- are when we last see its open-group operator. Similarly for a
- register's end. */
- const char **old_regstart = NULL, **old_regend = NULL;
-
- /* The is_active field of reg_info helps us keep track of which (possibly
- nested) subexpressions we are currently in. The matched_something
- field of reg_info[reg_num] helps us tell whether or not we have
- matched any of the pattern so far this time through the reg_num-th
- subexpression. These two fields get reset each time through any
- loop their register is in. */
- register_info_type *reg_info = NULL;
-
- /* The following record the register info as found in the above
- variables when we find a match better than any we've seen before.
- This happens as we backtrack through the failure points, which in
- turn happens only if we have not yet matched the entire string. */
- unsigned best_regs_set = false;
- const char **best_regstart = NULL, **best_regend = NULL;
-
- /* Logically, this is `best_regend[0]'. But we don't want to have to
- allocate space for that if we're not allocating space for anything
- else (see below). Also, we never need info about register 0 for
- any of the other register vectors, and it seems rather a kludge to
- treat `best_regend' differently than the rest. So we keep track of
- the end of the best match so far in a separate variable. We
- initialize this to NULL so that when we backtrack the first time
- and need to test it, it's not garbage. */
- const char *match_end = NULL;
-
- /* Used when we pop values we don't care about. */
- const char **reg_dummy = NULL;
- register_info_type *reg_info_dummy = NULL;
-
-#ifdef DEBUG
- /* Counts the total number of registers pushed. */
- unsigned num_regs_pushed = 0;
-#endif
-
- DEBUG_PRINT1 ("\n\nEntering re_match_2.\n");
-
- INIT_FAIL_STACK ();
-
- /* Do not bother to initialize all the register variables if there are
- no groups in the pattern, as it takes a fair amount of time. If
- there are groups, we include space for register 0 (the whole
- pattern), even though we never use it, since it simplifies the
- array indexing. We should fix this. */
- if (bufp->re_nsub)
- {
- regstart = REGEX_TALLOC (num_regs, const char *);
- regend = REGEX_TALLOC (num_regs, const char *);
- old_regstart = REGEX_TALLOC (num_regs, const char *);
- old_regend = REGEX_TALLOC (num_regs, const char *);
- best_regstart = REGEX_TALLOC (num_regs, const char *);
- best_regend = REGEX_TALLOC (num_regs, const char *);
- reg_info = REGEX_TALLOC (num_regs, register_info_type);
- reg_dummy = REGEX_TALLOC (num_regs, const char *);
- reg_info_dummy = REGEX_TALLOC (num_regs, register_info_type);
-
- if (!(regstart && regend && old_regstart && old_regend && reg_info
- && best_regstart && best_regend && reg_dummy && reg_info_dummy))
- {
- FREE_VARIABLES ();
- return -2;
- }
- }
-#ifdef REGEX_MALLOC
- else
- {
- /* We must initialize all our variables to NULL, so that
- `FREE_VARIABLES' doesn't try to free them. */
- regstart = regend = old_regstart = old_regend = best_regstart
- = best_regend = reg_dummy = NULL;
- reg_info = reg_info_dummy = (register_info_type *) NULL;
- }
-#endif /* REGEX_MALLOC */
-
- /* The starting position is bogus. */
- if (pos < 0 || pos > size1 + size2)
- {
- FREE_VARIABLES ();
- return -1;
- }
-
- /* Initialize subexpression text positions to -1 to mark ones that no
- start_memory/stop_memory has been seen for. Also initialize the
- register information struct. */
- for (mcnt = 1; mcnt < num_regs; mcnt++)
- {
- regstart[mcnt] = regend[mcnt]
- = old_regstart[mcnt] = old_regend[mcnt] = REG_UNSET_VALUE;
-
- REG_MATCH_NULL_STRING_P (reg_info[mcnt]) = MATCH_NULL_UNSET_VALUE;
- IS_ACTIVE (reg_info[mcnt]) = 0;
- MATCHED_SOMETHING (reg_info[mcnt]) = 0;
- EVER_MATCHED_SOMETHING (reg_info[mcnt]) = 0;
- }
-
- /* We move `string1' into `string2' if the latter's empty -- but not if
- `string1' is null. */
- if (size2 == 0 && string1 != NULL)
- {
- string2 = string1;
- size2 = size1;
- string1 = 0;
- size1 = 0;
- }
- end1 = string1 + size1;
- end2 = string2 + size2;
-
- /* Compute where to stop matching, within the two strings. */
- if (stop <= size1)
- {
- end_match_1 = string1 + stop;
- end_match_2 = string2;
- }
- else
- {
- end_match_1 = end1;
- end_match_2 = string2 + stop - size1;
- }
-
- /* `p' scans through the pattern as `d' scans through the data.
- `dend' is the end of the input string that `d' points within. `d'
- is advanced into the following input string whenever necessary, but
- this happens before fetching; therefore, at the beginning of the
- loop, `d' can be pointing at the end of a string, but it cannot
- equal `string2'. */
- if (size1 > 0 && pos <= size1)
- {
- d = string1 + pos;
- dend = end_match_1;
- }
- else
- {
- d = string2 + pos - size1;
- dend = end_match_2;
- }
-
- DEBUG_PRINT1 ("The compiled pattern is: ");
- DEBUG_PRINT_COMPILED_PATTERN (bufp, p, pend);
- DEBUG_PRINT1 ("The string to match is: `");
- DEBUG_PRINT_DOUBLE_STRING (d, string1, size1, string2, size2);
- DEBUG_PRINT1 ("'\n");
-
- /* This loops over pattern commands. It exits by returning from the
- function if the match is complete, or it drops through if the match
- fails at this starting point in the input data. */
- for (;;)
- {
- DEBUG_PRINT2 ("\n0x%x: ", p);
-
- if (p == pend)
- { /* End of pattern means we might have succeeded. */
- DEBUG_PRINT1 ("end of pattern ... ");
-
- /* If we haven't matched the entire string, and we want the
- longest match, try backtracking. */
- if (d != end_match_2)
- {
- DEBUG_PRINT1 ("backtracking.\n");
-
- if (!FAIL_STACK_EMPTY ())
- { /* More failure points to try. */
- boolean same_str_p = (FIRST_STRING_P (match_end)
- == MATCHING_IN_FIRST_STRING);
-
- /* If exceeds best match so far, save it. */
- if (!best_regs_set
- || (same_str_p && d > match_end)
- || (!same_str_p && !MATCHING_IN_FIRST_STRING))
- {
- best_regs_set = true;
- match_end = d;
-
- DEBUG_PRINT1 ("\nSAVING match as best so far.\n");
-
- for (mcnt = 1; mcnt < num_regs; mcnt++)
- {
- best_regstart[mcnt] = regstart[mcnt];
- best_regend[mcnt] = regend[mcnt];
- }
- }
- goto fail;
- }
-
- /* If no failure points, don't restore garbage. */
- else if (best_regs_set)
- {
- restore_best_regs:
- /* Restore best match. It may happen that `dend ==
- end_match_1' while the restored d is in string2.
- For example, the pattern `x.*y.*z' against the
- strings `x-' and `y-z-', if the two strings are
- not consecutive in memory. */
- DEBUG_PRINT1 ("Restoring best registers.\n");
-
- d = match_end;
- dend = ((d >= string1 && d <= end1)
- ? end_match_1 : end_match_2);
-
- for (mcnt = 1; mcnt < num_regs; mcnt++)
- {
- regstart[mcnt] = best_regstart[mcnt];
- regend[mcnt] = best_regend[mcnt];
- }
- }
- } /* d != end_match_2 */
-
- DEBUG_PRINT1 ("Accepting match.\n");
-
- /* If caller wants register contents data back, do it. */
- if (regs && !bufp->no_sub)
- {
- /* Have the register data arrays been allocated? */
- if (bufp->regs_allocated == REGS_UNALLOCATED)
- { /* No. So allocate them with malloc. We need one
- extra element beyond `num_regs' for the `-1' marker
- GNU code uses. */
- regs->num_regs = MAX (RE_NREGS, num_regs + 1);
- regs->start = TALLOC (regs->num_regs, regoff_t);
- regs->end = TALLOC (regs->num_regs, regoff_t);
- if (regs->start == NULL || regs->end == NULL)
- return -2;
- bufp->regs_allocated = REGS_REALLOCATE;
- }
- else if (bufp->regs_allocated == REGS_REALLOCATE)
- { /* Yes. If we need more elements than were already
- allocated, reallocate them. If we need fewer, just
- leave it alone. */
- if (regs->num_regs < num_regs + 1)
- {
- regs->num_regs = num_regs + 1;
- RETALLOC (regs->start, regs->num_regs, regoff_t);
- RETALLOC (regs->end, regs->num_regs, regoff_t);
- if (regs->start == NULL || regs->end == NULL)
- return -2;
- }
- }
- else
- assert (bufp->regs_allocated == REGS_FIXED);
-
- /* Convert the pointer data in `regstart' and `regend' to
- indices. Register zero has to be set differently,
- since we haven't kept track of any info for it. */
- if (regs->num_regs > 0)
- {
- regs->start[0] = pos;
- regs->end[0] = (MATCHING_IN_FIRST_STRING ? d - string1
- : d - string2 + size1);
- }
-
- /* Go through the first `min (num_regs, regs->num_regs)'
- registers, since that is all we initialized. */
- for (mcnt = 1; mcnt < MIN (num_regs, regs->num_regs); mcnt++)
- {
- if (REG_UNSET (regstart[mcnt]) || REG_UNSET (regend[mcnt]))
- regs->start[mcnt] = regs->end[mcnt] = -1;
- else
- {
- regs->start[mcnt] = POINTER_TO_OFFSET (regstart[mcnt]);
- regs->end[mcnt] = POINTER_TO_OFFSET (regend[mcnt]);
- }
- }
-
- /* If the regs structure we return has more elements than
- were in the pattern, set the extra elements to -1. If
- we (re)allocated the registers, this is the case,
- because we always allocate enough to have at least one
- -1 at the end. */
- for (mcnt = num_regs; mcnt < regs->num_regs; mcnt++)
- regs->start[mcnt] = regs->end[mcnt] = -1;
- } /* regs && !bufp->no_sub */
-
- FREE_VARIABLES ();
- DEBUG_PRINT4 ("%u failure points pushed, %u popped (%u remain).\n",
- nfailure_points_pushed, nfailure_points_popped,
- nfailure_points_pushed - nfailure_points_popped);
- DEBUG_PRINT2 ("%u registers pushed.\n", num_regs_pushed);
-
- mcnt = d - pos - (MATCHING_IN_FIRST_STRING
- ? string1
- : string2 - size1);
-
- DEBUG_PRINT2 ("Returning %d from re_match_2.\n", mcnt);
-
- return mcnt;
- }
-
- /* Otherwise match next pattern command. */
-#ifdef SWITCH_ENUM_BUG
- switch ((int) ((re_opcode_t) *p++))
-#else
- switch ((re_opcode_t) *p++)
-#endif
- {
- /* Ignore these. Used to ignore the n of succeed_n's which
- currently have n == 0. */
- case no_op:
- DEBUG_PRINT1 ("EXECUTING no_op.\n");
- break;
-
-
- /* Match the next n pattern characters exactly. The following
- byte in the pattern defines n, and the n bytes after that
- are the characters to match. */
- case exactn:
- mcnt = *p++;
- DEBUG_PRINT2 ("EXECUTING exactn %d.\n", mcnt);
-
- /* This is written out as an if-else so we don't waste time
- testing `translate' inside the loop. */
- if (translate)
- {
- do
- {
- PREFETCH ();
- if (translate[(unsigned char) *d++] != (char) *p++)
- goto fail;
- }
- while (--mcnt);
- }
- else
- {
- do
- {
- PREFETCH ();
- if (*d++ != (char) *p++) goto fail;
- }
- while (--mcnt);
- }
- SET_REGS_MATCHED ();
- break;
-
-
- /* Match any character except possibly a newline or a null. */
- case anychar:
- DEBUG_PRINT1 ("EXECUTING anychar.\n");
-
- PREFETCH ();
-
- if ((!(bufp->syntax & RE_DOT_NEWLINE) && TRANSLATE (*d) == '\n')
- || (bufp->syntax & RE_DOT_NOT_NULL && TRANSLATE (*d) == '\000'))
- goto fail;
-
- SET_REGS_MATCHED ();
- DEBUG_PRINT2 (" Matched `%d'.\n", *d);
- d++;
- break;
-
-
- case charset:
- case charset_not:
- {
- register unsigned char c;
- boolean not = (re_opcode_t) *(p - 1) == charset_not;
-
- DEBUG_PRINT2 ("EXECUTING charset%s.\n", not ? "_not" : "");
-
- PREFETCH ();
- c = TRANSLATE (*d); /* The character to match. */
-
- /* Cast to `unsigned' instead of `unsigned char' in case the
- bit list is a full 32 bytes long. */
- if (c < (unsigned) (*p * BYTEWIDTH)
- && p[1 + c / BYTEWIDTH] & (1 << (c % BYTEWIDTH)))
- not = !not;
-
- p += 1 + *p;
-
- if (!not) goto fail;
-
- SET_REGS_MATCHED ();
- d++;
- break;
- }
-
-
- /* The beginning of a group is represented by start_memory.
- The arguments are the register number in the next byte, and the
- number of groups inner to this one in the next. The text
- matched within the group is recorded (in the internal
- registers data structure) under the register number. */
- case start_memory:
- DEBUG_PRINT3 ("EXECUTING start_memory %d (%d):\n", *p, p[1]);
-
- /* Find out if this group can match the empty string. */
- p1 = p; /* To send to group_match_null_string_p. */
-
- if (REG_MATCH_NULL_STRING_P (reg_info[*p]) == MATCH_NULL_UNSET_VALUE)
- REG_MATCH_NULL_STRING_P (reg_info[*p])
- = group_match_null_string_p (&p1, pend, reg_info);
-
- /* Save the position in the string where we were the last time
- we were at this open-group operator in case the group is
- operated upon by a repetition operator, e.g., with `(a*)*b'
- against `ab'; then we want to ignore where we are now in
- the string in case this attempt to match fails. */
- old_regstart[*p] = REG_MATCH_NULL_STRING_P (reg_info[*p])
- ? REG_UNSET (regstart[*p]) ? d : regstart[*p]
- : regstart[*p];
- DEBUG_PRINT2 (" old_regstart: %d\n",
- POINTER_TO_OFFSET (old_regstart[*p]));
-
- regstart[*p] = d;
- DEBUG_PRINT2 (" regstart: %d\n", POINTER_TO_OFFSET (regstart[*p]));
-
- IS_ACTIVE (reg_info[*p]) = 1;
- MATCHED_SOMETHING (reg_info[*p]) = 0;
-
- /* This is the new highest active register. */
- highest_active_reg = *p;
-
- /* If nothing was active before, this is the new lowest active
- register. */
- if (lowest_active_reg == NO_LOWEST_ACTIVE_REG)
- lowest_active_reg = *p;
-
- /* Move past the register number and inner group count. */
- p += 2;
- break;
-
-
- /* The stop_memory opcode represents the end of a group. Its
- arguments are the same as start_memory's: the register
- number, and the number of inner groups. */
- case stop_memory:
- DEBUG_PRINT3 ("EXECUTING stop_memory %d (%d):\n", *p, p[1]);
-
- /* We need to save the string position the last time we were at
- this close-group operator in case the group is operated
- upon by a repetition operator, e.g., with `((a*)*(b*)*)*'
- against `aba'; then we want to ignore where we are now in
- the string in case this attempt to match fails. */
- old_regend[*p] = REG_MATCH_NULL_STRING_P (reg_info[*p])
- ? REG_UNSET (regend[*p]) ? d : regend[*p]
- : regend[*p];
- DEBUG_PRINT2 (" old_regend: %d\n",
- POINTER_TO_OFFSET (old_regend[*p]));
-
- regend[*p] = d;
- DEBUG_PRINT2 (" regend: %d\n", POINTER_TO_OFFSET (regend[*p]));
-
- /* This register isn't active anymore. */
- IS_ACTIVE (reg_info[*p]) = 0;
-
- /* If this was the only register active, nothing is active
- anymore. */
- if (lowest_active_reg == highest_active_reg)
- {
- lowest_active_reg = NO_LOWEST_ACTIVE_REG;
- highest_active_reg = NO_HIGHEST_ACTIVE_REG;
- }
- else
- { /* We must scan for the new highest active register, since
- it isn't necessarily one less than now: consider
- (a(b)c(d(e)f)g). When group 3 ends, after the f), the
- new highest active register is 1. */
- unsigned char r = *p - 1;
- while (r > 0 && !IS_ACTIVE (reg_info[r]))
- r--;
-
- /* If we end up at register zero, that means that we saved
- the registers as the result of an `on_failure_jump', not
- a `start_memory', and we jumped to past the innermost
- `stop_memory'. For example, in ((.)*) we save
- registers 1 and 2 as a result of the *, but when we pop
- back to the second ), we are at the stop_memory 1.
- Thus, nothing is active. */
- if (r == 0)
- {
- lowest_active_reg = NO_LOWEST_ACTIVE_REG;
- highest_active_reg = NO_HIGHEST_ACTIVE_REG;
- }
- else
- highest_active_reg = r;
- }
-
- /* If just failed to match something this time around with a
- group that's operated on by a repetition operator, try to
- force exit from the ``loop'', and restore the register
- information for this group that we had before trying this
- last match. */
- if ((!MATCHED_SOMETHING (reg_info[*p])
- || (re_opcode_t) p[-3] == start_memory)
- && (p + 2) < pend)
- {
- boolean is_a_jump_n = false;
-
- p1 = p + 2;
- mcnt = 0;
- switch ((re_opcode_t) *p1++)
- {
- case jump_n:
- is_a_jump_n = true;
- case pop_failure_jump:
- case maybe_pop_jump:
- case jump:
- case dummy_failure_jump:
- EXTRACT_NUMBER_AND_INCR (mcnt, p1);
- if (is_a_jump_n)
- p1 += 2;
- break;
-
- default:
- /* do nothing */ ;
- }
- p1 += mcnt;
-
- /* If the next operation is a jump backwards in the pattern
- to an on_failure_jump right before the start_memory
- corresponding to this stop_memory, exit from the loop
- by forcing a failure after pushing on the stack the
- on_failure_jump's jump in the pattern, and d. */
- if (mcnt < 0 && (re_opcode_t) *p1 == on_failure_jump
- && (re_opcode_t) p1[3] == start_memory && p1[4] == *p)
- {
- /* If this group ever matched anything, then restore
- what its registers were before trying this last
- failed match, e.g., with `(a*)*b' against `ab' for
- regstart[1], and, e.g., with `((a*)*(b*)*)*'
- against `aba' for regend[3].
-
- Also restore the registers for inner groups for,
- e.g., `((a*)(b*))*' against `aba' (register 3 would
- otherwise get trashed). */
-
- if (EVER_MATCHED_SOMETHING (reg_info[*p]))
- {
- unsigned r;
-
- EVER_MATCHED_SOMETHING (reg_info[*p]) = 0;
-
- /* Restore this and inner groups' (if any) registers. */
- for (r = *p; r < *p + *(p + 1); r++)
- {
- regstart[r] = old_regstart[r];
-
- /* xx why this test? */
- if ((int) old_regend[r] >= (int) regstart[r])
- regend[r] = old_regend[r];
- }
- }
- p1++;
- EXTRACT_NUMBER_AND_INCR (mcnt, p1);
- PUSH_FAILURE_POINT (p1 + mcnt, d, -2);
-
- goto fail;
- }
- }
-
- /* Move past the register number and the inner group count. */
- p += 2;
- break;
-
-
- /* \<digit> has been turned into a `duplicate' command which is
- followed by the numeric value of <digit> as the register number. */
- case duplicate:
- {
- register const char *d2, *dend2;
- int regno = *p++; /* Get which register to match against. */
- DEBUG_PRINT2 ("EXECUTING duplicate %d.\n", regno);
-
- /* Can't back reference a group which we've never matched. */
- if (REG_UNSET (regstart[regno]) || REG_UNSET (regend[regno]))
- goto fail;
-
- /* Where in input to try to start matching. */
- d2 = regstart[regno];
-
- /* Where to stop matching; if both the place to start and
- the place to stop matching are in the same string, then
- set to the place to stop, otherwise, for now have to use
- the end of the first string. */
-
- dend2 = ((FIRST_STRING_P (regstart[regno])
- == FIRST_STRING_P (regend[regno]))
- ? regend[regno] : end_match_1);
- for (;;)
- {
- /* If necessary, advance to next segment in register
- contents. */
- while (d2 == dend2)
- {
- if (dend2 == end_match_2) break;
- if (dend2 == regend[regno]) break;
-
- /* End of string1 => advance to string2. */
- d2 = string2;
- dend2 = regend[regno];
- }
- /* At end of register contents => success */
- if (d2 == dend2) break;
-
- /* If necessary, advance to next segment in data. */
- PREFETCH ();
-
- /* How many characters left in this segment to match. */
- mcnt = dend - d;
-
- /* Want how many consecutive characters we can match in
- one shot, so, if necessary, adjust the count. */
- if (mcnt > dend2 - d2)
- mcnt = dend2 - d2;
-
- /* Compare that many; failure if mismatch, else move
- past them. */
- if (translate
- ? bcmp_translate (d, d2, mcnt, translate)
- : bcmp (d, d2, mcnt))
- goto fail;
- d += mcnt, d2 += mcnt;
- }
- }
- break;
-
-
- /* begline matches the empty string at the beginning of the string
- (unless `not_bol' is set in `bufp'), and, if
- `newline_anchor' is set, after newlines. */
- case begline:
- DEBUG_PRINT1 ("EXECUTING begline.\n");
-
- if (AT_STRINGS_BEG (d))
- {
- if (!bufp->not_bol) break;
- }
- else if (d[-1] == '\n' && bufp->newline_anchor)
- {
- break;
- }
- /* In all other cases, we fail. */
- goto fail;
-
-
- /* endline is the dual of begline. */
- case endline:
- DEBUG_PRINT1 ("EXECUTING endline.\n");
-
- if (AT_STRINGS_END (d))
- {
- if (!bufp->not_eol) break;
- }
-
- /* We have to ``prefetch'' the next character. */
- else if ((d == end1 ? *string2 : *d) == '\n'
- && bufp->newline_anchor)
- {
- break;
- }
- goto fail;
-
-
- /* Match at the very beginning of the data. */
- case begbuf:
- DEBUG_PRINT1 ("EXECUTING begbuf.\n");
- if (AT_STRINGS_BEG (d))
- break;
- goto fail;
-
-
- /* Match at the very end of the data. */
- case endbuf:
- DEBUG_PRINT1 ("EXECUTING endbuf.\n");
- if (AT_STRINGS_END (d))
- break;
- goto fail;
-
-
- /* on_failure_keep_string_jump is used to optimize `.*\n'. It
- pushes NULL as the value for the string on the stack. Then
- `pop_failure_point' will keep the current value for the
- string, instead of restoring it. To see why, consider
- matching `foo\nbar' against `.*\n'. The .* matches the foo;
- then the . fails against the \n. But the next thing we want
- to do is match the \n against the \n; if we restored the
- string value, we would be back at the foo.
-
- Because this is used only in specific cases, we don't need to
- check all the things that `on_failure_jump' does, to make
- sure the right things get saved on the stack. Hence we don't
- share its code. The only reason to push anything on the
- stack at all is that otherwise we would have to change
- `anychar's code to do something besides goto fail in this
- case; that seems worse than this. */
- case on_failure_keep_string_jump:
- DEBUG_PRINT1 ("EXECUTING on_failure_keep_string_jump");
-
- EXTRACT_NUMBER_AND_INCR (mcnt, p);
- DEBUG_PRINT3 (" %d (to 0x%x):\n", mcnt, p + mcnt);
-
- PUSH_FAILURE_POINT (p + mcnt, NULL, -2);
- break;
-
-
- /* Uses of on_failure_jump:
-
- Each alternative starts with an on_failure_jump that points
- to the beginning of the next alternative. Each alternative
- except the last ends with a jump that in effect jumps past
- the rest of the alternatives. (They really jump to the
- ending jump of the following alternative, because tensioning
- these jumps is a hassle.)
-
- Repeats start with an on_failure_jump that points past both
- the repetition text and either the following jump or
- pop_failure_jump back to this on_failure_jump. */
- case on_failure_jump:
- on_failure:
- DEBUG_PRINT1 ("EXECUTING on_failure_jump");
-
- EXTRACT_NUMBER_AND_INCR (mcnt, p);
- DEBUG_PRINT3 (" %d (to 0x%x)", mcnt, p + mcnt);
-
- /* If this on_failure_jump comes right before a group (i.e.,
- the original * applied to a group), save the information
- for that group and all inner ones, so that if we fail back
- to this point, the group's information will be correct.
- For example, in \(a*\)*\1, we need the preceding group,
- and in \(\(a*\)b*\)\2, we need the inner group. */
-
- /* We can't use `p' to check ahead because we push
- a failure point to `p + mcnt' after we do this. */
- p1 = p;
-
- /* We need to skip no_op's before we look for the
- start_memory in case this on_failure_jump is happening as
- the result of a completed succeed_n, as in \(a\)\{1,3\}b\1
- against aba. */
- while (p1 < pend && (re_opcode_t) *p1 == no_op)
- p1++;
-
- if (p1 < pend && (re_opcode_t) *p1 == start_memory)
- {
- /* We have a new highest active register now. This will
- get reset at the start_memory we are about to get to,
- but we will have saved all the registers relevant to
- this repetition op, as described above. */
- highest_active_reg = *(p1 + 1) + *(p1 + 2);
- if (lowest_active_reg == NO_LOWEST_ACTIVE_REG)
- lowest_active_reg = *(p1 + 1);
- }
-
- DEBUG_PRINT1 (":\n");
- PUSH_FAILURE_POINT (p + mcnt, d, -2);
- break;
-
-
- /* A smart repeat ends with `maybe_pop_jump'.
- We change it to either `pop_failure_jump' or `jump'. */
- case maybe_pop_jump:
- EXTRACT_NUMBER_AND_INCR (mcnt, p);
- DEBUG_PRINT2 ("EXECUTING maybe_pop_jump %d.\n", mcnt);
- {
- register unsigned char *p2 = p;
-
- /* Compare the beginning of the repeat with what in the
- pattern follows its end. If we can establish that there
- is nothing that they would both match, i.e., that we
- would have to backtrack because of (as in, e.g., `a*a')
- then we can change to pop_failure_jump, because we'll
- never have to backtrack.
-
- This is not true in the case of alternatives: in
- `(a|ab)*' we do need to backtrack to the `ab' alternative
- (e.g., if the string was `ab'). But instead of trying to
- detect that here, the alternative has put on a dummy
- failure point which is what we will end up popping. */
-
- /* Skip over open/close-group commands. */
- while (p2 + 2 < pend
- && ((re_opcode_t) *p2 == stop_memory
- || (re_opcode_t) *p2 == start_memory))
- p2 += 3; /* Skip over args, too. */
-
- /* If we're at the end of the pattern, we can change. */
- if (p2 == pend)
- {
- /* Consider what happens when matching ":\(.*\)"
- against ":/". I don't really understand this code
- yet. */
- p[-3] = (unsigned char) pop_failure_jump;
- DEBUG_PRINT1
- (" End of pattern: change to `pop_failure_jump'.\n");
- }
-
- else if ((re_opcode_t) *p2 == exactn
- || (bufp->newline_anchor && (re_opcode_t) *p2 == endline))
- {
- register unsigned char c
- = *p2 == (unsigned char) endline ? '\n' : p2[2];
- p1 = p + mcnt;
-
- /* p1[0] ... p1[2] are the `on_failure_jump' corresponding
- to the `maybe_finalize_jump' of this case. Examine what
- follows. */
- if ((re_opcode_t) p1[3] == exactn && p1[5] != c)
- {
- p[-3] = (unsigned char) pop_failure_jump;
- DEBUG_PRINT3 (" %c != %c => pop_failure_jump.\n",
- c, p1[5]);
- }
-
- else if ((re_opcode_t) p1[3] == charset
- || (re_opcode_t) p1[3] == charset_not)
- {
- int not = (re_opcode_t) p1[3] == charset_not;
-
- if (c < (unsigned char) (p1[4] * BYTEWIDTH)
- && p1[5 + c / BYTEWIDTH] & (1 << (c % BYTEWIDTH)))
- not = !not;
-
- /* `not' is equal to 1 if c would match, which means
- that we can't change to pop_failure_jump. */
- if (!not)
- {
- p[-3] = (unsigned char) pop_failure_jump;
- DEBUG_PRINT1 (" No match => pop_failure_jump.\n");
- }
- }
- }
- }
- p -= 2; /* Point at relative address again. */
- if ((re_opcode_t) p[-1] != pop_failure_jump)
- {
- p[-1] = (unsigned char) jump;
- DEBUG_PRINT1 (" Match => jump.\n");
- goto unconditional_jump;
- }
- /* Note fall through. */
-
-
- /* The end of a simple repeat has a pop_failure_jump back to
- its matching on_failure_jump, where the latter will push a
- failure point. The pop_failure_jump takes off failure
- points put on by this pop_failure_jump's matching
- on_failure_jump; we got through the pattern to here from the
- matching on_failure_jump, so didn't fail. */
- case pop_failure_jump:
- {
- /* We need to pass separate storage for the lowest and
- highest registers, even though we don't care about the
- actual values. Otherwise, we will restore only one
- register from the stack, since lowest will == highest in
- `pop_failure_point'. */
- unsigned dummy_low_reg, dummy_high_reg;
- unsigned char *pdummy;
- const char *sdummy;
-
- DEBUG_PRINT1 ("EXECUTING pop_failure_jump.\n");
- POP_FAILURE_POINT (sdummy, pdummy,
- dummy_low_reg, dummy_high_reg,
- reg_dummy, reg_dummy, reg_info_dummy);
- }
- /* Note fall through. */
-
-
- /* Unconditionally jump (without popping any failure points). */
- case jump:
- unconditional_jump:
- EXTRACT_NUMBER_AND_INCR (mcnt, p); /* Get the amount to jump. */
- DEBUG_PRINT2 ("EXECUTING jump %d ", mcnt);
- p += mcnt; /* Do the jump. */
- DEBUG_PRINT2 ("(to 0x%x).\n", p);
- break;
-
-
- /* We need this opcode so we can detect where alternatives end
- in `group_match_null_string_p' et al. */
- case jump_past_alt:
- DEBUG_PRINT1 ("EXECUTING jump_past_alt.\n");
- goto unconditional_jump;
-
-
- /* Normally, the on_failure_jump pushes a failure point, which
- then gets popped at pop_failure_jump. We will end up at
- pop_failure_jump, also, and with a pattern of, say, `a+', we
- are skipping over the on_failure_jump, so we have to push
- something meaningless for pop_failure_jump to pop. */
- case dummy_failure_jump:
- DEBUG_PRINT1 ("EXECUTING dummy_failure_jump.\n");
- /* It doesn't matter what we push for the string here. What
- the code at `fail' tests is the value for the pattern. */
- PUSH_FAILURE_POINT (0, 0, -2);
- goto unconditional_jump;
-
-
- /* At the end of an alternative, we need to push a dummy failure
- point in case we are followed by a `pop_failure_jump', because
- we don't want the failure point for the alternative to be
- popped. For example, matching `(a|ab)*' against `aab'
- requires that we match the `ab' alternative. */
- case push_dummy_failure:
- DEBUG_PRINT1 ("EXECUTING push_dummy_failure.\n");
- /* See comments just above at `dummy_failure_jump' about the
- two zeroes. */
- PUSH_FAILURE_POINT (0, 0, -2);
- break;
-
- /* Have to succeed matching what follows at least n times.
- After that, handle like `on_failure_jump'. */
- case succeed_n:
- EXTRACT_NUMBER (mcnt, p + 2);
- DEBUG_PRINT2 ("EXECUTING succeed_n %d.\n", mcnt);
-
- assert (mcnt >= 0);
- /* Originally, this is how many times we HAVE to succeed. */
- if (mcnt > 0)
- {
- mcnt--;
- p += 2;
- STORE_NUMBER_AND_INCR (p, mcnt);
- DEBUG_PRINT3 (" Setting 0x%x to %d.\n", p, mcnt);
- }
- else if (mcnt == 0)
- {
- DEBUG_PRINT2 (" Setting two bytes from 0x%x to no_op.\n", p+2);
- p[2] = (unsigned char) no_op;
- p[3] = (unsigned char) no_op;
- goto on_failure;
- }
- break;
-
- case jump_n:
- EXTRACT_NUMBER (mcnt, p + 2);
- DEBUG_PRINT2 ("EXECUTING jump_n %d.\n", mcnt);
-
- /* Originally, this is how many times we CAN jump. */
- if (mcnt)
- {
- mcnt--;
- STORE_NUMBER (p + 2, mcnt);
- goto unconditional_jump;
- }
- /* If don't have to jump any more, skip over the rest of command. */
- else
- p += 4;
- break;
-
- case set_number_at:
- {
- DEBUG_PRINT1 ("EXECUTING set_number_at.\n");
-
- EXTRACT_NUMBER_AND_INCR (mcnt, p);
- p1 = p + mcnt;
- EXTRACT_NUMBER_AND_INCR (mcnt, p);
- DEBUG_PRINT3 (" Setting 0x%x to %d.\n", p1, mcnt);
- STORE_NUMBER (p1, mcnt);
- break;
- }
-
- case wordbound:
- DEBUG_PRINT1 ("EXECUTING wordbound.\n");
- if (AT_WORD_BOUNDARY (d))
- break;
- goto fail;
-
- case notwordbound:
- DEBUG_PRINT1 ("EXECUTING notwordbound.\n");
- if (AT_WORD_BOUNDARY (d))
- goto fail;
- break;
-
- case wordbeg:
- DEBUG_PRINT1 ("EXECUTING wordbeg.\n");
- if (WORDCHAR_P (d) && (AT_STRINGS_BEG (d) || !WORDCHAR_P (d - 1)))
- break;
- goto fail;
-
- case wordend:
- DEBUG_PRINT1 ("EXECUTING wordend.\n");
- if (!AT_STRINGS_BEG (d) && WORDCHAR_P (d - 1)
- && (!WORDCHAR_P (d) || AT_STRINGS_END (d)))
- break;
- goto fail;
-
-#ifdef emacs
-#ifdef emacs19
- case before_dot:
- DEBUG_PRINT1 ("EXECUTING before_dot.\n");
- if (PTR_CHAR_POS ((unsigned char *) d) >= point)
- goto fail;
- break;
-
- case at_dot:
- DEBUG_PRINT1 ("EXECUTING at_dot.\n");
- if (PTR_CHAR_POS ((unsigned char *) d) != point)
- goto fail;
- break;
-
- case after_dot:
- DEBUG_PRINT1 ("EXECUTING after_dot.\n");
- if (PTR_CHAR_POS ((unsigned char *) d) <= point)
- goto fail;
- break;
-#else /* not emacs19 */
- case at_dot:
- DEBUG_PRINT1 ("EXECUTING at_dot.\n");
- if (PTR_CHAR_POS ((unsigned char *) d) + 1 != point)
- goto fail;
- break;
-#endif /* not emacs19 */
-
- case syntaxspec:
- DEBUG_PRINT2 ("EXECUTING syntaxspec %d.\n", mcnt);
- mcnt = *p++;
- goto matchsyntax;
-
- case wordchar:
- DEBUG_PRINT1 ("EXECUTING Emacs wordchar.\n");
- mcnt = (int) Sword;
- matchsyntax:
- PREFETCH ();
- if (SYNTAX (*d++) != (enum syntaxcode) mcnt)
- goto fail;
- SET_REGS_MATCHED ();
- break;
-
- case notsyntaxspec:
- DEBUG_PRINT2 ("EXECUTING notsyntaxspec %d.\n", mcnt);
- mcnt = *p++;
- goto matchnotsyntax;
-
- case notwordchar:
- DEBUG_PRINT1 ("EXECUTING Emacs notwordchar.\n");
- mcnt = (int) Sword;
- matchnotsyntax:
- PREFETCH ();
- if (SYNTAX (*d++) == (enum syntaxcode) mcnt)
- goto fail;
- SET_REGS_MATCHED ();
- break;
-
-#else /* not emacs */
- case wordchar:
- DEBUG_PRINT1 ("EXECUTING non-Emacs wordchar.\n");
- PREFETCH ();
- if (!WORDCHAR_P (d))
- goto fail;
- SET_REGS_MATCHED ();
- d++;
- break;
-
- case notwordchar:
- DEBUG_PRINT1 ("EXECUTING non-Emacs notwordchar.\n");
- PREFETCH ();
- if (WORDCHAR_P (d))
- goto fail;
- SET_REGS_MATCHED ();
- d++;
- break;
-#endif /* not emacs */
-
- default:
- abort ();
- }
- continue; /* Successfully executed one pattern command; keep going. */
-
-
- /* We goto here if a matching operation fails. */
- fail:
- if (!FAIL_STACK_EMPTY ())
- { /* A restart point is known. Restore to that state. */
- DEBUG_PRINT1 ("\nFAIL:\n");
- POP_FAILURE_POINT (d, p,
- lowest_active_reg, highest_active_reg,
- regstart, regend, reg_info);
-
- /* If this failure point is a dummy, try the next one. */
- if (!p)
- goto fail;
-
- /* If we failed to the end of the pattern, don't examine *p. */
- assert (p <= pend);
- if (p < pend)
- {
- boolean is_a_jump_n = false;
-
- /* If failed to a backwards jump that's part of a repetition
- loop, need to pop this failure point and use the next one. */
- switch ((re_opcode_t) *p)
- {
- case jump_n:
- is_a_jump_n = true;
- case maybe_pop_jump:
- case pop_failure_jump:
- case jump:
- p1 = p + 1;
- EXTRACT_NUMBER_AND_INCR (mcnt, p1);
- p1 += mcnt;
-
- if ((is_a_jump_n && (re_opcode_t) *p1 == succeed_n)
- || (!is_a_jump_n
- && (re_opcode_t) *p1 == on_failure_jump))
- goto fail;
- break;
- default:
- /* do nothing */ ;
- }
- }
-
- if (d >= string1 && d <= end1)
- dend = end_match_1;
- }
- else
- break; /* Matching at this starting point really fails. */
- } /* for (;;) */
-
- if (best_regs_set)
- goto restore_best_regs;
-
- FREE_VARIABLES ();
-
- return -1; /* Failure to match. */
-} /* re_match_2 */
-\f
-/* Subroutine definitions for re_match_2. */
-
-
-/* We are passed P pointing to a register number after a start_memory.
-
- Return true if the pattern up to the corresponding stop_memory can
- match the empty string, and false otherwise.
-
- If we find the matching stop_memory, sets P to point to one past its number.
- Otherwise, sets P to an undefined byte less than or equal to END.
-
- We don't handle duplicates properly (yet). */
-
-static boolean
-group_match_null_string_p (p, end, reg_info)
- unsigned char **p, *end;
- register_info_type *reg_info;
-{
- int mcnt;
- /* Point to after the args to the start_memory. */
- unsigned char *p1 = *p + 2;
-
- while (p1 < end)
- {
- /* Skip over opcodes that can match nothing, and return true or
- false, as appropriate, when we get to one that can't, or to the
- matching stop_memory. */
-
- switch ((re_opcode_t) *p1)
- {
- /* Could be either a loop or a series of alternatives. */
- case on_failure_jump:
- p1++;
- EXTRACT_NUMBER_AND_INCR (mcnt, p1);
-
- /* If the next operation is not a jump backwards in the
- pattern. */
-
- if (mcnt >= 0)
- {
- /* Go through the on_failure_jumps of the alternatives,
- seeing if any of the alternatives cannot match nothing.
- The last alternative starts with only a jump,
- whereas the rest start with on_failure_jump and end
- with a jump, e.g., here is the pattern for `a|b|c':
-
- /on_failure_jump/0/6/exactn/1/a/jump_past_alt/0/6
- /on_failure_jump/0/6/exactn/1/b/jump_past_alt/0/3
- /exactn/1/c
-
- So, we have to first go through the first (n-1)
- alternatives and then deal with the last one separately. */
-
-
- /* Deal with the first (n-1) alternatives, which start
- with an on_failure_jump (see above) that jumps to right
- past a jump_past_alt. */
-
- while ((re_opcode_t) p1[mcnt-3] == jump_past_alt)
- {
- /* `mcnt' holds how many bytes long the alternative
- is, including the ending `jump_past_alt' and
- its number. */
-
- if (!alt_match_null_string_p (p1, p1 + mcnt - 3,
- reg_info))
- return false;
-
- /* Move to right after this alternative, including the
- jump_past_alt. */
- p1 += mcnt;
-
- /* Break if it's the beginning of an n-th alternative
- that doesn't begin with an on_failure_jump. */
- if ((re_opcode_t) *p1 != on_failure_jump)
- break;
-
- /* Still have to check that it's not an n-th
- alternative that starts with an on_failure_jump. */
- p1++;
- EXTRACT_NUMBER_AND_INCR (mcnt, p1);
- if ((re_opcode_t) p1[mcnt-3] != jump_past_alt)
- {
- /* Get to the beginning of the n-th alternative. */
- p1 -= 3;
- break;
- }
- }
-
- /* Deal with the last alternative: go back and get number
- of the `jump_past_alt' just before it. `mcnt' contains
- the length of the alternative. */
- EXTRACT_NUMBER (mcnt, p1 - 2);
-
- if (!alt_match_null_string_p (p1, p1 + mcnt, reg_info))
- return false;
-
- p1 += mcnt; /* Get past the n-th alternative. */
- } /* if mcnt > 0 */
- break;
-
-
- case stop_memory:
- assert (p1[1] == **p);
- *p = p1 + 2;
- return true;
-
-
- default:
- if (!common_op_match_null_string_p (&p1, end, reg_info))
- return false;
- }
- } /* while p1 < end */
-
- return false;
-} /* group_match_null_string_p */
-
-
-/* Similar to group_match_null_string_p, but doesn't deal with alternatives:
- It expects P to be the first byte of a single alternative and END one
- byte past the last. The alternative can contain groups. */
-
-static boolean
-alt_match_null_string_p (p, end, reg_info)
- unsigned char *p, *end;
- register_info_type *reg_info;
-{
- int mcnt;
- unsigned char *p1 = p;
-
- while (p1 < end)
- {
- /* Skip over opcodes that can match nothing, and break when we get
- to one that can't. */
-
- switch ((re_opcode_t) *p1)
- {
- /* It's a loop. */
- case on_failure_jump:
- p1++;
- EXTRACT_NUMBER_AND_INCR (mcnt, p1);
- p1 += mcnt;
- break;
-
- default:
- if (!common_op_match_null_string_p (&p1, end, reg_info))
- return false;
- }
- } /* while p1 < end */
-
- return true;
-} /* alt_match_null_string_p */
-
-
-/* Deals with the ops common to group_match_null_string_p and
- alt_match_null_string_p.
-
- Sets P to one after the op and its arguments, if any. */
-
-static boolean
-common_op_match_null_string_p (p, end, reg_info)
- unsigned char **p, *end;
- register_info_type *reg_info;
-{
- int mcnt;
- boolean ret;
- int reg_no;
- unsigned char *p1 = *p;
-
- switch ((re_opcode_t) *p1++)
- {
- case no_op:
- case begline:
- case endline:
- case begbuf:
- case endbuf:
- case wordbeg:
- case wordend:
- case wordbound:
- case notwordbound:
-#ifdef emacs
- case before_dot:
- case at_dot:
- case after_dot:
-#endif
- break;
-
- case start_memory:
- reg_no = *p1;
- assert (reg_no > 0 && reg_no <= MAX_REGNUM);
- ret = group_match_null_string_p (&p1, end, reg_info);
-
- /* Have to set this here in case we're checking a group which
- contains a group and a back reference to it. */
-
- if (REG_MATCH_NULL_STRING_P (reg_info[reg_no]) == MATCH_NULL_UNSET_VALUE)
- REG_MATCH_NULL_STRING_P (reg_info[reg_no]) = ret;
-
- if (!ret)
- return false;
- break;
-
- /* If this is an optimized succeed_n for zero times, make the jump. */
- case jump:
- EXTRACT_NUMBER_AND_INCR (mcnt, p1);
- if (mcnt >= 0)
- p1 += mcnt;
- else
- return false;
- break;
-
- case succeed_n:
- /* Get to the number of times to succeed. */
- p1 += 2;
- EXTRACT_NUMBER_AND_INCR (mcnt, p1);
-
- if (mcnt == 0)
- {
- p1 -= 4;
- EXTRACT_NUMBER_AND_INCR (mcnt, p1);
- p1 += mcnt;
- }
- else
- return false;
- break;
-
- case duplicate:
- if (!REG_MATCH_NULL_STRING_P (reg_info[*p1]))
- return false;
- break;
-
- case set_number_at:
- p1 += 4;
-
- default:
- /* All other opcodes mean we cannot match the empty string. */
- return false;
- }
-
- *p = p1;
- return true;
-} /* common_op_match_null_string_p */
-
-
-/* Return zero if TRANSLATE[S1] and TRANSLATE[S2] are identical for LEN
- bytes; nonzero otherwise. */
-
-static int
-bcmp_translate(
- unsigned char *s1,
- unsigned char *s2,
- int len,
- char *translate
-)
-{
- register unsigned char *p1 = s1, *p2 = s2;
- while (len)
- {
- if (translate[*p1++] != translate[*p2++]) return 1;
- len--;
- }
- return 0;
-}
-\f
-/* Entry points for GNU code. */
-
-/* re_compile_pattern is the GNU regular expression compiler: it
- compiles PATTERN (of length SIZE) and puts the result in BUFP.
- Returns 0 if the pattern was valid, otherwise an error string.
-
- Assumes the `allocated' (and perhaps `buffer') and `translate' fields
- are set in BUFP on entry.
-
- We call regex_compile to do the actual compilation. */
-
-const char *
-re_compile_pattern (pattern, length, bufp)
- const char *pattern;
- int length;
- struct re_pattern_buffer *bufp;
-{
- reg_errcode_t ret;
-
- /* GNU code is written to assume at least RE_NREGS registers will be set
- (and at least one extra will be -1). */
- bufp->regs_allocated = REGS_UNALLOCATED;
-
- /* And GNU code determines whether or not to get register information
- by passing null for the REGS argument to re_match, etc., not by
- setting no_sub. */
- bufp->no_sub = 0;
-
- /* Match anchors at newline. */
- bufp->newline_anchor = 1;
-
- ret = regex_compile (pattern, length, re_syntax_options, bufp);
-
- return re_error_msg[(int) ret];
-}
-\f
-/* Entry points compatible with 4.2 BSD regex library. We don't define
- them if this is an Emacs or POSIX compilation. */
-
-#if !defined (emacs) && !defined (_POSIX_SOURCE)
-
-/* BSD has one and only one pattern buffer. */
-static struct re_pattern_buffer re_comp_buf;
-
-char *
-re_comp (s)
- const char *s;
-{
- reg_errcode_t ret;
-
- if (!s)
- {
- if (!re_comp_buf.buffer)
- return "No previous regular expression";
- return 0;
- }
-
- if (!re_comp_buf.buffer)
- {
- re_comp_buf.buffer = (unsigned char *) malloc (200);
- if (re_comp_buf.buffer == NULL)
- return "Memory exhausted";
- re_comp_buf.allocated = 200;
-
- re_comp_buf.fastmap = (char *) malloc (1 << BYTEWIDTH);
- if (re_comp_buf.fastmap == NULL)
- return "Memory exhausted";
- }
-
- /* Since `re_exec' always passes NULL for the `regs' argument, we
- don't need to initialize the pattern buffer fields which affect it. */
-
- /* Match anchors at newlines. */
- re_comp_buf.newline_anchor = 1;
-
- ret = regex_compile (s, strlen (s), re_syntax_options, &re_comp_buf);
-
- /* Yes, we're discarding `const' here. */
- return (char *) re_error_msg[(int) ret];
-}
-
-
-int
-re_exec (s)
- const char *s;
-{
- const int len = strlen (s);
- return
- 0 <= re_search (&re_comp_buf, s, len, 0, len, (struct re_registers *) 0);
-}
-#endif /* not emacs and not _POSIX_SOURCE */
-\f
-/* POSIX.2 functions. Don't define these for Emacs. */
-
-#ifndef emacs
-
-/* regcomp takes a regular expression as a string and compiles it.
-
- PREG is a regex_t *. We do not expect any fields to be initialized,
- since POSIX says we shouldn't. Thus, we set
-
- `buffer' to the compiled pattern;
- `used' to the length of the compiled pattern;
- `syntax' to RE_SYNTAX_POSIX_EXTENDED if the
- REG_EXTENDED bit in CFLAGS is set; otherwise, to
- RE_SYNTAX_POSIX_BASIC;
- `newline_anchor' to REG_NEWLINE being set in CFLAGS;
- `fastmap' and `fastmap_accurate' to zero;
- `re_nsub' to the number of subexpressions in PATTERN.
-
- PATTERN is the address of the pattern string.
-
- CFLAGS is a series of bits which affect compilation.
-
- If REG_EXTENDED is set, we use POSIX extended syntax; otherwise, we
- use POSIX basic syntax.
-
- If REG_NEWLINE is set, then . and [^...] don't match newline.
- Also, regexec will try a match beginning after every newline.
-
- If REG_ICASE is set, then we considers upper- and lowercase
- versions of letters to be equivalent when matching.
-
- If REG_NOSUB is set, then when PREG is passed to regexec, that
- routine will report only success or failure, and nothing about the
- registers.
-
- It returns 0 if it succeeds, nonzero if it doesn't. (See regex.h for
- the return codes and their meanings.) */
-
-int
-regcomp (preg, pattern, cflags)
- regex_t *preg;
- const char *pattern;
- int cflags;
-{
- reg_errcode_t ret;
- unsigned syntax
- = (cflags & REG_EXTENDED) ?
- RE_SYNTAX_POSIX_EXTENDED : RE_SYNTAX_POSIX_BASIC;
-
- /* regex_compile will allocate the space for the compiled pattern. */
- preg->buffer = 0;
- preg->allocated = 0;
-
- /* Don't bother to use a fastmap when searching. This simplifies the
- REG_NEWLINE case: if we used a fastmap, we'd have to put all the
- characters after newlines into the fastmap. This way, we just try
- every character. */
- preg->fastmap = 0;
-
- if (cflags & REG_ICASE)
- {
- unsigned i;
-
- preg->translate = (char *) malloc (CHAR_SET_SIZE);
- if (preg->translate == NULL)
- return (int) REG_ESPACE;
-
- /* Map uppercase characters to corresponding lowercase ones. */
- for (i = 0; i < CHAR_SET_SIZE; i++)
- preg->translate[i] = ISUPPER (i) ? tolower (i) : i;
- }
- else
- preg->translate = NULL;
-
- /* If REG_NEWLINE is set, newlines are treated differently. */
- if (cflags & REG_NEWLINE)
- { /* REG_NEWLINE implies neither . nor [^...] match newline. */
- syntax &= ~RE_DOT_NEWLINE;
- syntax |= RE_HAT_LISTS_NOT_NEWLINE;
- /* It also changes the matching behavior. */
- preg->newline_anchor = 1;
- }
- else
- preg->newline_anchor = 0;
-
- preg->no_sub = !!(cflags & REG_NOSUB);
-
- /* POSIX says a null character in the pattern terminates it, so we
- can use strlen here in compiling the pattern. */
- ret = regex_compile (pattern, strlen (pattern), syntax, preg);
-
- /* POSIX doesn't distinguish between an unmatched open-group and an
- unmatched close-group: both are REG_EPAREN. */
- if (ret == REG_ERPAREN) ret = REG_EPAREN;
-
- return (int) ret;
-}
-
-
-/* regexec searches for a given pattern, specified by PREG, in the
- string STRING.
-
- If NMATCH is zero or REG_NOSUB was set in the cflags argument to
- `regcomp', we ignore PMATCH. Otherwise, we assume PMATCH has at
- least NMATCH elements, and we set them to the offsets of the
- corresponding matched substrings.
-
- EFLAGS specifies `execution flags' which affect matching: if
- REG_NOTBOL is set, then ^ does not match at the beginning of the
- string; if REG_NOTEOL is set, then $ does not match at the end.
-
- We return 0 if we find a match and REG_NOMATCH if not. */
-
-int
-regexec (preg, string, nmatch, pmatch, eflags)
- const regex_t *preg;
- const char *string;
- size_t nmatch;
- regmatch_t pmatch[];
- int eflags;
-{
- int ret;
- struct re_registers regs;
- regex_t private_preg;
- int len = strlen (string);
- boolean want_reg_info = !preg->no_sub && nmatch > 0;
-
- private_preg = *preg;
-
- private_preg.not_bol = !!(eflags & REG_NOTBOL);
- private_preg.not_eol = !!(eflags & REG_NOTEOL);
-
- /* The user has told us exactly how many registers to return
- information about, via `nmatch'. We have to pass that on to the
- matching routines. */
- private_preg.regs_allocated = REGS_FIXED;
-
- if (want_reg_info)
- {
- regs.num_regs = nmatch;
- regs.start = TALLOC (nmatch, regoff_t);
- regs.end = TALLOC (nmatch, regoff_t);
- if (regs.start == NULL || regs.end == NULL)
- return (int) REG_NOMATCH;
- }
-
- /* Perform the searching operation. */
- ret = re_search (&private_preg, string, len,
- /* start: */ 0, /* range: */ len,
- want_reg_info ? ®s : (struct re_registers *) 0);
-
- /* Copy the register information to the POSIX structure. */
- if (want_reg_info)
- {
- if (ret >= 0)
- {
- unsigned r;
-
- for (r = 0; r < nmatch; r++)
- {
- pmatch[r].rm_so = regs.start[r];
- pmatch[r].rm_eo = regs.end[r];
- }
- }
-
- /* If we needed the temporary register info, free the space now. */
- free (regs.start);
- free (regs.end);
- }
-
- /* We want zero return to mean success, unlike `re_search'. */
- return ret >= 0 ? (int) REG_NOERROR : (int) REG_NOMATCH;
-}
-
-
-/* Returns a message corresponding to an error code, ERRCODE, returned
- from either regcomp or regexec. We don't use PREG here. */
-
-size_t
-regerror (errcode, preg, errbuf, errbuf_size)
- int errcode;
- const regex_t *preg;
- char *errbuf;
- size_t errbuf_size;
-{
- const char *msg;
- size_t msg_size;
-
- if (errcode < 0
- || errcode >= (sizeof (re_error_msg) / sizeof (re_error_msg[0])))
- /* Only error codes returned by the rest of the code should be passed
- to this routine. If we are given anything else, or if other regex
- code generates an invalid error code, then the program has a bug.
- Dump core so we can fix it. */
- abort ();
-
- msg = re_error_msg[errcode];
-
- /* POSIX doesn't require that we do anything in this case, but why
- not be nice. */
- if (! msg)
- msg = "Success";
-
- msg_size = strlen (msg) + 1; /* Includes the null. */
-
- if (errbuf_size != 0)
- {
- if (msg_size > errbuf_size)
- {
- strncpy (errbuf, msg, errbuf_size - 1);
- errbuf[errbuf_size - 1] = 0;
- }
- else
- strcpy (errbuf, msg);
- }
-
- return msg_size;
-}
-
-
-/* Free dynamically allocated space used by PREG. */
-
-void
-regfree (preg)
- regex_t *preg;
-{
- if (preg->buffer != NULL)
- free (preg->buffer);
- preg->buffer = NULL;
-
- preg->allocated = 0;
- preg->used = 0;
-
- if (preg->fastmap != NULL)
- free (preg->fastmap);
- preg->fastmap = NULL;
- preg->fastmap_accurate = 0;
-
- if (preg->translate != NULL)
- free (preg->translate);
- preg->translate = NULL;
-}
-
-#endif /* not emacs */
-\f
-/*
-Local variables:
-make-backup-files: t
-version-control: t
-trim-versions-without-asking: nil
-End:
-*/
+++ /dev/null
-/* Definitions for data structures and routines for the regular
- expression library, version 0.12.
-
- Copyright (C) 1985, 1989, 1990, 1991, 1992, 1993 Free Software Foundation, Inc.
-
- This program is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 2, or (at your option)
- any later version.
-
- This program is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- GNU General Public License for more details.
-
- You should have received a copy of the GNU General Public License
- along with this program; if not, write to the Free Software
- Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */
-
-#ifndef __REGEXP_LIBRARY_H__
-#define __REGEXP_LIBRARY_H__
-
-/* POSIX says that <sys/types.h> must be included (by the caller) before
- <regex.h>. */
-
-#ifdef VMS
-/* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
- should be there. */
-#include <stddef.h>
-#endif
-
-
-/* The following bits are used to determine the regexp syntax we
- recognize. The set/not-set meanings are chosen so that Emacs syntax
- remains the value 0. The bits are given in alphabetical order, and
- the definitions shifted by one from the previous bit; thus, when we
- add or remove a bit, only one other definition need change. */
-typedef unsigned reg_syntax_t;
-
-/* If this bit is not set, then \ inside a bracket expression is literal.
- If set, then such a \ quotes the following character. */
-#define RE_BACKSLASH_ESCAPE_IN_LISTS (1)
-
-/* If this bit is not set, then + and ? are operators, and \+ and \? are
- literals.
- If set, then \+ and \? are operators and + and ? are literals. */
-#define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
-
-/* If this bit is set, then character classes are supported. They are:
- [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
- [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
- If not set, then character classes are not supported. */
-#define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
-
-/* If this bit is set, then ^ and $ are always anchors (outside bracket
- expressions, of course).
- If this bit is not set, then it depends:
- ^ is an anchor if it is at the beginning of a regular
- expression or after an open-group or an alternation operator;
- $ is an anchor if it is at the end of a regular expression, or
- before a close-group or an alternation operator.
-
- This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
- POSIX draft 11.2 says that * etc. in leading positions is undefined.
- We already implemented a previous draft which made those constructs
- invalid, though, so we haven't changed the code back. */
-#define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
-
-/* If this bit is set, then special characters are always special
- regardless of where they are in the pattern.
- If this bit is not set, then special characters are special only in
- some contexts; otherwise they are ordinary. Specifically,
- * + ? and intervals are only special when not after the beginning,
- open-group, or alternation operator. */
-#define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
-
-/* If this bit is set, then *, +, ?, and { cannot be first in an re or
- immediately after an alternation or begin-group operator. */
-#define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
-
-/* If this bit is set, then . matches newline.
- If not set, then it doesn't. */
-#define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
-
-/* If this bit is set, then . doesn't match NUL.
- If not set, then it does. */
-#define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
-
-/* If this bit is set, nonmatching lists [^...] do not match newline.
- If not set, they do. */
-#define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
-
-/* If this bit is set, either \{...\} or {...} defines an
- interval, depending on RE_NO_BK_BRACES.
- If not set, \{, \}, {, and } are literals. */
-#define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
-
-/* If this bit is set, +, ? and | aren't recognized as operators.
- If not set, they are. */
-#define RE_LIMITED_OPS (RE_INTERVALS << 1)
-
-/* If this bit is set, newline is an alternation operator.
- If not set, newline is literal. */
-#define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
-
-/* If this bit is set, then `{...}' defines an interval, and \{ and \}
- are literals.
- If not set, then `\{...\}' defines an interval. */
-#define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
-
-/* If this bit is set, (...) defines a group, and \( and \) are literals.
- If not set, \(...\) defines a group, and ( and ) are literals. */
-#define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
-
-/* If this bit is set, then \<digit> matches <digit>.
- If not set, then \<digit> is a back-reference. */
-#define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
-
-/* If this bit is set, then | is an alternation operator, and \| is literal.
- If not set, then \| is an alternation operator, and | is literal. */
-#define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
-
-/* If this bit is set, then an ending range point collating higher
- than the starting range point, as in [z-a], is invalid.
- If not set, then when ending range point collates higher than the
- starting range point, the range is ignored. */
-#define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
-
-/* If this bit is set, then an unmatched ) is ordinary.
- If not set, then an unmatched ) is invalid. */
-#define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
-
-/* This global variable defines the particular regexp syntax to use (for
- some interfaces). When a regexp is compiled, the syntax used is
- stored in the pattern buffer, so changing this does not affect
- already-compiled regexps. */
-extern reg_syntax_t re_syntax_options;
-\f
-/* Define combinations of the above bits for the standard possibilities.
- (The [[[ comments delimit what gets put into the Texinfo file, so
- don't delete them!) */
-/* [[[begin syntaxes]]] */
-#define RE_SYNTAX_EMACS 0
-
-#define RE_SYNTAX_AWK \
- (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
- | RE_NO_BK_PARENS | RE_NO_BK_REFS \
- | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
- | RE_UNMATCHED_RIGHT_PAREN_ORD)
-
-#define RE_SYNTAX_POSIX_AWK \
- (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS)
-
-#define RE_SYNTAX_GREP \
- (RE_BK_PLUS_QM | RE_CHAR_CLASSES \
- | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
- | RE_NEWLINE_ALT)
-
-#define RE_SYNTAX_EGREP \
- (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
- | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
- | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
- | RE_NO_BK_VBAR)
-
-#define RE_SYNTAX_POSIX_EGREP \
- (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
-
-/* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
-#define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
-
-#define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
-
-/* Syntax bits common to both basic and extended POSIX regex syntax. */
-#define _RE_SYNTAX_POSIX_COMMON \
- (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
- | RE_INTERVALS | RE_NO_EMPTY_RANGES)
-
-#define RE_SYNTAX_POSIX_BASIC \
- (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
-
-/* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
- RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
- isn't minimal, since other operators, such as \`, aren't disabled. */
-#define RE_SYNTAX_POSIX_MINIMAL_BASIC \
- (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
-
-#define RE_SYNTAX_POSIX_EXTENDED \
- (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
- | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
- | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
- | RE_UNMATCHED_RIGHT_PAREN_ORD)
-
-/* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INVALID_OPS
- replaces RE_CONTEXT_INDEP_OPS and RE_NO_BK_REFS is added. */
-#define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
- (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
- | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
- | RE_NO_BK_PARENS | RE_NO_BK_REFS \
- | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
-/* [[[end syntaxes]]] */
-\f
-/* Maximum number of duplicates an interval can allow. Some systems
- (erroneously) define this in other header files, but we want our
- value, so remove any previous define. */
-#ifdef RE_DUP_MAX
-#undef RE_DUP_MAX
-#endif
-#define RE_DUP_MAX ((1 << 15) - 1)
-
-
-/* POSIX `cflags' bits (i.e., information for `regcomp'). */
-
-/* If this bit is set, then use extended regular expression syntax.
- If not set, then use basic regular expression syntax. */
-#define REG_EXTENDED 1
-
-/* If this bit is set, then ignore case when matching.
- If not set, then case is significant. */
-#define REG_ICASE (REG_EXTENDED << 1)
-
-/* If this bit is set, then anchors do not match at newline
- characters in the string.
- If not set, then anchors do match at newlines. */
-#define REG_NEWLINE (REG_ICASE << 1)
-
-/* If this bit is set, then report only success or fail in regexec.
- If not set, then returns differ between not matching and errors. */
-#define REG_NOSUB (REG_NEWLINE << 1)
-
-
-/* POSIX `eflags' bits (i.e., information for regexec). */
-
-/* If this bit is set, then the beginning-of-line operator doesn't match
- the beginning of the string (presumably because it's not the
- beginning of a line).
- If not set, then the beginning-of-line operator does match the
- beginning of the string. */
-#define REG_NOTBOL 1
-
-/* Like REG_NOTBOL, except for the end-of-line. */
-#define REG_NOTEOL (1 << 1)
-
-
-/* If any error codes are removed, changed, or added, update the
- `re_error_msg' table in regex.c. */
-typedef enum
-{
- REG_NOERROR = 0, /* Success. */
- REG_NOMATCH, /* Didn't find a match (for regexec). */
-
- /* POSIX regcomp return error codes. (In the order listed in the
- standard.) */
- REG_BADPAT, /* Invalid pattern. */
- REG_ECOLLATE, /* Not implemented. */
- REG_ECTYPE, /* Invalid character class name. */
- REG_EESCAPE, /* Trailing backslash. */
- REG_ESUBREG, /* Invalid back reference. */
- REG_EBRACK, /* Unmatched left bracket. */
- REG_EPAREN, /* Parenthesis imbalance. */
- REG_EBRACE, /* Unmatched \{. */
- REG_BADBR, /* Invalid contents of \{\}. */
- REG_ERANGE, /* Invalid range end. */
- REG_ESPACE, /* Ran out of memory. */
- REG_BADRPT, /* No preceding re for repetition op. */
-
- /* Error codes we've added. */
- REG_EEND, /* Premature end. */
- REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
- REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
-} reg_errcode_t;
-\f
-/* This data structure represents a compiled pattern. Before calling
- the pattern compiler, the fields `buffer', `allocated', `fastmap',
- `translate', and `no_sub' can be set. After the pattern has been
- compiled, the `re_nsub' field is available. All other fields are
- private to the regex routines. */
-
-struct re_pattern_buffer
-{
-/* [[[begin pattern_buffer]]] */
- /* Space that holds the compiled pattern. It is declared as
- `unsigned char *' because its elements are
- sometimes used as array indexes. */
- unsigned char *buffer;
-
- /* Number of bytes to which `buffer' points. */
- unsigned long allocated;
-
- /* Number of bytes actually used in `buffer'. */
- unsigned long used;
-
- /* Syntax setting with which the pattern was compiled. */
- reg_syntax_t syntax;
-
- /* Pointer to a fastmap, if any, otherwise zero. re_search uses
- the fastmap, if there is one, to skip over impossible
- starting points for matches. */
- char *fastmap;
-
- /* Either a translate table to apply to all characters before
- comparing them, or zero for no translation. The translation
- is applied to a pattern when it is compiled and to a string
- when it is matched. */
- char *translate;
-
- /* Number of subexpressions found by the compiler. */
- size_t re_nsub;
-
- /* Zero if this pattern cannot match the empty string, one else.
- Well, in truth it's used only in `re_search_2', to see
- whether or not we should use the fastmap, so we don't set
- this absolutely perfectly; see `re_compile_fastmap' (the
- `duplicate' case). */
- unsigned can_be_null : 1;
-
- /* If REGS_UNALLOCATED, allocate space in the `regs' structure
- for `max (RE_NREGS, re_nsub + 1)' groups.
- If REGS_REALLOCATE, reallocate space if necessary.
- If REGS_FIXED, use what's there. */
-#define REGS_UNALLOCATED 0
-#define REGS_REALLOCATE 1
-#define REGS_FIXED 2
- unsigned regs_allocated : 2;
-
- /* Set to zero when `regex_compile' compiles a pattern; set to one
- by `re_compile_fastmap' if it updates the fastmap. */
- unsigned fastmap_accurate : 1;
-
- /* If set, `re_match_2' does not return information about
- subexpressions. */
- unsigned no_sub : 1;
-
- /* If set, a beginning-of-line anchor doesn't match at the
- beginning of the string. */
- unsigned not_bol : 1;
-
- /* Similarly for an end-of-line anchor. */
- unsigned not_eol : 1;
-
- /* If true, an anchor at a newline matches. */
- unsigned newline_anchor : 1;
-
-/* [[[end pattern_buffer]]] */
-};
-
-typedef struct re_pattern_buffer regex_t;
-
-
-/* search.c (search_buffer) in Emacs needs this one opcode value. It is
- defined both in `regex.c' and here. */
-#define RE_EXACTN_VALUE 1
-\f
-/* Type for byte offsets within the string. POSIX mandates this. */
-typedef int regoff_t;
-
-
-/* This is the structure we store register match data in. See
- regex.texinfo for a full description of what registers match. */
-struct re_registers
-{
- unsigned num_regs;
- regoff_t *start;
- regoff_t *end;
-};
-
-
-/* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
- `re_match_2' returns information about at least this many registers
- the first time a `regs' structure is passed. */
-#ifndef RE_NREGS
-#define RE_NREGS 30
-#endif
-
-
-/* POSIX specification for registers. Aside from the different names than
- `re_registers', POSIX uses an array of structures, instead of a
- structure of arrays. */
-typedef struct
-{
- regoff_t rm_so; /* Byte offset from string's start to substring's start. */
- regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
-} regmatch_t;
-\f
-/* Declarations for routines. */
-
-/* To avoid duplicating every routine declaration -- once with a
- prototype (if we are ANSI), and once without (if we aren't) -- we
- use the following macro to declare argument types. This
- unfortunately clutters up the declarations a bit, but I think it's
- worth it. */
-
-#if __STDC__
-
-#define _RE_ARGS(args) args
-
-#else /* not __STDC__ */
-
-#define _RE_ARGS(args) ()
-
-#endif /* not __STDC__ */
-
-/* Sets the current default syntax to SYNTAX, and return the old syntax.
- You can also simply assign to the `re_syntax_options' variable. */
-extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
-
-/* Compile the regular expression PATTERN, with length LENGTH
- and syntax given by the global `re_syntax_options', into the buffer
- BUFFER. Return NULL if successful, and an error string if not. */
-extern const char *re_compile_pattern
- _RE_ARGS ((const char *pattern, int length,
- struct re_pattern_buffer *buffer));
-
-
-/* Compile a fastmap for the compiled pattern in BUFFER; used to
- accelerate searches. Return 0 if successful and -2 if was an
- internal error. */
-extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
-
-
-/* Search in the string STRING (with length LENGTH) for the pattern
- compiled into BUFFER. Start searching at position START, for RANGE
- characters. Return the starting position of the match, -1 for no
- match, or -2 for an internal error. Also return register
- information in REGS (if REGS and BUFFER->no_sub are nonzero). */
-extern int re_search
- _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
- int length, int start, int range, struct re_registers *regs));
-
-
-/* Like `re_search', but search in the concatenation of STRING1 and
- STRING2. Also, stop searching at index START + STOP. */
-extern int re_search_2
- _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
- int length1, const char *string2, int length2,
- int start, int range, struct re_registers *regs, int stop));
-
-
-/* Like `re_search', but return how many characters in STRING the regexp
- in BUFFER matched, starting at position START. */
-extern int re_match
- _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
- int length, int start, struct re_registers *regs));
-
-
-/* Relates to `re_match' as `re_search_2' relates to `re_search'. */
-extern int re_match_2
- _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
- int length1, const char *string2, int length2,
- int start, struct re_registers *regs, int stop));
-
-
-/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
- ENDS. Subsequent matches using BUFFER and REGS will use this memory
- for recording register information. STARTS and ENDS must be
- allocated with malloc, and must each be at least `NUM_REGS * sizeof
- (regoff_t)' bytes long.
-
- If NUM_REGS == 0, then subsequent matches should allocate their own
- register data.
-
- Unless this function is called, the first search or match using
- PATTERN_BUFFER will allocate its own register data, without
- freeing the old data. */
-extern void re_set_registers
- _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
- unsigned num_regs, regoff_t *starts, regoff_t *ends));
-
-/* 4.2 bsd compatibility. */
-extern char *re_comp _RE_ARGS ((const char *));
-extern int re_exec _RE_ARGS ((const char *));
-
-/* POSIX compatibility. */
-extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags));
-extern int regexec
- _RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch,
- regmatch_t pmatch[], int eflags));
-extern size_t regerror
- _RE_ARGS ((int errcode, const regex_t *preg, char *errbuf,
- size_t errbuf_size));
-extern void regfree _RE_ARGS ((regex_t *preg));
-
-#endif /* not __REGEXP_LIBRARY_H__ */
-\f
-/*
-Local variables:
-make-backup-files: t
-version-control: t
-trim-versions-without-asking: nil
-End:
-*/
--- /dev/null
+/* Extended regular expression matching and search library,
+ version 0.12.
+ (Implements POSIX draft P10003.2/D11.2, except for
+ internationalization features.)
+
+ Copyright (C) 1993 Free Software Foundation, Inc.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software
+ Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */
+
+/* AIX requires this to be the first thing in the file. */
+#if defined (_AIX) && !defined (REGEX_MALLOC)
+ #pragma alloca
+#endif
+
+#define _GNU_SOURCE
+
+/* We need this for `regex.h', and perhaps for the Emacs include files. */
+#include <sys/types.h>
+
+/* We used to test for `BSTRING' here, but only GCC and Emacs define
+ `BSTRING', as far as I know, and neither of them use this code. */
+#include <string.h>
+#ifndef bcmp
+#define bcmp(s1, s2, n) memcmp ((s1), (s2), (n))
+#endif
+#ifndef bcopy
+#define bcopy(s, d, n) memcpy ((d), (s), (n))
+#endif
+#ifndef bzero
+#define bzero(s, n) memset ((s), 0, (n))
+#endif
+
+#include <stdlib.h>
+
+
+/* Define the syntax stuff for \<, \>, etc. */
+
+/* This must be nonzero for the wordchar and notwordchar pattern
+ commands in re_match_2. */
+#ifndef Sword
+#define Sword 1
+#endif
+
+#ifdef SYNTAX_TABLE
+
+extern char *re_syntax_table;
+
+#else /* not SYNTAX_TABLE */
+
+/* How many characters in the character set. */
+#define CHAR_SET_SIZE 256
+
+static char re_syntax_table[CHAR_SET_SIZE];
+
+static void
+init_syntax_once ()
+{
+ register int c;
+ static int done = 0;
+
+ if (done)
+ return;
+
+ bzero (re_syntax_table, sizeof re_syntax_table);
+
+ for (c = 'a'; c <= 'z'; c++)
+ re_syntax_table[c] = Sword;
+
+ for (c = 'A'; c <= 'Z'; c++)
+ re_syntax_table[c] = Sword;
+
+ for (c = '0'; c <= '9'; c++)
+ re_syntax_table[c] = Sword;
+
+ re_syntax_table['_'] = Sword;
+
+ done = 1;
+}
+
+#endif /* not SYNTAX_TABLE */
+
+#define SYNTAX(c) re_syntax_table[c]
+
+\f
+/* Get the interface, including the syntax bits. */
+#include "regex.h"
+
+/* isalpha etc. are used for the character classes. */
+#include <ctype.h>
+
+#ifndef isascii
+#define isascii(c) 1
+#endif
+
+#ifdef isblank
+#define ISBLANK(c) (isascii (c) && isblank (c))
+#else
+#define ISBLANK(c) ((c) == ' ' || (c) == '\t')
+#endif
+#ifdef isgraph
+#define ISGRAPH(c) (isascii (c) && isgraph (c))
+#else
+#define ISGRAPH(c) (isascii (c) && isprint (c) && !isspace (c))
+#endif
+
+#define ISPRINT(c) (isascii (c) && isprint (c))
+#define ISDIGIT(c) (isascii (c) && isdigit (c))
+#define ISALNUM(c) (isascii (c) && isalnum (c))
+#define ISALPHA(c) (isascii (c) && isalpha (c))
+#define ISCNTRL(c) (isascii (c) && iscntrl (c))
+#define ISLOWER(c) (isascii (c) && islower (c))
+#define ISPUNCT(c) (isascii (c) && ispunct (c))
+#define ISSPACE(c) (isascii (c) && isspace (c))
+#define ISUPPER(c) (isascii (c) && isupper (c))
+#define ISXDIGIT(c) (isascii (c) && isxdigit (c))
+
+#ifndef NULL
+#define NULL 0
+#endif
+
+/* We remove any previous definition of `SIGN_EXTEND_CHAR',
+ since ours (we hope) works properly with all combinations of
+ machines, compilers, `char' and `unsigned char' argument types.
+ (Per Bothner suggested the basic approach.) */
+#undef SIGN_EXTEND_CHAR
+#if __STDC__
+#define SIGN_EXTEND_CHAR(c) ((signed char) (c))
+#else /* not __STDC__ */
+/* As in Harbison and Steele. */
+#define SIGN_EXTEND_CHAR(c) ((((unsigned char) (c)) ^ 128) - 128)
+#endif
+\f
+/* Should we use malloc or alloca? If REGEX_MALLOC is not defined, we
+ use `alloca' instead of `malloc'. This is because using malloc in
+ re_search* or re_match* could cause memory leaks when C-g is used in
+ Emacs; also, malloc is slower and causes storage fragmentation. On
+ the other hand, malloc is more portable, and easier to debug.
+
+ Because we sometimes use alloca, some routines have to be macros,
+ not functions -- `alloca'-allocated space disappears at the end of the
+ function it is called in. */
+
+#ifdef REGEX_MALLOC
+
+#define REGEX_ALLOCATE malloc
+#define REGEX_REALLOCATE(source, osize, nsize) realloc (source, nsize)
+
+#else /* not REGEX_MALLOC */
+
+/* Emacs already defines alloca, sometimes. */
+#ifndef alloca
+
+/* Make alloca work the best possible way. */
+#ifdef __GNUC__
+#define alloca __builtin_alloca
+#else /* not __GNUC__ */
+#if HAVE_ALLOCA_H
+#include <alloca.h>
+#else /* not __GNUC__ or HAVE_ALLOCA_H */
+#ifndef _AIX /* Already did AIX, up at the top. */
+char *alloca ();
+#endif /* not _AIX */
+#endif /* not HAVE_ALLOCA_H */
+#endif /* not __GNUC__ */
+
+#endif /* not alloca */
+
+#define REGEX_ALLOCATE alloca
+
+/* Assumes a `char *destination' variable. */
+#define REGEX_REALLOCATE(source, osize, nsize) \
+ (destination = (char *) alloca (nsize), \
+ bcopy (source, destination, osize), \
+ destination)
+
+#endif /* not REGEX_MALLOC */
+
+
+/* True if `size1' is non-NULL and PTR is pointing anywhere inside
+ `string1' or just past its end. This works if PTR is NULL, which is
+ a good thing. */
+#define FIRST_STRING_P(ptr) \
+ (size1 && string1 <= (ptr) && (ptr) <= string1 + size1)
+
+/* (Re)Allocate N items of type T using malloc, or fail. */
+#define TALLOC(n, t) ((t *) malloc ((n) * sizeof (t)))
+#define RETALLOC(addr, n, t) ((addr) = (t *) realloc (addr, (n) * sizeof (t)))
+#define REGEX_TALLOC(n, t) ((t *) REGEX_ALLOCATE ((n) * sizeof (t)))
+
+#define BYTEWIDTH 8 /* In bits. */
+
+#define STREQ(s1, s2) ((strcmp (s1, s2) == 0))
+
+#define MAX(a, b) ((a) > (b) ? (a) : (b))
+#define MIN(a, b) ((a) < (b) ? (a) : (b))
+
+typedef char boolean;
+#define false 0
+#define true 1
+\f
+/* These are the command codes that appear in compiled regular
+ expressions. Some opcodes are followed by argument bytes. A
+ command code can specify any interpretation whatsoever for its
+ arguments. Zero bytes may appear in the compiled regular expression.
+
+ The value of `exactn' is needed in search.c (search_buffer) in Emacs.
+ So regex.h defines a symbol `RE_EXACTN_VALUE' to be 1; the value of
+ `exactn' we use here must also be 1. */
+
+typedef enum
+{
+ no_op = 0,
+
+ /* Followed by one byte giving n, then by n literal bytes. */
+ exactn = 1,
+
+ /* Matches any (more or less) character. */
+ anychar,
+
+ /* Matches any one char belonging to specified set. First
+ following byte is number of bitmap bytes. Then come bytes
+ for a bitmap saying which chars are in. Bits in each byte
+ are ordered low-bit-first. A character is in the set if its
+ bit is 1. A character too large to have a bit in the map is
+ automatically not in the set. */
+ charset,
+
+ /* Same parameters as charset, but match any character that is
+ not one of those specified. */
+ charset_not,
+
+ /* Start remembering the text that is matched, for storing in a
+ register. Followed by one byte with the register number, in
+ the range 0 to one less than the pattern buffer's re_nsub
+ field. Then followed by one byte with the number of groups
+ inner to this one. (This last has to be part of the
+ start_memory only because we need it in the on_failure_jump
+ of re_match_2.) */
+ start_memory,
+
+ /* Stop remembering the text that is matched and store it in a
+ memory register. Followed by one byte with the register
+ number, in the range 0 to one less than `re_nsub' in the
+ pattern buffer, and one byte with the number of inner groups,
+ just like `start_memory'. (We need the number of inner
+ groups here because we don't have any easy way of finding the
+ corresponding start_memory when we're at a stop_memory.) */
+ stop_memory,
+
+ /* Match a duplicate of something remembered. Followed by one
+ byte containing the register number. */
+ duplicate,
+
+ /* Fail unless at beginning of line. */
+ begline,
+
+ /* Fail unless at end of line. */
+ endline,
+
+ /* Succeeds if at beginning of buffer (if emacs) or at beginning
+ of string to be matched (if not). */
+ begbuf,
+
+ /* Analogously, for end of buffer/string. */
+ endbuf,
+
+ /* Followed by two byte relative address to which to jump. */
+ jump,
+
+ /* Same as jump, but marks the end of an alternative. */
+ jump_past_alt,
+
+ /* Followed by two-byte relative address of place to resume at
+ in case of failure. */
+ on_failure_jump,
+
+ /* Like on_failure_jump, but pushes a placeholder instead of the
+ current string position when executed. */
+ on_failure_keep_string_jump,
+
+ /* Throw away latest failure point and then jump to following
+ two-byte relative address. */
+ pop_failure_jump,
+
+ /* Change to pop_failure_jump if know won't have to backtrack to
+ match; otherwise change to jump. This is used to jump
+ back to the beginning of a repeat. If what follows this jump
+ clearly won't match what the repeat does, such that we can be
+ sure that there is no use backtracking out of repetitions
+ already matched, then we change it to a pop_failure_jump.
+ Followed by two-byte address. */
+ maybe_pop_jump,
+
+ /* Jump to following two-byte address, and push a dummy failure
+ point. This failure point will be thrown away if an attempt
+ is made to use it for a failure. A `+' construct makes this
+ before the first repeat. Also used as an intermediary kind
+ of jump when compiling an alternative. */
+ dummy_failure_jump,
+
+ /* Push a dummy failure point and continue. Used at the end of
+ alternatives. */
+ push_dummy_failure,
+
+ /* Followed by two-byte relative address and two-byte number n.
+ After matching N times, jump to the address upon failure. */
+ succeed_n,
+
+ /* Followed by two-byte relative address, and two-byte number n.
+ Jump to the address N times, then fail. */
+ jump_n,
+
+ /* Set the following two-byte relative address to the
+ subsequent two-byte number. The address *includes* the two
+ bytes of number. */
+ set_number_at,
+
+ wordchar, /* Matches any word-constituent character. */
+ notwordchar, /* Matches any char that is not a word-constituent. */
+
+ wordbeg, /* Succeeds if at word beginning. */
+ wordend, /* Succeeds if at word end. */
+
+ wordbound, /* Succeeds if at a word boundary. */
+ notwordbound /* Succeeds if not at a word boundary. */
+
+#ifdef emacs
+ ,before_dot, /* Succeeds if before point. */
+ at_dot, /* Succeeds if at point. */
+ after_dot, /* Succeeds if after point. */
+
+ /* Matches any character whose syntax is specified. Followed by
+ a byte which contains a syntax code, e.g., Sword. */
+ syntaxspec,
+
+ /* Matches any character whose syntax is not that specified. */
+ notsyntaxspec
+#endif /* emacs */
+} re_opcode_t;
+\f
+/* Common operations on the compiled pattern. */
+
+/* Store NUMBER in two contiguous bytes starting at DESTINATION. */
+
+#define STORE_NUMBER(destination, number) \
+ do { \
+ (destination)[0] = (number) & 0377; \
+ (destination)[1] = (number) >> 8; \
+ } while (0)
+
+/* Same as STORE_NUMBER, except increment DESTINATION to
+ the byte after where the number is stored. Therefore, DESTINATION
+ must be an lvalue. */
+
+#define STORE_NUMBER_AND_INCR(destination, number) \
+ do { \
+ STORE_NUMBER (destination, number); \
+ (destination) += 2; \
+ } while (0)
+
+/* Put into DESTINATION a number stored in two contiguous bytes starting
+ at SOURCE. */
+
+#define EXTRACT_NUMBER(destination, source) \
+ do { \
+ (destination) = *(source) & 0377; \
+ (destination) += SIGN_EXTEND_CHAR (*((source) + 1)) << 8; \
+ } while (0)
+
+#ifdef DEBUG
+static void
+extract_number (dest, source)
+ int *dest;
+ unsigned char *source;
+{
+ int temp = SIGN_EXTEND_CHAR (*(source + 1));
+ *dest = *source & 0377;
+ *dest += temp << 8;
+}
+
+#ifndef EXTRACT_MACROS /* To debug the macros. */
+#undef EXTRACT_NUMBER
+#define EXTRACT_NUMBER(dest, src) extract_number (&dest, src)
+#endif /* not EXTRACT_MACROS */
+
+#endif /* DEBUG */
+
+/* Same as EXTRACT_NUMBER, except increment SOURCE to after the number.
+ SOURCE must be an lvalue. */
+
+#define EXTRACT_NUMBER_AND_INCR(destination, source) \
+ do { \
+ EXTRACT_NUMBER (destination, source); \
+ (source) += 2; \
+ } while (0)
+
+#ifdef DEBUG
+static void
+extract_number_and_incr (destination, source)
+ int *destination;
+ unsigned char **source;
+{
+ extract_number (destination, *source);
+ *source += 2;
+}
+
+#ifndef EXTRACT_MACROS
+#undef EXTRACT_NUMBER_AND_INCR
+#define EXTRACT_NUMBER_AND_INCR(dest, src) \
+ extract_number_and_incr (&dest, &src)
+#endif /* not EXTRACT_MACROS */
+
+#endif /* DEBUG */
+\f
+/* If DEBUG is defined, Regex prints many voluminous messages about what
+ it is doing (if the variable `debug' is nonzero). If linked with the
+ main program in `iregex.c', you can enter patterns and strings
+ interactively. And if linked with the main program in `main.c' and
+ the other test files, you can run the already-written tests. */
+
+#ifdef DEBUG
+
+/* We use standard I/O for debugging. */
+#include <stdio.h>
+
+/* It is useful to test things that ``must'' be true when debugging. */
+#include <assert.h>
+
+static int debug = 0;
+
+#define DEBUG_STATEMENT(e) e
+#define DEBUG_PRINT1(x) if (debug) printf (x)
+#define DEBUG_PRINT2(x1, x2) if (debug) printf (x1, x2)
+#define DEBUG_PRINT3(x1, x2, x3) if (debug) printf (x1, x2, x3)
+#define DEBUG_PRINT4(x1, x2, x3, x4) if (debug) printf (x1, x2, x3, x4)
+#define DEBUG_PRINT_COMPILED_PATTERN(p, s, e) \
+ if (debug) print_partial_compiled_pattern (s, e)
+#define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2) \
+ if (debug) print_double_string (w, s1, sz1, s2, sz2)
+
+
+extern void printchar ();
+
+/* Print the fastmap in human-readable form. */
+
+void
+print_fastmap (fastmap)
+ char *fastmap;
+{
+ unsigned was_a_range = 0;
+ unsigned i = 0;
+
+ while (i < (1 << BYTEWIDTH))
+ {
+ if (fastmap[i++])
+ {
+ was_a_range = 0;
+ printchar (i - 1);
+ while (i < (1 << BYTEWIDTH) && fastmap[i])
+ {
+ was_a_range = 1;
+ i++;
+ }
+ if (was_a_range)
+ {
+ printf ("-");
+ printchar (i - 1);
+ }
+ }
+ }
+ putchar ('\n');
+}
+
+
+/* Print a compiled pattern string in human-readable form, starting at
+ the START pointer into it and ending just before the pointer END. */
+
+void
+print_partial_compiled_pattern (start, end)
+ unsigned char *start;
+ unsigned char *end;
+{
+ int mcnt, mcnt2;
+ unsigned char *p = start;
+ unsigned char *pend = end;
+
+ if (start == NULL)
+ {
+ printf ("(null)\n");
+ return;
+ }
+
+ /* Loop over pattern commands. */
+ while (p < pend)
+ {
+ switch ((re_opcode_t) *p++)
+ {
+ case no_op:
+ printf ("/no_op");
+ break;
+
+ case exactn:
+ mcnt = *p++;
+ printf ("/exactn/%d", mcnt);
+ do
+ {
+ putchar ('/');
+ printchar (*p++);
+ }
+ while (--mcnt);
+ break;
+
+ case start_memory:
+ mcnt = *p++;
+ printf ("/start_memory/%d/%d", mcnt, *p++);
+ break;
+
+ case stop_memory:
+ mcnt = *p++;
+ printf ("/stop_memory/%d/%d", mcnt, *p++);
+ break;
+
+ case duplicate:
+ printf ("/duplicate/%d", *p++);
+ break;
+
+ case anychar:
+ printf ("/anychar");
+ break;
+
+ case charset:
+ case charset_not:
+ {
+ register int c;
+
+ printf ("/charset%s",
+ (re_opcode_t) *(p - 1) == charset_not ? "_not" : "");
+
+ assert (p + *p < pend);
+
+ for (c = 0; c < *p; c++)
+ {
+ unsigned bit;
+ unsigned char map_byte = p[1 + c];
+
+ putchar ('/');
+
+ for (bit = 0; bit < BYTEWIDTH; bit++)
+ if (map_byte & (1 << bit))
+ printchar (c * BYTEWIDTH + bit);
+ }
+ p += 1 + *p;
+ break;
+ }
+
+ case begline:
+ printf ("/begline");
+ break;
+
+ case endline:
+ printf ("/endline");
+ break;
+
+ case on_failure_jump:
+ extract_number_and_incr (&mcnt, &p);
+ printf ("/on_failure_jump/0/%d", mcnt);
+ break;
+
+ case on_failure_keep_string_jump:
+ extract_number_and_incr (&mcnt, &p);
+ printf ("/on_failure_keep_string_jump/0/%d", mcnt);
+ break;
+
+ case dummy_failure_jump:
+ extract_number_and_incr (&mcnt, &p);
+ printf ("/dummy_failure_jump/0/%d", mcnt);
+ break;
+
+ case push_dummy_failure:
+ printf ("/push_dummy_failure");
+ break;
+
+ case maybe_pop_jump:
+ extract_number_and_incr (&mcnt, &p);
+ printf ("/maybe_pop_jump/0/%d", mcnt);
+ break;
+
+ case pop_failure_jump:
+ extract_number_and_incr (&mcnt, &p);
+ printf ("/pop_failure_jump/0/%d", mcnt);
+ break;
+
+ case jump_past_alt:
+ extract_number_and_incr (&mcnt, &p);
+ printf ("/jump_past_alt/0/%d", mcnt);
+ break;
+
+ case jump:
+ extract_number_and_incr (&mcnt, &p);
+ printf ("/jump/0/%d", mcnt);
+ break;
+
+ case succeed_n:
+ extract_number_and_incr (&mcnt, &p);
+ extract_number_and_incr (&mcnt2, &p);
+ printf ("/succeed_n/0/%d/0/%d", mcnt, mcnt2);
+ break;
+
+ case jump_n:
+ extract_number_and_incr (&mcnt, &p);
+ extract_number_and_incr (&mcnt2, &p);
+ printf ("/jump_n/0/%d/0/%d", mcnt, mcnt2);
+ break;
+
+ case set_number_at:
+ extract_number_and_incr (&mcnt, &p);
+ extract_number_and_incr (&mcnt2, &p);
+ printf ("/set_number_at/0/%d/0/%d", mcnt, mcnt2);
+ break;
+
+ case wordbound:
+ printf ("/wordbound");
+ break;
+
+ case notwordbound:
+ printf ("/notwordbound");
+ break;
+
+ case wordbeg:
+ printf ("/wordbeg");
+ break;
+
+ case wordend:
+ printf ("/wordend");
+
+#ifdef emacs
+ case before_dot:
+ printf ("/before_dot");
+ break;
+
+ case at_dot:
+ printf ("/at_dot");
+ break;
+
+ case after_dot:
+ printf ("/after_dot");
+ break;
+
+ case syntaxspec:
+ printf ("/syntaxspec");
+ mcnt = *p++;
+ printf ("/%d", mcnt);
+ break;
+
+ case notsyntaxspec:
+ printf ("/notsyntaxspec");
+ mcnt = *p++;
+ printf ("/%d", mcnt);
+ break;
+#endif /* emacs */
+
+ case wordchar:
+ printf ("/wordchar");
+ break;
+
+ case notwordchar:
+ printf ("/notwordchar");
+ break;
+
+ case begbuf:
+ printf ("/begbuf");
+ break;
+
+ case endbuf:
+ printf ("/endbuf");
+ break;
+
+ default:
+ printf ("?%d", *(p-1));
+ }
+ }
+ printf ("/\n");
+}
+
+
+void
+print_compiled_pattern (bufp)
+ struct re_pattern_buffer *bufp;
+{
+ unsigned char *buffer = bufp->buffer;
+
+ print_partial_compiled_pattern (buffer, buffer + bufp->used);
+ printf ("%d bytes used/%d bytes allocated.\n", bufp->used, bufp->allocated);
+
+ if (bufp->fastmap_accurate && bufp->fastmap)
+ {
+ printf ("fastmap: ");
+ print_fastmap (bufp->fastmap);
+ }
+
+ printf ("re_nsub: %d\t", bufp->re_nsub);
+ printf ("regs_alloc: %d\t", bufp->regs_allocated);
+ printf ("can_be_null: %d\t", bufp->can_be_null);
+ printf ("newline_anchor: %d\n", bufp->newline_anchor);
+ printf ("no_sub: %d\t", bufp->no_sub);
+ printf ("not_bol: %d\t", bufp->not_bol);
+ printf ("not_eol: %d\t", bufp->not_eol);
+ printf ("syntax: %d\n", bufp->syntax);
+ /* Perhaps we should print the translate table? */
+}
+
+
+void
+print_double_string (where, string1, size1, string2, size2)
+ const char *where;
+ const char *string1;
+ const char *string2;
+ int size1;
+ int size2;
+{
+ unsigned this_char;
+
+ if (where == NULL)
+ printf ("(null)");
+ else
+ {
+ if (FIRST_STRING_P (where))
+ {
+ for (this_char = where - string1; this_char < size1; this_char++)
+ printchar (string1[this_char]);
+
+ where = string2;
+ }
+
+ for (this_char = where - string2; this_char < size2; this_char++)
+ printchar (string2[this_char]);
+ }
+}
+
+#else /* not DEBUG */
+
+#undef assert
+#define assert(e)
+
+#define DEBUG_STATEMENT(e)
+#define DEBUG_PRINT1(x)
+#define DEBUG_PRINT2(x1, x2)
+#define DEBUG_PRINT3(x1, x2, x3)
+#define DEBUG_PRINT4(x1, x2, x3, x4)
+#define DEBUG_PRINT_COMPILED_PATTERN(p, s, e)
+#define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2)
+
+#endif /* not DEBUG */
+\f
+/* Set by `re_set_syntax' to the current regexp syntax to recognize. Can
+ also be assigned to arbitrarily: each pattern buffer stores its own
+ syntax, so it can be changed between regex compilations. */
+reg_syntax_t re_syntax_options = RE_SYNTAX_EMACS;
+
+
+/* Specify the precise syntax of regexps for compilation. This provides
+ for compatibility for various utilities which historically have
+ different, incompatible syntaxes.
+
+ The argument SYNTAX is a bit mask comprised of the various bits
+ defined in regex.h. We return the old syntax. */
+
+reg_syntax_t
+re_set_syntax (syntax)
+ reg_syntax_t syntax;
+{
+ reg_syntax_t ret = re_syntax_options;
+
+ re_syntax_options = syntax;
+ return ret;
+}
+\f
+/* This table gives an error message for each of the error codes listed
+ in regex.h. Obviously the order here has to be same as there. */
+
+static const char *re_error_msg[] =
+ { NULL, /* REG_NOERROR */
+ "No match", /* REG_NOMATCH */
+ "Invalid regular expression", /* REG_BADPAT */
+ "Invalid collation character", /* REG_ECOLLATE */
+ "Invalid character class name", /* REG_ECTYPE */
+ "Trailing backslash", /* REG_EESCAPE */
+ "Invalid back reference", /* REG_ESUBREG */
+ "Unmatched [ or [^", /* REG_EBRACK */
+ "Unmatched ( or \\(", /* REG_EPAREN */
+ "Unmatched \\{", /* REG_EBRACE */
+ "Invalid content of \\{\\}", /* REG_BADBR */
+ "Invalid range end", /* REG_ERANGE */
+ "Memory exhausted", /* REG_ESPACE */
+ "Invalid preceding regular expression", /* REG_BADRPT */
+ "Premature end of regular expression", /* REG_EEND */
+ "Regular expression too big", /* REG_ESIZE */
+ "Unmatched ) or \\)", /* REG_ERPAREN */
+ };
+\f
+/* Subroutine declarations and macros for regex_compile. */
+
+static void store_op1 (), store_op2 ();
+static void insert_op1 (), insert_op2 ();
+static boolean at_begline_loc_p (), at_endline_loc_p ();
+static boolean group_in_compile_stack ();
+static reg_errcode_t compile_range ();
+
+/* Fetch the next character in the uncompiled pattern---translating it
+ if necessary. Also cast from a signed character in the constant
+ string passed to us by the user to an unsigned char that we can use
+ as an array index (in, e.g., `translate'). */
+#define PATFETCH(c) \
+ do {if (p == pend) return REG_EEND; \
+ c = (unsigned char) *p++; \
+ if (translate) c = translate[c]; \
+ } while (0)
+
+/* Fetch the next character in the uncompiled pattern, with no
+ translation. */
+#define PATFETCH_RAW(c) \
+ do {if (p == pend) return REG_EEND; \
+ c = (unsigned char) *p++; \
+ } while (0)
+
+/* Go backwards one character in the pattern. */
+#define PATUNFETCH p--
+
+
+/* If `translate' is non-null, return translate[D], else just D. We
+ cast the subscript to translate because some data is declared as
+ `char *', to avoid warnings when a string constant is passed. But
+ when we use a character as a subscript we must make it unsigned. */
+#define TRANSLATE(d) (translate ? translate[(unsigned char) (d)] : (d))
+
+
+/* Macros for outputting the compiled pattern into `buffer'. */
+
+/* If the buffer isn't allocated when it comes in, use this. */
+#define INIT_BUF_SIZE 32
+
+/* Make sure we have at least N more bytes of space in buffer. */
+#define GET_BUFFER_SPACE(n) \
+ while (b - bufp->buffer + (n) > bufp->allocated) \
+ EXTEND_BUFFER ()
+
+/* Make sure we have one more byte of buffer space and then add C to it. */
+#define BUF_PUSH(c) \
+ do { \
+ GET_BUFFER_SPACE (1); \
+ *b++ = (unsigned char) (c); \
+ } while (0)
+
+
+/* Ensure we have two more bytes of buffer space and then append C1 and C2. */
+#define BUF_PUSH_2(c1, c2) \
+ do { \
+ GET_BUFFER_SPACE (2); \
+ *b++ = (unsigned char) (c1); \
+ *b++ = (unsigned char) (c2); \
+ } while (0)
+
+
+/* As with BUF_PUSH_2, except for three bytes. */
+#define BUF_PUSH_3(c1, c2, c3) \
+ do { \
+ GET_BUFFER_SPACE (3); \
+ *b++ = (unsigned char) (c1); \
+ *b++ = (unsigned char) (c2); \
+ *b++ = (unsigned char) (c3); \
+ } while (0)
+
+
+/* Store a jump with opcode OP at LOC to location TO. We store a
+ relative address offset by the three bytes the jump itself occupies. */
+#define STORE_JUMP(op, loc, to) \
+ store_op1 (op, loc, (to) - (loc) - 3)
+
+/* Likewise, for a two-argument jump. */
+#define STORE_JUMP2(op, loc, to, arg) \
+ store_op2 (op, loc, (to) - (loc) - 3, arg)
+
+/* Like `STORE_JUMP', but for inserting. Assume `b' is the buffer end. */
+#define INSERT_JUMP(op, loc, to) \
+ insert_op1 (op, loc, (to) - (loc) - 3, b)
+
+/* Like `STORE_JUMP2', but for inserting. Assume `b' is the buffer end. */
+#define INSERT_JUMP2(op, loc, to, arg) \
+ insert_op2 (op, loc, (to) - (loc) - 3, arg, b)
+
+
+/* This is not an arbitrary limit: the arguments which represent offsets
+ into the pattern are two bytes long. So if 2^16 bytes turns out to
+ be too small, many things would have to change. */
+#define MAX_BUF_SIZE (1L << 16)
+
+
+/* Extend the buffer by twice its current size via realloc and
+ reset the pointers that pointed into the old block to point to the
+ correct places in the new one. If extending the buffer results in it
+ being larger than MAX_BUF_SIZE, then flag memory exhausted. */
+#define EXTEND_BUFFER() \
+ do { \
+ unsigned char *old_buffer = bufp->buffer; \
+ if (bufp->allocated == MAX_BUF_SIZE) \
+ return REG_ESIZE; \
+ bufp->allocated <<= 1; \
+ if (bufp->allocated > MAX_BUF_SIZE) \
+ bufp->allocated = MAX_BUF_SIZE; \
+ bufp->buffer = (unsigned char *) realloc (bufp->buffer, bufp->allocated);\
+ if (bufp->buffer == NULL) \
+ return REG_ESPACE; \
+ /* If the buffer moved, move all the pointers into it. */ \
+ if (old_buffer != bufp->buffer) \
+ { \
+ b = (b - old_buffer) + bufp->buffer; \
+ begalt = (begalt - old_buffer) + bufp->buffer; \
+ if (fixup_alt_jump) \
+ fixup_alt_jump = (fixup_alt_jump - old_buffer) + bufp->buffer;\
+ if (laststart) \
+ laststart = (laststart - old_buffer) + bufp->buffer; \
+ if (pending_exact) \
+ pending_exact = (pending_exact - old_buffer) + bufp->buffer; \
+ } \
+ } while (0)
+
+
+/* Since we have one byte reserved for the register number argument to
+ {start,stop}_memory, the maximum number of groups we can report
+ things about is what fits in that byte. */
+#define MAX_REGNUM 255
+
+/* But patterns can have more than `MAX_REGNUM' registers. We just
+ ignore the excess. */
+typedef unsigned regnum_t;
+
+
+/* Macros for the compile stack. */
+
+/* Since offsets can go either forwards or backwards, this type needs to
+ be able to hold values from -(MAX_BUF_SIZE - 1) to MAX_BUF_SIZE - 1. */
+typedef int pattern_offset_t;
+
+typedef struct
+{
+ pattern_offset_t begalt_offset;
+ pattern_offset_t fixup_alt_jump;
+ pattern_offset_t inner_group_offset;
+ pattern_offset_t laststart_offset;
+ regnum_t regnum;
+} compile_stack_elt_t;
+
+
+typedef struct
+{
+ compile_stack_elt_t *stack;
+ unsigned size;
+ unsigned avail; /* Offset of next open position. */
+} compile_stack_type;
+
+
+#define INIT_COMPILE_STACK_SIZE 32
+
+#define COMPILE_STACK_EMPTY (compile_stack.avail == 0)
+#define COMPILE_STACK_FULL (compile_stack.avail == compile_stack.size)
+
+/* The next available element. */
+#define COMPILE_STACK_TOP (compile_stack.stack[compile_stack.avail])
+
+
+/* Set the bit for character C in a list. */
+#define SET_LIST_BIT(c) \
+ (b[((unsigned char) (c)) / BYTEWIDTH] \
+ |= 1 << (((unsigned char) c) % BYTEWIDTH))
+
+
+/* Get the next unsigned number in the uncompiled pattern. */
+#define GET_UNSIGNED_NUMBER(num) \
+ { if (p != pend) \
+ { \
+ PATFETCH (c); \
+ while (ISDIGIT (c)) \
+ { \
+ if (num < 0) \
+ num = 0; \
+ num = num * 10 + c - '0'; \
+ if (p == pend) \
+ break; \
+ PATFETCH (c); \
+ } \
+ } \
+ }
+
+#define CHAR_CLASS_MAX_LENGTH 6 /* Namely, `xdigit'. */
+
+#define IS_CHAR_CLASS(string) \
+ (STREQ (string, "alpha") || STREQ (string, "upper") \
+ || STREQ (string, "lower") || STREQ (string, "digit") \
+ || STREQ (string, "alnum") || STREQ (string, "xdigit") \
+ || STREQ (string, "space") || STREQ (string, "print") \
+ || STREQ (string, "punct") || STREQ (string, "graph") \
+ || STREQ (string, "cntrl") || STREQ (string, "blank"))
+\f
+/* `regex_compile' compiles PATTERN (of length SIZE) according to SYNTAX.
+ Returns one of error codes defined in `regex.h', or zero for success.
+
+ Assumes the `allocated' (and perhaps `buffer') and `translate'
+ fields are set in BUFP on entry.
+
+ If it succeeds, results are put in BUFP (if it returns an error, the
+ contents of BUFP are undefined):
+ `buffer' is the compiled pattern;
+ `syntax' is set to SYNTAX;
+ `used' is set to the length of the compiled pattern;
+ `fastmap_accurate' is zero;
+ `re_nsub' is the number of subexpressions in PATTERN;
+ `not_bol' and `not_eol' are zero;
+
+ The `fastmap' and `newline_anchor' fields are neither
+ examined nor set. */
+
+static reg_errcode_t
+regex_compile (pattern, size, syntax, bufp)
+ const char *pattern;
+ int size;
+ reg_syntax_t syntax;
+ struct re_pattern_buffer *bufp;
+{
+ /* We fetch characters from PATTERN here. Even though PATTERN is
+ `char *' (i.e., signed), we declare these variables as unsigned, so
+ they can be reliably used as array indices. */
+ register unsigned char c, c1;
+
+ /* A random tempory spot in PATTERN. */
+ const char *p1;
+
+ /* Points to the end of the buffer, where we should append. */
+ register unsigned char *b;
+
+ /* Keeps track of unclosed groups. */
+ compile_stack_type compile_stack;
+
+ /* Points to the current (ending) position in the pattern. */
+ const char *p = pattern;
+ const char *pend = pattern + size;
+
+ /* How to translate the characters in the pattern. */
+ char *translate = bufp->translate;
+
+ /* Address of the count-byte of the most recently inserted `exactn'
+ command. This makes it possible to tell if a new exact-match
+ character can be added to that command or if the character requires
+ a new `exactn' command. */
+ unsigned char *pending_exact = 0;
+
+ /* Address of start of the most recently finished expression.
+ This tells, e.g., postfix * where to find the start of its
+ operand. Reset at the beginning of groups and alternatives. */
+ unsigned char *laststart = 0;
+
+ /* Address of beginning of regexp, or inside of last group. */
+ unsigned char *begalt;
+
+ /* Place in the uncompiled pattern (i.e., the {) to
+ which to go back if the interval is invalid. */
+ const char *beg_interval;
+
+ /* Address of the place where a forward jump should go to the end of
+ the containing expression. Each alternative of an `or' -- except the
+ last -- ends with a forward jump of this sort. */
+ unsigned char *fixup_alt_jump = 0;
+
+ /* Counts open-groups as they are encountered. Remembered for the
+ matching close-group on the compile stack, so the same register
+ number is put in the stop_memory as the start_memory. */
+ regnum_t regnum = 0;
+
+#ifdef DEBUG
+ DEBUG_PRINT1 ("\nCompiling pattern: ");
+ if (debug)
+ {
+ unsigned debug_count;
+
+ for (debug_count = 0; debug_count < size; debug_count++)
+ printchar (pattern[debug_count]);
+ putchar ('\n');
+ }
+#endif /* DEBUG */
+
+ /* Initialize the compile stack. */
+ compile_stack.stack = TALLOC (INIT_COMPILE_STACK_SIZE, compile_stack_elt_t);
+ if (compile_stack.stack == NULL)
+ return REG_ESPACE;
+
+ compile_stack.size = INIT_COMPILE_STACK_SIZE;
+ compile_stack.avail = 0;
+
+ /* Initialize the pattern buffer. */
+ bufp->syntax = syntax;
+ bufp->fastmap_accurate = 0;
+ bufp->not_bol = bufp->not_eol = 0;
+
+ /* Set `used' to zero, so that if we return an error, the pattern
+ printer (for debugging) will think there's no pattern. We reset it
+ at the end. */
+ bufp->used = 0;
+
+ /* Always count groups, whether or not bufp->no_sub is set. */
+ bufp->re_nsub = 0;
+
+#if !defined (emacs) && !defined (SYNTAX_TABLE)
+ /* Initialize the syntax table. */
+ init_syntax_once ();
+#endif
+
+ if (bufp->allocated == 0)
+ {
+ if (bufp->buffer)
+ { /* If zero allocated, but buffer is non-null, try to realloc
+ enough space. This loses if buffer's address is bogus, but
+ that is the user's responsibility. */
+ RETALLOC (bufp->buffer, INIT_BUF_SIZE, unsigned char);
+ }
+ else
+ { /* Caller did not allocate a buffer. Do it for them. */
+ bufp->buffer = TALLOC (INIT_BUF_SIZE, unsigned char);
+ }
+ if (!bufp->buffer) return REG_ESPACE;
+
+ bufp->allocated = INIT_BUF_SIZE;
+ }
+
+ begalt = b = bufp->buffer;
+
+ /* Loop through the uncompiled pattern until we're at the end. */
+ while (p != pend)
+ {
+ PATFETCH (c);
+
+ switch (c)
+ {
+ case '^':
+ {
+ if ( /* If at start of pattern, it's an operator. */
+ p == pattern + 1
+ /* If context independent, it's an operator. */
+ || syntax & RE_CONTEXT_INDEP_ANCHORS
+ /* Otherwise, depends on what's come before. */
+ || at_begline_loc_p (pattern, p, syntax))
+ BUF_PUSH (begline);
+ else
+ goto normal_char;
+ }
+ break;
+
+
+ case '$':
+ {
+ if ( /* If at end of pattern, it's an operator. */
+ p == pend
+ /* If context independent, it's an operator. */
+ || syntax & RE_CONTEXT_INDEP_ANCHORS
+ /* Otherwise, depends on what's next. */
+ || at_endline_loc_p (p, pend, syntax))
+ BUF_PUSH (endline);
+ else
+ goto normal_char;
+ }
+ break;
+
+
+ case '+':
+ case '?':
+ if ((syntax & RE_BK_PLUS_QM)
+ || (syntax & RE_LIMITED_OPS))
+ goto normal_char;
+ handle_plus:
+ case '*':
+ /* If there is no previous pattern... */
+ if (!laststart)
+ {
+ if (syntax & RE_CONTEXT_INVALID_OPS)
+ return REG_BADRPT;
+ else if (!(syntax & RE_CONTEXT_INDEP_OPS))
+ goto normal_char;
+ }
+
+ {
+ /* Are we optimizing this jump? */
+ boolean keep_string_p = false;
+
+ /* 1 means zero (many) matches is allowed. */
+ char zero_times_ok = 0, many_times_ok = 0;
+
+ /* If there is a sequence of repetition chars, collapse it
+ down to just one (the right one). We can't combine
+ interval operators with these because of, e.g., `a{2}*',
+ which should only match an even number of `a's. */
+
+ for (;;)
+ {
+ zero_times_ok |= c != '+';
+ many_times_ok |= c != '?';
+
+ if (p == pend)
+ break;
+
+ PATFETCH (c);
+
+ if (c == '*'
+ || (!(syntax & RE_BK_PLUS_QM) && (c == '+' || c == '?')))
+ ;
+
+ else if (syntax & RE_BK_PLUS_QM && c == '\\')
+ {
+ if (p == pend) return REG_EESCAPE;
+
+ PATFETCH (c1);
+ if (!(c1 == '+' || c1 == '?'))
+ {
+ PATUNFETCH;
+ PATUNFETCH;
+ break;
+ }
+
+ c = c1;
+ }
+ else
+ {
+ PATUNFETCH;
+ break;
+ }
+
+ /* If we get here, we found another repeat character. */
+ }
+
+ /* Star, etc. applied to an empty pattern is equivalent
+ to an empty pattern. */
+ if (!laststart)
+ break;
+
+ /* Now we know whether or not zero matches is allowed
+ and also whether or not two or more matches is allowed. */
+ if (many_times_ok)
+ { /* More than one repetition is allowed, so put in at the
+ end a backward relative jump from `b' to before the next
+ jump we're going to put in below (which jumps from
+ laststart to after this jump).
+
+ But if we are at the `*' in the exact sequence `.*\n',
+ insert an unconditional jump backwards to the .,
+ instead of the beginning of the loop. This way we only
+ push a failure point once, instead of every time
+ through the loop. */
+ assert (p - 1 > pattern);
+
+ /* Allocate the space for the jump. */
+ GET_BUFFER_SPACE (3);
+
+ /* We know we are not at the first character of the pattern,
+ because laststart was nonzero. And we've already
+ incremented `p', by the way, to be the character after
+ the `*'. Do we have to do something analogous here
+ for null bytes, because of RE_DOT_NOT_NULL? */
+ if (TRANSLATE (*(p - 2)) == TRANSLATE ('.')
+ && zero_times_ok
+ && p < pend && TRANSLATE (*p) == TRANSLATE ('\n')
+ && !(syntax & RE_DOT_NEWLINE))
+ { /* We have .*\n. */
+ STORE_JUMP (jump, b, laststart);
+ keep_string_p = true;
+ }
+ else
+ /* Anything else. */
+ STORE_JUMP (maybe_pop_jump, b, laststart - 3);
+
+ /* We've added more stuff to the buffer. */
+ b += 3;
+ }
+
+ /* On failure, jump from laststart to b + 3, which will be the
+ end of the buffer after this jump is inserted. */
+ GET_BUFFER_SPACE (3);
+ INSERT_JUMP (keep_string_p ? on_failure_keep_string_jump
+ : on_failure_jump,
+ laststart, b + 3);
+ pending_exact = 0;
+ b += 3;
+
+ if (!zero_times_ok)
+ {
+ /* At least one repetition is required, so insert a
+ `dummy_failure_jump' before the initial
+ `on_failure_jump' instruction of the loop. This
+ effects a skip over that instruction the first time
+ we hit that loop. */
+ GET_BUFFER_SPACE (3);
+ INSERT_JUMP (dummy_failure_jump, laststart, laststart + 6);
+ b += 3;
+ }
+ }
+ break;
+
+
+ case '.':
+ laststart = b;
+ BUF_PUSH (anychar);
+ break;
+
+
+ case '[':
+ {
+ boolean had_char_class = false;
+
+ if (p == pend) return REG_EBRACK;
+
+ /* Ensure that we have enough space to push a charset: the
+ opcode, the length count, and the bitset; 34 bytes in all. */
+ GET_BUFFER_SPACE (34);
+
+ laststart = b;
+
+ /* We test `*p == '^' twice, instead of using an if
+ statement, so we only need one BUF_PUSH. */
+ BUF_PUSH (*p == '^' ? charset_not : charset);
+ if (*p == '^')
+ p++;
+
+ /* Remember the first position in the bracket expression. */
+ p1 = p;
+
+ /* Push the number of bytes in the bitmap. */
+ BUF_PUSH ((1 << BYTEWIDTH) / BYTEWIDTH);
+
+ /* Clear the whole map. */
+ bzero (b, (1 << BYTEWIDTH) / BYTEWIDTH);
+
+ /* charset_not matches newline according to a syntax bit. */
+ if ((re_opcode_t) b[-2] == charset_not
+ && (syntax & RE_HAT_LISTS_NOT_NEWLINE))
+ SET_LIST_BIT ('\n');
+
+ /* Read in characters and ranges, setting map bits. */
+ for (;;)
+ {
+ if (p == pend) return REG_EBRACK;
+
+ PATFETCH (c);
+
+ /* \ might escape characters inside [...] and [^...]. */
+ if ((syntax & RE_BACKSLASH_ESCAPE_IN_LISTS) && c == '\\')
+ {
+ if (p == pend) return REG_EESCAPE;
+
+ PATFETCH (c1);
+ SET_LIST_BIT (c1);
+ continue;
+ }
+
+ /* Could be the end of the bracket expression. If it's
+ not (i.e., when the bracket expression is `[]' so
+ far), the ']' character bit gets set way below. */
+ if (c == ']' && p != p1 + 1)
+ break;
+
+ /* Look ahead to see if it's a range when the last thing
+ was a character class. */
+ if (had_char_class && c == '-' && *p != ']')
+ return REG_ERANGE;
+
+ /* Look ahead to see if it's a range when the last thing
+ was a character: if this is a hyphen not at the
+ beginning or the end of a list, then it's the range
+ operator. */
+ if (c == '-'
+ && !(p - 2 >= pattern && p[-2] == '[')
+ && !(p - 3 >= pattern && p[-3] == '[' && p[-2] == '^')
+ && *p != ']')
+ {
+ reg_errcode_t ret
+ = compile_range (&p, pend, translate, syntax, b);
+ if (ret != REG_NOERROR) return ret;
+ }
+
+ else if (p[0] == '-' && p[1] != ']')
+ { /* This handles ranges made up of characters only. */
+ reg_errcode_t ret;
+
+ /* Move past the `-'. */
+ PATFETCH (c1);
+
+ ret = compile_range (&p, pend, translate, syntax, b);
+ if (ret != REG_NOERROR) return ret;
+ }
+
+ /* See if we're at the beginning of a possible character
+ class. */
+
+ else if (syntax & RE_CHAR_CLASSES && c == '[' && *p == ':')
+ { /* Leave room for the null. */
+ char str[CHAR_CLASS_MAX_LENGTH + 1];
+
+ PATFETCH (c);
+ c1 = 0;
+
+ /* If pattern is `[[:'. */
+ if (p == pend) return REG_EBRACK;
+
+ for (;;)
+ {
+ PATFETCH (c);
+ if (c == ':' || c == ']' || p == pend
+ || c1 == CHAR_CLASS_MAX_LENGTH)
+ break;
+ str[c1++] = c;
+ }
+ str[c1] = '\0';
+
+ /* If isn't a word bracketed by `[:' and:`]':
+ undo the ending character, the letters, and leave
+ the leading `:' and `[' (but set bits for them). */
+ if (c == ':' && *p == ']')
+ {
+ int ch;
+ boolean is_alnum = STREQ (str, "alnum");
+ boolean is_alpha = STREQ (str, "alpha");
+ boolean is_blank = STREQ (str, "blank");
+ boolean is_cntrl = STREQ (str, "cntrl");
+ boolean is_digit = STREQ (str, "digit");
+ boolean is_graph = STREQ (str, "graph");
+ boolean is_lower = STREQ (str, "lower");
+ boolean is_print = STREQ (str, "print");
+ boolean is_punct = STREQ (str, "punct");
+ boolean is_space = STREQ (str, "space");
+ boolean is_upper = STREQ (str, "upper");
+ boolean is_xdigit = STREQ (str, "xdigit");
+
+ if (!IS_CHAR_CLASS (str)) return REG_ECTYPE;
+
+ /* Throw away the ] at the end of the character
+ class. */
+ PATFETCH (c);
+
+ if (p == pend) return REG_EBRACK;
+
+ for (ch = 0; ch < 1 << BYTEWIDTH; ch++)
+ {
+ if ( (is_alnum && ISALNUM (ch))
+ || (is_alpha && ISALPHA (ch))
+ || (is_blank && ISBLANK (ch))
+ || (is_cntrl && ISCNTRL (ch))
+ || (is_digit && ISDIGIT (ch))
+ || (is_graph && ISGRAPH (ch))
+ || (is_lower && ISLOWER (ch))
+ || (is_print && ISPRINT (ch))
+ || (is_punct && ISPUNCT (ch))
+ || (is_space && ISSPACE (ch))
+ || (is_upper && ISUPPER (ch))
+ || (is_xdigit && ISXDIGIT (ch)))
+ SET_LIST_BIT (ch);
+ }
+ had_char_class = true;
+ }
+ else
+ {
+ c1++;
+ while (c1--)
+ PATUNFETCH;
+ SET_LIST_BIT ('[');
+ SET_LIST_BIT (':');
+ had_char_class = false;
+ }
+ }
+ else
+ {
+ had_char_class = false;
+ SET_LIST_BIT (c);
+ }
+ }
+
+ /* Discard any (non)matching list bytes that are all 0 at the
+ end of the map. Decrease the map-length byte too. */
+ while ((int) b[-1] > 0 && b[b[-1] - 1] == 0)
+ b[-1]--;
+ b += b[-1];
+ }
+ break;
+
+
+ case '(':
+ if (syntax & RE_NO_BK_PARENS)
+ goto handle_open;
+ else
+ goto normal_char;
+
+
+ case ')':
+ if (syntax & RE_NO_BK_PARENS)
+ goto handle_close;
+ else
+ goto normal_char;
+
+
+ case '\n':
+ if (syntax & RE_NEWLINE_ALT)
+ goto handle_alt;
+ else
+ goto normal_char;
+
+
+ case '|':
+ if (syntax & RE_NO_BK_VBAR)
+ goto handle_alt;
+ else
+ goto normal_char;
+
+
+ case '{':
+ if (syntax & RE_INTERVALS && syntax & RE_NO_BK_BRACES)
+ goto handle_interval;
+ else
+ goto normal_char;
+
+
+ case '\\':
+ if (p == pend) return REG_EESCAPE;
+
+ /* Do not translate the character after the \, so that we can
+ distinguish, e.g., \B from \b, even if we normally would
+ translate, e.g., B to b. */
+ PATFETCH_RAW (c);
+
+ switch (c)
+ {
+ case '(':
+ if (syntax & RE_NO_BK_PARENS)
+ goto normal_backslash;
+
+ handle_open:
+ bufp->re_nsub++;
+ regnum++;
+
+ if (COMPILE_STACK_FULL)
+ {
+ RETALLOC (compile_stack.stack, compile_stack.size << 1,
+ compile_stack_elt_t);
+ if (compile_stack.stack == NULL) return REG_ESPACE;
+
+ compile_stack.size <<= 1;
+ }
+
+ /* These are the values to restore when we hit end of this
+ group. They are all relative offsets, so that if the
+ whole pattern moves because of realloc, they will still
+ be valid. */
+ COMPILE_STACK_TOP.begalt_offset = begalt - bufp->buffer;
+ COMPILE_STACK_TOP.fixup_alt_jump
+ = fixup_alt_jump ? fixup_alt_jump - bufp->buffer + 1 : 0;
+ COMPILE_STACK_TOP.laststart_offset = b - bufp->buffer;
+ COMPILE_STACK_TOP.regnum = regnum;
+
+ /* We will eventually replace the 0 with the number of
+ groups inner to this one. But do not push a
+ start_memory for groups beyond the last one we can
+ represent in the compiled pattern. */
+ if (regnum <= MAX_REGNUM)
+ {
+ COMPILE_STACK_TOP.inner_group_offset = b - bufp->buffer + 2;
+ BUF_PUSH_3 (start_memory, regnum, 0);
+ }
+
+ compile_stack.avail++;
+
+ fixup_alt_jump = 0;
+ laststart = 0;
+ begalt = b;
+ /* If we've reached MAX_REGNUM groups, then this open
+ won't actually generate any code, so we'll have to
+ clear pending_exact explicitly. */
+ pending_exact = 0;
+ break;
+
+
+ case ')':
+ if (syntax & RE_NO_BK_PARENS) goto normal_backslash;
+
+ if (COMPILE_STACK_EMPTY)
+ {
+ if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)
+ goto normal_backslash;
+ else
+ return REG_ERPAREN;
+ }
+
+ handle_close:
+ if (fixup_alt_jump)
+ { /* Push a dummy failure point at the end of the
+ alternative for a possible future
+ `pop_failure_jump' to pop. See comments at
+ `push_dummy_failure' in `re_match_2'. */
+ BUF_PUSH (push_dummy_failure);
+
+ /* We allocated space for this jump when we assigned
+ to `fixup_alt_jump', in the `handle_alt' case below. */
+ STORE_JUMP (jump_past_alt, fixup_alt_jump, b - 1);
+ }
+
+ /* See similar code for backslashed left paren above. */
+ if (COMPILE_STACK_EMPTY)
+ {
+ if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)
+ goto normal_char;
+ else
+ return REG_ERPAREN;
+ }
+
+ /* Since we just checked for an empty stack above, this
+ ``can't happen''. */
+ assert (compile_stack.avail != 0);
+ {
+ /* We don't just want to restore into `regnum', because
+ later groups should continue to be numbered higher,
+ as in `(ab)c(de)' -- the second group is #2. */
+ regnum_t this_group_regnum;
+
+ compile_stack.avail--;
+ begalt = bufp->buffer + COMPILE_STACK_TOP.begalt_offset;
+ fixup_alt_jump
+ = COMPILE_STACK_TOP.fixup_alt_jump
+ ? bufp->buffer + COMPILE_STACK_TOP.fixup_alt_jump - 1
+ : 0;
+ laststart = bufp->buffer + COMPILE_STACK_TOP.laststart_offset;
+ this_group_regnum = COMPILE_STACK_TOP.regnum;
+ /* If we've reached MAX_REGNUM groups, then this open
+ won't actually generate any code, so we'll have to
+ clear pending_exact explicitly. */
+ pending_exact = 0;
+
+ /* We're at the end of the group, so now we know how many
+ groups were inside this one. */
+ if (this_group_regnum <= MAX_REGNUM)
+ {
+ unsigned char *inner_group_loc
+ = bufp->buffer + COMPILE_STACK_TOP.inner_group_offset;
+
+ *inner_group_loc = regnum - this_group_regnum;
+ BUF_PUSH_3 (stop_memory, this_group_regnum,
+ regnum - this_group_regnum);
+ }
+ }
+ break;
+
+
+ case '|': /* `\|'. */
+ if (syntax & RE_LIMITED_OPS || syntax & RE_NO_BK_VBAR)
+ goto normal_backslash;
+ handle_alt:
+ if (syntax & RE_LIMITED_OPS)
+ goto normal_char;
+
+ /* Insert before the previous alternative a jump which
+ jumps to this alternative if the former fails. */
+ GET_BUFFER_SPACE (3);
+ INSERT_JUMP (on_failure_jump, begalt, b + 6);
+ pending_exact = 0;
+ b += 3;
+
+ /* The alternative before this one has a jump after it
+ which gets executed if it gets matched. Adjust that
+ jump so it will jump to this alternative's analogous
+ jump (put in below, which in turn will jump to the next
+ (if any) alternative's such jump, etc.). The last such
+ jump jumps to the correct final destination. A picture:
+ _____ _____
+ | | | |
+ | v | v
+ a | b | c
+
+ If we are at `b', then fixup_alt_jump right now points to a
+ three-byte space after `a'. We'll put in the jump, set
+ fixup_alt_jump to right after `b', and leave behind three
+ bytes which we'll fill in when we get to after `c'. */
+
+ if (fixup_alt_jump)
+ STORE_JUMP (jump_past_alt, fixup_alt_jump, b);
+
+ /* Mark and leave space for a jump after this alternative,
+ to be filled in later either by next alternative or
+ when know we're at the end of a series of alternatives. */
+ fixup_alt_jump = b;
+ GET_BUFFER_SPACE (3);
+ b += 3;
+
+ laststart = 0;
+ begalt = b;
+ break;
+
+
+ case '{':
+ /* If \{ is a literal. */
+ if (!(syntax & RE_INTERVALS)
+ /* If we're at `\{' and it's not the open-interval
+ operator. */
+ || ((syntax & RE_INTERVALS) && (syntax & RE_NO_BK_BRACES))
+ || (p - 2 == pattern && p == pend))
+ goto normal_backslash;
+
+ handle_interval:
+ {
+ /* If got here, then the syntax allows intervals. */
+
+ /* At least (most) this many matches must be made. */
+ int lower_bound = -1, upper_bound = -1;
+
+ beg_interval = p - 1;
+
+ if (p == pend)
+ {
+ if (syntax & RE_NO_BK_BRACES)
+ goto unfetch_interval;
+ else
+ return REG_EBRACE;
+ }
+
+ GET_UNSIGNED_NUMBER (lower_bound);
+
+ if (c == ',')
+ {
+ GET_UNSIGNED_NUMBER (upper_bound);
+ if (upper_bound < 0) upper_bound = RE_DUP_MAX;
+ }
+ else
+ /* Interval such as `{1}' => match exactly once. */
+ upper_bound = lower_bound;
+
+ if (lower_bound < 0 || upper_bound > RE_DUP_MAX
+ || lower_bound > upper_bound)
+ {
+ if (syntax & RE_NO_BK_BRACES)
+ goto unfetch_interval;
+ else
+ return REG_BADBR;
+ }
+
+ if (!(syntax & RE_NO_BK_BRACES))
+ {
+ if (c != '\\') return REG_EBRACE;
+
+ PATFETCH (c);
+ }
+
+ if (c != '}')
+ {
+ if (syntax & RE_NO_BK_BRACES)
+ goto unfetch_interval;
+ else
+ return REG_BADBR;
+ }
+
+ /* We just parsed a valid interval. */
+
+ /* If it's invalid to have no preceding re. */
+ if (!laststart)
+ {
+ if (syntax & RE_CONTEXT_INVALID_OPS)
+ return REG_BADRPT;
+ else if (syntax & RE_CONTEXT_INDEP_OPS)
+ laststart = b;
+ else
+ goto unfetch_interval;
+ }
+
+ /* If the upper bound is zero, don't want to succeed at
+ all; jump from `laststart' to `b + 3', which will be
+ the end of the buffer after we insert the jump. */
+ if (upper_bound == 0)
+ {
+ GET_BUFFER_SPACE (3);
+ INSERT_JUMP (jump, laststart, b + 3);
+ b += 3;
+ }
+
+ /* Otherwise, we have a nontrivial interval. When
+ we're all done, the pattern will look like:
+ set_number_at <jump count> <upper bound>
+ set_number_at <succeed_n count> <lower bound>
+ succeed_n <after jump addr> <succed_n count>
+ <body of loop>
+ jump_n <succeed_n addr> <jump count>
+ (The upper bound and `jump_n' are omitted if
+ `upper_bound' is 1, though.) */
+ else
+ { /* If the upper bound is > 1, we need to insert
+ more at the end of the loop. */
+ unsigned nbytes = 10 + (upper_bound > 1) * 10;
+
+ GET_BUFFER_SPACE (nbytes);
+
+ /* Initialize lower bound of the `succeed_n', even
+ though it will be set during matching by its
+ attendant `set_number_at' (inserted next),
+ because `re_compile_fastmap' needs to know.
+ Jump to the `jump_n' we might insert below. */
+ INSERT_JUMP2 (succeed_n, laststart,
+ b + 5 + (upper_bound > 1) * 5,
+ lower_bound);
+ b += 5;
+
+ /* Code to initialize the lower bound. Insert
+ before the `succeed_n'. The `5' is the last two
+ bytes of this `set_number_at', plus 3 bytes of
+ the following `succeed_n'. */
+ insert_op2 (set_number_at, laststart, 5, lower_bound, b);
+ b += 5;
+
+ if (upper_bound > 1)
+ { /* More than one repetition is allowed, so
+ append a backward jump to the `succeed_n'
+ that starts this interval.
+
+ When we've reached this during matching,
+ we'll have matched the interval once, so
+ jump back only `upper_bound - 1' times. */
+ STORE_JUMP2 (jump_n, b, laststart + 5,
+ upper_bound - 1);
+ b += 5;
+
+ /* The location we want to set is the second
+ parameter of the `jump_n'; that is `b-2' as
+ an absolute address. `laststart' will be
+ the `set_number_at' we're about to insert;
+ `laststart+3' the number to set, the source
+ for the relative address. But we are
+ inserting into the middle of the pattern --
+ so everything is getting moved up by 5.
+ Conclusion: (b - 2) - (laststart + 3) + 5,
+ i.e., b - laststart.
+
+ We insert this at the beginning of the loop
+ so that if we fail during matching, we'll
+ reinitialize the bounds. */
+ insert_op2 (set_number_at, laststart, b - laststart,
+ upper_bound - 1, b);
+ b += 5;
+ }
+ }
+ pending_exact = 0;
+ beg_interval = NULL;
+ }
+ break;
+
+ unfetch_interval:
+ /* If an invalid interval, match the characters as literals. */
+ assert (beg_interval);
+ p = beg_interval;
+ beg_interval = NULL;
+
+ /* normal_char and normal_backslash need `c'. */
+ PATFETCH (c);
+
+ if (!(syntax & RE_NO_BK_BRACES))
+ {
+ if (p > pattern && p[-1] == '\\')
+ goto normal_backslash;
+ }
+ goto normal_char;
+
+#ifdef emacs
+ /* There is no way to specify the before_dot and after_dot
+ operators. rms says this is ok. --karl */
+ case '=':
+ BUF_PUSH (at_dot);
+ break;
+
+ case 's':
+ laststart = b;
+ PATFETCH (c);
+ BUF_PUSH_2 (syntaxspec, syntax_spec_code[c]);
+ break;
+
+ case 'S':
+ laststart = b;
+ PATFETCH (c);
+ BUF_PUSH_2 (notsyntaxspec, syntax_spec_code[c]);
+ break;
+#endif /* emacs */
+
+
+ case 'w':
+ laststart = b;
+ BUF_PUSH (wordchar);
+ break;
+
+
+ case 'W':
+ laststart = b;
+ BUF_PUSH (notwordchar);
+ break;
+
+
+ case '<':
+ BUF_PUSH (wordbeg);
+ break;
+
+ case '>':
+ BUF_PUSH (wordend);
+ break;
+
+ case 'b':
+ BUF_PUSH (wordbound);
+ break;
+
+ case 'B':
+ BUF_PUSH (notwordbound);
+ break;
+
+ case '`':
+ BUF_PUSH (begbuf);
+ break;
+
+ case '\'':
+ BUF_PUSH (endbuf);
+ break;
+
+ case '1': case '2': case '3': case '4': case '5':
+ case '6': case '7': case '8': case '9':
+ if (syntax & RE_NO_BK_REFS)
+ goto normal_char;
+
+ c1 = c - '0';
+
+ if (c1 > regnum)
+ return REG_ESUBREG;
+
+ /* Can't back reference to a subexpression if inside of it. */
+ if (group_in_compile_stack (compile_stack, c1))
+ goto normal_char;
+
+ laststart = b;
+ BUF_PUSH_2 (duplicate, c1);
+ break;
+
+
+ case '+':
+ case '?':
+ if (syntax & RE_BK_PLUS_QM)
+ goto handle_plus;
+ else
+ goto normal_backslash;
+
+ default:
+ normal_backslash:
+ /* You might think it would be useful for \ to mean
+ not to translate; but if we don't translate it
+ it will never match anything. */
+ c = TRANSLATE (c);
+ goto normal_char;
+ }
+ break;
+
+
+ default:
+ /* Expects the character in `c'. */
+ normal_char:
+ /* If no exactn currently being built. */
+ if (!pending_exact
+
+ /* If last exactn not at current position. */
+ || pending_exact + *pending_exact + 1 != b
+
+ /* We have only one byte following the exactn for the count. */
+ || *pending_exact == (1 << BYTEWIDTH) - 1
+
+ /* If followed by a repetition operator. */
+ || *p == '*' || *p == '^'
+ || ((syntax & RE_BK_PLUS_QM)
+ ? *p == '\\' && (p[1] == '+' || p[1] == '?')
+ : (*p == '+' || *p == '?'))
+ || ((syntax & RE_INTERVALS)
+ && ((syntax & RE_NO_BK_BRACES)
+ ? *p == '{'
+ : (p[0] == '\\' && p[1] == '{'))))
+ {
+ /* Start building a new exactn. */
+
+ laststart = b;
+
+ BUF_PUSH_2 (exactn, 0);
+ pending_exact = b - 1;
+ }
+
+ BUF_PUSH (c);
+ (*pending_exact)++;
+ break;
+ } /* switch (c) */
+ } /* while p != pend */
+
+
+ /* Through the pattern now. */
+
+ if (fixup_alt_jump)
+ STORE_JUMP (jump_past_alt, fixup_alt_jump, b);
+
+ if (!COMPILE_STACK_EMPTY)
+ return REG_EPAREN;
+
+ free (compile_stack.stack);
+
+ /* We have succeeded; set the length of the buffer. */
+ bufp->used = b - bufp->buffer;
+
+#ifdef DEBUG
+ if (debug)
+ {
+ DEBUG_PRINT1 ("\nCompiled pattern: ");
+ print_compiled_pattern (bufp);
+ }
+#endif /* DEBUG */
+
+ return REG_NOERROR;
+} /* regex_compile */
+\f
+/* Subroutines for `regex_compile'. */
+
+/* Store OP at LOC followed by two-byte integer parameter ARG. */
+
+static void
+store_op1 (op, loc, arg)
+ re_opcode_t op;
+ unsigned char *loc;
+ int arg;
+{
+ *loc = (unsigned char) op;
+ STORE_NUMBER (loc + 1, arg);
+}
+
+
+/* Like `store_op1', but for two two-byte parameters ARG1 and ARG2. */
+
+static void
+store_op2 (op, loc, arg1, arg2)
+ re_opcode_t op;
+ unsigned char *loc;
+ int arg1, arg2;
+{
+ *loc = (unsigned char) op;
+ STORE_NUMBER (loc + 1, arg1);
+ STORE_NUMBER (loc + 3, arg2);
+}
+
+
+/* Copy the bytes from LOC to END to open up three bytes of space at LOC
+ for OP followed by two-byte integer parameter ARG. */
+
+static void
+insert_op1 (op, loc, arg, end)
+ re_opcode_t op;
+ unsigned char *loc;
+ int arg;
+ unsigned char *end;
+{
+ register unsigned char *pfrom = end;
+ register unsigned char *pto = end + 3;
+
+ while (pfrom != loc)
+ *--pto = *--pfrom;
+
+ store_op1 (op, loc, arg);
+}
+
+
+/* Like `insert_op1', but for two two-byte parameters ARG1 and ARG2. */
+
+static void
+insert_op2 (op, loc, arg1, arg2, end)
+ re_opcode_t op;
+ unsigned char *loc;
+ int arg1, arg2;
+ unsigned char *end;
+{
+ register unsigned char *pfrom = end;
+ register unsigned char *pto = end + 5;
+
+ while (pfrom != loc)
+ *--pto = *--pfrom;
+
+ store_op2 (op, loc, arg1, arg2);
+}
+
+
+/* P points to just after a ^ in PATTERN. Return true if that ^ comes
+ after an alternative or a begin-subexpression. We assume there is at
+ least one character before the ^. */
+
+static boolean
+at_begline_loc_p (pattern, p, syntax)
+ const char *pattern, *p;
+ reg_syntax_t syntax;
+{
+ const char *prev = p - 2;
+ boolean prev_prev_backslash = prev > pattern && prev[-1] == '\\';
+
+ return
+ /* After a subexpression? */
+ (*prev == '(' && (syntax & RE_NO_BK_PARENS || prev_prev_backslash))
+ /* After an alternative? */
+ || (*prev == '|' && (syntax & RE_NO_BK_VBAR || prev_prev_backslash));
+}
+
+
+/* The dual of at_begline_loc_p. This one is for $. We assume there is
+ at least one character after the $, i.e., `P < PEND'. */
+
+static boolean
+at_endline_loc_p (p, pend, syntax)
+ const char *p, *pend;
+ int syntax;
+{
+ const char *next = p;
+ boolean next_backslash = *next == '\\';
+ const char *next_next = p + 1 < pend ? p + 1 : NULL;
+
+ return
+ /* Before a subexpression? */
+ (syntax & RE_NO_BK_PARENS ? *next == ')'
+ : next_backslash && next_next && *next_next == ')')
+ /* Before an alternative? */
+ || (syntax & RE_NO_BK_VBAR ? *next == '|'
+ : next_backslash && next_next && *next_next == '|');
+}
+
+
+/* Returns true if REGNUM is in one of COMPILE_STACK's elements and
+ false if it's not. */
+
+static boolean
+group_in_compile_stack (compile_stack, regnum)
+ compile_stack_type compile_stack;
+ regnum_t regnum;
+{
+ int this_element;
+
+ for (this_element = compile_stack.avail - 1;
+ this_element >= 0;
+ this_element--)
+ if (compile_stack.stack[this_element].regnum == regnum)
+ return true;
+
+ return false;
+}
+
+
+/* Read the ending character of a range (in a bracket expression) from the
+ uncompiled pattern *P_PTR (which ends at PEND). We assume the
+ starting character is in `P[-2]'. (`P[-1]' is the character `-'.)
+ Then we set the translation of all bits between the starting and
+ ending characters (inclusive) in the compiled pattern B.
+
+ Return an error code.
+
+ We use these short variable names so we can use the same macros as
+ `regex_compile' itself. */
+
+static reg_errcode_t
+compile_range (p_ptr, pend, translate, syntax, b)
+ const char **p_ptr, *pend;
+ char *translate;
+ reg_syntax_t syntax;
+ unsigned char *b;
+{
+ unsigned this_char;
+
+ const char *p = *p_ptr;
+ int range_start, range_end;
+
+ if (p == pend)
+ return REG_ERANGE;
+
+ /* Even though the pattern is a signed `char *', we need to fetch
+ with unsigned char *'s; if the high bit of the pattern character
+ is set, the range endpoints will be negative if we fetch using a
+ signed char *.
+
+ We also want to fetch the endpoints without translating them; the
+ appropriate translation is done in the bit-setting loop below. */
+ range_start = ((unsigned char *) p)[-2];
+ range_end = ((unsigned char *) p)[0];
+
+ /* Have to increment the pointer into the pattern string, so the
+ caller isn't still at the ending character. */
+ (*p_ptr)++;
+
+ /* If the start is after the end, the range is empty. */
+ if (range_start > range_end)
+ return syntax & RE_NO_EMPTY_RANGES ? REG_ERANGE : REG_NOERROR;
+
+ /* Here we see why `this_char' has to be larger than an `unsigned
+ char' -- the range is inclusive, so if `range_end' == 0xff
+ (assuming 8-bit characters), we would otherwise go into an infinite
+ loop, since all characters <= 0xff. */
+ for (this_char = range_start; this_char <= range_end; this_char++)
+ {
+ SET_LIST_BIT (TRANSLATE (this_char));
+ }
+
+ return REG_NOERROR;
+}
+\f
+/* Failure stack declarations and macros; both re_compile_fastmap and
+ re_match_2 use a failure stack. These have to be macros because of
+ REGEX_ALLOCATE. */
+
+
+/* Number of failure points for which to initially allocate space
+ when matching. If this number is exceeded, we allocate more
+ space, so it is not a hard limit. */
+#ifndef INIT_FAILURE_ALLOC
+#define INIT_FAILURE_ALLOC 5
+#endif
+
+/* Roughly the maximum number of failure points on the stack. Would be
+ exactly that if always used MAX_FAILURE_SPACE each time we failed.
+ This is a variable only so users of regex can assign to it; we never
+ change it ourselves. */
+int re_max_failures = 2000;
+
+typedef const unsigned char *fail_stack_elt_t;
+
+typedef struct
+{
+ fail_stack_elt_t *stack;
+ unsigned size;
+ unsigned avail; /* Offset of next open position. */
+} fail_stack_type;
+
+#define FAIL_STACK_EMPTY() (fail_stack.avail == 0)
+#define FAIL_STACK_PTR_EMPTY() (fail_stack_ptr->avail == 0)
+#define FAIL_STACK_FULL() (fail_stack.avail == fail_stack.size)
+#define FAIL_STACK_TOP() (fail_stack.stack[fail_stack.avail])
+
+
+/* Initialize `fail_stack'. Do `return -2' if the alloc fails. */
+
+#define INIT_FAIL_STACK() \
+ do { \
+ fail_stack.stack = (fail_stack_elt_t *) \
+ REGEX_ALLOCATE (INIT_FAILURE_ALLOC * sizeof (fail_stack_elt_t)); \
+ \
+ if (fail_stack.stack == NULL) \
+ return -2; \
+ \
+ fail_stack.size = INIT_FAILURE_ALLOC; \
+ fail_stack.avail = 0; \
+ } while (0)
+
+
+/* Double the size of FAIL_STACK, up to approximately `re_max_failures' items.
+
+ Return 1 if succeeds, and 0 if either ran out of memory
+ allocating space for it or it was already too large.
+
+ REGEX_REALLOCATE requires `destination' be declared. */
+
+#define DOUBLE_FAIL_STACK(fail_stack) \
+ ((fail_stack).size > re_max_failures * MAX_FAILURE_ITEMS \
+ ? 0 \
+ : ((fail_stack).stack = (fail_stack_elt_t *) \
+ REGEX_REALLOCATE ((fail_stack).stack, \
+ (fail_stack).size * sizeof (fail_stack_elt_t), \
+ ((fail_stack).size << 1) * sizeof (fail_stack_elt_t)), \
+ \
+ (fail_stack).stack == NULL \
+ ? 0 \
+ : ((fail_stack).size <<= 1, \
+ 1)))
+
+
+/* Push PATTERN_OP on FAIL_STACK.
+
+ Return 1 if was able to do so and 0 if ran out of memory allocating
+ space to do so. */
+#define PUSH_PATTERN_OP(pattern_op, fail_stack) \
+ ((FAIL_STACK_FULL () \
+ && !DOUBLE_FAIL_STACK (fail_stack)) \
+ ? 0 \
+ : ((fail_stack).stack[(fail_stack).avail++] = pattern_op, \
+ 1))
+
+/* This pushes an item onto the failure stack. Must be a four-byte
+ value. Assumes the variable `fail_stack'. Probably should only
+ be called from within `PUSH_FAILURE_POINT'. */
+#define PUSH_FAILURE_ITEM(item) \
+ fail_stack.stack[fail_stack.avail++] = (fail_stack_elt_t) item
+
+/* The complement operation. Assumes `fail_stack' is nonempty. */
+#define POP_FAILURE_ITEM() fail_stack.stack[--fail_stack.avail]
+
+/* Used to omit pushing failure point id's when we're not debugging. */
+#ifdef DEBUG
+#define DEBUG_PUSH PUSH_FAILURE_ITEM
+#define DEBUG_POP(item_addr) *(item_addr) = POP_FAILURE_ITEM ()
+#else
+#define DEBUG_PUSH(item)
+#define DEBUG_POP(item_addr)
+#endif
+
+
+/* Push the information about the state we will need
+ if we ever fail back to it.
+
+ Requires variables fail_stack, regstart, regend, reg_info, and
+ num_regs be declared. DOUBLE_FAIL_STACK requires `destination' be
+ declared.
+
+ Does `return FAILURE_CODE' if runs out of memory. */
+
+#define PUSH_FAILURE_POINT(pattern_place, string_place, failure_code) \
+ do { \
+ char *destination; \
+ /* Must be int, so when we don't save any registers, the arithmetic \
+ of 0 + -1 isn't done as unsigned. */ \
+ int this_reg; \
+ \
+ DEBUG_STATEMENT (failure_id++); \
+ DEBUG_STATEMENT (nfailure_points_pushed++); \
+ DEBUG_PRINT2 ("\nPUSH_FAILURE_POINT #%u:\n", failure_id); \
+ DEBUG_PRINT2 (" Before push, next avail: %d\n", (fail_stack).avail);\
+ DEBUG_PRINT2 (" size: %d\n", (fail_stack).size);\
+ \
+ DEBUG_PRINT2 (" slots needed: %d\n", NUM_FAILURE_ITEMS); \
+ DEBUG_PRINT2 (" available: %d\n", REMAINING_AVAIL_SLOTS); \
+ \
+ /* Ensure we have enough space allocated for what we will push. */ \
+ while (REMAINING_AVAIL_SLOTS < NUM_FAILURE_ITEMS) \
+ { \
+ if (!DOUBLE_FAIL_STACK (fail_stack)) \
+ return failure_code; \
+ \
+ DEBUG_PRINT2 ("\n Doubled stack; size now: %d\n", \
+ (fail_stack).size); \
+ DEBUG_PRINT2 (" slots available: %d\n", REMAINING_AVAIL_SLOTS);\
+ } \
+ \
+ /* Push the info, starting with the registers. */ \
+ DEBUG_PRINT1 ("\n"); \
+ \
+ for (this_reg = lowest_active_reg; this_reg <= highest_active_reg; \
+ this_reg++) \
+ { \
+ DEBUG_PRINT2 (" Pushing reg: %d\n", this_reg); \
+ DEBUG_STATEMENT (num_regs_pushed++); \
+ \
+ DEBUG_PRINT2 (" start: 0x%x\n", regstart[this_reg]); \
+ PUSH_FAILURE_ITEM (regstart[this_reg]); \
+ \
+ DEBUG_PRINT2 (" end: 0x%x\n", regend[this_reg]); \
+ PUSH_FAILURE_ITEM (regend[this_reg]); \
+ \
+ DEBUG_PRINT2 (" info: 0x%x\n ", reg_info[this_reg]); \
+ DEBUG_PRINT2 (" match_null=%d", \
+ REG_MATCH_NULL_STRING_P (reg_info[this_reg])); \
+ DEBUG_PRINT2 (" active=%d", IS_ACTIVE (reg_info[this_reg])); \
+ DEBUG_PRINT2 (" matched_something=%d", \
+ MATCHED_SOMETHING (reg_info[this_reg])); \
+ DEBUG_PRINT2 (" ever_matched=%d", \
+ EVER_MATCHED_SOMETHING (reg_info[this_reg])); \
+ DEBUG_PRINT1 ("\n"); \
+ PUSH_FAILURE_ITEM (reg_info[this_reg].word); \
+ } \
+ \
+ DEBUG_PRINT2 (" Pushing low active reg: %d\n", lowest_active_reg);\
+ PUSH_FAILURE_ITEM (lowest_active_reg); \
+ \
+ DEBUG_PRINT2 (" Pushing high active reg: %d\n", highest_active_reg);\
+ PUSH_FAILURE_ITEM (highest_active_reg); \
+ \
+ DEBUG_PRINT2 (" Pushing pattern 0x%x: ", pattern_place); \
+ DEBUG_PRINT_COMPILED_PATTERN (bufp, pattern_place, pend); \
+ PUSH_FAILURE_ITEM (pattern_place); \
+ \
+ DEBUG_PRINT2 (" Pushing string 0x%x: `", string_place); \
+ DEBUG_PRINT_DOUBLE_STRING (string_place, string1, size1, string2, \
+ size2); \
+ DEBUG_PRINT1 ("'\n"); \
+ PUSH_FAILURE_ITEM (string_place); \
+ \
+ DEBUG_PRINT2 (" Pushing failure id: %u\n", failure_id); \
+ DEBUG_PUSH (failure_id); \
+ } while (0)
+
+/* This is the number of items that are pushed and popped on the stack
+ for each register. */
+#define NUM_REG_ITEMS 3
+
+/* Individual items aside from the registers. */
+#ifdef DEBUG
+#define NUM_NONREG_ITEMS 5 /* Includes failure point id. */
+#else
+#define NUM_NONREG_ITEMS 4
+#endif
+
+/* We push at most this many items on the stack. */
+#define MAX_FAILURE_ITEMS ((num_regs - 1) * NUM_REG_ITEMS + NUM_NONREG_ITEMS)
+
+/* We actually push this many items. */
+#define NUM_FAILURE_ITEMS \
+ ((highest_active_reg - lowest_active_reg + 1) * NUM_REG_ITEMS \
+ + NUM_NONREG_ITEMS)
+
+/* How many items can still be added to the stack without overflowing it. */
+#define REMAINING_AVAIL_SLOTS ((fail_stack).size - (fail_stack).avail)
+
+
+/* Pops what PUSH_FAIL_STACK pushes.
+
+ We restore into the parameters, all of which should be lvalues:
+ STR -- the saved data position.
+ PAT -- the saved pattern position.
+ LOW_REG, HIGH_REG -- the highest and lowest active registers.
+ REGSTART, REGEND -- arrays of string positions.
+ REG_INFO -- array of information about each subexpression.
+
+ Also assumes the variables `fail_stack' and (if debugging), `bufp',
+ `pend', `string1', `size1', `string2', and `size2'. */
+
+#define POP_FAILURE_POINT(str, pat, low_reg, high_reg, regstart, regend, reg_info)\
+{ \
+ DEBUG_STATEMENT (fail_stack_elt_t failure_id;) \
+ int this_reg; \
+ const unsigned char *string_temp; \
+ \
+ assert (!FAIL_STACK_EMPTY ()); \
+ \
+ /* Remove failure points and point to how many regs pushed. */ \
+ DEBUG_PRINT1 ("POP_FAILURE_POINT:\n"); \
+ DEBUG_PRINT2 (" Before pop, next avail: %d\n", fail_stack.avail); \
+ DEBUG_PRINT2 (" size: %d\n", fail_stack.size); \
+ \
+ assert (fail_stack.avail >= NUM_NONREG_ITEMS); \
+ \
+ DEBUG_POP (&failure_id); \
+ DEBUG_PRINT2 (" Popping failure id: %u\n", failure_id); \
+ \
+ /* If the saved string location is NULL, it came from an \
+ on_failure_keep_string_jump opcode, and we want to throw away the \
+ saved NULL, thus retaining our current position in the string. */ \
+ string_temp = POP_FAILURE_ITEM (); \
+ if (string_temp != NULL) \
+ str = (const char *) string_temp; \
+ \
+ DEBUG_PRINT2 (" Popping string 0x%x: `", str); \
+ DEBUG_PRINT_DOUBLE_STRING (str, string1, size1, string2, size2); \
+ DEBUG_PRINT1 ("'\n"); \
+ \
+ pat = (unsigned char *) POP_FAILURE_ITEM (); \
+ DEBUG_PRINT2 (" Popping pattern 0x%x: ", pat); \
+ DEBUG_PRINT_COMPILED_PATTERN (bufp, pat, pend); \
+ \
+ /* Restore register info. */ \
+ high_reg = (unsigned) POP_FAILURE_ITEM (); \
+ DEBUG_PRINT2 (" Popping high active reg: %d\n", high_reg); \
+ \
+ low_reg = (unsigned) POP_FAILURE_ITEM (); \
+ DEBUG_PRINT2 (" Popping low active reg: %d\n", low_reg); \
+ \
+ for (this_reg = high_reg; this_reg >= low_reg; this_reg--) \
+ { \
+ DEBUG_PRINT2 (" Popping reg: %d\n", this_reg); \
+ \
+ reg_info[this_reg].word = POP_FAILURE_ITEM (); \
+ DEBUG_PRINT2 (" info: 0x%x\n", reg_info[this_reg]); \
+ \
+ regend[this_reg] = (const char *) POP_FAILURE_ITEM (); \
+ DEBUG_PRINT2 (" end: 0x%x\n", regend[this_reg]); \
+ \
+ regstart[this_reg] = (const char *) POP_FAILURE_ITEM (); \
+ DEBUG_PRINT2 (" start: 0x%x\n", regstart[this_reg]); \
+ } \
+ \
+ DEBUG_STATEMENT (nfailure_points_popped++); \
+} /* POP_FAILURE_POINT */
+\f
+/* re_compile_fastmap computes a ``fastmap'' for the compiled pattern in
+ BUFP. A fastmap records which of the (1 << BYTEWIDTH) possible
+ characters can start a string that matches the pattern. This fastmap
+ is used by re_search to skip quickly over impossible starting points.
+
+ The caller must supply the address of a (1 << BYTEWIDTH)-byte data
+ area as BUFP->fastmap.
+
+ We set the `fastmap', `fastmap_accurate', and `can_be_null' fields in
+ the pattern buffer.
+
+ Returns 0 if we succeed, -2 if an internal error. */
+
+int
+re_compile_fastmap (bufp)
+ struct re_pattern_buffer *bufp;
+{
+ int j, k;
+ fail_stack_type fail_stack;
+#ifndef REGEX_MALLOC
+ char *destination;
+#endif
+ /* We don't push any register information onto the failure stack. */
+ unsigned num_regs = 0;
+
+ register char *fastmap = bufp->fastmap;
+ unsigned char *pattern = bufp->buffer;
+ unsigned long size = bufp->used;
+ const unsigned char *p = pattern;
+ register unsigned char *pend = pattern + size;
+
+ /* Assume that each path through the pattern can be null until
+ proven otherwise. We set this false at the bottom of switch
+ statement, to which we get only if a particular path doesn't
+ match the empty string. */
+ boolean path_can_be_null = true;
+
+ /* We aren't doing a `succeed_n' to begin with. */
+ boolean succeed_n_p = false;
+
+ assert (fastmap != NULL && p != NULL);
+
+ INIT_FAIL_STACK ();
+ bzero (fastmap, 1 << BYTEWIDTH); /* Assume nothing's valid. */
+ bufp->fastmap_accurate = 1; /* It will be when we're done. */
+ bufp->can_be_null = 0;
+
+ while (p != pend || !FAIL_STACK_EMPTY ())
+ {
+ if (p == pend)
+ {
+ bufp->can_be_null |= path_can_be_null;
+
+ /* Reset for next path. */
+ path_can_be_null = true;
+
+ p = fail_stack.stack[--fail_stack.avail];
+ }
+
+ /* We should never be about to go beyond the end of the pattern. */
+ assert (p < pend);
+
+#ifdef SWITCH_ENUM_BUG
+ switch ((int) ((re_opcode_t) *p++))
+#else
+ switch ((re_opcode_t) *p++)
+#endif
+ {
+
+ /* I guess the idea here is to simply not bother with a fastmap
+ if a backreference is used, since it's too hard to figure out
+ the fastmap for the corresponding group. Setting
+ `can_be_null' stops `re_search_2' from using the fastmap, so
+ that is all we do. */
+ case duplicate:
+ bufp->can_be_null = 1;
+ return 0;
+
+
+ /* Following are the cases which match a character. These end
+ with `break'. */
+
+ case exactn:
+ fastmap[p[1]] = 1;
+ break;
+
+
+ case charset:
+ for (j = *p++ * BYTEWIDTH - 1; j >= 0; j--)
+ if (p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH)))
+ fastmap[j] = 1;
+ break;
+
+
+ case charset_not:
+ /* Chars beyond end of map must be allowed. */
+ for (j = *p * BYTEWIDTH; j < (1 << BYTEWIDTH); j++)
+ fastmap[j] = 1;
+
+ for (j = *p++ * BYTEWIDTH - 1; j >= 0; j--)
+ if (!(p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH))))
+ fastmap[j] = 1;
+ break;
+
+
+ case wordchar:
+ for (j = 0; j < (1 << BYTEWIDTH); j++)
+ if (SYNTAX (j) == Sword)
+ fastmap[j] = 1;
+ break;
+
+
+ case notwordchar:
+ for (j = 0; j < (1 << BYTEWIDTH); j++)
+ if (SYNTAX (j) != Sword)
+ fastmap[j] = 1;
+ break;
+
+
+ case anychar:
+ /* `.' matches anything ... */
+ for (j = 0; j < (1 << BYTEWIDTH); j++)
+ fastmap[j] = 1;
+
+ /* ... except perhaps newline. */
+ if (!(bufp->syntax & RE_DOT_NEWLINE))
+ fastmap['\n'] = 0;
+
+ /* Return if we have already set `can_be_null'; if we have,
+ then the fastmap is irrelevant. Something's wrong here. */
+ else if (bufp->can_be_null)
+ return 0;
+
+ /* Otherwise, have to check alternative paths. */
+ break;
+
+
+#ifdef emacs
+ case syntaxspec:
+ k = *p++;
+ for (j = 0; j < (1 << BYTEWIDTH); j++)
+ if (SYNTAX (j) == (enum syntaxcode) k)
+ fastmap[j] = 1;
+ break;
+
+
+ case notsyntaxspec:
+ k = *p++;
+ for (j = 0; j < (1 << BYTEWIDTH); j++)
+ if (SYNTAX (j) != (enum syntaxcode) k)
+ fastmap[j] = 1;
+ break;
+
+
+ /* All cases after this match the empty string. These end with
+ `continue'. */
+
+
+ case before_dot:
+ case at_dot:
+ case after_dot:
+ continue;
+#endif /* not emacs */
+
+
+ case no_op:
+ case begline:
+ case endline:
+ case begbuf:
+ case endbuf:
+ case wordbound:
+ case notwordbound:
+ case wordbeg:
+ case wordend:
+ case push_dummy_failure:
+ continue;
+
+
+ case jump_n:
+ case pop_failure_jump:
+ case maybe_pop_jump:
+ case jump:
+ case jump_past_alt:
+ case dummy_failure_jump:
+ EXTRACT_NUMBER_AND_INCR (j, p);
+ p += j;
+ if (j > 0)
+ continue;
+
+ /* Jump backward implies we just went through the body of a
+ loop and matched nothing. Opcode jumped to should be
+ `on_failure_jump' or `succeed_n'. Just treat it like an
+ ordinary jump. For a * loop, it has pushed its failure
+ point already; if so, discard that as redundant. */
+ if ((re_opcode_t) *p != on_failure_jump
+ && (re_opcode_t) *p != succeed_n)
+ continue;
+
+ p++;
+ EXTRACT_NUMBER_AND_INCR (j, p);
+ p += j;
+
+ /* If what's on the stack is where we are now, pop it. */
+ if (!FAIL_STACK_EMPTY ()
+ && fail_stack.stack[fail_stack.avail - 1] == p)
+ fail_stack.avail--;
+
+ continue;
+
+
+ case on_failure_jump:
+ case on_failure_keep_string_jump:
+ handle_on_failure_jump:
+ EXTRACT_NUMBER_AND_INCR (j, p);
+
+ /* For some patterns, e.g., `(a?)?', `p+j' here points to the
+ end of the pattern. We don't want to push such a point,
+ since when we restore it above, entering the switch will
+ increment `p' past the end of the pattern. We don't need
+ to push such a point since we obviously won't find any more
+ fastmap entries beyond `pend'. Such a pattern can match
+ the null string, though. */
+ if (p + j < pend)
+ {
+ if (!PUSH_PATTERN_OP (p + j, fail_stack))
+ return -2;
+ }
+ else
+ bufp->can_be_null = 1;
+
+ if (succeed_n_p)
+ {
+ EXTRACT_NUMBER_AND_INCR (k, p); /* Skip the n. */
+ succeed_n_p = false;
+ }
+
+ continue;
+
+
+ case succeed_n:
+ /* Get to the number of times to succeed. */
+ p += 2;
+
+ /* Increment p past the n for when k != 0. */
+ EXTRACT_NUMBER_AND_INCR (k, p);
+ if (k == 0)
+ {
+ p -= 4;
+ succeed_n_p = true; /* Spaghetti code alert. */
+ goto handle_on_failure_jump;
+ }
+ continue;
+
+
+ case set_number_at:
+ p += 4;
+ continue;
+
+
+ case start_memory:
+ case stop_memory:
+ p += 2;
+ continue;
+
+
+ default:
+ abort (); /* We have listed all the cases. */
+ } /* switch *p++ */
+
+ /* Getting here means we have found the possible starting
+ characters for one path of the pattern -- and that the empty
+ string does not match. We need not follow this path further.
+ Instead, look at the next alternative (remembered on the
+ stack), or quit if no more. The test at the top of the loop
+ does these things. */
+ path_can_be_null = false;
+ p = pend;
+ } /* while p */
+
+ /* Set `can_be_null' for the last path (also the first path, if the
+ pattern is empty). */
+ bufp->can_be_null |= path_can_be_null;
+ return 0;
+} /* re_compile_fastmap */
+\f
+/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
+ ENDS. Subsequent matches using PATTERN_BUFFER and REGS will use
+ this memory for recording register information. STARTS and ENDS
+ must be allocated using the malloc library routine, and must each
+ be at least NUM_REGS * sizeof (regoff_t) bytes long.
+
+ If NUM_REGS == 0, then subsequent matches should allocate their own
+ register data.
+
+ Unless this function is called, the first search or match using
+ PATTERN_BUFFER will allocate its own register data, without
+ freeing the old data. */
+
+void
+re_set_registers (bufp, regs, num_regs, starts, ends)
+ struct re_pattern_buffer *bufp;
+ struct re_registers *regs;
+ unsigned num_regs;
+ regoff_t *starts, *ends;
+{
+ if (num_regs)
+ {
+ bufp->regs_allocated = REGS_REALLOCATE;
+ regs->num_regs = num_regs;
+ regs->start = starts;
+ regs->end = ends;
+ }
+ else
+ {
+ bufp->regs_allocated = REGS_UNALLOCATED;
+ regs->num_regs = 0;
+ regs->start = regs->end = (regoff_t) 0;
+ }
+}
+\f
+/* Searching routines. */
+
+/* Like re_search_2, below, but only one string is specified, and
+ doesn't let you say where to stop matching. */
+
+int
+re_search (bufp, string, size, startpos, range, regs)
+ struct re_pattern_buffer *bufp;
+ const char *string;
+ int size, startpos, range;
+ struct re_registers *regs;
+{
+ return re_search_2 (bufp, NULL, 0, string, size, startpos, range,
+ regs, size);
+}
+
+
+/* Using the compiled pattern in BUFP->buffer, first tries to match the
+ virtual concatenation of STRING1 and STRING2, starting first at index
+ STARTPOS, then at STARTPOS + 1, and so on.
+
+ STRING1 and STRING2 have length SIZE1 and SIZE2, respectively.
+
+ RANGE is how far to scan while trying to match. RANGE = 0 means try
+ only at STARTPOS; in general, the last start tried is STARTPOS +
+ RANGE.
+
+ In REGS, return the indices of the virtual concatenation of STRING1
+ and STRING2 that matched the entire BUFP->buffer and its contained
+ subexpressions.
+
+ Do not consider matching one past the index STOP in the virtual
+ concatenation of STRING1 and STRING2.
+
+ We return either the position in the strings at which the match was
+ found, -1 if no match, or -2 if error (such as failure
+ stack overflow). */
+
+int
+re_search_2 (bufp, string1, size1, string2, size2, startpos, range, regs, stop)
+ struct re_pattern_buffer *bufp;
+ const char *string1, *string2;
+ int size1, size2;
+ int startpos;
+ int range;
+ struct re_registers *regs;
+ int stop;
+{
+ int val;
+ register char *fastmap = bufp->fastmap;
+ register char *translate = bufp->translate;
+ int total_size = size1 + size2;
+ int endpos = startpos + range;
+
+ /* Check for out-of-range STARTPOS. */
+ if (startpos < 0 || startpos > total_size)
+ return -1;
+
+ /* Fix up RANGE if it might eventually take us outside
+ the virtual concatenation of STRING1 and STRING2. */
+ if (endpos < -1)
+ range = -1 - startpos;
+ else if (endpos > total_size)
+ range = total_size - startpos;
+
+ /* If the search isn't to be a backwards one, don't waste time in a
+ search for a pattern that must be anchored. */
+ if (bufp->used > 0 && (re_opcode_t) bufp->buffer[0] == begbuf && range > 0)
+ {
+ if (startpos > 0)
+ return -1;
+ else
+ range = 1;
+ }
+
+ /* Update the fastmap now if not correct already. */
+ if (fastmap && !bufp->fastmap_accurate)
+ if (re_compile_fastmap (bufp) == -2)
+ return -2;
+
+ /* Loop through the string, looking for a place to start matching. */
+ for (;;)
+ {
+ /* If a fastmap is supplied, skip quickly over characters that
+ cannot be the start of a match. If the pattern can match the
+ null string, however, we don't need to skip characters; we want
+ the first null string. */
+ if (fastmap && startpos < total_size && !bufp->can_be_null)
+ {
+ if (range > 0) /* Searching forwards. */
+ {
+ register const char *d;
+ register int lim = 0;
+ int irange = range;
+
+ if (startpos < size1 && startpos + range >= size1)
+ lim = range - (size1 - startpos);
+
+ d = (startpos >= size1 ? string2 - size1 : string1) + startpos;
+
+ /* Written out as an if-else to avoid testing `translate'
+ inside the loop. */
+ if (translate)
+ while (range > lim
+ && !fastmap[(unsigned char)
+ translate[(unsigned char) *d++]])
+ range--;
+ else
+ while (range > lim && !fastmap[(unsigned char) *d++])
+ range--;
+
+ startpos += irange - range;
+ }
+ else /* Searching backwards. */
+ {
+ register char c = (size1 == 0 || startpos >= size1
+ ? string2[startpos - size1]
+ : string1[startpos]);
+
+ if (!fastmap[(unsigned char) TRANSLATE (c)])
+ goto advance;
+ }
+ }
+
+ /* If can't match the null string, and that's all we have left, fail. */
+ if (range >= 0 && startpos == total_size && fastmap
+ && !bufp->can_be_null)
+ return -1;
+
+ val = re_match_2 (bufp, string1, size1, string2, size2,
+ startpos, regs, stop);
+ if (val >= 0)
+ return startpos;
+
+ if (val == -2)
+ return -2;
+
+ advance:
+ if (!range)
+ break;
+ else if (range > 0)
+ {
+ range--;
+ startpos++;
+ }
+ else
+ {
+ range++;
+ startpos--;
+ }
+ }
+ return -1;
+} /* re_search_2 */
+\f
+/* Declarations and macros for re_match_2. */
+
+static int bcmp_translate ();
+static boolean alt_match_null_string_p (),
+ common_op_match_null_string_p (),
+ group_match_null_string_p ();
+
+/* Structure for per-register (a.k.a. per-group) information.
+ This must not be longer than one word, because we push this value
+ onto the failure stack. Other register information, such as the
+ starting and ending positions (which are addresses), and the list of
+ inner groups (which is a bits list) are maintained in separate
+ variables.
+
+ We are making a (strictly speaking) nonportable assumption here: that
+ the compiler will pack our bit fields into something that fits into
+ the type of `word', i.e., is something that fits into one item on the
+ failure stack. */
+typedef union
+{
+ fail_stack_elt_t word;
+ struct
+ {
+ /* This field is one if this group can match the empty string,
+ zero if not. If not yet determined, `MATCH_NULL_UNSET_VALUE'. */
+#define MATCH_NULL_UNSET_VALUE 3
+ unsigned match_null_string_p : 2;
+ unsigned is_active : 1;
+ unsigned matched_something : 1;
+ unsigned ever_matched_something : 1;
+ } bits;
+} register_info_type;
+
+#define REG_MATCH_NULL_STRING_P(R) ((R).bits.match_null_string_p)
+#define IS_ACTIVE(R) ((R).bits.is_active)
+#define MATCHED_SOMETHING(R) ((R).bits.matched_something)
+#define EVER_MATCHED_SOMETHING(R) ((R).bits.ever_matched_something)
+
+
+/* Call this when have matched a real character; it sets `matched' flags
+ for the subexpressions which we are currently inside. Also records
+ that those subexprs have matched. */
+#define SET_REGS_MATCHED() \
+ do \
+ { \
+ unsigned r; \
+ for (r = lowest_active_reg; r <= highest_active_reg; r++) \
+ { \
+ MATCHED_SOMETHING (reg_info[r]) \
+ = EVER_MATCHED_SOMETHING (reg_info[r]) \
+ = 1; \
+ } \
+ } \
+ while (0)
+
+
+/* This converts PTR, a pointer into one of the search strings `string1'
+ and `string2' into an offset from the beginning of that string. */
+#define POINTER_TO_OFFSET(ptr) \
+ (FIRST_STRING_P (ptr) ? (ptr) - string1 : (ptr) - string2 + size1)
+
+/* Registers are set to a sentinel when they haven't yet matched. */
+#define REG_UNSET_VALUE ((char *) -1)
+#define REG_UNSET(e) ((e) == REG_UNSET_VALUE)
+
+
+/* Macros for dealing with the split strings in re_match_2. */
+
+#define MATCHING_IN_FIRST_STRING (dend == end_match_1)
+
+/* Call before fetching a character with *d. This switches over to
+ string2 if necessary. */
+#define PREFETCH() \
+ while (d == dend) \
+ { \
+ /* End of string2 => fail. */ \
+ if (dend == end_match_2) \
+ goto fail; \
+ /* End of string1 => advance to string2. */ \
+ d = string2; \
+ dend = end_match_2; \
+ }
+
+
+/* Test if at very beginning or at very end of the virtual concatenation
+ of `string1' and `string2'. If only one string, it's `string2'. */
+#define AT_STRINGS_BEG(d) ((d) == (size1 ? string1 : string2) || !size2)
+#define AT_STRINGS_END(d) ((d) == end2)
+
+
+/* Test if D points to a character which is word-constituent. We have
+ two special cases to check for: if past the end of string1, look at
+ the first character in string2; and if before the beginning of
+ string2, look at the last character in string1. */
+#define WORDCHAR_P(d) \
+ (SYNTAX ((d) == end1 ? *string2 \
+ : (d) == string2 - 1 ? *(end1 - 1) : *(d)) \
+ == Sword)
+
+/* Test if the character before D and the one at D differ with respect
+ to being word-constituent. */
+#define AT_WORD_BOUNDARY(d) \
+ (AT_STRINGS_BEG (d) || AT_STRINGS_END (d) \
+ || WORDCHAR_P (d - 1) != WORDCHAR_P (d))
+
+
+/* Free everything we malloc. */
+#ifdef REGEX_MALLOC
+#define FREE_VAR(var) if (var) free (var); var = NULL
+#define FREE_VARIABLES() \
+ do { \
+ FREE_VAR (fail_stack.stack); \
+ FREE_VAR (regstart); \
+ FREE_VAR (regend); \
+ FREE_VAR (old_regstart); \
+ FREE_VAR (old_regend); \
+ FREE_VAR (best_regstart); \
+ FREE_VAR (best_regend); \
+ FREE_VAR (reg_info); \
+ FREE_VAR (reg_dummy); \
+ FREE_VAR (reg_info_dummy); \
+ } while (0)
+#else /* not REGEX_MALLOC */
+/* Some MIPS systems (at least) want this to free alloca'd storage. */
+#define FREE_VARIABLES() alloca (0)
+#endif /* not REGEX_MALLOC */
+
+
+/* These values must meet several constraints. They must not be valid
+ register values; since we have a limit of 255 registers (because
+ we use only one byte in the pattern for the register number), we can
+ use numbers larger than 255. They must differ by 1, because of
+ NUM_FAILURE_ITEMS above. And the value for the lowest register must
+ be larger than the value for the highest register, so we do not try
+ to actually save any registers when none are active. */
+#define NO_HIGHEST_ACTIVE_REG (1 << BYTEWIDTH)
+#define NO_LOWEST_ACTIVE_REG (NO_HIGHEST_ACTIVE_REG + 1)
+\f
+/* Matching routines. */
+
+#ifndef emacs /* Emacs never uses this. */
+/* re_match is like re_match_2 except it takes only a single string. */
+
+int
+re_match (bufp, string, size, pos, regs)
+ struct re_pattern_buffer *bufp;
+ const char *string;
+ int size, pos;
+ struct re_registers *regs;
+ {
+ return re_match_2 (bufp, NULL, 0, string, size, pos, regs, size);
+}
+#endif /* not emacs */
+
+
+/* re_match_2 matches the compiled pattern in BUFP against the
+ the (virtual) concatenation of STRING1 and STRING2 (of length SIZE1
+ and SIZE2, respectively). We start matching at POS, and stop
+ matching at STOP.
+
+ If REGS is non-null and the `no_sub' field of BUFP is nonzero, we
+ store offsets for the substring each group matched in REGS. See the
+ documentation for exactly how many groups we fill.
+
+ We return -1 if no match, -2 if an internal error (such as the
+ failure stack overflowing). Otherwise, we return the length of the
+ matched substring. */
+
+int
+re_match_2 (bufp, string1, size1, string2, size2, pos, regs, stop)
+ struct re_pattern_buffer *bufp;
+ const char *string1, *string2;
+ int size1, size2;
+ int pos;
+ struct re_registers *regs;
+ int stop;
+{
+ /* General temporaries. */
+ int mcnt;
+ unsigned char *p1;
+
+ /* Just past the end of the corresponding string. */
+ const char *end1, *end2;
+
+ /* Pointers into string1 and string2, just past the last characters in
+ each to consider matching. */
+ const char *end_match_1, *end_match_2;
+
+ /* Where we are in the data, and the end of the current string. */
+ const char *d, *dend;
+
+ /* Where we are in the pattern, and the end of the pattern. */
+ unsigned char *p = bufp->buffer;
+ register unsigned char *pend = p + bufp->used;
+
+ /* We use this to map every character in the string. */
+ char *translate = bufp->translate;
+
+ /* Failure point stack. Each place that can handle a failure further
+ down the line pushes a failure point on this stack. It consists of
+ restart, regend, and reg_info for all registers corresponding to
+ the subexpressions we're currently inside, plus the number of such
+ registers, and, finally, two char *'s. The first char * is where
+ to resume scanning the pattern; the second one is where to resume
+ scanning the strings. If the latter is zero, the failure point is
+ a ``dummy''; if a failure happens and the failure point is a dummy,
+ it gets discarded and the next next one is tried. */
+ fail_stack_type fail_stack;
+#ifdef DEBUG
+ static unsigned failure_id = 0;
+ unsigned nfailure_points_pushed = 0, nfailure_points_popped = 0;
+#endif
+
+ /* We fill all the registers internally, independent of what we
+ return, for use in backreferences. The number here includes
+ an element for register zero. */
+ unsigned num_regs = bufp->re_nsub + 1;
+
+ /* The currently active registers. */
+ unsigned lowest_active_reg = NO_LOWEST_ACTIVE_REG;
+ unsigned highest_active_reg = NO_HIGHEST_ACTIVE_REG;
+
+ /* Information on the contents of registers. These are pointers into
+ the input strings; they record just what was matched (on this
+ attempt) by a subexpression part of the pattern, that is, the
+ regnum-th regstart pointer points to where in the pattern we began
+ matching and the regnum-th regend points to right after where we
+ stopped matching the regnum-th subexpression. (The zeroth register
+ keeps track of what the whole pattern matches.) */
+ const char **regstart = NULL, **regend = NULL;
+
+ /* If a group that's operated upon by a repetition operator fails to
+ match anything, then the register for its start will need to be
+ restored because it will have been set to wherever in the string we
+ are when we last see its open-group operator. Similarly for a
+ register's end. */
+ const char **old_regstart = NULL, **old_regend = NULL;
+
+ /* The is_active field of reg_info helps us keep track of which (possibly
+ nested) subexpressions we are currently in. The matched_something
+ field of reg_info[reg_num] helps us tell whether or not we have
+ matched any of the pattern so far this time through the reg_num-th
+ subexpression. These two fields get reset each time through any
+ loop their register is in. */
+ register_info_type *reg_info = NULL;
+
+ /* The following record the register info as found in the above
+ variables when we find a match better than any we've seen before.
+ This happens as we backtrack through the failure points, which in
+ turn happens only if we have not yet matched the entire string. */
+ unsigned best_regs_set = false;
+ const char **best_regstart = NULL, **best_regend = NULL;
+
+ /* Logically, this is `best_regend[0]'. But we don't want to have to
+ allocate space for that if we're not allocating space for anything
+ else (see below). Also, we never need info about register 0 for
+ any of the other register vectors, and it seems rather a kludge to
+ treat `best_regend' differently than the rest. So we keep track of
+ the end of the best match so far in a separate variable. We
+ initialize this to NULL so that when we backtrack the first time
+ and need to test it, it's not garbage. */
+ const char *match_end = NULL;
+
+ /* Used when we pop values we don't care about. */
+ const char **reg_dummy = NULL;
+ register_info_type *reg_info_dummy = NULL;
+
+#ifdef DEBUG
+ /* Counts the total number of registers pushed. */
+ unsigned num_regs_pushed = 0;
+#endif
+
+ DEBUG_PRINT1 ("\n\nEntering re_match_2.\n");
+
+ INIT_FAIL_STACK ();
+
+ /* Do not bother to initialize all the register variables if there are
+ no groups in the pattern, as it takes a fair amount of time. If
+ there are groups, we include space for register 0 (the whole
+ pattern), even though we never use it, since it simplifies the
+ array indexing. We should fix this. */
+ if (bufp->re_nsub)
+ {
+ regstart = REGEX_TALLOC (num_regs, const char *);
+ regend = REGEX_TALLOC (num_regs, const char *);
+ old_regstart = REGEX_TALLOC (num_regs, const char *);
+ old_regend = REGEX_TALLOC (num_regs, const char *);
+ best_regstart = REGEX_TALLOC (num_regs, const char *);
+ best_regend = REGEX_TALLOC (num_regs, const char *);
+ reg_info = REGEX_TALLOC (num_regs, register_info_type);
+ reg_dummy = REGEX_TALLOC (num_regs, const char *);
+ reg_info_dummy = REGEX_TALLOC (num_regs, register_info_type);
+
+ if (!(regstart && regend && old_regstart && old_regend && reg_info
+ && best_regstart && best_regend && reg_dummy && reg_info_dummy))
+ {
+ FREE_VARIABLES ();
+ return -2;
+ }
+ }
+#ifdef REGEX_MALLOC
+ else
+ {
+ /* We must initialize all our variables to NULL, so that
+ `FREE_VARIABLES' doesn't try to free them. */
+ regstart = regend = old_regstart = old_regend = best_regstart
+ = best_regend = reg_dummy = NULL;
+ reg_info = reg_info_dummy = (register_info_type *) NULL;
+ }
+#endif /* REGEX_MALLOC */
+
+ /* The starting position is bogus. */
+ if (pos < 0 || pos > size1 + size2)
+ {
+ FREE_VARIABLES ();
+ return -1;
+ }
+
+ /* Initialize subexpression text positions to -1 to mark ones that no
+ start_memory/stop_memory has been seen for. Also initialize the
+ register information struct. */
+ for (mcnt = 1; mcnt < num_regs; mcnt++)
+ {
+ regstart[mcnt] = regend[mcnt]
+ = old_regstart[mcnt] = old_regend[mcnt] = REG_UNSET_VALUE;
+
+ REG_MATCH_NULL_STRING_P (reg_info[mcnt]) = MATCH_NULL_UNSET_VALUE;
+ IS_ACTIVE (reg_info[mcnt]) = 0;
+ MATCHED_SOMETHING (reg_info[mcnt]) = 0;
+ EVER_MATCHED_SOMETHING (reg_info[mcnt]) = 0;
+ }
+
+ /* We move `string1' into `string2' if the latter's empty -- but not if
+ `string1' is null. */
+ if (size2 == 0 && string1 != NULL)
+ {
+ string2 = string1;
+ size2 = size1;
+ string1 = 0;
+ size1 = 0;
+ }
+ end1 = string1 + size1;
+ end2 = string2 + size2;
+
+ /* Compute where to stop matching, within the two strings. */
+ if (stop <= size1)
+ {
+ end_match_1 = string1 + stop;
+ end_match_2 = string2;
+ }
+ else
+ {
+ end_match_1 = end1;
+ end_match_2 = string2 + stop - size1;
+ }
+
+ /* `p' scans through the pattern as `d' scans through the data.
+ `dend' is the end of the input string that `d' points within. `d'
+ is advanced into the following input string whenever necessary, but
+ this happens before fetching; therefore, at the beginning of the
+ loop, `d' can be pointing at the end of a string, but it cannot
+ equal `string2'. */
+ if (size1 > 0 && pos <= size1)
+ {
+ d = string1 + pos;
+ dend = end_match_1;
+ }
+ else
+ {
+ d = string2 + pos - size1;
+ dend = end_match_2;
+ }
+
+ DEBUG_PRINT1 ("The compiled pattern is: ");
+ DEBUG_PRINT_COMPILED_PATTERN (bufp, p, pend);
+ DEBUG_PRINT1 ("The string to match is: `");
+ DEBUG_PRINT_DOUBLE_STRING (d, string1, size1, string2, size2);
+ DEBUG_PRINT1 ("'\n");
+
+ /* This loops over pattern commands. It exits by returning from the
+ function if the match is complete, or it drops through if the match
+ fails at this starting point in the input data. */
+ for (;;)
+ {
+ DEBUG_PRINT2 ("\n0x%x: ", p);
+
+ if (p == pend)
+ { /* End of pattern means we might have succeeded. */
+ DEBUG_PRINT1 ("end of pattern ... ");
+
+ /* If we haven't matched the entire string, and we want the
+ longest match, try backtracking. */
+ if (d != end_match_2)
+ {
+ DEBUG_PRINT1 ("backtracking.\n");
+
+ if (!FAIL_STACK_EMPTY ())
+ { /* More failure points to try. */
+ boolean same_str_p = (FIRST_STRING_P (match_end)
+ == MATCHING_IN_FIRST_STRING);
+
+ /* If exceeds best match so far, save it. */
+ if (!best_regs_set
+ || (same_str_p && d > match_end)
+ || (!same_str_p && !MATCHING_IN_FIRST_STRING))
+ {
+ best_regs_set = true;
+ match_end = d;
+
+ DEBUG_PRINT1 ("\nSAVING match as best so far.\n");
+
+ for (mcnt = 1; mcnt < num_regs; mcnt++)
+ {
+ best_regstart[mcnt] = regstart[mcnt];
+ best_regend[mcnt] = regend[mcnt];
+ }
+ }
+ goto fail;
+ }
+
+ /* If no failure points, don't restore garbage. */
+ else if (best_regs_set)
+ {
+ restore_best_regs:
+ /* Restore best match. It may happen that `dend ==
+ end_match_1' while the restored d is in string2.
+ For example, the pattern `x.*y.*z' against the
+ strings `x-' and `y-z-', if the two strings are
+ not consecutive in memory. */
+ DEBUG_PRINT1 ("Restoring best registers.\n");
+
+ d = match_end;
+ dend = ((d >= string1 && d <= end1)
+ ? end_match_1 : end_match_2);
+
+ for (mcnt = 1; mcnt < num_regs; mcnt++)
+ {
+ regstart[mcnt] = best_regstart[mcnt];
+ regend[mcnt] = best_regend[mcnt];
+ }
+ }
+ } /* d != end_match_2 */
+
+ DEBUG_PRINT1 ("Accepting match.\n");
+
+ /* If caller wants register contents data back, do it. */
+ if (regs && !bufp->no_sub)
+ {
+ /* Have the register data arrays been allocated? */
+ if (bufp->regs_allocated == REGS_UNALLOCATED)
+ { /* No. So allocate them with malloc. We need one
+ extra element beyond `num_regs' for the `-1' marker
+ GNU code uses. */
+ regs->num_regs = MAX (RE_NREGS, num_regs + 1);
+ regs->start = TALLOC (regs->num_regs, regoff_t);
+ regs->end = TALLOC (regs->num_regs, regoff_t);
+ if (regs->start == NULL || regs->end == NULL)
+ return -2;
+ bufp->regs_allocated = REGS_REALLOCATE;
+ }
+ else if (bufp->regs_allocated == REGS_REALLOCATE)
+ { /* Yes. If we need more elements than were already
+ allocated, reallocate them. If we need fewer, just
+ leave it alone. */
+ if (regs->num_regs < num_regs + 1)
+ {
+ regs->num_regs = num_regs + 1;
+ RETALLOC (regs->start, regs->num_regs, regoff_t);
+ RETALLOC (regs->end, regs->num_regs, regoff_t);
+ if (regs->start == NULL || regs->end == NULL)
+ return -2;
+ }
+ }
+ else
+ assert (bufp->regs_allocated == REGS_FIXED);
+
+ /* Convert the pointer data in `regstart' and `regend' to
+ indices. Register zero has to be set differently,
+ since we haven't kept track of any info for it. */
+ if (regs->num_regs > 0)
+ {
+ regs->start[0] = pos;
+ regs->end[0] = (MATCHING_IN_FIRST_STRING ? d - string1
+ : d - string2 + size1);
+ }
+
+ /* Go through the first `min (num_regs, regs->num_regs)'
+ registers, since that is all we initialized. */
+ for (mcnt = 1; mcnt < MIN (num_regs, regs->num_regs); mcnt++)
+ {
+ if (REG_UNSET (regstart[mcnt]) || REG_UNSET (regend[mcnt]))
+ regs->start[mcnt] = regs->end[mcnt] = -1;
+ else
+ {
+ regs->start[mcnt] = POINTER_TO_OFFSET (regstart[mcnt]);
+ regs->end[mcnt] = POINTER_TO_OFFSET (regend[mcnt]);
+ }
+ }
+
+ /* If the regs structure we return has more elements than
+ were in the pattern, set the extra elements to -1. If
+ we (re)allocated the registers, this is the case,
+ because we always allocate enough to have at least one
+ -1 at the end. */
+ for (mcnt = num_regs; mcnt < regs->num_regs; mcnt++)
+ regs->start[mcnt] = regs->end[mcnt] = -1;
+ } /* regs && !bufp->no_sub */
+
+ FREE_VARIABLES ();
+ DEBUG_PRINT4 ("%u failure points pushed, %u popped (%u remain).\n",
+ nfailure_points_pushed, nfailure_points_popped,
+ nfailure_points_pushed - nfailure_points_popped);
+ DEBUG_PRINT2 ("%u registers pushed.\n", num_regs_pushed);
+
+ mcnt = d - pos - (MATCHING_IN_FIRST_STRING
+ ? string1
+ : string2 - size1);
+
+ DEBUG_PRINT2 ("Returning %d from re_match_2.\n", mcnt);
+
+ return mcnt;
+ }
+
+ /* Otherwise match next pattern command. */
+#ifdef SWITCH_ENUM_BUG
+ switch ((int) ((re_opcode_t) *p++))
+#else
+ switch ((re_opcode_t) *p++)
+#endif
+ {
+ /* Ignore these. Used to ignore the n of succeed_n's which
+ currently have n == 0. */
+ case no_op:
+ DEBUG_PRINT1 ("EXECUTING no_op.\n");
+ break;
+
+
+ /* Match the next n pattern characters exactly. The following
+ byte in the pattern defines n, and the n bytes after that
+ are the characters to match. */
+ case exactn:
+ mcnt = *p++;
+ DEBUG_PRINT2 ("EXECUTING exactn %d.\n", mcnt);
+
+ /* This is written out as an if-else so we don't waste time
+ testing `translate' inside the loop. */
+ if (translate)
+ {
+ do
+ {
+ PREFETCH ();
+ if (translate[(unsigned char) *d++] != (char) *p++)
+ goto fail;
+ }
+ while (--mcnt);
+ }
+ else
+ {
+ do
+ {
+ PREFETCH ();
+ if (*d++ != (char) *p++) goto fail;
+ }
+ while (--mcnt);
+ }
+ SET_REGS_MATCHED ();
+ break;
+
+
+ /* Match any character except possibly a newline or a null. */
+ case anychar:
+ DEBUG_PRINT1 ("EXECUTING anychar.\n");
+
+ PREFETCH ();
+
+ if ((!(bufp->syntax & RE_DOT_NEWLINE) && TRANSLATE (*d) == '\n')
+ || (bufp->syntax & RE_DOT_NOT_NULL && TRANSLATE (*d) == '\000'))
+ goto fail;
+
+ SET_REGS_MATCHED ();
+ DEBUG_PRINT2 (" Matched `%d'.\n", *d);
+ d++;
+ break;
+
+
+ case charset:
+ case charset_not:
+ {
+ register unsigned char c;
+ boolean not = (re_opcode_t) *(p - 1) == charset_not;
+
+ DEBUG_PRINT2 ("EXECUTING charset%s.\n", not ? "_not" : "");
+
+ PREFETCH ();
+ c = TRANSLATE (*d); /* The character to match. */
+
+ /* Cast to `unsigned' instead of `unsigned char' in case the
+ bit list is a full 32 bytes long. */
+ if (c < (unsigned) (*p * BYTEWIDTH)
+ && p[1 + c / BYTEWIDTH] & (1 << (c % BYTEWIDTH)))
+ not = !not;
+
+ p += 1 + *p;
+
+ if (!not) goto fail;
+
+ SET_REGS_MATCHED ();
+ d++;
+ break;
+ }
+
+
+ /* The beginning of a group is represented by start_memory.
+ The arguments are the register number in the next byte, and the
+ number of groups inner to this one in the next. The text
+ matched within the group is recorded (in the internal
+ registers data structure) under the register number. */
+ case start_memory:
+ DEBUG_PRINT3 ("EXECUTING start_memory %d (%d):\n", *p, p[1]);
+
+ /* Find out if this group can match the empty string. */
+ p1 = p; /* To send to group_match_null_string_p. */
+
+ if (REG_MATCH_NULL_STRING_P (reg_info[*p]) == MATCH_NULL_UNSET_VALUE)
+ REG_MATCH_NULL_STRING_P (reg_info[*p])
+ = group_match_null_string_p (&p1, pend, reg_info);
+
+ /* Save the position in the string where we were the last time
+ we were at this open-group operator in case the group is
+ operated upon by a repetition operator, e.g., with `(a*)*b'
+ against `ab'; then we want to ignore where we are now in
+ the string in case this attempt to match fails. */
+ old_regstart[*p] = REG_MATCH_NULL_STRING_P (reg_info[*p])
+ ? REG_UNSET (regstart[*p]) ? d : regstart[*p]
+ : regstart[*p];
+ DEBUG_PRINT2 (" old_regstart: %d\n",
+ POINTER_TO_OFFSET (old_regstart[*p]));
+
+ regstart[*p] = d;
+ DEBUG_PRINT2 (" regstart: %d\n", POINTER_TO_OFFSET (regstart[*p]));
+
+ IS_ACTIVE (reg_info[*p]) = 1;
+ MATCHED_SOMETHING (reg_info[*p]) = 0;
+
+ /* This is the new highest active register. */
+ highest_active_reg = *p;
+
+ /* If nothing was active before, this is the new lowest active
+ register. */
+ if (lowest_active_reg == NO_LOWEST_ACTIVE_REG)
+ lowest_active_reg = *p;
+
+ /* Move past the register number and inner group count. */
+ p += 2;
+ break;
+
+
+ /* The stop_memory opcode represents the end of a group. Its
+ arguments are the same as start_memory's: the register
+ number, and the number of inner groups. */
+ case stop_memory:
+ DEBUG_PRINT3 ("EXECUTING stop_memory %d (%d):\n", *p, p[1]);
+
+ /* We need to save the string position the last time we were at
+ this close-group operator in case the group is operated
+ upon by a repetition operator, e.g., with `((a*)*(b*)*)*'
+ against `aba'; then we want to ignore where we are now in
+ the string in case this attempt to match fails. */
+ old_regend[*p] = REG_MATCH_NULL_STRING_P (reg_info[*p])
+ ? REG_UNSET (regend[*p]) ? d : regend[*p]
+ : regend[*p];
+ DEBUG_PRINT2 (" old_regend: %d\n",
+ POINTER_TO_OFFSET (old_regend[*p]));
+
+ regend[*p] = d;
+ DEBUG_PRINT2 (" regend: %d\n", POINTER_TO_OFFSET (regend[*p]));
+
+ /* This register isn't active anymore. */
+ IS_ACTIVE (reg_info[*p]) = 0;
+
+ /* If this was the only register active, nothing is active
+ anymore. */
+ if (lowest_active_reg == highest_active_reg)
+ {
+ lowest_active_reg = NO_LOWEST_ACTIVE_REG;
+ highest_active_reg = NO_HIGHEST_ACTIVE_REG;
+ }
+ else
+ { /* We must scan for the new highest active register, since
+ it isn't necessarily one less than now: consider
+ (a(b)c(d(e)f)g). When group 3 ends, after the f), the
+ new highest active register is 1. */
+ unsigned char r = *p - 1;
+ while (r > 0 && !IS_ACTIVE (reg_info[r]))
+ r--;
+
+ /* If we end up at register zero, that means that we saved
+ the registers as the result of an `on_failure_jump', not
+ a `start_memory', and we jumped to past the innermost
+ `stop_memory'. For example, in ((.)*) we save
+ registers 1 and 2 as a result of the *, but when we pop
+ back to the second ), we are at the stop_memory 1.
+ Thus, nothing is active. */
+ if (r == 0)
+ {
+ lowest_active_reg = NO_LOWEST_ACTIVE_REG;
+ highest_active_reg = NO_HIGHEST_ACTIVE_REG;
+ }
+ else
+ highest_active_reg = r;
+ }
+
+ /* If just failed to match something this time around with a
+ group that's operated on by a repetition operator, try to
+ force exit from the ``loop'', and restore the register
+ information for this group that we had before trying this
+ last match. */
+ if ((!MATCHED_SOMETHING (reg_info[*p])
+ || (re_opcode_t) p[-3] == start_memory)
+ && (p + 2) < pend)
+ {
+ boolean is_a_jump_n = false;
+
+ p1 = p + 2;
+ mcnt = 0;
+ switch ((re_opcode_t) *p1++)
+ {
+ case jump_n:
+ is_a_jump_n = true;
+ case pop_failure_jump:
+ case maybe_pop_jump:
+ case jump:
+ case dummy_failure_jump:
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ if (is_a_jump_n)
+ p1 += 2;
+ break;
+
+ default:
+ /* do nothing */ ;
+ }
+ p1 += mcnt;
+
+ /* If the next operation is a jump backwards in the pattern
+ to an on_failure_jump right before the start_memory
+ corresponding to this stop_memory, exit from the loop
+ by forcing a failure after pushing on the stack the
+ on_failure_jump's jump in the pattern, and d. */
+ if (mcnt < 0 && (re_opcode_t) *p1 == on_failure_jump
+ && (re_opcode_t) p1[3] == start_memory && p1[4] == *p)
+ {
+ /* If this group ever matched anything, then restore
+ what its registers were before trying this last
+ failed match, e.g., with `(a*)*b' against `ab' for
+ regstart[1], and, e.g., with `((a*)*(b*)*)*'
+ against `aba' for regend[3].
+
+ Also restore the registers for inner groups for,
+ e.g., `((a*)(b*))*' against `aba' (register 3 would
+ otherwise get trashed). */
+
+ if (EVER_MATCHED_SOMETHING (reg_info[*p]))
+ {
+ unsigned r;
+
+ EVER_MATCHED_SOMETHING (reg_info[*p]) = 0;
+
+ /* Restore this and inner groups' (if any) registers. */
+ for (r = *p; r < *p + *(p + 1); r++)
+ {
+ regstart[r] = old_regstart[r];
+
+ /* xx why this test? */
+ if ((int) old_regend[r] >= (int) regstart[r])
+ regend[r] = old_regend[r];
+ }
+ }
+ p1++;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ PUSH_FAILURE_POINT (p1 + mcnt, d, -2);
+
+ goto fail;
+ }
+ }
+
+ /* Move past the register number and the inner group count. */
+ p += 2;
+ break;
+
+
+ /* \<digit> has been turned into a `duplicate' command which is
+ followed by the numeric value of <digit> as the register number. */
+ case duplicate:
+ {
+ register const char *d2, *dend2;
+ int regno = *p++; /* Get which register to match against. */
+ DEBUG_PRINT2 ("EXECUTING duplicate %d.\n", regno);
+
+ /* Can't back reference a group which we've never matched. */
+ if (REG_UNSET (regstart[regno]) || REG_UNSET (regend[regno]))
+ goto fail;
+
+ /* Where in input to try to start matching. */
+ d2 = regstart[regno];
+
+ /* Where to stop matching; if both the place to start and
+ the place to stop matching are in the same string, then
+ set to the place to stop, otherwise, for now have to use
+ the end of the first string. */
+
+ dend2 = ((FIRST_STRING_P (regstart[regno])
+ == FIRST_STRING_P (regend[regno]))
+ ? regend[regno] : end_match_1);
+ for (;;)
+ {
+ /* If necessary, advance to next segment in register
+ contents. */
+ while (d2 == dend2)
+ {
+ if (dend2 == end_match_2) break;
+ if (dend2 == regend[regno]) break;
+
+ /* End of string1 => advance to string2. */
+ d2 = string2;
+ dend2 = regend[regno];
+ }
+ /* At end of register contents => success */
+ if (d2 == dend2) break;
+
+ /* If necessary, advance to next segment in data. */
+ PREFETCH ();
+
+ /* How many characters left in this segment to match. */
+ mcnt = dend - d;
+
+ /* Want how many consecutive characters we can match in
+ one shot, so, if necessary, adjust the count. */
+ if (mcnt > dend2 - d2)
+ mcnt = dend2 - d2;
+
+ /* Compare that many; failure if mismatch, else move
+ past them. */
+ if (translate
+ ? bcmp_translate (d, d2, mcnt, translate)
+ : bcmp (d, d2, mcnt))
+ goto fail;
+ d += mcnt, d2 += mcnt;
+ }
+ }
+ break;
+
+
+ /* begline matches the empty string at the beginning of the string
+ (unless `not_bol' is set in `bufp'), and, if
+ `newline_anchor' is set, after newlines. */
+ case begline:
+ DEBUG_PRINT1 ("EXECUTING begline.\n");
+
+ if (AT_STRINGS_BEG (d))
+ {
+ if (!bufp->not_bol) break;
+ }
+ else if (d[-1] == '\n' && bufp->newline_anchor)
+ {
+ break;
+ }
+ /* In all other cases, we fail. */
+ goto fail;
+
+
+ /* endline is the dual of begline. */
+ case endline:
+ DEBUG_PRINT1 ("EXECUTING endline.\n");
+
+ if (AT_STRINGS_END (d))
+ {
+ if (!bufp->not_eol) break;
+ }
+
+ /* We have to ``prefetch'' the next character. */
+ else if ((d == end1 ? *string2 : *d) == '\n'
+ && bufp->newline_anchor)
+ {
+ break;
+ }
+ goto fail;
+
+
+ /* Match at the very beginning of the data. */
+ case begbuf:
+ DEBUG_PRINT1 ("EXECUTING begbuf.\n");
+ if (AT_STRINGS_BEG (d))
+ break;
+ goto fail;
+
+
+ /* Match at the very end of the data. */
+ case endbuf:
+ DEBUG_PRINT1 ("EXECUTING endbuf.\n");
+ if (AT_STRINGS_END (d))
+ break;
+ goto fail;
+
+
+ /* on_failure_keep_string_jump is used to optimize `.*\n'. It
+ pushes NULL as the value for the string on the stack. Then
+ `pop_failure_point' will keep the current value for the
+ string, instead of restoring it. To see why, consider
+ matching `foo\nbar' against `.*\n'. The .* matches the foo;
+ then the . fails against the \n. But the next thing we want
+ to do is match the \n against the \n; if we restored the
+ string value, we would be back at the foo.
+
+ Because this is used only in specific cases, we don't need to
+ check all the things that `on_failure_jump' does, to make
+ sure the right things get saved on the stack. Hence we don't
+ share its code. The only reason to push anything on the
+ stack at all is that otherwise we would have to change
+ `anychar's code to do something besides goto fail in this
+ case; that seems worse than this. */
+ case on_failure_keep_string_jump:
+ DEBUG_PRINT1 ("EXECUTING on_failure_keep_string_jump");
+
+ EXTRACT_NUMBER_AND_INCR (mcnt, p);
+ DEBUG_PRINT3 (" %d (to 0x%x):\n", mcnt, p + mcnt);
+
+ PUSH_FAILURE_POINT (p + mcnt, NULL, -2);
+ break;
+
+
+ /* Uses of on_failure_jump:
+
+ Each alternative starts with an on_failure_jump that points
+ to the beginning of the next alternative. Each alternative
+ except the last ends with a jump that in effect jumps past
+ the rest of the alternatives. (They really jump to the
+ ending jump of the following alternative, because tensioning
+ these jumps is a hassle.)
+
+ Repeats start with an on_failure_jump that points past both
+ the repetition text and either the following jump or
+ pop_failure_jump back to this on_failure_jump. */
+ case on_failure_jump:
+ on_failure:
+ DEBUG_PRINT1 ("EXECUTING on_failure_jump");
+
+ EXTRACT_NUMBER_AND_INCR (mcnt, p);
+ DEBUG_PRINT3 (" %d (to 0x%x)", mcnt, p + mcnt);
+
+ /* If this on_failure_jump comes right before a group (i.e.,
+ the original * applied to a group), save the information
+ for that group and all inner ones, so that if we fail back
+ to this point, the group's information will be correct.
+ For example, in \(a*\)*\1, we need the preceding group,
+ and in \(\(a*\)b*\)\2, we need the inner group. */
+
+ /* We can't use `p' to check ahead because we push
+ a failure point to `p + mcnt' after we do this. */
+ p1 = p;
+
+ /* We need to skip no_op's before we look for the
+ start_memory in case this on_failure_jump is happening as
+ the result of a completed succeed_n, as in \(a\)\{1,3\}b\1
+ against aba. */
+ while (p1 < pend && (re_opcode_t) *p1 == no_op)
+ p1++;
+
+ if (p1 < pend && (re_opcode_t) *p1 == start_memory)
+ {
+ /* We have a new highest active register now. This will
+ get reset at the start_memory we are about to get to,
+ but we will have saved all the registers relevant to
+ this repetition op, as described above. */
+ highest_active_reg = *(p1 + 1) + *(p1 + 2);
+ if (lowest_active_reg == NO_LOWEST_ACTIVE_REG)
+ lowest_active_reg = *(p1 + 1);
+ }
+
+ DEBUG_PRINT1 (":\n");
+ PUSH_FAILURE_POINT (p + mcnt, d, -2);
+ break;
+
+
+ /* A smart repeat ends with `maybe_pop_jump'.
+ We change it to either `pop_failure_jump' or `jump'. */
+ case maybe_pop_jump:
+ EXTRACT_NUMBER_AND_INCR (mcnt, p);
+ DEBUG_PRINT2 ("EXECUTING maybe_pop_jump %d.\n", mcnt);
+ {
+ register unsigned char *p2 = p;
+
+ /* Compare the beginning of the repeat with what in the
+ pattern follows its end. If we can establish that there
+ is nothing that they would both match, i.e., that we
+ would have to backtrack because of (as in, e.g., `a*a')
+ then we can change to pop_failure_jump, because we'll
+ never have to backtrack.
+
+ This is not true in the case of alternatives: in
+ `(a|ab)*' we do need to backtrack to the `ab' alternative
+ (e.g., if the string was `ab'). But instead of trying to
+ detect that here, the alternative has put on a dummy
+ failure point which is what we will end up popping. */
+
+ /* Skip over open/close-group commands. */
+ while (p2 + 2 < pend
+ && ((re_opcode_t) *p2 == stop_memory
+ || (re_opcode_t) *p2 == start_memory))
+ p2 += 3; /* Skip over args, too. */
+
+ /* If we're at the end of the pattern, we can change. */
+ if (p2 == pend)
+ {
+ /* Consider what happens when matching ":\(.*\)"
+ against ":/". I don't really understand this code
+ yet. */
+ p[-3] = (unsigned char) pop_failure_jump;
+ DEBUG_PRINT1
+ (" End of pattern: change to `pop_failure_jump'.\n");
+ }
+
+ else if ((re_opcode_t) *p2 == exactn
+ || (bufp->newline_anchor && (re_opcode_t) *p2 == endline))
+ {
+ register unsigned char c
+ = *p2 == (unsigned char) endline ? '\n' : p2[2];
+ p1 = p + mcnt;
+
+ /* p1[0] ... p1[2] are the `on_failure_jump' corresponding
+ to the `maybe_finalize_jump' of this case. Examine what
+ follows. */
+ if ((re_opcode_t) p1[3] == exactn && p1[5] != c)
+ {
+ p[-3] = (unsigned char) pop_failure_jump;
+ DEBUG_PRINT3 (" %c != %c => pop_failure_jump.\n",
+ c, p1[5]);
+ }
+
+ else if ((re_opcode_t) p1[3] == charset
+ || (re_opcode_t) p1[3] == charset_not)
+ {
+ int not = (re_opcode_t) p1[3] == charset_not;
+
+ if (c < (unsigned char) (p1[4] * BYTEWIDTH)
+ && p1[5 + c / BYTEWIDTH] & (1 << (c % BYTEWIDTH)))
+ not = !not;
+
+ /* `not' is equal to 1 if c would match, which means
+ that we can't change to pop_failure_jump. */
+ if (!not)
+ {
+ p[-3] = (unsigned char) pop_failure_jump;
+ DEBUG_PRINT1 (" No match => pop_failure_jump.\n");
+ }
+ }
+ }
+ }
+ p -= 2; /* Point at relative address again. */
+ if ((re_opcode_t) p[-1] != pop_failure_jump)
+ {
+ p[-1] = (unsigned char) jump;
+ DEBUG_PRINT1 (" Match => jump.\n");
+ goto unconditional_jump;
+ }
+ /* Note fall through. */
+
+
+ /* The end of a simple repeat has a pop_failure_jump back to
+ its matching on_failure_jump, where the latter will push a
+ failure point. The pop_failure_jump takes off failure
+ points put on by this pop_failure_jump's matching
+ on_failure_jump; we got through the pattern to here from the
+ matching on_failure_jump, so didn't fail. */
+ case pop_failure_jump:
+ {
+ /* We need to pass separate storage for the lowest and
+ highest registers, even though we don't care about the
+ actual values. Otherwise, we will restore only one
+ register from the stack, since lowest will == highest in
+ `pop_failure_point'. */
+ unsigned dummy_low_reg, dummy_high_reg;
+ unsigned char *pdummy;
+ const char *sdummy;
+
+ DEBUG_PRINT1 ("EXECUTING pop_failure_jump.\n");
+ POP_FAILURE_POINT (sdummy, pdummy,
+ dummy_low_reg, dummy_high_reg,
+ reg_dummy, reg_dummy, reg_info_dummy);
+ }
+ /* Note fall through. */
+
+
+ /* Unconditionally jump (without popping any failure points). */
+ case jump:
+ unconditional_jump:
+ EXTRACT_NUMBER_AND_INCR (mcnt, p); /* Get the amount to jump. */
+ DEBUG_PRINT2 ("EXECUTING jump %d ", mcnt);
+ p += mcnt; /* Do the jump. */
+ DEBUG_PRINT2 ("(to 0x%x).\n", p);
+ break;
+
+
+ /* We need this opcode so we can detect where alternatives end
+ in `group_match_null_string_p' et al. */
+ case jump_past_alt:
+ DEBUG_PRINT1 ("EXECUTING jump_past_alt.\n");
+ goto unconditional_jump;
+
+
+ /* Normally, the on_failure_jump pushes a failure point, which
+ then gets popped at pop_failure_jump. We will end up at
+ pop_failure_jump, also, and with a pattern of, say, `a+', we
+ are skipping over the on_failure_jump, so we have to push
+ something meaningless for pop_failure_jump to pop. */
+ case dummy_failure_jump:
+ DEBUG_PRINT1 ("EXECUTING dummy_failure_jump.\n");
+ /* It doesn't matter what we push for the string here. What
+ the code at `fail' tests is the value for the pattern. */
+ PUSH_FAILURE_POINT (0, 0, -2);
+ goto unconditional_jump;
+
+
+ /* At the end of an alternative, we need to push a dummy failure
+ point in case we are followed by a `pop_failure_jump', because
+ we don't want the failure point for the alternative to be
+ popped. For example, matching `(a|ab)*' against `aab'
+ requires that we match the `ab' alternative. */
+ case push_dummy_failure:
+ DEBUG_PRINT1 ("EXECUTING push_dummy_failure.\n");
+ /* See comments just above at `dummy_failure_jump' about the
+ two zeroes. */
+ PUSH_FAILURE_POINT (0, 0, -2);
+ break;
+
+ /* Have to succeed matching what follows at least n times.
+ After that, handle like `on_failure_jump'. */
+ case succeed_n:
+ EXTRACT_NUMBER (mcnt, p + 2);
+ DEBUG_PRINT2 ("EXECUTING succeed_n %d.\n", mcnt);
+
+ assert (mcnt >= 0);
+ /* Originally, this is how many times we HAVE to succeed. */
+ if (mcnt > 0)
+ {
+ mcnt--;
+ p += 2;
+ STORE_NUMBER_AND_INCR (p, mcnt);
+ DEBUG_PRINT3 (" Setting 0x%x to %d.\n", p, mcnt);
+ }
+ else if (mcnt == 0)
+ {
+ DEBUG_PRINT2 (" Setting two bytes from 0x%x to no_op.\n", p+2);
+ p[2] = (unsigned char) no_op;
+ p[3] = (unsigned char) no_op;
+ goto on_failure;
+ }
+ break;
+
+ case jump_n:
+ EXTRACT_NUMBER (mcnt, p + 2);
+ DEBUG_PRINT2 ("EXECUTING jump_n %d.\n", mcnt);
+
+ /* Originally, this is how many times we CAN jump. */
+ if (mcnt)
+ {
+ mcnt--;
+ STORE_NUMBER (p + 2, mcnt);
+ goto unconditional_jump;
+ }
+ /* If don't have to jump any more, skip over the rest of command. */
+ else
+ p += 4;
+ break;
+
+ case set_number_at:
+ {
+ DEBUG_PRINT1 ("EXECUTING set_number_at.\n");
+
+ EXTRACT_NUMBER_AND_INCR (mcnt, p);
+ p1 = p + mcnt;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p);
+ DEBUG_PRINT3 (" Setting 0x%x to %d.\n", p1, mcnt);
+ STORE_NUMBER (p1, mcnt);
+ break;
+ }
+
+ case wordbound:
+ DEBUG_PRINT1 ("EXECUTING wordbound.\n");
+ if (AT_WORD_BOUNDARY (d))
+ break;
+ goto fail;
+
+ case notwordbound:
+ DEBUG_PRINT1 ("EXECUTING notwordbound.\n");
+ if (AT_WORD_BOUNDARY (d))
+ goto fail;
+ break;
+
+ case wordbeg:
+ DEBUG_PRINT1 ("EXECUTING wordbeg.\n");
+ if (WORDCHAR_P (d) && (AT_STRINGS_BEG (d) || !WORDCHAR_P (d - 1)))
+ break;
+ goto fail;
+
+ case wordend:
+ DEBUG_PRINT1 ("EXECUTING wordend.\n");
+ if (!AT_STRINGS_BEG (d) && WORDCHAR_P (d - 1)
+ && (!WORDCHAR_P (d) || AT_STRINGS_END (d)))
+ break;
+ goto fail;
+
+#ifdef emacs
+#ifdef emacs19
+ case before_dot:
+ DEBUG_PRINT1 ("EXECUTING before_dot.\n");
+ if (PTR_CHAR_POS ((unsigned char *) d) >= point)
+ goto fail;
+ break;
+
+ case at_dot:
+ DEBUG_PRINT1 ("EXECUTING at_dot.\n");
+ if (PTR_CHAR_POS ((unsigned char *) d) != point)
+ goto fail;
+ break;
+
+ case after_dot:
+ DEBUG_PRINT1 ("EXECUTING after_dot.\n");
+ if (PTR_CHAR_POS ((unsigned char *) d) <= point)
+ goto fail;
+ break;
+#else /* not emacs19 */
+ case at_dot:
+ DEBUG_PRINT1 ("EXECUTING at_dot.\n");
+ if (PTR_CHAR_POS ((unsigned char *) d) + 1 != point)
+ goto fail;
+ break;
+#endif /* not emacs19 */
+
+ case syntaxspec:
+ DEBUG_PRINT2 ("EXECUTING syntaxspec %d.\n", mcnt);
+ mcnt = *p++;
+ goto matchsyntax;
+
+ case wordchar:
+ DEBUG_PRINT1 ("EXECUTING Emacs wordchar.\n");
+ mcnt = (int) Sword;
+ matchsyntax:
+ PREFETCH ();
+ if (SYNTAX (*d++) != (enum syntaxcode) mcnt)
+ goto fail;
+ SET_REGS_MATCHED ();
+ break;
+
+ case notsyntaxspec:
+ DEBUG_PRINT2 ("EXECUTING notsyntaxspec %d.\n", mcnt);
+ mcnt = *p++;
+ goto matchnotsyntax;
+
+ case notwordchar:
+ DEBUG_PRINT1 ("EXECUTING Emacs notwordchar.\n");
+ mcnt = (int) Sword;
+ matchnotsyntax:
+ PREFETCH ();
+ if (SYNTAX (*d++) == (enum syntaxcode) mcnt)
+ goto fail;
+ SET_REGS_MATCHED ();
+ break;
+
+#else /* not emacs */
+ case wordchar:
+ DEBUG_PRINT1 ("EXECUTING non-Emacs wordchar.\n");
+ PREFETCH ();
+ if (!WORDCHAR_P (d))
+ goto fail;
+ SET_REGS_MATCHED ();
+ d++;
+ break;
+
+ case notwordchar:
+ DEBUG_PRINT1 ("EXECUTING non-Emacs notwordchar.\n");
+ PREFETCH ();
+ if (WORDCHAR_P (d))
+ goto fail;
+ SET_REGS_MATCHED ();
+ d++;
+ break;
+#endif /* not emacs */
+
+ default:
+ abort ();
+ }
+ continue; /* Successfully executed one pattern command; keep going. */
+
+
+ /* We goto here if a matching operation fails. */
+ fail:
+ if (!FAIL_STACK_EMPTY ())
+ { /* A restart point is known. Restore to that state. */
+ DEBUG_PRINT1 ("\nFAIL:\n");
+ POP_FAILURE_POINT (d, p,
+ lowest_active_reg, highest_active_reg,
+ regstart, regend, reg_info);
+
+ /* If this failure point is a dummy, try the next one. */
+ if (!p)
+ goto fail;
+
+ /* If we failed to the end of the pattern, don't examine *p. */
+ assert (p <= pend);
+ if (p < pend)
+ {
+ boolean is_a_jump_n = false;
+
+ /* If failed to a backwards jump that's part of a repetition
+ loop, need to pop this failure point and use the next one. */
+ switch ((re_opcode_t) *p)
+ {
+ case jump_n:
+ is_a_jump_n = true;
+ case maybe_pop_jump:
+ case pop_failure_jump:
+ case jump:
+ p1 = p + 1;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ p1 += mcnt;
+
+ if ((is_a_jump_n && (re_opcode_t) *p1 == succeed_n)
+ || (!is_a_jump_n
+ && (re_opcode_t) *p1 == on_failure_jump))
+ goto fail;
+ break;
+ default:
+ /* do nothing */ ;
+ }
+ }
+
+ if (d >= string1 && d <= end1)
+ dend = end_match_1;
+ }
+ else
+ break; /* Matching at this starting point really fails. */
+ } /* for (;;) */
+
+ if (best_regs_set)
+ goto restore_best_regs;
+
+ FREE_VARIABLES ();
+
+ return -1; /* Failure to match. */
+} /* re_match_2 */
+\f
+/* Subroutine definitions for re_match_2. */
+
+
+/* We are passed P pointing to a register number after a start_memory.
+
+ Return true if the pattern up to the corresponding stop_memory can
+ match the empty string, and false otherwise.
+
+ If we find the matching stop_memory, sets P to point to one past its number.
+ Otherwise, sets P to an undefined byte less than or equal to END.
+
+ We don't handle duplicates properly (yet). */
+
+static boolean
+group_match_null_string_p (p, end, reg_info)
+ unsigned char **p, *end;
+ register_info_type *reg_info;
+{
+ int mcnt;
+ /* Point to after the args to the start_memory. */
+ unsigned char *p1 = *p + 2;
+
+ while (p1 < end)
+ {
+ /* Skip over opcodes that can match nothing, and return true or
+ false, as appropriate, when we get to one that can't, or to the
+ matching stop_memory. */
+
+ switch ((re_opcode_t) *p1)
+ {
+ /* Could be either a loop or a series of alternatives. */
+ case on_failure_jump:
+ p1++;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+
+ /* If the next operation is not a jump backwards in the
+ pattern. */
+
+ if (mcnt >= 0)
+ {
+ /* Go through the on_failure_jumps of the alternatives,
+ seeing if any of the alternatives cannot match nothing.
+ The last alternative starts with only a jump,
+ whereas the rest start with on_failure_jump and end
+ with a jump, e.g., here is the pattern for `a|b|c':
+
+ /on_failure_jump/0/6/exactn/1/a/jump_past_alt/0/6
+ /on_failure_jump/0/6/exactn/1/b/jump_past_alt/0/3
+ /exactn/1/c
+
+ So, we have to first go through the first (n-1)
+ alternatives and then deal with the last one separately. */
+
+
+ /* Deal with the first (n-1) alternatives, which start
+ with an on_failure_jump (see above) that jumps to right
+ past a jump_past_alt. */
+
+ while ((re_opcode_t) p1[mcnt-3] == jump_past_alt)
+ {
+ /* `mcnt' holds how many bytes long the alternative
+ is, including the ending `jump_past_alt' and
+ its number. */
+
+ if (!alt_match_null_string_p (p1, p1 + mcnt - 3,
+ reg_info))
+ return false;
+
+ /* Move to right after this alternative, including the
+ jump_past_alt. */
+ p1 += mcnt;
+
+ /* Break if it's the beginning of an n-th alternative
+ that doesn't begin with an on_failure_jump. */
+ if ((re_opcode_t) *p1 != on_failure_jump)
+ break;
+
+ /* Still have to check that it's not an n-th
+ alternative that starts with an on_failure_jump. */
+ p1++;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ if ((re_opcode_t) p1[mcnt-3] != jump_past_alt)
+ {
+ /* Get to the beginning of the n-th alternative. */
+ p1 -= 3;
+ break;
+ }
+ }
+
+ /* Deal with the last alternative: go back and get number
+ of the `jump_past_alt' just before it. `mcnt' contains
+ the length of the alternative. */
+ EXTRACT_NUMBER (mcnt, p1 - 2);
+
+ if (!alt_match_null_string_p (p1, p1 + mcnt, reg_info))
+ return false;
+
+ p1 += mcnt; /* Get past the n-th alternative. */
+ } /* if mcnt > 0 */
+ break;
+
+
+ case stop_memory:
+ assert (p1[1] == **p);
+ *p = p1 + 2;
+ return true;
+
+
+ default:
+ if (!common_op_match_null_string_p (&p1, end, reg_info))
+ return false;
+ }
+ } /* while p1 < end */
+
+ return false;
+} /* group_match_null_string_p */
+
+
+/* Similar to group_match_null_string_p, but doesn't deal with alternatives:
+ It expects P to be the first byte of a single alternative and END one
+ byte past the last. The alternative can contain groups. */
+
+static boolean
+alt_match_null_string_p (p, end, reg_info)
+ unsigned char *p, *end;
+ register_info_type *reg_info;
+{
+ int mcnt;
+ unsigned char *p1 = p;
+
+ while (p1 < end)
+ {
+ /* Skip over opcodes that can match nothing, and break when we get
+ to one that can't. */
+
+ switch ((re_opcode_t) *p1)
+ {
+ /* It's a loop. */
+ case on_failure_jump:
+ p1++;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ p1 += mcnt;
+ break;
+
+ default:
+ if (!common_op_match_null_string_p (&p1, end, reg_info))
+ return false;
+ }
+ } /* while p1 < end */
+
+ return true;
+} /* alt_match_null_string_p */
+
+
+/* Deals with the ops common to group_match_null_string_p and
+ alt_match_null_string_p.
+
+ Sets P to one after the op and its arguments, if any. */
+
+static boolean
+common_op_match_null_string_p (p, end, reg_info)
+ unsigned char **p, *end;
+ register_info_type *reg_info;
+{
+ int mcnt;
+ boolean ret;
+ int reg_no;
+ unsigned char *p1 = *p;
+
+ switch ((re_opcode_t) *p1++)
+ {
+ case no_op:
+ case begline:
+ case endline:
+ case begbuf:
+ case endbuf:
+ case wordbeg:
+ case wordend:
+ case wordbound:
+ case notwordbound:
+#ifdef emacs
+ case before_dot:
+ case at_dot:
+ case after_dot:
+#endif
+ break;
+
+ case start_memory:
+ reg_no = *p1;
+ assert (reg_no > 0 && reg_no <= MAX_REGNUM);
+ ret = group_match_null_string_p (&p1, end, reg_info);
+
+ /* Have to set this here in case we're checking a group which
+ contains a group and a back reference to it. */
+
+ if (REG_MATCH_NULL_STRING_P (reg_info[reg_no]) == MATCH_NULL_UNSET_VALUE)
+ REG_MATCH_NULL_STRING_P (reg_info[reg_no]) = ret;
+
+ if (!ret)
+ return false;
+ break;
+
+ /* If this is an optimized succeed_n for zero times, make the jump. */
+ case jump:
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ if (mcnt >= 0)
+ p1 += mcnt;
+ else
+ return false;
+ break;
+
+ case succeed_n:
+ /* Get to the number of times to succeed. */
+ p1 += 2;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+
+ if (mcnt == 0)
+ {
+ p1 -= 4;
+ EXTRACT_NUMBER_AND_INCR (mcnt, p1);
+ p1 += mcnt;
+ }
+ else
+ return false;
+ break;
+
+ case duplicate:
+ if (!REG_MATCH_NULL_STRING_P (reg_info[*p1]))
+ return false;
+ break;
+
+ case set_number_at:
+ p1 += 4;
+
+ default:
+ /* All other opcodes mean we cannot match the empty string. */
+ return false;
+ }
+
+ *p = p1;
+ return true;
+} /* common_op_match_null_string_p */
+
+
+/* Return zero if TRANSLATE[S1] and TRANSLATE[S2] are identical for LEN
+ bytes; nonzero otherwise. */
+
+static int
+bcmp_translate(
+ unsigned char *s1,
+ unsigned char *s2,
+ int len,
+ char *translate
+)
+{
+ register unsigned char *p1 = s1, *p2 = s2;
+ while (len)
+ {
+ if (translate[*p1++] != translate[*p2++]) return 1;
+ len--;
+ }
+ return 0;
+}
+\f
+/* Entry points for GNU code. */
+
+/* re_compile_pattern is the GNU regular expression compiler: it
+ compiles PATTERN (of length SIZE) and puts the result in BUFP.
+ Returns 0 if the pattern was valid, otherwise an error string.
+
+ Assumes the `allocated' (and perhaps `buffer') and `translate' fields
+ are set in BUFP on entry.
+
+ We call regex_compile to do the actual compilation. */
+
+const char *
+re_compile_pattern (pattern, length, bufp)
+ const char *pattern;
+ int length;
+ struct re_pattern_buffer *bufp;
+{
+ reg_errcode_t ret;
+
+ /* GNU code is written to assume at least RE_NREGS registers will be set
+ (and at least one extra will be -1). */
+ bufp->regs_allocated = REGS_UNALLOCATED;
+
+ /* And GNU code determines whether or not to get register information
+ by passing null for the REGS argument to re_match, etc., not by
+ setting no_sub. */
+ bufp->no_sub = 0;
+
+ /* Match anchors at newline. */
+ bufp->newline_anchor = 1;
+
+ ret = regex_compile (pattern, length, re_syntax_options, bufp);
+
+ return re_error_msg[(int) ret];
+}
+\f
+/* Entry points compatible with 4.2 BSD regex library. We don't define
+ them if this is an Emacs or POSIX compilation. */
+
+#if !defined (emacs) && !defined (_POSIX_SOURCE)
+
+/* BSD has one and only one pattern buffer. */
+static struct re_pattern_buffer re_comp_buf;
+
+char *
+re_comp (s)
+ const char *s;
+{
+ reg_errcode_t ret;
+
+ if (!s)
+ {
+ if (!re_comp_buf.buffer)
+ return "No previous regular expression";
+ return 0;
+ }
+
+ if (!re_comp_buf.buffer)
+ {
+ re_comp_buf.buffer = (unsigned char *) malloc (200);
+ if (re_comp_buf.buffer == NULL)
+ return "Memory exhausted";
+ re_comp_buf.allocated = 200;
+
+ re_comp_buf.fastmap = (char *) malloc (1 << BYTEWIDTH);
+ if (re_comp_buf.fastmap == NULL)
+ return "Memory exhausted";
+ }
+
+ /* Since `re_exec' always passes NULL for the `regs' argument, we
+ don't need to initialize the pattern buffer fields which affect it. */
+
+ /* Match anchors at newlines. */
+ re_comp_buf.newline_anchor = 1;
+
+ ret = regex_compile (s, strlen (s), re_syntax_options, &re_comp_buf);
+
+ /* Yes, we're discarding `const' here. */
+ return (char *) re_error_msg[(int) ret];
+}
+
+
+int
+re_exec (s)
+ const char *s;
+{
+ const int len = strlen (s);
+ return
+ 0 <= re_search (&re_comp_buf, s, len, 0, len, (struct re_registers *) 0);
+}
+#endif /* not emacs and not _POSIX_SOURCE */
+\f
+/* POSIX.2 functions. Don't define these for Emacs. */
+
+#ifndef emacs
+
+/* regcomp takes a regular expression as a string and compiles it.
+
+ PREG is a regex_t *. We do not expect any fields to be initialized,
+ since POSIX says we shouldn't. Thus, we set
+
+ `buffer' to the compiled pattern;
+ `used' to the length of the compiled pattern;
+ `syntax' to RE_SYNTAX_POSIX_EXTENDED if the
+ REG_EXTENDED bit in CFLAGS is set; otherwise, to
+ RE_SYNTAX_POSIX_BASIC;
+ `newline_anchor' to REG_NEWLINE being set in CFLAGS;
+ `fastmap' and `fastmap_accurate' to zero;
+ `re_nsub' to the number of subexpressions in PATTERN.
+
+ PATTERN is the address of the pattern string.
+
+ CFLAGS is a series of bits which affect compilation.
+
+ If REG_EXTENDED is set, we use POSIX extended syntax; otherwise, we
+ use POSIX basic syntax.
+
+ If REG_NEWLINE is set, then . and [^...] don't match newline.
+ Also, regexec will try a match beginning after every newline.
+
+ If REG_ICASE is set, then we considers upper- and lowercase
+ versions of letters to be equivalent when matching.
+
+ If REG_NOSUB is set, then when PREG is passed to regexec, that
+ routine will report only success or failure, and nothing about the
+ registers.
+
+ It returns 0 if it succeeds, nonzero if it doesn't. (See regex.h for
+ the return codes and their meanings.) */
+
+int
+regcomp (preg, pattern, cflags)
+ regex_t *preg;
+ const char *pattern;
+ int cflags;
+{
+ reg_errcode_t ret;
+ unsigned syntax
+ = (cflags & REG_EXTENDED) ?
+ RE_SYNTAX_POSIX_EXTENDED : RE_SYNTAX_POSIX_BASIC;
+
+ /* regex_compile will allocate the space for the compiled pattern. */
+ preg->buffer = 0;
+ preg->allocated = 0;
+
+ /* Don't bother to use a fastmap when searching. This simplifies the
+ REG_NEWLINE case: if we used a fastmap, we'd have to put all the
+ characters after newlines into the fastmap. This way, we just try
+ every character. */
+ preg->fastmap = 0;
+
+ if (cflags & REG_ICASE)
+ {
+ unsigned i;
+
+ preg->translate = (char *) malloc (CHAR_SET_SIZE);
+ if (preg->translate == NULL)
+ return (int) REG_ESPACE;
+
+ /* Map uppercase characters to corresponding lowercase ones. */
+ for (i = 0; i < CHAR_SET_SIZE; i++)
+ preg->translate[i] = ISUPPER (i) ? tolower (i) : i;
+ }
+ else
+ preg->translate = NULL;
+
+ /* If REG_NEWLINE is set, newlines are treated differently. */
+ if (cflags & REG_NEWLINE)
+ { /* REG_NEWLINE implies neither . nor [^...] match newline. */
+ syntax &= ~RE_DOT_NEWLINE;
+ syntax |= RE_HAT_LISTS_NOT_NEWLINE;
+ /* It also changes the matching behavior. */
+ preg->newline_anchor = 1;
+ }
+ else
+ preg->newline_anchor = 0;
+
+ preg->no_sub = !!(cflags & REG_NOSUB);
+
+ /* POSIX says a null character in the pattern terminates it, so we
+ can use strlen here in compiling the pattern. */
+ ret = regex_compile (pattern, strlen (pattern), syntax, preg);
+
+ /* POSIX doesn't distinguish between an unmatched open-group and an
+ unmatched close-group: both are REG_EPAREN. */
+ if (ret == REG_ERPAREN) ret = REG_EPAREN;
+
+ return (int) ret;
+}
+
+
+/* regexec searches for a given pattern, specified by PREG, in the
+ string STRING.
+
+ If NMATCH is zero or REG_NOSUB was set in the cflags argument to
+ `regcomp', we ignore PMATCH. Otherwise, we assume PMATCH has at
+ least NMATCH elements, and we set them to the offsets of the
+ corresponding matched substrings.
+
+ EFLAGS specifies `execution flags' which affect matching: if
+ REG_NOTBOL is set, then ^ does not match at the beginning of the
+ string; if REG_NOTEOL is set, then $ does not match at the end.
+
+ We return 0 if we find a match and REG_NOMATCH if not. */
+
+int
+regexec (preg, string, nmatch, pmatch, eflags)
+ const regex_t *preg;
+ const char *string;
+ size_t nmatch;
+ regmatch_t pmatch[];
+ int eflags;
+{
+ int ret;
+ struct re_registers regs;
+ regex_t private_preg;
+ int len = strlen (string);
+ boolean want_reg_info = !preg->no_sub && nmatch > 0;
+
+ private_preg = *preg;
+
+ private_preg.not_bol = !!(eflags & REG_NOTBOL);
+ private_preg.not_eol = !!(eflags & REG_NOTEOL);
+
+ /* The user has told us exactly how many registers to return
+ information about, via `nmatch'. We have to pass that on to the
+ matching routines. */
+ private_preg.regs_allocated = REGS_FIXED;
+
+ if (want_reg_info)
+ {
+ regs.num_regs = nmatch;
+ regs.start = TALLOC (nmatch, regoff_t);
+ regs.end = TALLOC (nmatch, regoff_t);
+ if (regs.start == NULL || regs.end == NULL)
+ return (int) REG_NOMATCH;
+ }
+
+ /* Perform the searching operation. */
+ ret = re_search (&private_preg, string, len,
+ /* start: */ 0, /* range: */ len,
+ want_reg_info ? ®s : (struct re_registers *) 0);
+
+ /* Copy the register information to the POSIX structure. */
+ if (want_reg_info)
+ {
+ if (ret >= 0)
+ {
+ unsigned r;
+
+ for (r = 0; r < nmatch; r++)
+ {
+ pmatch[r].rm_so = regs.start[r];
+ pmatch[r].rm_eo = regs.end[r];
+ }
+ }
+
+ /* If we needed the temporary register info, free the space now. */
+ free (regs.start);
+ free (regs.end);
+ }
+
+ /* We want zero return to mean success, unlike `re_search'. */
+ return ret >= 0 ? (int) REG_NOERROR : (int) REG_NOMATCH;
+}
+
+
+/* Returns a message corresponding to an error code, ERRCODE, returned
+ from either regcomp or regexec. We don't use PREG here. */
+
+size_t
+regerror (errcode, preg, errbuf, errbuf_size)
+ int errcode;
+ const regex_t *preg;
+ char *errbuf;
+ size_t errbuf_size;
+{
+ const char *msg;
+ size_t msg_size;
+
+ if (errcode < 0
+ || errcode >= (sizeof (re_error_msg) / sizeof (re_error_msg[0])))
+ /* Only error codes returned by the rest of the code should be passed
+ to this routine. If we are given anything else, or if other regex
+ code generates an invalid error code, then the program has a bug.
+ Dump core so we can fix it. */
+ abort ();
+
+ msg = re_error_msg[errcode];
+
+ /* POSIX doesn't require that we do anything in this case, but why
+ not be nice. */
+ if (! msg)
+ msg = "Success";
+
+ msg_size = strlen (msg) + 1; /* Includes the null. */
+
+ if (errbuf_size != 0)
+ {
+ if (msg_size > errbuf_size)
+ {
+ strncpy (errbuf, msg, errbuf_size - 1);
+ errbuf[errbuf_size - 1] = 0;
+ }
+ else
+ strcpy (errbuf, msg);
+ }
+
+ return msg_size;
+}
+
+
+/* Free dynamically allocated space used by PREG. */
+
+void
+regfree (preg)
+ regex_t *preg;
+{
+ if (preg->buffer != NULL)
+ free (preg->buffer);
+ preg->buffer = NULL;
+
+ preg->allocated = 0;
+ preg->used = 0;
+
+ if (preg->fastmap != NULL)
+ free (preg->fastmap);
+ preg->fastmap = NULL;
+ preg->fastmap_accurate = 0;
+
+ if (preg->translate != NULL)
+ free (preg->translate);
+ preg->translate = NULL;
+}
+
+#endif /* not emacs */
+\f
+/*
+Local variables:
+make-backup-files: t
+version-control: t
+trim-versions-without-asking: nil
+End:
+*/
--- /dev/null
+/* Definitions for data structures and routines for the regular
+ expression library, version 0.12.
+
+ Copyright (C) 1985, 1989, 1990, 1991, 1992, 1993 Free Software Foundation, Inc.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program; if not, write to the Free Software
+ Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */
+
+#ifndef __REGEXP_LIBRARY_H__
+#define __REGEXP_LIBRARY_H__
+
+/* POSIX says that <sys/types.h> must be included (by the caller) before
+ <regex.h>. */
+
+#ifdef VMS
+/* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
+ should be there. */
+#include <stddef.h>
+#endif
+
+
+/* The following bits are used to determine the regexp syntax we
+ recognize. The set/not-set meanings are chosen so that Emacs syntax
+ remains the value 0. The bits are given in alphabetical order, and
+ the definitions shifted by one from the previous bit; thus, when we
+ add or remove a bit, only one other definition need change. */
+typedef unsigned reg_syntax_t;
+
+/* If this bit is not set, then \ inside a bracket expression is literal.
+ If set, then such a \ quotes the following character. */
+#define RE_BACKSLASH_ESCAPE_IN_LISTS (1)
+
+/* If this bit is not set, then + and ? are operators, and \+ and \? are
+ literals.
+ If set, then \+ and \? are operators and + and ? are literals. */
+#define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
+
+/* If this bit is set, then character classes are supported. They are:
+ [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
+ [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
+ If not set, then character classes are not supported. */
+#define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
+
+/* If this bit is set, then ^ and $ are always anchors (outside bracket
+ expressions, of course).
+ If this bit is not set, then it depends:
+ ^ is an anchor if it is at the beginning of a regular
+ expression or after an open-group or an alternation operator;
+ $ is an anchor if it is at the end of a regular expression, or
+ before a close-group or an alternation operator.
+
+ This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
+ POSIX draft 11.2 says that * etc. in leading positions is undefined.
+ We already implemented a previous draft which made those constructs
+ invalid, though, so we haven't changed the code back. */
+#define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
+
+/* If this bit is set, then special characters are always special
+ regardless of where they are in the pattern.
+ If this bit is not set, then special characters are special only in
+ some contexts; otherwise they are ordinary. Specifically,
+ * + ? and intervals are only special when not after the beginning,
+ open-group, or alternation operator. */
+#define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
+
+/* If this bit is set, then *, +, ?, and { cannot be first in an re or
+ immediately after an alternation or begin-group operator. */
+#define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
+
+/* If this bit is set, then . matches newline.
+ If not set, then it doesn't. */
+#define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
+
+/* If this bit is set, then . doesn't match NUL.
+ If not set, then it does. */
+#define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
+
+/* If this bit is set, nonmatching lists [^...] do not match newline.
+ If not set, they do. */
+#define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
+
+/* If this bit is set, either \{...\} or {...} defines an
+ interval, depending on RE_NO_BK_BRACES.
+ If not set, \{, \}, {, and } are literals. */
+#define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
+
+/* If this bit is set, +, ? and | aren't recognized as operators.
+ If not set, they are. */
+#define RE_LIMITED_OPS (RE_INTERVALS << 1)
+
+/* If this bit is set, newline is an alternation operator.
+ If not set, newline is literal. */
+#define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
+
+/* If this bit is set, then `{...}' defines an interval, and \{ and \}
+ are literals.
+ If not set, then `\{...\}' defines an interval. */
+#define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
+
+/* If this bit is set, (...) defines a group, and \( and \) are literals.
+ If not set, \(...\) defines a group, and ( and ) are literals. */
+#define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
+
+/* If this bit is set, then \<digit> matches <digit>.
+ If not set, then \<digit> is a back-reference. */
+#define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
+
+/* If this bit is set, then | is an alternation operator, and \| is literal.
+ If not set, then \| is an alternation operator, and | is literal. */
+#define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
+
+/* If this bit is set, then an ending range point collating higher
+ than the starting range point, as in [z-a], is invalid.
+ If not set, then when ending range point collates higher than the
+ starting range point, the range is ignored. */
+#define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
+
+/* If this bit is set, then an unmatched ) is ordinary.
+ If not set, then an unmatched ) is invalid. */
+#define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
+
+/* This global variable defines the particular regexp syntax to use (for
+ some interfaces). When a regexp is compiled, the syntax used is
+ stored in the pattern buffer, so changing this does not affect
+ already-compiled regexps. */
+extern reg_syntax_t re_syntax_options;
+\f
+/* Define combinations of the above bits for the standard possibilities.
+ (The [[[ comments delimit what gets put into the Texinfo file, so
+ don't delete them!) */
+/* [[[begin syntaxes]]] */
+#define RE_SYNTAX_EMACS 0
+
+#define RE_SYNTAX_AWK \
+ (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
+ | RE_NO_BK_PARENS | RE_NO_BK_REFS \
+ | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
+ | RE_UNMATCHED_RIGHT_PAREN_ORD)
+
+#define RE_SYNTAX_POSIX_AWK \
+ (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS)
+
+#define RE_SYNTAX_GREP \
+ (RE_BK_PLUS_QM | RE_CHAR_CLASSES \
+ | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
+ | RE_NEWLINE_ALT)
+
+#define RE_SYNTAX_EGREP \
+ (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
+ | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
+ | RE_NO_BK_VBAR)
+
+#define RE_SYNTAX_POSIX_EGREP \
+ (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
+
+/* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
+#define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
+
+#define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
+
+/* Syntax bits common to both basic and extended POSIX regex syntax. */
+#define _RE_SYNTAX_POSIX_COMMON \
+ (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
+ | RE_INTERVALS | RE_NO_EMPTY_RANGES)
+
+#define RE_SYNTAX_POSIX_BASIC \
+ (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
+
+/* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
+ RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
+ isn't minimal, since other operators, such as \`, aren't disabled. */
+#define RE_SYNTAX_POSIX_MINIMAL_BASIC \
+ (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
+
+#define RE_SYNTAX_POSIX_EXTENDED \
+ (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
+ | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
+ | RE_UNMATCHED_RIGHT_PAREN_ORD)
+
+/* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INVALID_OPS
+ replaces RE_CONTEXT_INDEP_OPS and RE_NO_BK_REFS is added. */
+#define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
+ (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
+ | RE_NO_BK_PARENS | RE_NO_BK_REFS \
+ | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
+/* [[[end syntaxes]]] */
+\f
+/* Maximum number of duplicates an interval can allow. Some systems
+ (erroneously) define this in other header files, but we want our
+ value, so remove any previous define. */
+#ifdef RE_DUP_MAX
+#undef RE_DUP_MAX
+#endif
+#define RE_DUP_MAX ((1 << 15) - 1)
+
+
+/* POSIX `cflags' bits (i.e., information for `regcomp'). */
+
+/* If this bit is set, then use extended regular expression syntax.
+ If not set, then use basic regular expression syntax. */
+#define REG_EXTENDED 1
+
+/* If this bit is set, then ignore case when matching.
+ If not set, then case is significant. */
+#define REG_ICASE (REG_EXTENDED << 1)
+
+/* If this bit is set, then anchors do not match at newline
+ characters in the string.
+ If not set, then anchors do match at newlines. */
+#define REG_NEWLINE (REG_ICASE << 1)
+
+/* If this bit is set, then report only success or fail in regexec.
+ If not set, then returns differ between not matching and errors. */
+#define REG_NOSUB (REG_NEWLINE << 1)
+
+
+/* POSIX `eflags' bits (i.e., information for regexec). */
+
+/* If this bit is set, then the beginning-of-line operator doesn't match
+ the beginning of the string (presumably because it's not the
+ beginning of a line).
+ If not set, then the beginning-of-line operator does match the
+ beginning of the string. */
+#define REG_NOTBOL 1
+
+/* Like REG_NOTBOL, except for the end-of-line. */
+#define REG_NOTEOL (1 << 1)
+
+
+/* If any error codes are removed, changed, or added, update the
+ `re_error_msg' table in regex.c. */
+typedef enum
+{
+ REG_NOERROR = 0, /* Success. */
+ REG_NOMATCH, /* Didn't find a match (for regexec). */
+
+ /* POSIX regcomp return error codes. (In the order listed in the
+ standard.) */
+ REG_BADPAT, /* Invalid pattern. */
+ REG_ECOLLATE, /* Not implemented. */
+ REG_ECTYPE, /* Invalid character class name. */
+ REG_EESCAPE, /* Trailing backslash. */
+ REG_ESUBREG, /* Invalid back reference. */
+ REG_EBRACK, /* Unmatched left bracket. */
+ REG_EPAREN, /* Parenthesis imbalance. */
+ REG_EBRACE, /* Unmatched \{. */
+ REG_BADBR, /* Invalid contents of \{\}. */
+ REG_ERANGE, /* Invalid range end. */
+ REG_ESPACE, /* Ran out of memory. */
+ REG_BADRPT, /* No preceding re for repetition op. */
+
+ /* Error codes we've added. */
+ REG_EEND, /* Premature end. */
+ REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
+ REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
+} reg_errcode_t;
+\f
+/* This data structure represents a compiled pattern. Before calling
+ the pattern compiler, the fields `buffer', `allocated', `fastmap',
+ `translate', and `no_sub' can be set. After the pattern has been
+ compiled, the `re_nsub' field is available. All other fields are
+ private to the regex routines. */
+
+struct re_pattern_buffer
+{
+/* [[[begin pattern_buffer]]] */
+ /* Space that holds the compiled pattern. It is declared as
+ `unsigned char *' because its elements are
+ sometimes used as array indexes. */
+ unsigned char *buffer;
+
+ /* Number of bytes to which `buffer' points. */
+ unsigned long allocated;
+
+ /* Number of bytes actually used in `buffer'. */
+ unsigned long used;
+
+ /* Syntax setting with which the pattern was compiled. */
+ reg_syntax_t syntax;
+
+ /* Pointer to a fastmap, if any, otherwise zero. re_search uses
+ the fastmap, if there is one, to skip over impossible
+ starting points for matches. */
+ char *fastmap;
+
+ /* Either a translate table to apply to all characters before
+ comparing them, or zero for no translation. The translation
+ is applied to a pattern when it is compiled and to a string
+ when it is matched. */
+ char *translate;
+
+ /* Number of subexpressions found by the compiler. */
+ size_t re_nsub;
+
+ /* Zero if this pattern cannot match the empty string, one else.
+ Well, in truth it's used only in `re_search_2', to see
+ whether or not we should use the fastmap, so we don't set
+ this absolutely perfectly; see `re_compile_fastmap' (the
+ `duplicate' case). */
+ unsigned can_be_null : 1;
+
+ /* If REGS_UNALLOCATED, allocate space in the `regs' structure
+ for `max (RE_NREGS, re_nsub + 1)' groups.
+ If REGS_REALLOCATE, reallocate space if necessary.
+ If REGS_FIXED, use what's there. */
+#define REGS_UNALLOCATED 0
+#define REGS_REALLOCATE 1
+#define REGS_FIXED 2
+ unsigned regs_allocated : 2;
+
+ /* Set to zero when `regex_compile' compiles a pattern; set to one
+ by `re_compile_fastmap' if it updates the fastmap. */
+ unsigned fastmap_accurate : 1;
+
+ /* If set, `re_match_2' does not return information about
+ subexpressions. */
+ unsigned no_sub : 1;
+
+ /* If set, a beginning-of-line anchor doesn't match at the
+ beginning of the string. */
+ unsigned not_bol : 1;
+
+ /* Similarly for an end-of-line anchor. */
+ unsigned not_eol : 1;
+
+ /* If true, an anchor at a newline matches. */
+ unsigned newline_anchor : 1;
+
+/* [[[end pattern_buffer]]] */
+};
+
+typedef struct re_pattern_buffer regex_t;
+
+
+/* search.c (search_buffer) in Emacs needs this one opcode value. It is
+ defined both in `regex.c' and here. */
+#define RE_EXACTN_VALUE 1
+\f
+/* Type for byte offsets within the string. POSIX mandates this. */
+typedef int regoff_t;
+
+
+/* This is the structure we store register match data in. See
+ regex.texinfo for a full description of what registers match. */
+struct re_registers
+{
+ unsigned num_regs;
+ regoff_t *start;
+ regoff_t *end;
+};
+
+
+/* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
+ `re_match_2' returns information about at least this many registers
+ the first time a `regs' structure is passed. */
+#ifndef RE_NREGS
+#define RE_NREGS 30
+#endif
+
+
+/* POSIX specification for registers. Aside from the different names than
+ `re_registers', POSIX uses an array of structures, instead of a
+ structure of arrays. */
+typedef struct
+{
+ regoff_t rm_so; /* Byte offset from string's start to substring's start. */
+ regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
+} regmatch_t;
+\f
+/* Declarations for routines. */
+
+/* To avoid duplicating every routine declaration -- once with a
+ prototype (if we are ANSI), and once without (if we aren't) -- we
+ use the following macro to declare argument types. This
+ unfortunately clutters up the declarations a bit, but I think it's
+ worth it. */
+
+#if __STDC__
+
+#define _RE_ARGS(args) args
+
+#else /* not __STDC__ */
+
+#define _RE_ARGS(args) ()
+
+#endif /* not __STDC__ */
+
+/* Sets the current default syntax to SYNTAX, and return the old syntax.
+ You can also simply assign to the `re_syntax_options' variable. */
+extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
+
+/* Compile the regular expression PATTERN, with length LENGTH
+ and syntax given by the global `re_syntax_options', into the buffer
+ BUFFER. Return NULL if successful, and an error string if not. */
+extern const char *re_compile_pattern
+ _RE_ARGS ((const char *pattern, int length,
+ struct re_pattern_buffer *buffer));
+
+
+/* Compile a fastmap for the compiled pattern in BUFFER; used to
+ accelerate searches. Return 0 if successful and -2 if was an
+ internal error. */
+extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
+
+
+/* Search in the string STRING (with length LENGTH) for the pattern
+ compiled into BUFFER. Start searching at position START, for RANGE
+ characters. Return the starting position of the match, -1 for no
+ match, or -2 for an internal error. Also return register
+ information in REGS (if REGS and BUFFER->no_sub are nonzero). */
+extern int re_search
+ _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
+ int length, int start, int range, struct re_registers *regs));
+
+
+/* Like `re_search', but search in the concatenation of STRING1 and
+ STRING2. Also, stop searching at index START + STOP. */
+extern int re_search_2
+ _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
+ int length1, const char *string2, int length2,
+ int start, int range, struct re_registers *regs, int stop));
+
+
+/* Like `re_search', but return how many characters in STRING the regexp
+ in BUFFER matched, starting at position START. */
+extern int re_match
+ _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
+ int length, int start, struct re_registers *regs));
+
+
+/* Relates to `re_match' as `re_search_2' relates to `re_search'. */
+extern int re_match_2
+ _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
+ int length1, const char *string2, int length2,
+ int start, struct re_registers *regs, int stop));
+
+
+/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
+ ENDS. Subsequent matches using BUFFER and REGS will use this memory
+ for recording register information. STARTS and ENDS must be
+ allocated with malloc, and must each be at least `NUM_REGS * sizeof
+ (regoff_t)' bytes long.
+
+ If NUM_REGS == 0, then subsequent matches should allocate their own
+ register data.
+
+ Unless this function is called, the first search or match using
+ PATTERN_BUFFER will allocate its own register data, without
+ freeing the old data. */
+extern void re_set_registers
+ _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
+ unsigned num_regs, regoff_t *starts, regoff_t *ends));
+
+/* 4.2 bsd compatibility. */
+extern char *re_comp _RE_ARGS ((const char *));
+extern int re_exec _RE_ARGS ((const char *));
+
+/* POSIX compatibility. */
+extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags));
+extern int regexec
+ _RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch,
+ regmatch_t pmatch[], int eflags));
+extern size_t regerror
+ _RE_ARGS ((int errcode, const regex_t *preg, char *errbuf,
+ size_t errbuf_size));
+extern void regfree _RE_ARGS ((regex_t *preg));
+
+#endif /* not __REGEXP_LIBRARY_H__ */
+\f
+/*
+Local variables:
+make-backup-files: t
+version-control: t
+trim-versions-without-asking: nil
+End:
+*/
CC = @CC@
CFLAGS = @CFLAGS@
LDFLAGS = @LDFLAGS@
+CC_LD_DYNPATH = @CC_LD_DYNPATH@
AR = @AR@
TAR = @TAR@
#INSTALL = @INSTALL@ # needs install-sh or install.sh in sources
AC_MSG_NOTICE([CHECKS for programs])
#
AC_PROG_CC([cc gcc])
+# which switch to pass runtime path to dynamic libraries to the linker
+AC_CACHE_CHECK([if linker supports -R], ld_dashr, [
+ SAVE_LDFLAGS="${LDFLAGS}"
+ LDFLAGS="${SAVE_LDFLAGS} -R /"
+ AC_LINK_IFELSE(AC_LANG_PROGRAM([], []), [ld_dashr=yes], [ld_dashr=no])
+ LDFLAGS="${SAVE_LDFLAGS}"
+])
+if test "$ld_dashr" = "yes"; then
+ AC_SUBST(CC_LD_DYNPATH, [-R])
+else
+ AC_CACHE_CHECK([if linker supports -Wl,-rpath,], ld_wl_rpath, [
+ SAVE_LDFLAGS="${LDFLAGS}"
+ LDFLAGS="${SAVE_LDFLAGS} -Wl,-rpath,/"
+ AC_LINK_IFELSE(AC_LANG_PROGRAM([], []), [ld_wl_rpath=yes], [ld_wl_rpath=no])
+ LDFLAGS="${SAVE_LD_FLAGS}"
+ ])
+ if test "$ld_wl_rpath" = "yes"; then
+ AC_SUBST(CC_LD_DYNPATH, [-Wl,-rpath,])
+ else
+ AC_CACHE_CHECK([if linker supports -rpath], ld_rpath, [
+ SAVE_LDFLAGS="${LDFLAGS}"
+ LDFLAGS="${SAVE_LDFLAGS} -rpath /"
+ AC_LINK_IFELSE(AC_LANG_PROGRAM([], []), [ld_rpath=yes], [ld_rpath=no])
+ LDFLAGS="${SAVE_LD_FLAGS}"
+ ])
+ if test "$ld_rpath" = "yes"; then
+ AC_SUBST(CC_LD_DYNPATH, [-rpath])
+ else
+ AC_MSG_WARN([linker does not support runtime path to dynamic libraries])
+ fi
+ fi
+fi
#AC_PROG_INSTALL # needs install-sh or install.sh in sources
AC_CHECK_TOOLS(AR, [gar ar], :)
AC_CHECK_PROGS(TAR, [gtar tar])
static void emit_line(FILE *file, const char *set, const char *reset, const char *line, int len)
{
- int has_trailing_newline = (len > 0 && line[len-1] == '\n');
+ int has_trailing_newline, has_trailing_carriage_return;
+
+ has_trailing_newline = (len > 0 && line[len-1] == '\n');
if (has_trailing_newline)
len--;
+ has_trailing_carriage_return = (len > 0 && line[len-1] == '\r');
+ if (has_trailing_carriage_return)
+ len--;
fputs(set, file);
fwrite(line, len, 1, file);
fputs(reset, file);
+ if (has_trailing_carriage_return)
+ fputc('\r', file);
if (has_trailing_newline)
fputc('\n', file);
}
"\\|"
"^\\(.*=[ \t]*\\(class\\|record\\).*\\)$"
},
+ { "php", "^[\t ]*\\(\\(function\\|class\\).*\\)" },
{ "python", "^\\s*\\(\\(class\\|def\\)\\s.*\\)$" },
{ "ruby", "^\\s*\\(\\(class\\|module\\|def\\)\\s.*\\)$" },
{ "tex", "^\\(\\\\\\(\\(sub\\)*section\\|chapter\\|part\\)\\*\\{0,1\\}{.*\\)$" },
DIFF_OPT_SET(options, EXIT_WITH_STATUS);
}
- /*
- * If we postprocess in diffcore, we cannot simply return
- * upon the first hit. We need to run diff as usual.
- */
- if (options->pickaxe || options->filter)
- DIFF_OPT_CLR(options, QUIET);
-
return 0;
}
void diffcore_std(struct diff_options *options)
{
- if (DIFF_OPT_TST(options, QUIET))
- return;
-
- if (options->skip_stat_unmatch && !DIFF_OPT_TST(options, FIND_COPIES_HARDER))
+ if (options->skip_stat_unmatch)
diffcore_skip_stat_unmatch(options);
if (options->break_opt != -1)
diffcore_break(options->break_opt);
}
set default_config(branch.autosetupmerge) true
+set default_config(merge.tool) {}
+set default_config(merge.keepbackup) true
set default_config(merge.diffstat) true
set default_config(merge.summary) false
set default_config(merge.verbosity) 2
set default_config(gui.trustmtime) false
set default_config(gui.fastcopyblame) false
set default_config(gui.copyblamethreshold) 40
+set default_config(gui.blamehistoryctx) 7
set default_config(gui.diffcontext) 5
set default_config(gui.commitmsgwidth) 75
set default_config(gui.newbranchtemplate) {}
unlock_index
display_all_files
if {$current_diff_path ne {}} reshow_diff
+ if {$current_diff_path eq {}} select_first_diff
+
uplevel #0 $after
}
0xfa, 0x17, 0x02, 0x10, 0xfe, 0x1f};
} -maskdata $filemask
+image create bitmap file_statechange -background white -foreground green -data {
+#define file_merge_width 14
+#define file_merge_height 15
+static unsigned char file_statechange_bits[] = {
+ 0xfe, 0x01, 0x02, 0x03, 0x02, 0x05, 0x02, 0x09, 0x02, 0x1f, 0x62, 0x10,
+ 0x62, 0x10, 0xba, 0x11, 0xba, 0x11, 0x62, 0x10, 0x62, 0x10, 0x02, 0x10,
+ 0x02, 0x10, 0x02, 0x10, 0xfe, 0x1f};
+} -maskdata $filemask
+
set ui_index .vpane.files.index.list
set ui_workdir .vpane.files.workdir.list
set all_icons(M$ui_index) file_fulltick
set all_icons(D$ui_index) file_removed
set all_icons(U$ui_index) file_merge
+set all_icons(T$ui_index) file_statechange
set all_icons(_$ui_workdir) file_plain
set all_icons(M$ui_workdir) file_mod
set all_icons(D$ui_workdir) file_question
set all_icons(U$ui_workdir) file_merge
set all_icons(O$ui_workdir) file_plain
+set all_icons(T$ui_workdir) file_statechange
set max_status_desc 0
foreach i {
{MM {mc "Portions staged for commit"}}
{MD {mc "Staged for commit, missing"}}
+ {_T {mc "File type changed, not staged"}}
+ {T_ {mc "File type changed, staged"}}
+
{_O {mc "Untracked, not staged"}}
{A_ {mc "Staged for commit"}}
{AM {mc "Portions staged for commit"}}
{D_ {mc "Staged for removal"}}
{DO {mc "Staged for removal, still present"}}
+ {_U {mc "Requires merge resolution"}}
{U_ {mc "Requires merge resolution"}}
{UU {mc "Requires merge resolution"}}
{UM {mc "Requires merge resolution"}}
{UD {mc "Requires merge resolution"}}
+ {UT {mc "Requires merge resolution"}}
} {
set text [eval [lindex $i 1]]
if {$max_status_desc < [string length $text]} {
rescan ui_ready
}
+proc ui_do_rescan {} {
+ rescan {force_first_diff; ui_ready}
+}
+
proc do_commit {} {
commit_tree
}
proc next_diff {} {
global next_diff_p next_diff_w next_diff_i
- show_diff $next_diff_p $next_diff_w $next_diff_i
+ show_diff $next_diff_p $next_diff_w {}
+}
+
+proc find_anchor_pos {lst name} {
+ set lid [lsearch -sorted -exact $lst $name]
+
+ if {$lid == -1} {
+ set lid 0
+ foreach lname $lst {
+ if {$lname >= $name} break
+ incr lid
+ }
+ }
+
+ return $lid
+}
+
+proc find_file_from {flist idx delta path mmask} {
+ global file_states
+
+ set len [llength $flist]
+ while {$idx >= 0 && $idx < $len} {
+ set name [lindex $flist $idx]
+
+ if {$name ne $path && [info exists file_states($name)]} {
+ set state [lindex $file_states($name) 0]
+
+ if {$mmask eq {} || [regexp $mmask $state]} {
+ return $idx
+ }
+ }
+
+ incr idx $delta
+ }
+
+ return {}
+}
+
+proc find_next_diff {w path {lno {}} {mmask {}}} {
+ global next_diff_p next_diff_w next_diff_i
+ global file_lists ui_index ui_workdir
+
+ set flist $file_lists($w)
+ if {$lno eq {}} {
+ set lno [find_anchor_pos $flist $path]
+ } else {
+ incr lno -1
+ }
+
+ if {$mmask ne {} && ![regexp {(^\^)|(\$$)} $mmask]} {
+ if {$w eq $ui_index} {
+ set mmask "^$mmask"
+ } else {
+ set mmask "$mmask\$"
+ }
+ }
+
+ set idx [find_file_from $flist $lno 1 $path $mmask]
+ if {$idx eq {}} {
+ incr lno -1
+ set idx [find_file_from $flist $lno -1 $path $mmask]
+ }
+
+ if {$idx ne {}} {
+ set next_diff_w $w
+ set next_diff_p [lindex $flist $idx]
+ set next_diff_i [expr {$idx+1}]
+ return 1
+ } else {
+ return 0
+ }
+}
+
+proc next_diff_after_action {w path {lno {}} {mmask {}}} {
+ global current_diff_path
+
+ if {$path ne $current_diff_path} {
+ return {}
+ } elseif {[find_next_diff $w $path $lno $mmask]} {
+ return {next_diff;}
+ } else {
+ return {reshow_diff;}
+ }
+}
+
+proc select_first_diff {} {
+ global ui_workdir
+
+ if {[find_next_diff $ui_workdir {} 1 {^_?U}] ||
+ [find_next_diff $ui_workdir {} 1 {[^O]$}]} {
+ next_diff
+ }
+}
+
+proc force_first_diff {} {
+ global current_diff_path
+
+ if {[info exists file_states($current_diff_path)]} {
+ set state [lindex $file_states($current_diff_path) 0]
+
+ if {[string index $state 1] ne {O}} return
+ }
+
+ select_first_diff
}
proc toggle_or_diff {w x y} {
$ui_index tag remove in_sel 0.0 end
$ui_workdir tag remove in_sel 0.0 end
- if {$col == 0 && $y > 1} {
- set i [expr {$lno-1}]
- set ll [expr {[llength $file_lists($w)]-1}]
-
- if {$i == $ll && $i == 0} {
- set after {reshow_diff;}
- } else {
- global next_diff_p next_diff_w next_diff_i
-
- set next_diff_w $w
-
- if {$i < $ll} {
- set i [expr {$i + 1}]
- set next_diff_i $i
- } else {
- set next_diff_i $i
- set i [expr {$i - 1}]
- }
+ # Do not stage files with conflicts
+ if {[info exists file_states($path)]} {
+ set state [lindex $file_states($path) 0]
+ } else {
+ set state {__}
+ }
- set next_diff_p [lindex $file_lists($w) $i]
+ if {[string first {U} $state] >= 0} {
+ set col 1
+ }
- if {$next_diff_p ne {} && $current_diff_path ne {}} {
- set after {next_diff;}
- } else {
- set after {}
- }
+ # Restage the file, or simply show the diff
+ if {$col == 0 && $y > 1} {
+ if {[string index $state 1] eq {O}} {
+ set mmask {}
+ } else {
+ set mmask {[^O]}
}
+ set after [next_diff_after_action $w $path $lno $mmask]
+
if {$w eq $ui_index} {
update_indexinfo \
"Unstaging [short_path $path] from commit" \
proc show_less_context {} {
global repo_config
- if {$repo_config(gui.diffcontext) >= 1} {
+ if {$repo_config(gui.diffcontext) > 1} {
incr repo_config(gui.diffcontext) -1
reshow_diff
}
.mbar.commit add separator
.mbar.commit add command -label [mc Rescan] \
- -command do_rescan \
+ -command ui_do_rescan \
-accelerator F5
lappend disable_on_lock \
[list .mbar.commit entryconf [.mbar.commit index last] -state]
switch -- $subcommand {
browser -
blame {
- set subcommand_args {rev? path}
+ if {$subcommand eq "blame"} {
+ set subcommand_args {[--line=<num>] rev? path}
+ } else {
+ set subcommand_args {rev? path}
+ }
if {$argv eq {}} usage
set head {}
set path {}
+ set jump_spec {}
set is_path 0
foreach a $argv {
if {$is_path || [file exists $_prefix$a]} {
set path {}
}
set is_path 1
+ } elseif {[regexp {^--line=(\d+)$} $a a lnum]} {
+ if {$jump_spec ne {} || $head ne {}} usage
+ set jump_spec [list $lnum]
} elseif {$head eq {}} {
if {$head ne {}} usage
set head $a
switch -- $subcommand {
browser {
+ if {$jump_spec ne {}} usage
if {$head eq {}} {
if {$path ne {} && [file isdirectory $path]} {
set head $current_branch
puts stderr [mc "fatal: cannot stat path %s: No such file or directory" $path]
exit 1
}
- blame::new $head $path
+ blame::new $head $path $jump_spec
}
}
return
pack .vpane.lower.commarea.buttons -side left -fill y
button .vpane.lower.commarea.buttons.rescan -text [mc Rescan] \
- -command do_rescan
+ -command ui_do_rescan
pack .vpane.lower.commarea.buttons.rescan -side top -fill x
lappend disable_on_lock \
{.vpane.lower.commarea.buttons.rescan conf -state}
# -- Diff Body Context Menu
#
+
+proc create_common_diff_popup {ctxm} {
+ $ctxm add command \
+ -label [mc "Show Less Context"] \
+ -command show_less_context
+ lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
+ $ctxm add command \
+ -label [mc "Show More Context"] \
+ -command show_more_context
+ lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
+ $ctxm add separator
+ $ctxm add command \
+ -label [mc Refresh] \
+ -command reshow_diff
+ lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
+ $ctxm add command \
+ -label [mc Copy] \
+ -command {tk_textCopy $ui_diff}
+ lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
+ $ctxm add command \
+ -label [mc "Select All"] \
+ -command {focus $ui_diff;$ui_diff tag add sel 0.0 end}
+ lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
+ $ctxm add command \
+ -label [mc "Copy All"] \
+ -command {
+ $ui_diff tag add sel 0.0 end
+ tk_textCopy $ui_diff
+ $ui_diff tag remove sel 0.0 end
+ }
+ lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
+ $ctxm add separator
+ $ctxm add command \
+ -label [mc "Decrease Font Size"] \
+ -command {incr_font_size font_diff -1}
+ lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
+ $ctxm add command \
+ -label [mc "Increase Font Size"] \
+ -command {incr_font_size font_diff 1}
+ lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
+ $ctxm add separator
+ $ctxm add command -label [mc "Options..."] \
+ -command do_options
+}
+
set ctxm .vpane.lower.diff.body.ctxm
menu $ctxm -tearoff 0
$ctxm add command \
set ui_diff_applyline [$ctxm index last]
lappend diff_actions [list $ctxm entryconf $ui_diff_applyline -state]
$ctxm add separator
-$ctxm add command \
- -label [mc "Show Less Context"] \
- -command show_less_context
-lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
-$ctxm add command \
- -label [mc "Show More Context"] \
- -command show_more_context
-lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
-$ctxm add separator
-$ctxm add command \
- -label [mc Refresh] \
- -command reshow_diff
-lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
-$ctxm add command \
- -label [mc Copy] \
- -command {tk_textCopy $ui_diff}
-lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
-$ctxm add command \
- -label [mc "Select All"] \
- -command {focus $ui_diff;$ui_diff tag add sel 0.0 end}
-lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
-$ctxm add command \
- -label [mc "Copy All"] \
- -command {
- $ui_diff tag add sel 0.0 end
- tk_textCopy $ui_diff
- $ui_diff tag remove sel 0.0 end
- }
-lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
-$ctxm add separator
-$ctxm add command \
- -label [mc "Decrease Font Size"] \
- -command {incr_font_size font_diff -1}
-lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
-$ctxm add command \
- -label [mc "Increase Font Size"] \
- -command {incr_font_size font_diff 1}
-lappend diff_actions [list $ctxm entryconf [$ctxm index last] -state]
-$ctxm add separator
-$ctxm add command -label [mc "Options..."] \
- -command do_options
-proc popup_diff_menu {ctxm x y X Y} {
+create_common_diff_popup $ctxm
+
+set ctxmmg .vpane.lower.diff.body.ctxmmg
+menu $ctxmmg -tearoff 0
+$ctxmmg add command \
+ -label [mc "Run Merge Tool"] \
+ -command {merge_resolve_tool}
+lappend diff_actions [list $ctxmmg entryconf [$ctxmmg index last] -state]
+$ctxmmg add separator
+$ctxmmg add command \
+ -label [mc "Use Remote Version"] \
+ -command {merge_resolve_one 3}
+lappend diff_actions [list $ctxmmg entryconf [$ctxmmg index last] -state]
+$ctxmmg add command \
+ -label [mc "Use Local Version"] \
+ -command {merge_resolve_one 2}
+lappend diff_actions [list $ctxmmg entryconf [$ctxmmg index last] -state]
+$ctxmmg add command \
+ -label [mc "Revert To Base"] \
+ -command {merge_resolve_one 1}
+lappend diff_actions [list $ctxmmg entryconf [$ctxmmg index last] -state]
+$ctxmmg add separator
+create_common_diff_popup $ctxmmg
+
+proc popup_diff_menu {ctxm ctxmmg x y X Y} {
global current_diff_path file_states
set ::cursorX $x
set ::cursorY $y
- if {$::ui_index eq $::current_diff_side} {
- set l [mc "Unstage Hunk From Commit"]
- set t [mc "Unstage Line From Commit"]
+ if {[info exists file_states($current_diff_path)]} {
+ set state [lindex $file_states($current_diff_path) 0]
} else {
- set l [mc "Stage Hunk For Commit"]
- set t [mc "Stage Line For Commit"]
- }
- if {$::is_3way_diff
- || $current_diff_path eq {}
- || ![info exists file_states($current_diff_path)]
- || {_O} eq [lindex $file_states($current_diff_path) 0]} {
- set s disabled
+ set state {__}
+ }
+ if {[string first {U} $state] >= 0} {
+ tk_popup $ctxmmg $X $Y
} else {
- set s normal
+ if {$::ui_index eq $::current_diff_side} {
+ set l [mc "Unstage Hunk From Commit"]
+ set t [mc "Unstage Line From Commit"]
+ } else {
+ set l [mc "Stage Hunk For Commit"]
+ set t [mc "Stage Line For Commit"]
+ }
+ if {$::is_3way_diff
+ || $current_diff_path eq {}
+ || {__} eq $state
+ || {_O} eq $state
+ || {_T} eq $state
+ || {T_} eq $state} {
+ set s disabled
+ } else {
+ set s normal
+ }
+ $ctxm entryconf $::ui_diff_applyhunk -state $s -label $l
+ $ctxm entryconf $::ui_diff_applyline -state $s -label $t
+ tk_popup $ctxm $X $Y
}
- $ctxm entryconf $::ui_diff_applyhunk -state $s -label $l
- $ctxm entryconf $::ui_diff_applyline -state $s -label $t
- tk_popup $ctxm $X $Y
}
-bind_button3 $ui_diff [list popup_diff_menu $ctxm %x %y %X %Y]
+bind_button3 $ui_diff [list popup_diff_menu $ctxm $ctxmmg %x %y %X %Y]
# -- Status Bar
#
bind . <$M1B-Key-P> do_push_anywhere
}
-bind . <Key-F5> do_rescan
-bind . <$M1B-Key-r> do_rescan
-bind . <$M1B-Key-R> do_rescan
+bind . <Key-F5> ui_do_rescan
+bind . <$M1B-Key-r> ui_do_rescan
+bind . <$M1B-Key-R> ui_do_rescan
bind . <$M1B-Key-s> do_signoff
bind . <$M1B-Key-S> do_signoff
bind . <$M1B-Key-t> do_add_selection
field tooltip_timer {} ; # Current timer event for our tooltip
field tooltip_commit {} ; # Commit(s) in tooltip
-constructor new {i_commit i_path} {
+constructor new {i_commit i_path i_jump} {
global cursor_ptr
variable active_color
variable group_colors
$w.ctxm add command \
-label [mc "Do Full Copy Detection"] \
-command [cb _fullcopyblame]
+ $w.ctxm add command \
+ -label [mc "Show History Context"] \
+ -command [cb _gitkcommit]
+ $w.ctxm add command \
+ -label [mc "Blame Parent Commit"] \
+ -command [cb _blameparent]
foreach i $w_columns {
for {set g 0} {$g < [llength $group_colors]} {incr g} {
wm protocol $top WM_DELETE_WINDOW "destroy $top"
bind $top <Destroy> [cb _kill]
- _load $this {}
+ _load $this $i_jump
}
method _kill {} {
set lno [lindex [split [$cur_w index $pos] .] 0]
set dat [lindex $line_data $lno]
if {$dat ne {}} {
- lappend history [list \
- $commit $path \
- $highlight_column \
- $highlight_line \
- [lindex [$w_file xview] 0] \
- [lindex [$w_file yview] 0] \
- ]
- set commit [lindex $dat 0]
- set path [lindex $dat 1]
- _load $this [list [lindex $dat 2]]
+ _load_new_commit $this \
+ [lindex $dat 0] \
+ [lindex $dat 1] \
+ [list [lindex $dat 2]]
}
}
+method _load_new_commit {new_commit new_path jump} {
+ lappend history [list \
+ $commit $path \
+ $highlight_column \
+ $highlight_line \
+ [lindex [$w_file xview] 0] \
+ [lindex [$w_file yview] 0] \
+ ]
+
+ set commit $new_commit
+ set path $new_path
+ _load $this $jump
+}
+
method _showcommit {cur_w lno} {
global repo_config
variable active_color
}
}
-method _copycommit {} {
+method _get_click_amov_info {} {
set pos @$::cursorX,$::cursorY
set lno [lindex [split [$::cursorW index $pos] .] 0]
- set dat [lindex $amov_data $lno]
+ return [lindex $amov_data $lno]
+}
+
+method _copycommit {} {
+ set dat [_get_click_amov_info $this]
if {$dat ne {}} {
clipboard clear
clipboard append \
}
}
+method _format_offset_date {base offset} {
+ set exval [expr {$base + $offset*24*60*60}]
+ return [clock format $exval -format {%Y-%m-%d}]
+}
+
+method _gitkcommit {} {
+ set dat [_get_click_amov_info $this]
+ if {$dat ne {}} {
+ set cmit [lindex $dat 0]
+ set radius [get_config gui.blamehistoryctx]
+ set cmdline [list --select-commit=$cmit]
+
+ if {$radius > 0} {
+ set author_time {}
+ set committer_time {}
+
+ catch {set author_time $header($cmit,author-time)}
+ catch {set committer_time $header($cmit,committer-time)}
+
+ if {$committer_time eq {}} {
+ set committer_time $author_time
+ }
+
+ set after_time [_format_offset_date $this $committer_time [expr {-$radius}]]
+ set before_time [_format_offset_date $this $committer_time $radius]
+
+ lappend cmdline --after=$after_time --before=$before_time
+ }
+
+ lappend cmdline $cmit
+
+ set base_rev "HEAD"
+ if {$commit ne {}} {
+ set base_rev $commit
+ }
+
+ if {$base_rev ne $cmit} {
+ lappend cmdline $base_rev
+ }
+
+ do_gitk $cmdline
+ }
+}
+
+method _blameparent {} {
+ set dat [_get_click_amov_info $this]
+ if {$dat ne {}} {
+ set cmit [lindex $dat 0]
+ set new_path [lindex $dat 1]
+
+ if {[catch {set cparent [git rev-parse --verify "$cmit^"]}]} {
+ error_popup [strcat [mc "Cannot find parent commit:"] "\n\n$err"]
+ return;
+ }
+
+ _kill $this
+
+ # Generate a diff between the commit and its parent,
+ # and use the hunks to update the line number.
+ # Request zero context to simplify calculations.
+ if {[catch {set fd [eval git_read diff-tree \
+ --unified=0 $cparent $cmit $new_path]} err]} {
+ $status stop [mc "Unable to display parent"]
+ error_popup [strcat [mc "Error loading diff:"] "\n\n$err"]
+ return
+ }
+
+ set r_orig_line [lindex $dat 2]
+
+ fconfigure $fd \
+ -blocking 0 \
+ -encoding binary \
+ -translation binary
+ fileevent $fd readable [cb _read_diff_load_commit \
+ $fd $cparent $new_path $r_orig_line]
+ set current_fd $fd
+ }
+}
+
+method _read_diff_load_commit {fd cparent new_path tline} {
+ if {$fd ne $current_fd} {
+ catch {close $fd}
+ return
+ }
+
+ while {[gets $fd line] >= 0} {
+ if {[regexp {^@@ -(\d+)(,(\d+))? \+(\d+)(,(\d+))? @@} $line line \
+ old_line osz old_size new_line nsz new_size]} {
+
+ if {$osz eq {}} { set old_size 1 }
+ if {$nsz eq {}} { set new_size 1 }
+
+ if {$new_line <= $tline} {
+ if {[expr {$new_line + $new_size}] > $tline} {
+ # Target line within the hunk
+ set line_shift [expr {
+ ($new_size-$old_size)*($tline-$new_line)/$new_size
+ }]
+ } else {
+ set line_shift [expr {$new_size-$old_size}]
+ }
+
+ set r_orig_line [expr {$r_orig_line - $line_shift}]
+ }
+ }
+ }
+
+ if {[eof $fd]} {
+ close $fd;
+ set current_fd {}
+
+ _load_new_commit $this \
+ $cparent \
+ $new_path \
+ [list $r_orig_line]
+ }
+} ifdeleted { catch {close $fd} }
+
method _show_tooltip {cur_w pos} {
if {$tooltip_wm ne {}} {
_open_tooltip $this $cur_w
append p [lindex $n 1]
}
append p $name
- blame::new $browser_commit $p
+ blame::new $browser_commit $p {}
}
}
}
_? {continue}
A? -
D? -
+ T_ -
M? {set files_ready 1}
+ _U -
U? {
error_popup [mc "Unmerged files cannot be committed.
__ -
A_ -
M_ -
+ T_ -
D_ {
unset file_states($path)
catch {unset selected_paths($path)}
set p $current_diff_path
if {$p eq {}} {
# No diff is being shown.
- } elseif {$current_diff_side eq {}
- || [catch {set s $file_states($p)}]
- || [lsearch -sorted -exact $file_lists($current_diff_side) $p] == -1} {
+ } elseif {$current_diff_side eq {}} {
clear_diff
+ } elseif {[catch {set s $file_states($p)}]
+ || [lsearch -sorted -exact $file_lists($current_diff_side) $p] == -1} {
+
+ if {[find_next_diff $current_diff_side $p {} {[^O]}]} {
+ next_diff
+ } else {
+ clear_diff
+ }
} else {
set save_pos [lindex [$ui_diff yview] 0]
show_diff $p $current_diff_side {} $save_pos
global is_3way_diff diff_active repo_config
global ui_diff ui_index ui_workdir
global current_diff_path current_diff_side current_diff_header
+ global current_diff_queue
if {$diff_active || ![lock_index read]} return
}
if {$lno >= 1} {
$w tag add in_diff $lno.0 [expr {$lno + 1}].0
+ $w see $lno.0
}
set s $file_states($path)
set m [lindex $s 0]
- set is_3way_diff 0
- set diff_active 1
set current_diff_path $path
set current_diff_side $w
- set current_diff_header {}
+ set current_diff_queue {}
ui_status [mc "Loading diff of %s..." [escape_path $path]]
+ if {[string first {U} $m] >= 0} {
+ merge_load_stages $path [list show_unmerged_diff $scroll_pos]
+ } elseif {$m eq {_O}} {
+ show_other_diff $path $w $m $scroll_pos
+ } else {
+ start_show_diff $scroll_pos
+ }
+}
+
+proc show_unmerged_diff {scroll_pos} {
+ global current_diff_path current_diff_side
+ global merge_stages ui_diff
+ global current_diff_queue
+
+ if {$merge_stages(2) eq {}} {
+ lappend current_diff_queue \
+ [list "LOCAL: deleted\nREMOTE:\n" d======= \
+ [list ":1:$current_diff_path" ":3:$current_diff_path"]]
+ } elseif {$merge_stages(3) eq {}} {
+ lappend current_diff_queue \
+ [list "REMOTE: deleted\nLOCAL:\n" d======= \
+ [list ":1:$current_diff_path" ":2:$current_diff_path"]]
+ } elseif {[lindex $merge_stages(1) 0] eq {120000}
+ || [lindex $merge_stages(2) 0] eq {120000}
+ || [lindex $merge_stages(3) 0] eq {120000}} {
+ lappend current_diff_queue \
+ [list "LOCAL:\n" d======= \
+ [list ":1:$current_diff_path" ":2:$current_diff_path"]]
+ lappend current_diff_queue \
+ [list "REMOTE:\n" d======= \
+ [list ":1:$current_diff_path" ":3:$current_diff_path"]]
+ } else {
+ start_show_diff $scroll_pos
+ return
+ }
+
+ advance_diff_queue $scroll_pos
+}
+
+proc advance_diff_queue {scroll_pos} {
+ global current_diff_queue ui_diff
+
+ set item [lindex $current_diff_queue 0]
+ set current_diff_queue [lrange $current_diff_queue 1 end]
+
+ $ui_diff conf -state normal
+ $ui_diff insert end [lindex $item 0] [lindex $item 1]
+ $ui_diff conf -state disabled
+
+ start_show_diff $scroll_pos [lindex $item 2]
+}
+
+proc show_other_diff {path w m scroll_pos} {
+ global file_states file_lists
+ global is_3way_diff diff_active repo_config
+ global ui_diff ui_index ui_workdir
+ global current_diff_path current_diff_side current_diff_header
+
# - Git won't give us the diff, there's nothing to compare to!
#
if {$m eq {_O}} {
ui_ready
return
}
+}
+
+proc start_show_diff {scroll_pos {add_opts {}}} {
+ global file_states file_lists
+ global is_3way_diff diff_active repo_config
+ global ui_diff ui_index ui_workdir
+ global current_diff_path current_diff_side current_diff_header
+
+ set path $current_diff_path
+ set w $current_diff_side
+
+ set s $file_states($path)
+ set m [lindex $s 0]
+ set is_3way_diff 0
+ set diff_active 1
+ set current_diff_header {}
set cmd [list]
if {$w eq $ui_index} {
lappend cmd diff-index
lappend cmd --cached
} elseif {$w eq $ui_workdir} {
- if {[string index $m 0] eq {U}} {
+ if {[string first {U} $m] >= 0} {
lappend cmd diff
} else {
lappend cmd diff-files
lappend cmd -p
lappend cmd --no-color
- if {$repo_config(gui.diffcontext) >= 0} {
+ if {$repo_config(gui.diffcontext) >= 1} {
lappend cmd "-U$repo_config(gui.diffcontext)"
}
if {$w eq $ui_index} {
lappend cmd [PARENT]
}
- lappend cmd --
- lappend cmd $path
+ if {$add_opts ne {}} {
+ eval lappend cmd $add_opts
+ } else {
+ lappend cmd --
+ lappend cmd $path
+ }
if {[catch {set fd [eval git_read --nice $cmd]} err]} {
set diff_active 0
return
}
+ set ::current_diff_inheader 1
fconfigure $fd \
-blocking 0 \
-encoding binary \
proc read_diff {fd scroll_pos} {
global ui_diff diff_active
global is_3way_diff current_diff_header
+ global current_diff_queue
$ui_diff conf -state normal
while {[gets $fd line] >= 0} {
# -- Cleanup uninteresting diff header lines.
#
- if { [string match {diff --git *} $line]
- || [string match {diff --cc *} $line]
- || [string match {diff --combined *} $line]
- || [string match {--- *} $line]
- || [string match {+++ *} $line]} {
- append current_diff_header $line "\n"
- continue
+ if {$::current_diff_inheader} {
+ if { [string match {diff --git *} $line]
+ || [string match {diff --cc *} $line]
+ || [string match {diff --combined *} $line]
+ || [string match {--- *} $line]
+ || [string match {+++ *} $line]} {
+ append current_diff_header $line "\n"
+ continue
+ }
}
if {[string match {index *} $line]} continue
if {$line eq {deleted file mode 120000}} {
set line "deleted symlink"
}
+ set ::current_diff_inheader 0
# -- Automatically detect if this is a 3 way diff.
#
if {[eof $fd]} {
close $fd
+
+ if {$current_diff_queue ne {}} {
+ advance_diff_queue $scroll_pos
+ return
+ }
+
set diff_active 0
unlock_index
if {$scroll_pos ne {}} {
}
unlock_index
display_file $current_diff_path $mi
+ # This should trigger shift to the next changed file
if {$o eq {_}} {
- clear_diff
- } else {
- set current_diff_path $current_diff_path
+ reshow_diff
}
}
switch -glob -- [lindex $s 0] {
A? {set new _O}
M? {set new _M}
+ T_ {set new _T}
D_ {set new _D}
D? {set new _?}
?? {continue}
?D {set new D_}
_O -
AM {set new A_}
+ _T {set new T_}
+ _U -
U? {
if {[file exists $path]} {
set new M_
switch -glob -- [lindex $file_states($path) 0] {
U? {continue}
?M -
+ ?T -
?D {
puts -nonewline $fd "[encoding convertto $path]\0"
display_file $path ?_
switch -glob -- [lindex $file_states($path) 0] {
A? -
M? -
+ T_ -
D? {
lappend pathList $path
if {$path eq $current_diff_path} {
_O -
?M -
?D -
+ ?T -
U? {
lappend pathList $path
if {$path eq $current_diff_path} {
switch -glob -- [lindex $file_states($path) 0] {
U? {continue}
?M -
+ ?T -
?D {lappend paths $path}
}
}
switch -glob -- [lindex $file_states($path) 0] {
U? {continue}
?M -
+ ?T -
?D {
lappend pathList $path
if {$path eq $current_diff_path} {
if {[array size selected_paths] > 0} {
revert_helper \
- {Reverting selected files} \
+ [mc "Reverting selected files"] \
[array names selected_paths]
} elseif {$current_diff_path ne {}} {
revert_helper \
- "Reverting [short_path $current_diff_path]" \
+ [mc "Reverting %s" [short_path $current_diff_path]] \
[list $current_diff_path]
}
}
--- /dev/null
+# git-gui merge conflict resolution
+# parts based on git-mergetool (c) 2006 Theodore Y. Ts'o
+
+proc merge_resolve_one {stage} {
+ global current_diff_path
+
+ switch -- $stage {
+ 1 { set target [mc "the base version"] }
+ 2 { set target [mc "this branch"] }
+ 3 { set target [mc "the other branch"] }
+ }
+
+ set op_question [mc "Force resolution to %s?
+Note that the diff shows only conflicting changes.
+
+%s will be overwritten.
+
+This operation can be undone only by restarting the merge." \
+ $target [short_path $current_diff_path]]
+
+ if {[ask_popup $op_question] eq {yes}} {
+ merge_load_stages $current_diff_path [list merge_force_stage $stage]
+ }
+}
+
+proc merge_add_resolution {path} {
+ global current_diff_path ui_workdir
+
+ set after [next_diff_after_action $ui_workdir $path {} {^_?U}]
+
+ update_index \
+ [mc "Adding resolution for %s" [short_path $path]] \
+ [list $path] \
+ [concat $after [list ui_ready]]
+}
+
+proc merge_force_stage {stage} {
+ global current_diff_path merge_stages
+
+ if {$merge_stages($stage) ne {}} {
+ git checkout-index -f --stage=$stage -- $current_diff_path
+ } else {
+ file delete -- $current_diff_path
+ }
+
+ merge_add_resolution $current_diff_path
+}
+
+proc merge_load_stages {path cont} {
+ global merge_stages_fd merge_stages merge_stages_buf
+
+ if {[info exists merge_stages_fd]} {
+ catch { kill_file_process $merge_stages_fd }
+ catch { close $merge_stages_fd }
+ }
+
+ set merge_stages(0) {}
+ set merge_stages(1) {}
+ set merge_stages(2) {}
+ set merge_stages(3) {}
+ set merge_stages_buf {}
+
+ set merge_stages_fd [eval git_read ls-files -u -z -- $path]
+
+ fconfigure $merge_stages_fd -blocking 0 -translation binary -encoding binary
+ fileevent $merge_stages_fd readable [list read_merge_stages $merge_stages_fd $cont]
+}
+
+proc read_merge_stages {fd cont} {
+ global merge_stages_buf merge_stages_fd merge_stages
+
+ append merge_stages_buf [read $fd]
+ set pck [split $merge_stages_buf "\0"]
+ set merge_stages_buf [lindex $pck end]
+
+ if {[eof $fd] && $merge_stages_buf ne {}} {
+ lappend pck {}
+ set merge_stages_buf {}
+ }
+
+ foreach p [lrange $pck 0 end-1] {
+ set fcols [split $p "\t"]
+ set cols [split [lindex $fcols 0] " "]
+ set stage [lindex $cols 2]
+
+ set merge_stages($stage) [lrange $cols 0 1]
+ }
+
+ if {[eof $fd]} {
+ close $fd
+ unset merge_stages_fd
+ eval $cont
+ }
+}
+
+proc merge_resolve_tool {} {
+ global current_diff_path
+
+ merge_load_stages $current_diff_path [list merge_resolve_tool2]
+}
+
+proc merge_resolve_tool2 {} {
+ global current_diff_path merge_stages
+
+ # Validate the stages
+ if {$merge_stages(2) eq {} ||
+ [lindex $merge_stages(2) 0] eq {120000} ||
+ [lindex $merge_stages(2) 0] eq {160000} ||
+ $merge_stages(3) eq {} ||
+ [lindex $merge_stages(3) 0] eq {120000} ||
+ [lindex $merge_stages(3) 0] eq {160000}
+ } {
+ error_popup [mc "Cannot resolve deletion or link conflicts using a tool"]
+ return
+ }
+
+ if {![file exists $current_diff_path]} {
+ error_popup [mc "Conflict file does not exist"]
+ return
+ }
+
+ # Determine the tool to use
+ set tool [get_config merge.tool]
+ if {$tool eq {}} { set tool meld }
+
+ set merge_tool_path [get_config "mergetool.$tool.path"]
+ if {$merge_tool_path eq {}} {
+ switch -- $tool {
+ emerge { set merge_tool_path "emacs" }
+ araxis { set merge_tool_path "compare" }
+ default { set merge_tool_path $tool }
+ }
+ }
+
+ # Make file names
+ set filebase [file rootname $current_diff_path]
+ set fileext [file extension $current_diff_path]
+ set basename [lindex [file split $current_diff_path] end]
+
+ set MERGED $current_diff_path
+ set BASE "./$MERGED.BASE$fileext"
+ set LOCAL "./$MERGED.LOCAL$fileext"
+ set REMOTE "./$MERGED.REMOTE$fileext"
+ set BACKUP "./$MERGED.BACKUP$fileext"
+
+ set base_stage $merge_stages(1)
+
+ # Build the command line
+ switch -- $tool {
+ kdiff3 {
+ if {$base_stage ne {}} {
+ set cmdline [list "$merge_tool_path" --auto --L1 "$MERGED (Base)" \
+ --L2 "$MERGED (Local)" --L3 "$MERGED (Remote)" -o "$MERGED" "$BASE" "$LOCAL" "$REMOTE"]
+ } else {
+ set cmdline [list "$merge_tool_path" --auto --L1 "$MERGED (Local)" \
+ --L2 "$MERGED (Remote)" -o "$MERGED" "$LOCAL" "$REMOTE"]
+ }
+ }
+ tkdiff {
+ if {$base_stage ne {}} {
+ set cmdline [list "$merge_tool_path" -a "$BASE" -o "$MERGED" "$LOCAL" "$REMOTE"]
+ } else {
+ set cmdline [list "$merge_tool_path" -o "$MERGED" "$LOCAL" "$REMOTE"]
+ }
+ }
+ meld {
+ set cmdline [list "$merge_tool_path" "$LOCAL" "$MERGED" "$REMOTE"]
+ }
+ gvimdiff {
+ set cmdline [list "$merge_tool_path" -f "$LOCAL" "$MERGED" "$REMOTE"]
+ }
+ xxdiff {
+ if {$base_stage ne {}} {
+ set cmdline [list "$merge_tool_path" -X --show-merged-pane \
+ -R {Accel.SaveAsMerged: "Ctrl-S"} \
+ -R {Accel.Search: "Ctrl+F"} \
+ -R {Accel.SearchForward: "Ctrl-G"} \
+ --merged-file "$MERGED" "$LOCAL" "$BASE" "$REMOTE"]
+ } else {
+ set cmdline [list "$merge_tool_path" -X --show-merged-pane \
+ -R {Accel.SaveAsMerged: "Ctrl-S"} \
+ -R {Accel.Search: "Ctrl+F"} \
+ -R {Accel.SearchForward: "Ctrl-G"} \
+ --merged-file "$MERGED" "$LOCAL" "$REMOTE"]
+ }
+ }
+ opendiff {
+ if {$base_stage ne {}} {
+ set cmdline [list "$merge_tool_path" "$LOCAL" "$REMOTE" -ancestor "$BASE" -merge "$MERGED"]
+ } else {
+ set cmdline [list "$merge_tool_path" "$LOCAL" "$REMOTE" -merge "$MERGED"]
+ }
+ }
+ ecmerge {
+ if {$base_stage ne {}} {
+ set cmdline [list "$merge_tool_path" "$BASE" "$LOCAL" "$REMOTE" --default --mode=merge3 --to="$MERGED"]
+ } else {
+ set cmdline [list "$merge_tool_path" "$LOCAL" "$REMOTE" --default --mode=merge2 --to="$MERGED"]
+ }
+ }
+ emerge {
+ if {$base_stage ne {}} {
+ set cmdline [list "$merge_tool_path" -f emerge-files-with-ancestor-command \
+ "$LOCAL" "$REMOTE" "$BASE" "$basename"]
+ } else {
+ set cmdline [list "$merge_tool_path" -f emerge-files-command \
+ "$LOCAL" "$REMOTE" "$basename"]
+ }
+ }
+ winmerge {
+ if {$base_stage ne {}} {
+ # This tool does not support 3-way merges.
+ # Use the 'conflict file' resolution feature instead.
+ set cmdline [list "$merge_tool_path" -e -ub "$MERGED"]
+ } else {
+ set cmdline [list "$merge_tool_path" -e -ub -wl \
+ -dl "Theirs File" -dr "Mine File" "$REMOTE" "$LOCAL" "$MERGED"]
+ }
+ }
+ araxis {
+ if {$base_stage ne {}} {
+ set cmdline [list "$merge_tool_path" -wait -merge -3 -a1 \
+ -title1:"'$MERGED (Base)'" -title2:"'$MERGED (Local)'" \
+ -title3:"'$MERGED (Remote)'" \
+ "$BASE" "$LOCAL" "$REMOTE" "$MERGED"]
+ } else {
+ set cmdline [list "$merge_tool_path" -wait -2 \
+ -title1:"'$MERGED (Local)'" -title2:"'$MERGED (Remote)'" \
+ "$LOCAL" "$REMOTE" "$MERGED"]
+ }
+ }
+ p4merge {
+ set cmdline [list "$merge_tool_path" "$BASE" "$REMOTE" "$LOCAL" "$MERGED"]
+ }
+ vimdiff {
+ error_popup [mc "Not a GUI merge tool: '%s'" $tool]
+ return
+ }
+ default {
+ error_popup [mc "Unsupported merge tool '%s'" $tool]
+ return
+ }
+ }
+
+ merge_tool_start $cmdline $MERGED $BACKUP [list $BASE $LOCAL $REMOTE]
+}
+
+proc delete_temp_files {files} {
+ foreach fname $files {
+ file delete $fname
+ }
+}
+
+proc merge_tool_get_stages {target stages} {
+ global merge_stages
+
+ set i 1
+ foreach fname $stages {
+ if {$merge_stages($i) eq {}} {
+ file delete $fname
+ catch { close [open $fname w] }
+ } else {
+ # A hack to support autocrlf properly
+ git checkout-index -f --stage=$i -- $target
+ file rename -force -- $target $fname
+ }
+ incr i
+ }
+}
+
+proc merge_tool_start {cmdline target backup stages} {
+ global merge_stages mtool_target mtool_tmpfiles mtool_fd mtool_mtime
+
+ if {[info exists mtool_fd]} {
+ if {[ask_popup [mc "Merge tool is already running, terminate it?"]] eq {yes}} {
+ catch { kill_file_process $mtool_fd }
+ catch { close $mtool_fd }
+ unset mtool_fd
+
+ set old_backup [lindex $mtool_tmpfiles end]
+ file rename -force -- $old_backup $mtool_target
+ delete_temp_files $mtool_tmpfiles
+ } else {
+ return
+ }
+ }
+
+ # Save the original file
+ file rename -force -- $target $backup
+
+ # Get the blobs; it destroys $target
+ if {[catch {merge_tool_get_stages $target $stages} err]} {
+ file rename -force -- $backup $target
+ delete_temp_files $stages
+ error_popup [mc "Error retrieving versions:\n%s" $err]
+ return
+ }
+
+ # Restore the conflict file
+ file copy -force -- $backup $target
+
+ # Initialize global state
+ set mtool_target $target
+ set mtool_mtime [file mtime $target]
+ set mtool_tmpfiles $stages
+
+ lappend mtool_tmpfiles $backup
+
+ # Force redirection to avoid interpreting output on stderr
+ # as an error, and launch the tool
+ lappend cmdline {2>@1}
+
+ if {[catch { set mtool_fd [_open_stdout_stderr $cmdline] } err]} {
+ delete_temp_files $mtool_tmpfiles
+ error_popup [mc "Could not start the merge tool:\n\n%s" $err]
+ return
+ }
+
+ ui_status [mc "Running merge tool..."]
+
+ fconfigure $mtool_fd -blocking 0 -translation binary -encoding binary
+ fileevent $mtool_fd readable [list read_mtool_output $mtool_fd]
+}
+
+proc read_mtool_output {fd} {
+ global mtool_fd mtool_tmpfiles
+
+ read $fd
+ if {[eof $fd]} {
+ unset mtool_fd
+
+ fconfigure $fd -blocking 1
+ merge_tool_finish $fd
+ }
+}
+
+proc merge_tool_finish {fd} {
+ global mtool_tmpfiles mtool_target mtool_mtime
+
+ set backup [lindex $mtool_tmpfiles end]
+ set failed 0
+
+ # Check the return code
+ if {[catch {close $fd} err]} {
+ set failed 1
+ if {$err ne {child process exited abnormally}} {
+ error_popup [strcat [mc "Merge tool failed."] "\n\n$err"]
+ }
+ }
+
+ # Check the modification time of the target file
+ if {!$failed && [file mtime $mtool_target] eq $mtool_mtime} {
+ if {[ask_popup [mc "File %s unchanged, still accept as resolved?" \
+ [short_path $mtool_target]]] ne {yes}} {
+ set failed 1
+ }
+ }
+
+ # Finish
+ if {$failed} {
+ file rename -force -- $backup $mtool_target
+ delete_temp_files $mtool_tmpfiles
+ ui_status [mc "Merge tool failed."]
+ } else {
+ if {[is_config_true merge.keepbackup]} {
+ file rename -force -- $backup "$mtool_target.orig"
+ }
+
+ delete_temp_files $mtool_tmpfiles
+
+ merge_add_resolution $mtool_target
+ }
+}
{b merge.summary {mc "Summarize Merge Commits"}}
{i-1..5 merge.verbosity {mc "Merge Verbosity"}}
{b merge.diffstat {mc "Show Diffstat After Merge"}}
+ {t merge.tool {mc "Use Merge Tool"}}
{b gui.trustmtime {mc "Trust File Modification Timestamps"}}
{b gui.pruneduringfetch {mc "Prune Tracking Branches During Fetch"}}
{b gui.matchtrackingbranch {mc "Match Tracking Branches"}}
{b gui.fastcopyblame {mc "Blame Copy Only On Changed Files"}}
{i-20..200 gui.copyblamethreshold {mc "Minimum Letters To Blame Copy On"}}
- {i-0..99 gui.diffcontext {mc "Number of Diff Context Lines"}}
+ {i-0..300 gui.blamehistoryctx {mc "Blame History Context Radius (days)"}}
+ {i-1..99 gui.diffcontext {mc "Number of Diff Context Lines"}}
{i-0..99 gui.commitmsgwidth {mc "Commit Message Text Width"}}
{t gui.newbranchtemplate {mc "New Branch Name Template"}}
} {
-# translation of fr.po to French
+# translation of fr.po to Français
# Translation of git-gui to French.
# Copyright (C) 2008 Shawn Pearce, et al.
# This file is distributed under the same license as the git package.
#
# Christian Couder <chriscool@tuxfamily.org>, 2008.
+# Alexandre Bourget <alexandre.bourget@savoirfairelinux.com>, 2008.
msgid ""
msgstr ""
"Project-Id-Version: fr\n"
"Report-Msgid-Bugs-To: \n"
-"POT-Creation-Date: 2008-03-14 07:18+0100\n"
-"PO-Revision-Date: 2008-04-04 22:05+0200\n"
-"Last-Translator: Christian Couder <chriscool@tuxfamily.org>\n"
-"Language-Team: French\n"
+"POT-Creation-Date: 2008-08-02 14:45-0700\n"
+"PO-Revision-Date: 2008-08-11 17:12-0400\n"
+"Last-Translator: Alexandre Bourget <alexandre.bourget@savoirfairelinux.com>\n"
+"Language-Team: Français <fr@li.org>\n"
"MIME-Version: 1.0\n"
"Content-Type: text/plain; charset=UTF-8\n"
"Content-Transfer-Encoding: 8bit\n"
"X-Generator: KBabel 1.11.4\n"
"Plural-Forms: nplurals=2; plural=(n > 1);\n"
-#: git-gui.sh:41 git-gui.sh:634 git-gui.sh:648 git-gui.sh:661 git-gui.sh:744
-#: git-gui.sh:763
+#: git-gui.sh:41 git-gui.sh:688 git-gui.sh:702 git-gui.sh:715 git-gui.sh:798
+#: git-gui.sh:817
msgid "git-gui: fatal error"
msgstr "git-gui: erreur fatale"
-#: git-gui.sh:593
+#: git-gui.sh:644
#, tcl-format
msgid "Invalid font specified in %s:"
-msgstr "Invalide fonte spécifiée dans %s :"
+msgstr "Police invalide spécifiée dans %s :"
-#: git-gui.sh:620
+#: git-gui.sh:674
msgid "Main Font"
-msgstr "Fonte principale"
+msgstr "Police principale"
-#: git-gui.sh:621
+#: git-gui.sh:675
msgid "Diff/Console Font"
-msgstr "Fonte diff/console"
+msgstr "Police diff/console"
-#: git-gui.sh:635
+#: git-gui.sh:689
msgid "Cannot find git in PATH."
msgstr "Impossible de trouver git dans PATH."
-#: git-gui.sh:662
+#: git-gui.sh:716
msgid "Cannot parse Git version string:"
msgstr "Impossible de parser la version de Git :"
-#: git-gui.sh:680
+#: git-gui.sh:734
#, tcl-format
msgid ""
"Git version cannot be determined.\n"
"\n"
"Peut'on considérer que '%s' est en version 1.5.0 ?\n"
-#: git-gui.sh:918
+#: git-gui.sh:972
msgid "Git directory not found:"
-msgstr "Impossible de trouver le répertoire de Git :"
+msgstr "Impossible de trouver le répertoire git :"
-#: git-gui.sh:925
+#: git-gui.sh:979
msgid "Cannot move to top of working directory:"
msgstr "Impossible d'aller à la racine du répertoire de travail :"
-#: git-gui.sh:932
+#: git-gui.sh:986
msgid "Cannot use funny .git directory:"
-msgstr "Impossible d'utiliser un drôle de répertoire git :"
+msgstr "Impossible d'utiliser le répertoire .git:"
-#: git-gui.sh:937
+#: git-gui.sh:991
msgid "No working directory"
-msgstr "Pas de répertoire de travail"
+msgstr "Aucun répertoire de travail"
-#: git-gui.sh:1084 lib/checkout_op.tcl:283
+#: git-gui.sh:1138 lib/checkout_op.tcl:305
msgid "Refreshing file status..."
msgstr "Rafraichissement du status des fichiers..."
-#: git-gui.sh:1149
+#: git-gui.sh:1194
msgid "Scanning for modified files ..."
msgstr "Recherche de fichiers modifiés..."
-#: git-gui.sh:1324 lib/browser.tcl:246
+#: git-gui.sh:1369 lib/browser.tcl:246
msgid "Ready."
msgstr "Prêt."
-#: git-gui.sh:1590
+#: git-gui.sh:1635
msgid "Unmodified"
msgstr "Non modifié"
-#: git-gui.sh:1592
+#: git-gui.sh:1637
msgid "Modified, not staged"
-msgstr "Modifié, non pré-commité"
+msgstr "Modifié, pas indexé"
-#: git-gui.sh:1593 git-gui.sh:1598
+#: git-gui.sh:1638 git-gui.sh:1643
msgid "Staged for commit"
-msgstr "Pré-commité"
+msgstr "Indexé"
-#: git-gui.sh:1594 git-gui.sh:1599
+#: git-gui.sh:1639 git-gui.sh:1644
msgid "Portions staged for commit"
-msgstr "En partie pré-commité"
+msgstr "Portions indexées"
-#: git-gui.sh:1595 git-gui.sh:1600
+#: git-gui.sh:1640 git-gui.sh:1645
msgid "Staged for commit, missing"
-msgstr "Pré-commité, manquant"
+msgstr "Indexés, manquant"
-#: git-gui.sh:1597
+#: git-gui.sh:1642
msgid "Untracked, not staged"
-msgstr "Non suivi, non pré-commité"
+msgstr "Non versionné, non indexé"
-#: git-gui.sh:1602
+#: git-gui.sh:1647
msgid "Missing"
msgstr "Manquant"
-#: git-gui.sh:1603
+#: git-gui.sh:1648
msgid "Staged for removal"
-msgstr "Pré-commité pour suppression"
+msgstr "Indexé pour suppression"
-#: git-gui.sh:1604
+#: git-gui.sh:1649
msgid "Staged for removal, still present"
-msgstr "Pré-commité pour suppression, toujours présent"
+msgstr "Indexé pour suppression, toujours présent"
-#: git-gui.sh:1606 git-gui.sh:1607 git-gui.sh:1608 git-gui.sh:1609
+#: git-gui.sh:1651 git-gui.sh:1652 git-gui.sh:1653 git-gui.sh:1654
msgid "Requires merge resolution"
msgstr "Nécessite la résolution d'une fusion"
-#: git-gui.sh:1644
+#: git-gui.sh:1689
msgid "Starting gitk... please wait..."
-msgstr "Lancement de gitk... merci de patienter..."
+msgstr "Lancement de gitk... un instant..."
-#: git-gui.sh:1653
-#, tcl-format
-msgid ""
-"Unable to start gitk:\n"
-"\n"
-"%s does not exist"
-msgstr ""
-"Impossible de lancer gitk :\n"
-"\n"
-"%s inexistant"
+#: git-gui.sh:1698
+msgid "Couldn't find gitk in PATH"
+msgstr "Impossible de trouver gitk dans PATH."
-#: git-gui.sh:1860 lib/choose_repository.tcl:36
+#: git-gui.sh:1948 lib/choose_repository.tcl:36
msgid "Repository"
-msgstr "Référentiel"
+msgstr "Dépôt"
-#: git-gui.sh:1861
+#: git-gui.sh:1949
msgid "Edit"
-msgstr "Editer"
+msgstr "Edition"
-#: git-gui.sh:1863 lib/choose_rev.tcl:561
+#: git-gui.sh:1951 lib/choose_rev.tcl:561
msgid "Branch"
msgstr "Branche"
-#: git-gui.sh:1866 lib/choose_rev.tcl:548
+#: git-gui.sh:1954 lib/choose_rev.tcl:548
msgid "Commit@@noun"
msgstr "Commit"
-#: git-gui.sh:1869 lib/merge.tcl:120 lib/merge.tcl:149 lib/merge.tcl:167
+#: git-gui.sh:1957 lib/merge.tcl:120 lib/merge.tcl:149 lib/merge.tcl:167
msgid "Merge"
msgstr "Fusionner"
-#: git-gui.sh:1870 lib/choose_rev.tcl:557
+#: git-gui.sh:1958 lib/choose_rev.tcl:557
msgid "Remote"
-msgstr "Référentiel distant"
+msgstr "Dépôt distant"
-#: git-gui.sh:1879
+#: git-gui.sh:1967
msgid "Browse Current Branch's Files"
-msgstr "Visionner fichiers dans branche courante"
+msgstr "Naviguer dans la branche courante"
-#: git-gui.sh:1883
+#: git-gui.sh:1971
msgid "Browse Branch Files..."
-msgstr "Visionner fichiers de branche"
+msgstr "Naviguer dans la branche..."
-#: git-gui.sh:1888
+#: git-gui.sh:1976
msgid "Visualize Current Branch's History"
msgstr "Visualiser historique branche courante"
-#: git-gui.sh:1892
+#: git-gui.sh:1980
msgid "Visualize All Branch History"
-msgstr "Visualiser historique toutes branches"
+msgstr "Voir l'historique de toutes les branches"
-#: git-gui.sh:1899
+#: git-gui.sh:1987
#, tcl-format
msgid "Browse %s's Files"
-msgstr "Visionner fichiers de %s"
+msgstr "Naviguer l'arborescence de %s"
-#: git-gui.sh:1901
+#: git-gui.sh:1989
#, tcl-format
msgid "Visualize %s's History"
-msgstr "Visualiser historique de %s"
+msgstr "Voir l'historique de la branche: %s"
-#: git-gui.sh:1906 lib/database.tcl:27 lib/database.tcl:67
+#: git-gui.sh:1994 lib/database.tcl:27 lib/database.tcl:67
msgid "Database Statistics"
-msgstr "Statistiques base de donnée"
+msgstr "Statistiques du dépôt"
-#: git-gui.sh:1909 lib/database.tcl:34
+#: git-gui.sh:1997 lib/database.tcl:34
msgid "Compress Database"
-msgstr "Comprimer base de donnée"
+msgstr "Comprimer le dépôt"
-#: git-gui.sh:1912
+#: git-gui.sh:2000
msgid "Verify Database"
-msgstr "Vérifier base de donnée"
+msgstr "Vérifier le dépôt"
-#: git-gui.sh:1919 git-gui.sh:1923 git-gui.sh:1927 lib/shortcut.tcl:7
+#: git-gui.sh:2007 git-gui.sh:2011 git-gui.sh:2015 lib/shortcut.tcl:7
#: lib/shortcut.tcl:39 lib/shortcut.tcl:71
msgid "Create Desktop Icon"
msgstr "Créer icône sur bureau"
-#: git-gui.sh:1932 lib/choose_repository.tcl:177 lib/choose_repository.tcl:185
+#: git-gui.sh:2023 lib/choose_repository.tcl:177 lib/choose_repository.tcl:185
msgid "Quit"
msgstr "Quitter"
-#: git-gui.sh:1939
+#: git-gui.sh:2031
msgid "Undo"
msgstr "Défaire"
-#: git-gui.sh:1942
+#: git-gui.sh:2034
msgid "Redo"
msgstr "Refaire"
-#: git-gui.sh:1946 git-gui.sh:2443
+#: git-gui.sh:2038 git-gui.sh:2545
msgid "Cut"
msgstr "Couper"
-#: git-gui.sh:1949 git-gui.sh:2446 git-gui.sh:2520 git-gui.sh:2614
+#: git-gui.sh:2041 git-gui.sh:2548 git-gui.sh:2622 git-gui.sh:2715
#: lib/console.tcl:69
msgid "Copy"
msgstr "Copier"
-#: git-gui.sh:1952 git-gui.sh:2449
+#: git-gui.sh:2044 git-gui.sh:2551
msgid "Paste"
msgstr "Coller"
-#: git-gui.sh:1955 git-gui.sh:2452 lib/branch_delete.tcl:26
+#: git-gui.sh:2047 git-gui.sh:2554 lib/branch_delete.tcl:26
#: lib/remote_branch_delete.tcl:38
msgid "Delete"
msgstr "Supprimer"
-#: git-gui.sh:1959 git-gui.sh:2456 git-gui.sh:2618 lib/console.tcl:71
+#: git-gui.sh:2051 git-gui.sh:2558 git-gui.sh:2719 lib/console.tcl:71
msgid "Select All"
msgstr "Tout sélectionner"
-#: git-gui.sh:1968
+#: git-gui.sh:2060
msgid "Create..."
msgstr "Créer..."
-#: git-gui.sh:1974
+#: git-gui.sh:2066
msgid "Checkout..."
-msgstr "Emprunter... "
+msgstr "Charger (checkout)..."
-#: git-gui.sh:1980
+#: git-gui.sh:2072
msgid "Rename..."
msgstr "Renommer..."
-#: git-gui.sh:1985 git-gui.sh:2085
+#: git-gui.sh:2077 git-gui.sh:2187
msgid "Delete..."
msgstr "Supprimer..."
-#: git-gui.sh:1990
+#: git-gui.sh:2082
msgid "Reset..."
msgstr "Réinitialiser..."
-#: git-gui.sh:2002 git-gui.sh:2389
+#: git-gui.sh:2094 git-gui.sh:2491
msgid "New Commit"
msgstr "Nouveau commit"
-#: git-gui.sh:2010 git-gui.sh:2396
+#: git-gui.sh:2102 git-gui.sh:2498
msgid "Amend Last Commit"
msgstr "Corriger dernier commit"
-#: git-gui.sh:2019 git-gui.sh:2356 lib/remote_branch_delete.tcl:99
+#: git-gui.sh:2111 git-gui.sh:2458 lib/remote_branch_delete.tcl:99
msgid "Rescan"
-msgstr "Resynchroniser"
+msgstr "Recharger modifs."
-#: git-gui.sh:2025
+#: git-gui.sh:2117
msgid "Stage To Commit"
-msgstr "Commiter un pré-commit"
+msgstr "Indexer"
-#: git-gui.sh:2031
+#: git-gui.sh:2123
msgid "Stage Changed Files To Commit"
-msgstr "Commiter fichiers modifiés dans pré-commit"
+msgstr "Indexer toutes modifications"
-#: git-gui.sh:2037
+#: git-gui.sh:2129
msgid "Unstage From Commit"
-msgstr "Commit vers pré-commit"
+msgstr "Désindexer"
-#: git-gui.sh:2042 lib/index.tcl:395
+#: git-gui.sh:2134 lib/index.tcl:395
msgid "Revert Changes"
-msgstr "Inverser modification"
+msgstr "Annuler les modifications (revert)"
-#: git-gui.sh:2049 git-gui.sh:2368 git-gui.sh:2467
+#: git-gui.sh:2141 git-gui.sh:2702
+msgid "Show Less Context"
+msgstr "Montrer moins de contexte"
+
+#: git-gui.sh:2145 git-gui.sh:2706
+msgid "Show More Context"
+msgstr "Montrer plus de contexte"
+
+#: git-gui.sh:2151 git-gui.sh:2470 git-gui.sh:2569
msgid "Sign Off"
-msgstr "Se désinscrire"
+msgstr "Signer"
-#: git-gui.sh:2053 git-gui.sh:2372
+#: git-gui.sh:2155 git-gui.sh:2474
msgid "Commit@@verb"
msgstr "Commiter"
-#: git-gui.sh:2064
+#: git-gui.sh:2166
msgid "Local Merge..."
msgstr "Fusion locale..."
-#: git-gui.sh:2069
+#: git-gui.sh:2171
msgid "Abort Merge..."
msgstr "Abandonner fusion..."
-#: git-gui.sh:2081
+#: git-gui.sh:2183
msgid "Push..."
msgstr "Pousser..."
-#: git-gui.sh:2092 lib/choose_repository.tcl:41
-msgid "Apple"
-msgstr "Pomme"
-
-#: git-gui.sh:2095 git-gui.sh:2117 lib/about.tcl:14
+#: git-gui.sh:2197 git-gui.sh:2219 lib/about.tcl:14
#: lib/choose_repository.tcl:44 lib/choose_repository.tcl:50
#, tcl-format
msgid "About %s"
-msgstr "A propos de %s"
+msgstr "À propos de %s"
-#: git-gui.sh:2099
+#: git-gui.sh:2201
msgid "Preferences..."
msgstr "Préférences..."
-#: git-gui.sh:2107 git-gui.sh:2639
+#: git-gui.sh:2209 git-gui.sh:2740
msgid "Options..."
msgstr "Options..."
-#: git-gui.sh:2113 lib/choose_repository.tcl:47
+#: git-gui.sh:2215 lib/choose_repository.tcl:47
msgid "Help"
msgstr "Aide"
-#: git-gui.sh:2154
+#: git-gui.sh:2256
msgid "Online Documentation"
msgstr "Documentation en ligne"
-#: git-gui.sh:2238
+#: git-gui.sh:2340
#, tcl-format
msgid "fatal: cannot stat path %s: No such file or directory"
-msgstr "erreur fatale : pas d'infos sur le chemin %s : Fichier ou répertoire inexistant"
+msgstr ""
+"erreur fatale : pas d'infos sur le chemin %s : Fichier ou répertoire "
+"inexistant"
-#: git-gui.sh:2271
+#: git-gui.sh:2373
msgid "Current Branch:"
msgstr "Branche courante :"
-#: git-gui.sh:2292
+#: git-gui.sh:2394
msgid "Staged Changes (Will Commit)"
-msgstr "Modifications pré-commitées"
+msgstr "Modifs. indexées (pour commit)"
-#: git-gui.sh:2312
+#: git-gui.sh:2414
msgid "Unstaged Changes"
-msgstr "Modifications non pré-commitées"
+msgstr "Modifs. non indexées"
-#: git-gui.sh:2362
+#: git-gui.sh:2464
msgid "Stage Changed"
-msgstr "Pré-commit modifié"
+msgstr "Indexer modifs."
-#: git-gui.sh:2378 lib/transport.tcl:93 lib/transport.tcl:182
+#: git-gui.sh:2480 lib/transport.tcl:93 lib/transport.tcl:182
msgid "Push"
msgstr "Pousser"
-#: git-gui.sh:2408
+#: git-gui.sh:2510
msgid "Initial Commit Message:"
msgstr "Message de commit initial :"
-#: git-gui.sh:2409
+#: git-gui.sh:2511
msgid "Amended Commit Message:"
msgstr "Message de commit corrigé :"
-#: git-gui.sh:2410
+#: git-gui.sh:2512
msgid "Amended Initial Commit Message:"
msgstr "Message de commit initial corrigé :"
-#: git-gui.sh:2411
+#: git-gui.sh:2513
msgid "Amended Merge Commit Message:"
msgstr "Message de commit de fusion corrigé :"
-#: git-gui.sh:2412
+#: git-gui.sh:2514
msgid "Merge Commit Message:"
msgstr "Message de commit de fusion :"
-#: git-gui.sh:2413
+#: git-gui.sh:2515
msgid "Commit Message:"
msgstr "Message de commit :"
-#: git-gui.sh:2459 git-gui.sh:2622 lib/console.tcl:73
+#: git-gui.sh:2561 git-gui.sh:2723 lib/console.tcl:73
msgid "Copy All"
msgstr "Copier tout"
-#: git-gui.sh:2483 lib/blame.tcl:107
+#: git-gui.sh:2585 lib/blame.tcl:100
msgid "File:"
msgstr "Fichier :"
-#: git-gui.sh:2589
+#: git-gui.sh:2691
msgid "Apply/Reverse Hunk"
msgstr "Appliquer/Inverser section"
-#: git-gui.sh:2595
-msgid "Show Less Context"
-msgstr "Montrer moins de contexte"
-
-#: git-gui.sh:2602
-msgid "Show More Context"
-msgstr "Montrer plus de contexte"
+#: git-gui.sh:2696
+msgid "Apply/Reverse Line"
+msgstr "Appliquer/Inverser la ligne"
-#: git-gui.sh:2610
+#: git-gui.sh:2711
msgid "Refresh"
msgstr "Rafraichir"
-#: git-gui.sh:2631
+#: git-gui.sh:2732
msgid "Decrease Font Size"
-msgstr "Réduire fonte"
+msgstr "Diminuer la police"
-#: git-gui.sh:2635
+#: git-gui.sh:2736
msgid "Increase Font Size"
-msgstr "Agrandir fonte"
+msgstr "Agrandir la police"
-#: git-gui.sh:2646
+#: git-gui.sh:2747
msgid "Unstage Hunk From Commit"
-msgstr "Enlever section pré-commitée"
+msgstr "Désindexer la section"
+
+#: git-gui.sh:2748
+msgid "Unstage Line From Commit"
+msgstr "Désindexer la ligne"
-#: git-gui.sh:2648
+#: git-gui.sh:2750
msgid "Stage Hunk For Commit"
-msgstr "Pré-commiter section"
+msgstr "Indexer la section"
-#: git-gui.sh:2667
+#: git-gui.sh:2751
+msgid "Stage Line For Commit"
+msgstr "Indexer la ligne"
+
+#: git-gui.sh:2771
msgid "Initializing..."
msgstr "Initialisation..."
-#: git-gui.sh:2762
+#: git-gui.sh:2876
#, tcl-format
msgid ""
"Possible environment issues exist.\n"
"sous-processus de Git lancés par %s\n"
"\n"
-#: git-gui.sh:2792
+#: git-gui.sh:2906
msgid ""
"\n"
"This is due to a known issue with the\n"
"Ceci est du à un problème connu avec\n"
"le binaire Tcl distribué par Cygwin."
-#: git-gui.sh:2797
+#: git-gui.sh:2911
#, tcl-format
msgid ""
"\n"
msgid "git-gui - a graphical user interface for Git."
msgstr "git-gui - une interface graphique utilisateur pour Git"
-#: lib/blame.tcl:77
+#: lib/blame.tcl:70
msgid "File Viewer"
msgstr "Visionneur de fichier"
-#: lib/blame.tcl:81
+#: lib/blame.tcl:74
msgid "Commit:"
msgstr "Commit :"
-#: lib/blame.tcl:264
+#: lib/blame.tcl:257
msgid "Copy Commit"
msgstr "Copier commit"
-#: lib/blame.tcl:384
+#: lib/blame.tcl:260
+msgid "Do Full Copy Detection"
+msgstr "Lancer la détection approfondie des copies"
+
+#: lib/blame.tcl:388
#, tcl-format
msgid "Reading %s..."
msgstr "Lecture de %s..."
-#: lib/blame.tcl:488
+#: lib/blame.tcl:492
msgid "Loading copy/move tracking annotations..."
msgstr "Chargement des annotations de suivi des copies/déplacements..."
-#: lib/blame.tcl:508
+#: lib/blame.tcl:512
msgid "lines annotated"
msgstr "lignes annotées"
-#: lib/blame.tcl:689
+#: lib/blame.tcl:704
msgid "Loading original location annotations..."
msgstr "Chargement des annotations d'emplacement original"
-#: lib/blame.tcl:692
+#: lib/blame.tcl:707
msgid "Annotation complete."
msgstr "Annotation terminée."
-#: lib/blame.tcl:746
+#: lib/blame.tcl:737
+msgid "Busy"
+msgstr "Occupé"
+
+#: lib/blame.tcl:738
+msgid "Annotation process is already running."
+msgstr "Annotation en cours d'exécution."
+
+#: lib/blame.tcl:777
+msgid "Running thorough copy detection..."
+msgstr "Recherche de copie approfondie en cours..."
+
+#: lib/blame.tcl:827
msgid "Loading annotation..."
msgstr "Chargement des annotations..."
-#: lib/blame.tcl:802
+#: lib/blame.tcl:883
msgid "Author:"
msgstr "Auteur :"
-#: lib/blame.tcl:806
+#: lib/blame.tcl:887
msgid "Committer:"
msgstr "Commiteur :"
-#: lib/blame.tcl:811
+#: lib/blame.tcl:892
msgid "Original File:"
msgstr "Fichier original :"
-#: lib/blame.tcl:925
+#: lib/blame.tcl:1006
msgid "Originally By:"
msgstr "A l'origine par :"
-#: lib/blame.tcl:931
+#: lib/blame.tcl:1012
msgid "In File:"
msgstr "Dans le fichier :"
-#: lib/blame.tcl:936
+#: lib/blame.tcl:1017
msgid "Copied Or Moved Here By:"
msgstr "Copié ou déplacé ici par :"
#: lib/branch_checkout.tcl:14 lib/branch_checkout.tcl:19
msgid "Checkout Branch"
-msgstr "Emprunter branche"
+msgstr "Charger la branche (checkout)"
#: lib/branch_checkout.tcl:23
msgid "Checkout"
-msgstr "Emprunter"
+msgstr "Charger (checkout)"
#: lib/branch_checkout.tcl:27 lib/branch_create.tcl:35
#: lib/branch_delete.tcl:32 lib/branch_rename.tcl:30 lib/browser.tcl:282
-#: lib/checkout_op.tcl:522 lib/choose_font.tcl:43 lib/merge.tcl:171
+#: lib/checkout_op.tcl:544 lib/choose_font.tcl:43 lib/merge.tcl:171
#: lib/option.tcl:103 lib/remote_branch_delete.tcl:42 lib/transport.tcl:97
msgid "Cancel"
msgstr "Annuler"
msgid "Revision"
msgstr "Révision"
-#: lib/branch_checkout.tcl:36 lib/branch_create.tcl:69 lib/option.tcl:242
+#: lib/branch_checkout.tcl:36 lib/branch_create.tcl:69 lib/option.tcl:244
msgid "Options"
msgstr "Options"
#: lib/branch_checkout.tcl:39 lib/branch_create.tcl:92
msgid "Fetch Tracking Branch"
-msgstr "Branche suivant récupération"
+msgstr "Récupérer la branche de suivi"
#: lib/branch_checkout.tcl:44
msgid "Detach From Local Branch"
-msgstr "Détacher de branche locale"
+msgstr "Détacher de la branche locale"
#: lib/branch_create.tcl:22
msgid "Create Branch"
#: lib/branch_create.tcl:66
msgid "Starting Revision"
-msgstr "Début de révision"
+msgstr "Révision initiale"
#: lib/branch_create.tcl:72
msgid "Update Existing Branch:"
#: lib/branch_create.tcl:80
msgid "Fast Forward Only"
-msgstr "Avance rapide seulement"
+msgstr "Mise-à-jour rectiligne seulement (fast-forward)"
-#: lib/branch_create.tcl:85 lib/checkout_op.tcl:514
+#: lib/branch_create.tcl:85 lib/checkout_op.tcl:536
msgid "Reset"
msgstr "Réinitialiser"
#: lib/branch_create.tcl:97
msgid "Checkout After Creation"
-msgstr "Emprunt après création"
+msgstr "Charger (checkout) après création"
#: lib/branch_create.tcl:131
msgid "Please select a tracking branch."
-msgstr "Merci de choisir une branche de suivi"
+msgstr "Choisissez une branche de suivi"
#: lib/branch_create.tcl:140
#, tcl-format
msgid "Tracking branch %s is not a branch in the remote repository."
-msgstr "La branche de suivi %s n'est pas une branche dans le référentiel distant."
+msgstr "La branche de suivi %s n'est pas une branche dans le dépôt distant."
#: lib/branch_create.tcl:153 lib/branch_rename.tcl:86
msgid "Please supply a branch name."
-msgstr "Merci de fournir un nom de branche."
+msgstr "Fournissez un nom de branche."
#: lib/branch_create.tcl:164 lib/branch_rename.tcl:106
#, tcl-format
#: lib/branch_delete.tcl:52
msgid "Delete Only If Merged Into"
-msgstr "Supprimer ssi fusion dedans"
+msgstr "Supprimer seulement si fusionnée dans:"
#: lib/branch_delete.tcl:54
msgid "Always (Do not perform merge test.)"
msgid "Please select a branch to rename."
msgstr "Merci de sélectionner une branche à renommer."
-#: lib/branch_rename.tcl:96 lib/checkout_op.tcl:179
+#: lib/branch_rename.tcl:96 lib/checkout_op.tcl:201
#, tcl-format
msgid "Branch '%s' already exists."
msgstr "La branche '%s' existe déjà."
#: lib/branch_rename.tcl:117
#, tcl-format
msgid "Failed to rename '%s'."
-msgstr "Le renommage de '%s' a échoué."
+msgstr "Échec pour renommer '%s'."
#: lib/browser.tcl:17
msgid "Starting..."
#: lib/browser.tcl:267 lib/browser.tcl:273
msgid "Browse Branch Files"
-msgstr "Visionner fichiers de branches"
+msgstr "Naviguer dans les fichiers de le branche"
#: lib/browser.tcl:278 lib/choose_repository.tcl:387
-#: lib/choose_repository.tcl:474 lib/choose_repository.tcl:484
-#: lib/choose_repository.tcl:987
+#: lib/choose_repository.tcl:472 lib/choose_repository.tcl:482
+#: lib/choose_repository.tcl:985
msgid "Browse"
-msgstr "Visionner"
+msgstr "Naviguer"
-#: lib/checkout_op.tcl:79
+#: lib/checkout_op.tcl:84
#, tcl-format
msgid "Fetching %s from %s"
msgstr "Récupération de %s à partir de %s"
-#: lib/checkout_op.tcl:127
+#: lib/checkout_op.tcl:132
#, tcl-format
msgid "fatal: Cannot resolve %s"
msgstr "erreur fatale : Impossible de résoudre %s"
-#: lib/checkout_op.tcl:140 lib/console.tcl:81 lib/database.tcl:31
+#: lib/checkout_op.tcl:145 lib/console.tcl:81 lib/database.tcl:31
msgid "Close"
msgstr "Fermer"
-#: lib/checkout_op.tcl:169
+#: lib/checkout_op.tcl:174
#, tcl-format
msgid "Branch '%s' does not exist."
msgstr "La branche '%s' n'existe pas."
-#: lib/checkout_op.tcl:206
+#: lib/checkout_op.tcl:193
+#, tcl-format
+msgid "Failed to configure simplified git-pull for '%s'."
+msgstr "Échec de la configuration simplifiée de git-pull pour '%s'."
+
+#: lib/checkout_op.tcl:228
#, tcl-format
msgid ""
"Branch '%s' already exists.\n"
msgstr ""
"La branche '%s' existe déjà.\n"
"\n"
-"Impossible d'avancer rapidement à %s.\n"
+"Impossible de faire une avance rapide (fast forward) vers %s.\n"
"Une fusion est nécessaire."
-#: lib/checkout_op.tcl:220
+#: lib/checkout_op.tcl:242
#, tcl-format
msgid "Merge strategy '%s' not supported."
msgstr "La stratégie de fusion '%s' n'est pas supportée."
-#: lib/checkout_op.tcl:239
+#: lib/checkout_op.tcl:261
#, tcl-format
msgid "Failed to update '%s'."
msgstr "La mise à jour de '%s' a échouée."
-#: lib/checkout_op.tcl:251
+#: lib/checkout_op.tcl:273
msgid "Staging area (index) is already locked."
-msgstr "L'espace de pré-commit ('index' ou 'staging') est déjà vérouillé."
+msgstr "L'index (staging area) est déjà vérouillé"
-#: lib/checkout_op.tcl:266
+#: lib/checkout_op.tcl:288
msgid ""
"Last scanned state does not match repository state.\n"
"\n"
"\n"
"The rescan will be automatically started now.\n"
msgstr ""
-"L'état lors de la dernière synchronisation ne correspond plus à l'état du référentiel.\n"
+"L'état lors de la dernière synchronisation ne correspond plus à l'état du "
+"dépôt\n"
"\n"
-"Un autre programme Git a modifié ce référentiel depuis la dernière synchronisation. Une resynchronisation doit être effectuée avant de pouvoir modifier la branche courante.\n"
+"Un autre programme Git a modifié ce dépôt depuis la dernière "
+"synchronisation. Une resynchronisation doit être effectuée avant de pouvoir "
+"modifier la branche courante.\n"
"\n"
"Cela va être fait tout de suite automatiquement.\n"
-#: lib/checkout_op.tcl:322
+#: lib/checkout_op.tcl:344
#, tcl-format
msgid "Updating working directory to '%s'..."
msgstr "Mise à jour du répertoire courant avec '%s'..."
-#: lib/checkout_op.tcl:323
+#: lib/checkout_op.tcl:345
msgid "files checked out"
-msgstr "fichiers empruntés"
+msgstr "fichiers chargés"
-#: lib/checkout_op.tcl:353
+#: lib/checkout_op.tcl:375
#, tcl-format
msgid "Aborted checkout of '%s' (file level merging is required)."
-msgstr "Emprunt de '%s' abandonné. (Il est nécessaire de fusionner des fichiers.)"
+msgstr "Chargement de '%s' abandonné (il est nécessaire de fusionner des fichiers)."
-#: lib/checkout_op.tcl:354
+#: lib/checkout_op.tcl:376
msgid "File level merge required."
msgstr "Il est nécessaire de fusionner des fichiers."
-#: lib/checkout_op.tcl:358
+#: lib/checkout_op.tcl:380
#, tcl-format
msgid "Staying on branch '%s'."
msgstr "Le répertoire de travail reste sur la branche '%s'."
-#: lib/checkout_op.tcl:429
+#: lib/checkout_op.tcl:451
msgid ""
"You are no longer on a local branch.\n"
"\n"
"Si vous vouliez être sur une branche, créez en une maintenant en partant de "
"'Cet emprunt détaché'."
-#: lib/checkout_op.tcl:446 lib/checkout_op.tcl:450
+#: lib/checkout_op.tcl:468 lib/checkout_op.tcl:472
#, tcl-format
msgid "Checked out '%s'."
-msgstr "'%s' emprunté."
+msgstr "'%s' chargé."
-#: lib/checkout_op.tcl:478
+#: lib/checkout_op.tcl:500
#, tcl-format
msgid "Resetting '%s' to '%s' will lose the following commits:"
msgstr "Réinitialiser '%s' à '%s' va faire perdre les commits suivants :"
-#: lib/checkout_op.tcl:500
+#: lib/checkout_op.tcl:522
msgid "Recovering lost commits may not be easy."
msgstr "Récupérer les commits perdus ne sera peut être pas facile."
-#: lib/checkout_op.tcl:505
+#: lib/checkout_op.tcl:527
#, tcl-format
msgid "Reset '%s'?"
msgstr "Réinitialiser '%s' ?"
-#: lib/checkout_op.tcl:510 lib/merge.tcl:163
+#: lib/checkout_op.tcl:532 lib/merge.tcl:163
msgid "Visualize"
msgstr "Visualiser"
-#: lib/checkout_op.tcl:578
+#: lib/checkout_op.tcl:600
#, tcl-format
msgid ""
"Failed to set current branch.\n"
#: lib/choose_font.tcl:53
msgid "Font Family"
-msgstr "Famille de fonte"
+msgstr "Familles de polices"
#: lib/choose_font.tcl:74
msgid "Font Size"
-msgstr "Taille de fonte"
+msgstr "Taille de police"
#: lib/choose_font.tcl:91
msgid "Font Example"
-msgstr "Exemple de fonte"
+msgstr "Exemple de police"
#: lib/choose_font.tcl:103
msgid ""
"If you like this text, it can be your font."
msgstr ""
"C'est un texte d'exemple.\n"
-"Si vous aimez ce texte, vous pouvez choisir cette fonte."
+"Si vous aimez ce texte, vous pouvez choisir cette police"
#: lib/choose_repository.tcl:28
msgid "Git Gui"
#: lib/choose_repository.tcl:81 lib/choose_repository.tcl:376
msgid "Create New Repository"
-msgstr "Créer nouveau référentiel"
+msgstr "Créer nouveau dépôt"
#: lib/choose_repository.tcl:87
msgid "New..."
msgstr "Nouveau..."
-#: lib/choose_repository.tcl:94 lib/choose_repository.tcl:460
+#: lib/choose_repository.tcl:94 lib/choose_repository.tcl:458
msgid "Clone Existing Repository"
-msgstr "Cloner référentiel existant"
+msgstr "Cloner dépôt existant"
#: lib/choose_repository.tcl:100
msgid "Clone..."
msgstr "Cloner..."
-#: lib/choose_repository.tcl:107 lib/choose_repository.tcl:976
+#: lib/choose_repository.tcl:107 lib/choose_repository.tcl:974
msgid "Open Existing Repository"
-msgstr "Ouvrir référentiel existant"
+msgstr "Ouvrir dépôt existant"
#: lib/choose_repository.tcl:113
msgid "Open..."
#: lib/choose_repository.tcl:126
msgid "Recent Repositories"
-msgstr "Référentiels récents"
+msgstr "Dépôt récemment utilisés"
#: lib/choose_repository.tcl:132
msgid "Open Recent Repository:"
-msgstr "Ouvrir référentiel récent :"
+msgstr "Ouvrir dépôt récent :"
#: lib/choose_repository.tcl:296 lib/choose_repository.tcl:303
#: lib/choose_repository.tcl:310
#, tcl-format
msgid "Failed to create repository %s:"
-msgstr "La création du référentiel %s a échouée :"
+msgstr "La création du dépôt %s a échouée :"
-#: lib/choose_repository.tcl:381 lib/choose_repository.tcl:478
+#: lib/choose_repository.tcl:381 lib/choose_repository.tcl:476
msgid "Directory:"
msgstr "Répertoire :"
-#: lib/choose_repository.tcl:412 lib/choose_repository.tcl:537
-#: lib/choose_repository.tcl:1011
+#: lib/choose_repository.tcl:410 lib/choose_repository.tcl:535
+#: lib/choose_repository.tcl:1007
msgid "Git Repository"
-msgstr "Référentiel Git"
+msgstr "Dépôt Git"
-#: lib/choose_repository.tcl:437
+#: lib/choose_repository.tcl:435
#, tcl-format
msgid "Directory %s already exists."
msgstr "Le répertoire %s existe déjà."
-#: lib/choose_repository.tcl:441
+#: lib/choose_repository.tcl:439
#, tcl-format
msgid "File %s already exists."
msgstr "Le fichier %s existe déjà."
-#: lib/choose_repository.tcl:455
+#: lib/choose_repository.tcl:453
msgid "Clone"
msgstr "Cloner"
-#: lib/choose_repository.tcl:468
+#: lib/choose_repository.tcl:466
msgid "URL:"
msgstr "URL :"
-#: lib/choose_repository.tcl:489
+#: lib/choose_repository.tcl:487
msgid "Clone Type:"
msgstr "Type de clonage :"
-#: lib/choose_repository.tcl:495
+#: lib/choose_repository.tcl:493
msgid "Standard (Fast, Semi-Redundant, Hardlinks)"
msgstr "Standard (rapide, semi-redondant, liens durs)"
-#: lib/choose_repository.tcl:501
+#: lib/choose_repository.tcl:499
msgid "Full Copy (Slower, Redundant Backup)"
msgstr "Copy complète (plus lent, sauvegarde redondante)"
-#: lib/choose_repository.tcl:507
+#: lib/choose_repository.tcl:505
msgid "Shared (Fastest, Not Recommended, No Backup)"
msgstr "Partagé (le plus rapide, non recommandé, pas de sauvegarde)"
-#: lib/choose_repository.tcl:543 lib/choose_repository.tcl:590
-#: lib/choose_repository.tcl:736 lib/choose_repository.tcl:806
-#: lib/choose_repository.tcl:1017 lib/choose_repository.tcl:1025
+#: lib/choose_repository.tcl:541 lib/choose_repository.tcl:588
+#: lib/choose_repository.tcl:734 lib/choose_repository.tcl:804
+#: lib/choose_repository.tcl:1013 lib/choose_repository.tcl:1021
#, tcl-format
msgid "Not a Git repository: %s"
-msgstr "'%s' n'est pas un référentiel Git."
+msgstr "'%s' n'est pas un dépôt Git."
-#: lib/choose_repository.tcl:579
+#: lib/choose_repository.tcl:577
msgid "Standard only available for local repository."
-msgstr "Standard n'est disponible que pour un référentiel local."
+msgstr "Standard n'est disponible que pour un dépôt local."
-#: lib/choose_repository.tcl:583
+#: lib/choose_repository.tcl:581
msgid "Shared only available for local repository."
-msgstr "Partagé n'est disponible que pour un référentiel local."
+msgstr "Partagé n'est disponible que pour un dépôt local."
-#: lib/choose_repository.tcl:604
+#: lib/choose_repository.tcl:602
#, tcl-format
msgid "Location %s already exists."
msgstr "L'emplacement %s existe déjà."
-#: lib/choose_repository.tcl:615
+#: lib/choose_repository.tcl:613
msgid "Failed to configure origin"
msgstr "La configuration de l'origine a échouée."
-#: lib/choose_repository.tcl:627
+#: lib/choose_repository.tcl:625
msgid "Counting objects"
-msgstr "Comptage des objets"
+msgstr "Décompte des objets"
-#: lib/choose_repository.tcl:628
+#: lib/choose_repository.tcl:626
msgid "buckets"
msgstr "paniers"
-#: lib/choose_repository.tcl:652
+#: lib/choose_repository.tcl:650
#, tcl-format
msgid "Unable to copy objects/info/alternates: %s"
msgstr "Impossible de copier 'objects/info/alternates' : %s"
-#: lib/choose_repository.tcl:688
+#: lib/choose_repository.tcl:686
#, tcl-format
msgid "Nothing to clone from %s."
msgstr "Il n'y a rien à cloner depuis %s."
-#: lib/choose_repository.tcl:690 lib/choose_repository.tcl:904
-#: lib/choose_repository.tcl:916
+#: lib/choose_repository.tcl:688 lib/choose_repository.tcl:902
+#: lib/choose_repository.tcl:914
msgid "The 'master' branch has not been initialized."
-msgstr "Cette branche 'master' n'a pas été initialisée."
+msgstr "La branche 'master' n'a pas été initialisée."
-#: lib/choose_repository.tcl:703
+#: lib/choose_repository.tcl:701
msgid "Hardlinks are unavailable. Falling back to copying."
-msgstr "Les liens durs ne sont pas disponibles. On se résoud à copier."
+msgstr "Les liens durs ne sont pas supportés. Une copie sera effectuée à la place."
-#: lib/choose_repository.tcl:715
+#: lib/choose_repository.tcl:713
#, tcl-format
msgid "Cloning from %s"
msgstr "Clonage depuis %s"
-#: lib/choose_repository.tcl:746
+#: lib/choose_repository.tcl:744
msgid "Copying objects"
msgstr "Copie des objets"
-#: lib/choose_repository.tcl:747
+#: lib/choose_repository.tcl:745
msgid "KiB"
msgstr "KiB"
-#: lib/choose_repository.tcl:771
+#: lib/choose_repository.tcl:769
#, tcl-format
msgid "Unable to copy object: %s"
msgstr "Impossible de copier l'objet : %s"
-#: lib/choose_repository.tcl:781
+#: lib/choose_repository.tcl:779
msgid "Linking objects"
msgstr "Liaison des objets"
-#: lib/choose_repository.tcl:782
+#: lib/choose_repository.tcl:780
msgid "objects"
msgstr "objets"
-#: lib/choose_repository.tcl:790
+#: lib/choose_repository.tcl:788
#, tcl-format
msgid "Unable to hardlink object: %s"
msgstr "Impossible créer un lien dur pour l'objet : %s"
-#: lib/choose_repository.tcl:845
+#: lib/choose_repository.tcl:843
msgid "Cannot fetch branches and objects. See console output for details."
msgstr ""
"Impossible de récupérer les branches et objets. Voir la sortie console pour "
"plus de détails."
-#: lib/choose_repository.tcl:856
+#: lib/choose_repository.tcl:854
msgid "Cannot fetch tags. See console output for details."
msgstr ""
-"Impossible de récupérer les marques. Voir la sortie console pour plus de "
-"détails."
+"Impossible de récupérer les marques (tags). Voir la sortie console pour plus "
+"de détails."
-#: lib/choose_repository.tcl:880
+#: lib/choose_repository.tcl:878
msgid "Cannot determine HEAD. See console output for details."
msgstr "Impossible de déterminer HEAD. Voir la sortie console pour plus de détails."
-#: lib/choose_repository.tcl:889
+#: lib/choose_repository.tcl:887
#, tcl-format
msgid "Unable to cleanup %s"
msgstr "Impossible de nettoyer %s"
-#: lib/choose_repository.tcl:895
+#: lib/choose_repository.tcl:893
msgid "Clone failed."
msgstr "Le clonage a échoué."
-#: lib/choose_repository.tcl:902
+#: lib/choose_repository.tcl:900
msgid "No default branch obtained."
msgstr "Aucune branche par défaut n'a été obtenue."
-#: lib/choose_repository.tcl:913
+#: lib/choose_repository.tcl:911
#, tcl-format
msgid "Cannot resolve %s as a commit."
msgstr "Impossible de résoudre %s comme commit."
-#: lib/choose_repository.tcl:925
+#: lib/choose_repository.tcl:923
msgid "Creating working directory"
msgstr "Création du répertoire de travail"
-#: lib/choose_repository.tcl:926 lib/index.tcl:65 lib/index.tcl:127
+#: lib/choose_repository.tcl:924 lib/index.tcl:65 lib/index.tcl:127
#: lib/index.tcl:193
msgid "files"
msgstr "fichiers"
-#: lib/choose_repository.tcl:955
+#: lib/choose_repository.tcl:953
msgid "Initial file checkout failed."
-msgstr "L'emprunt initial de fichier a échoué."
+msgstr "Chargement initial du fichier échoué."
-#: lib/choose_repository.tcl:971
+#: lib/choose_repository.tcl:969
msgid "Open"
msgstr "Ouvrir"
-#: lib/choose_repository.tcl:981
+#: lib/choose_repository.tcl:979
msgid "Repository:"
-msgstr "Référentiel :"
+msgstr "Dépôt :"
-#: lib/choose_repository.tcl:1031
+#: lib/choose_repository.tcl:1027
#, tcl-format
msgid "Failed to open repository %s:"
-msgstr "Impossible d'ouvrir le référentiel %s :"
+msgstr "Impossible d'ouvrir le dépôt %s :"
#: lib/choose_rev.tcl:53
msgid "This Detached Checkout"
#: lib/choose_rev.tcl:79
msgid "Tracking Branch"
-msgstr "Suivi de branche"
+msgstr "Branche de suivi"
#: lib/choose_rev.tcl:84 lib/choose_rev.tcl:538
msgid "Tag"
-msgstr "Marque"
+msgstr "Marque (tag)"
#: lib/choose_rev.tcl:317
#, tcl-format
#: lib/choose_rev.tcl:531
msgid "Updated"
-msgstr "Misa à jour"
+msgstr "Mise-à-jour:"
#: lib/choose_rev.tcl:559
msgid "URL"
"The rescan will be automatically started now.\n"
msgstr ""
"L'état lors de la dernière synchronisation ne correspond plus à l'état du "
-"référentiel.\n"
+"dépôt.\n"
"\n"
-"Un autre programme Git a modifié ce référentiel depuis la dernière "
+"Un autre programme Git a modifié ce dépôt depuis la dernière "
"synchronisation. Une resynshronisation doit être effectuée avant de pouvoir "
"créer un nouveau commit.\n"
"\n"
msgstr ""
"Pas de modification à commiter.\n"
"\n"
-"Vous devez pré-commiter au moins 1 fichier avant de pouvoir commiter.\n"
+"Vous devez indexer au moins 1 fichier avant de pouvoir commiter.\n"
#: lib/commit.tcl:183
msgid ""
#: lib/commit.tcl:221
msgid "Calling pre-commit hook..."
-msgstr "Appel du programme externe d'avant commit..."
+msgstr "Lancement de l'action d'avant-commit..."
#: lib/commit.tcl:236
msgid "Commit declined by pre-commit hook."
-msgstr "Commit refusé par le programme externe d'avant commit."
+msgstr "Commit refusé par l'action d'avant-commit."
#: lib/commit.tcl:259
msgid "Calling commit-msg hook..."
-msgstr "Appel du programme externe de message de commit..."
+msgstr "Lancement de l'action \"message de commit\"..."
#: lib/commit.tcl:274
msgid "Commit declined by commit-msg hook."
-msgstr "Commit refusé par le programme externe de message de commit."
+msgstr "Commit refusé par l'action \"message de commit\"."
#: lib/commit.tcl:287
msgid "Committing changes..."
"\n"
"Compress the database now?"
msgstr ""
-"Ce référentiel comprend actuellement environ %i objets ayant leur fichier "
+"Ce dépôt comprend actuellement environ %i objets ayant leur fichier "
"particulier.\n"
"\n"
"Pour conserver une performance optimale, il est fortement recommandé de "
msgid "Invalid date from Git: %s"
msgstr "Date invalide de Git : %s"
-#: lib/diff.tcl:42
+#: lib/diff.tcl:44
#, tcl-format
msgid ""
"No differences detected.\n"
"Une resynchronisation va être lancée automatiquement pour trouver d'autres "
"fichiers qui pourraient se trouver dans le même état."
-#: lib/diff.tcl:81
+#: lib/diff.tcl:83
#, tcl-format
msgid "Loading diff of %s..."
msgstr "Chargement des différences de %s..."
-#: lib/diff.tcl:114 lib/diff.tcl:184
+#: lib/diff.tcl:116 lib/diff.tcl:190
#, tcl-format
msgid "Unable to display %s"
msgstr "Impossible d'afficher %s"
-#: lib/diff.tcl:115
+#: lib/diff.tcl:117
msgid "Error loading file:"
msgstr "Erreur lors du chargement du fichier :"
-#: lib/diff.tcl:122
+#: lib/diff.tcl:124
msgid "Git Repository (subproject)"
-msgstr "Référentiel Git (sous projet)"
+msgstr "Dépôt Git (sous projet)"
-#: lib/diff.tcl:134
+#: lib/diff.tcl:136
msgid "* Binary file (not showing content)."
msgstr "* Fichier binaire (pas d'apperçu du contenu)."
-#: lib/diff.tcl:185
+#: lib/diff.tcl:191
msgid "Error loading diff:"
msgstr "Erreur lors du chargement des différences :"
-#: lib/diff.tcl:303
+#: lib/diff.tcl:313
msgid "Failed to unstage selected hunk."
-msgstr "La suppression dans le pré-commit de la section sélectionnée a échouée."
+msgstr "Échec lors de la désindexation de la section sélectionnée."
-#: lib/diff.tcl:310
+#: lib/diff.tcl:320
msgid "Failed to stage selected hunk."
-msgstr "Le pré-commit de la section sélectionnée a échoué."
+msgstr "Échec lors de l'indexation de la section."
+
+#: lib/diff.tcl:386
+msgid "Failed to unstage selected line."
+msgstr "Échec lors de la désindexation de la ligne sélectionnée."
+
+#: lib/diff.tcl:394
+msgid "Failed to stage selected line."
+msgstr "Échec lors de l'indexation de la ligne."
#: lib/error.tcl:20 lib/error.tcl:114
msgid "error"
#: lib/index.tcl:6
msgid "Unable to unlock the index."
-msgstr "Impossible de dévérouiller le pré-commit."
+msgstr "Impossible de dévérouiller l'index."
#: lib/index.tcl:15
msgid "Index Error"
-msgstr "Erreur de pré-commit"
+msgstr "Erreur de l'index"
#: lib/index.tcl:21
msgid ""
"Updating the Git index failed. A rescan will be automatically started to "
"resynchronize git-gui."
-msgstr "Le pré-commit a échoué. Une resynchronisation va être lancée automatiquement."
+msgstr ""
+"Échec de la mise à jour de l'index. Une resynchronisation va être lancée "
+"automatiquement."
#: lib/index.tcl:27
msgid "Continue"
#: lib/index.tcl:31
msgid "Unlock Index"
-msgstr "Dévérouiller le pré-commit"
+msgstr "Déverouiller l'index"
#: lib/index.tcl:282
#, tcl-format
msgid "Unstaging %s from commit"
-msgstr "Supprimer %s du commit"
+msgstr "Désindexation de: %s"
#: lib/index.tcl:313
msgid "Ready to commit."
#: lib/index.tcl:326
#, tcl-format
msgid "Adding %s"
-msgstr "Ajouter %s"
+msgstr "Ajout de %s"
#: lib/index.tcl:381
#, tcl-format
msgid "Revert changes in file %s?"
-msgstr "Inverser les modifications dans le fichier %s ? "
+msgstr "Annuler les modifications dans le fichier %s ? "
#: lib/index.tcl:383
#, tcl-format
msgid "Revert changes in these %i files?"
-msgstr "Inverser les modifications dans ces %i fichiers ?"
+msgstr "Annuler les modifications dans ces %i fichiers ?"
#: lib/index.tcl:391
msgid "Any unstaged changes will be permanently lost by the revert."
msgstr ""
-"Toutes les modifications non pré-commitées seront définitivement perdues "
-"lors de l'inversion."
+"Toutes les modifications non-indexées seront définitivement perdues par "
+"l'annulation."
#: lib/index.tcl:394
msgid "Do Nothing"
"\n"
"You must finish amending this commit before starting any type of merge.\n"
msgstr ""
-"Impossible de fucionner pendant une correction.\n"
+"Impossible de fusionner pendant une correction.\n"
"\n"
"Vous devez finir de corriger ce commit avant de lancer une quelconque "
"fusion.\n"
"The rescan will be automatically started now.\n"
msgstr ""
"L'état lors de la dernière synchronisation ne correspond plus à l'état du "
-"référentiel.\n"
+"dépôt.\n"
"\n"
-"Un autre programme Git a modifié ce référentiel depuis la dernière "
+"Un autre programme Git a modifié ce dépôt depuis la dernière "
"synchronisation. Une resynchronisation doit être effectuée avant de pouvoir "
"fusionner de nouveau.\n"
"\n"
"\n"
"Le fichier %s a des conflicts de fusion.\n"
"\n"
-"Vous devez les résoudre, puis pré-commiter le fichier, et enfin commiter "
-"pour terminer la fusion courante. Seulementà ce moment là, il sera possible "
+"Vous devez les résoudre, puis indexer le fichier, et enfin commiter pour "
+"terminer la fusion courante. Seulement à ce moment là sera-t-il possible "
"d'effectuer une nouvelle fusion.\n"
#: lib/merge.tcl:54
msgid "files reset"
msgstr "fichiers réinitialisés"
-#: lib/merge.tcl:265
+#: lib/merge.tcl:266
msgid "Abort failed."
msgstr "L'abandon a échoué."
-#: lib/merge.tcl:267
+#: lib/merge.tcl:268
msgid "Abort completed. Ready."
msgstr "Abandon teminé. Prêt."
#: lib/option.tcl:109
#, tcl-format
msgid "%s Repository"
-msgstr "Référentiel de %s"
+msgstr "Dépôt: %s"
#: lib/option.tcl:110
msgid "Global (All Repositories)"
-msgstr "Globales (tous les référentiels)"
+msgstr "Globales (tous les dépôts)"
#: lib/option.tcl:116
msgid "User Name"
#: lib/option.tcl:124
msgid "Prune Tracking Branches During Fetch"
-msgstr "Nettoyer les branches de suivi pendant la récupération"
+msgstr "Purger les branches de suivi pendant la récupération"
#: lib/option.tcl:125
msgid "Match Tracking Branches"
msgstr "Faire correspondre les branches de suivi"
#: lib/option.tcl:126
+msgid "Blame Copy Only On Changed Files"
+msgstr "Annoter les copies seulement sur fichiers modifiés"
+
+#: lib/option.tcl:127
+msgid "Minimum Letters To Blame Copy On"
+msgstr "Minimum de caratères pour annoter une copie"
+
+#: lib/option.tcl:128
msgid "Number of Diff Context Lines"
msgstr "Nombre de lignes de contexte dans les diffs"
-#: lib/option.tcl:127
+#: lib/option.tcl:129
msgid "Commit Message Text Width"
msgstr "Largeur du texte de message de commit"
-#: lib/option.tcl:128
+#: lib/option.tcl:130
msgid "New Branch Name Template"
msgstr "Nouveau modèle de nom de branche"
-#: lib/option.tcl:192
+#: lib/option.tcl:194
msgid "Spelling Dictionary:"
msgstr "Dictionnaire d'orthographe :"
-#: lib/option.tcl:216
+#: lib/option.tcl:218
msgid "Change Font"
-msgstr "Modifier les fontes"
+msgstr "Modifier les polices"
-#: lib/option.tcl:220
+#: lib/option.tcl:222
#, tcl-format
msgid "Choose %s"
msgstr "Choisir %s"
-#: lib/option.tcl:226
+#: lib/option.tcl:228
msgid "pt."
msgstr "pt."
-#: lib/option.tcl:240
+#: lib/option.tcl:242
msgid "Preferences"
msgstr "Préférences"
-#: lib/option.tcl:275
+#: lib/option.tcl:277
msgid "Failed to completely save options:"
msgstr "La sauvegarde complète des options a échouée :"
+#: lib/remote.tcl:165
+msgid "Prune from"
+msgstr "Purger de"
+
+#: lib/remote.tcl:170
+msgid "Fetch from"
+msgstr "Récupérer de"
+
+#: lib/remote.tcl:213
+msgid "Push to"
+msgstr "Pousser vers"
+
#: lib/remote_branch_delete.tcl:29 lib/remote_branch_delete.tcl:34
msgid "Delete Remote Branch"
msgstr "Supprimer branche distante"
#: lib/remote_branch_delete.tcl:47
msgid "From Repository"
-msgstr "Référentiel"
+msgstr "Dépôt source"
#: lib/remote_branch_delete.tcl:50 lib/transport.tcl:123
msgid "Remote:"
#: lib/remote_branch_delete.tcl:286
msgid "No repository selected."
-msgstr "Aucun référentiel n'est sélectionné."
+msgstr "Aucun dépôt n'est sélectionné."
#: lib/remote_branch_delete.tcl:291
#, tcl-format
msgid "Scanning %s..."
msgstr "Synchronisation de %s..."
-#: lib/remote.tcl:165
-msgid "Prune from"
-msgstr "Nettoyer de"
-
-#: lib/remote.tcl:170
-msgid "Fetch from"
-msgstr "Récupérer de"
-
-#: lib/remote.tcl:213
-msgid "Push to"
-msgstr "Pousser vers"
-
#: lib/shortcut.tcl:20 lib/shortcut.tcl:61
msgid "Cannot write shortcut:"
msgstr "Impossible d'écrire le raccourcis :"
msgid "Unrecognized spell checker"
msgstr "Vérificateur d'orthographe non reconnu"
-#: lib/spellcheck.tcl:180
+#: lib/spellcheck.tcl:186
msgid "No Suggestions"
msgstr "Aucune suggestion"
-#: lib/spellcheck.tcl:381
+#: lib/spellcheck.tcl:387
msgid "Unexpected EOF from spell checker"
-msgstr "Fin de fichier innatendue envoyée par le vérificateur d'orthographe"
+msgstr "EOF inattendue envoyée par le vérificateur d'orthographe"
-#: lib/spellcheck.tcl:385
+#: lib/spellcheck.tcl:391
msgid "Spell Checker Failed"
msgstr "Le vérificateur d'orthographe a échoué"
#: lib/transport.tcl:18
#, tcl-format
msgid "remote prune %s"
-msgstr "nettoyer à distance %s"
+msgstr "purger à distance %s"
#: lib/transport.tcl:19
#, tcl-format
#: lib/transport.tcl:120
msgid "Destination Repository"
-msgstr "Référentiel de destination"
+msgstr "Dépôt de destination"
#: lib/transport.tcl:158
msgid "Transfer Options"
-msgstr "Transférer options"
+msgstr "Options de transfert"
#: lib/transport.tcl:160
msgid "Force overwrite existing branch (may discard changes)"
#: lib/transport.tcl:168
msgid "Include tags"
-msgstr "Inclure les marques"
+msgstr "Inclure les marques (tags)"
foreach i [split $s ""] {
scan $i %c c
if {$c<128} {
- # escape '[', '\' and ']'
- if {$c == 0x5b || $c == 0x5d} {
+ # escape '[', '\', '$' and ']'
+ if {$c == 0x5b || $c == 0x5d || $c == 0x24} {
append res "\\"
}
append res $i
pick_one $sha1 ||
die_with_patch $sha1 "Could not apply $sha1... $rest"
make_patch $sha1
- : > "$DOTEST"/amend
+ git rev-parse --verify HEAD > "$DOTEST"/amend
warn "Stopped at $sha1... $rest"
warn "You can amend the commit now, with"
warn
else
. "$DOTEST"/author-script ||
die "Cannot find the author identity"
+ amend=
if test -f "$DOTEST"/amend
then
+ amend=$(git rev-parse --verify HEAD)
+ test "$amend" = $(cat "$DOTEST"/amend) ||
+ die "\
+You have uncommitted changes in your working tree. Please, commit them
+first and then run 'git rebase --continue' again."
git reset --soft HEAD^ ||
die "Cannot rewind the HEAD"
fi
export GIT_AUTHOR_NAME GIT_AUTHOR_EMAIL GIT_AUTHOR_DATE &&
- git commit --no-verify -F "$DOTEST"/message -e ||
- die "Could not commit staged changes."
+ git commit --no-verify -F "$DOTEST"/message -e || {
+ test -n "$amend" && git reset --soft $amend
+ die "Could not commit staged changes."
+ }
fi
require_clean_work_tree
create_stash () {
stash_msg="$1"
+ git update-index -q --refresh
if no_changes
then
exit 0
stash_msg="$*"
+ git update-index -q --refresh
if no_changes
then
echo 'No local changes to save'
}
apply_stash () {
+ git update-index -q --refresh &&
git diff-files --quiet --ignore-submodules ||
die 'Cannot restore on top of a dirty state'
$head ||= 'HEAD';
my @refs;
my ($url, $rev, $uuid, $gs) = working_head_info($head, \@refs);
+ unless ($gs) {
+ die "Unable to determine upstream SVN information from ",
+ "$head history.\nPerhaps the repository is empty.";
+ }
$url = defined $_commit_url ? $_commit_url : $gs->full_url;
my $last_rev = $_revision if defined $_revision;
if ($url) {
print "Committing to $url ...\n";
}
- unless ($gs) {
- die "Unable to determine upstream SVN information from ",
- "$head history.\nPerhaps the repository is empty.";
- }
my ($linear_refs, $parents) = linearize_history($gs, \@refs);
if ($_no_rebase && scalar(@$linear_refs) > 1) {
warn "Attempting to commit more than one change while ",
}
}
+sub escape_uri_only {
+ my ($uri) = @_;
+ my @tmp;
+ foreach (split m{/}, $uri) {
+ s/([^\w.%+-]|%(?![a-fA-F0-9]{2}))/sprintf("%%%02X",ord($1))/eg;
+ push @tmp, $_;
+ }
+ join('/', @tmp);
+}
+
+sub escape_url {
+ my ($url) = @_;
+ if ($url =~ m#^([^:]+)://([^/]*)(.*)$#) {
+ my ($scheme, $domain, $uri) = ($1, $2, escape_uri_only($3));
+ $url = "$scheme://$domain$uri";
+ }
+ $url;
+}
+
sub cmd_info {
my $path = canonicalize_path(defined($_[0]) ? $_[0] : ".");
+ my $fullpath = canonicalize_path($cmd_dir_prefix . $path);
if (exists $_[1]) {
die "Too many arguments specified\n";
}
my ($file_type, $diff_status) = find_file_type_and_diff_status($path);
if (!$file_type && !$diff_status) {
- print STDERR "$path: (Not a versioned resource)\n\n";
- return;
+ print STDERR "svn: '$path' is not under version control\n";
+ exit 1;
}
my ($url, $rev, $uuid, $gs) = working_head_info('HEAD');
# canonicalize_path() will return "" to make libsvn 1.5.x happy,
$path = "." if $path eq "";
- my $full_url = $url . ($path eq "." ? "" : "/$path");
+ my $full_url = $url . ($fullpath eq "" ? "" : "/$fullpath");
if ($_url) {
- print $full_url, "\n";
+ print escape_url($full_url), "\n";
return;
}
my $result = "Path: $path\n";
$result .= "Name: " . basename($path) . "\n" if $file_type ne "dir";
- $result .= "URL: " . $full_url . "\n";
+ $result .= "URL: " . escape_url($full_url) . "\n";
eval {
my $repos_root = $gs->repos_root;
Git::SVN::remove_username($repos_root);
- $result .= "Repository Root: $repos_root\n";
+ $result .= "Repository Root: " . escape_url($repos_root) . "\n";
};
if ($@) {
$result .= "Repository Root: (offline)\n";
}
my ($lc_author, $lc_rev, $lc_date_utc);
- my @args = Git::SVN::Log::git_svn_log_cmd($rev, $rev, "--", $path);
+ my @args = Git::SVN::Log::git_svn_log_cmd($rev, $rev, "--", $fullpath);
my $log = command_output_pipe(@args);
my $esc_color = qr/(?:\033\[(?:(?:\d+;)*\d*)?m)*/;
while (<$log>) {
my $out = syswrite($tmp_fh, $str, $res);
defined($out) && $out == $res
or croak("write ",
- $tmp_fh->filename,
+ Git::temp_path($tmp_fh),
": $!\n");
}
defined $res or croak $!;
}
$hash = $::_repository->hash_and_insert_object(
- $fh->filename);
+ Git::temp_path($fh));
$hash =~ /^[a-f\d]{40}$/ or die "not a sha1: $hash\n";
Git::temp_release($fb->{base}, 1);
while (<$diff_fh>) {
chomp $_; # this gets rid of the trailing "\0"
if ($state eq 'meta' && /^:(\d{6})\s(\d{6})\s
- $::sha1\s($::sha1)\s
+ ($::sha1)\s($::sha1)\s
([MTCRAD])\d*$/xo) {
push @mods, { mode_a => $1, mode_b => $2,
- sha1_b => $3, chg => $4 };
- if ($4 =~ /^(?:C|R)$/) {
+ sha1_a => $3, sha1_b => $4,
+ chg => $5 };
+ if ($5 =~ /^(?:C|R)$/) {
$state = 'file_a';
} else {
$state = 'file_b';
my $fbat = $self->add_file($self->repo_path($m->{file_b}), $pbat,
$self->url_path($m->{file_a}), $self->{r});
print "\tR\t$m->{file_a} => $m->{file_b}\n" unless $::_q;
+ $self->apply_autoprops($file, $fbat);
$self->chg_file($fbat, $m);
$self->close_file($fbat,undef,$self->{pool});
$self->SUPER::change_file_prop($fbat, $pname, $pval, $self->{pool});
}
-sub chg_file {
- my ($self, $fbat, $m) = @_;
- if ($m->{mode_b} =~ /755$/ && $m->{mode_a} !~ /755$/) {
- $self->change_file_prop($fbat,'svn:executable','*');
- } elsif ($m->{mode_b} !~ /755$/ && $m->{mode_a} =~ /755$/) {
- $self->change_file_prop($fbat,'svn:executable',undef);
- }
- my $fh = Git::temp_acquire('git_blob');
- if ($m->{mode_b} =~ /^120/) {
+sub _chg_file_get_blob ($$$$) {
+ my ($self, $fbat, $m, $which) = @_;
+ my $fh = Git::temp_acquire("git_blob_$which");
+ if ($m->{"mode_$which"} =~ /^120/) {
print $fh 'link ' or croak $!;
$self->change_file_prop($fbat,'svn:special','*');
- } elsif ($m->{mode_a} =~ /^120/ && $m->{mode_b} !~ /^120/) {
+ } elsif ($m->{mode_a} =~ /^120/ && $m->{"mode_$which"} !~ /^120/) {
$self->change_file_prop($fbat,'svn:special',undef);
}
- my $size = $::_repository->cat_blob($m->{sha1_b}, $fh);
- croak "Failed to read object $m->{sha1_b}" if ($size < 0);
+ my $blob = $m->{"sha1_$which"};
+ return ($fh,) if ($blob =~ /^0{40}$/);
+ my $size = $::_repository->cat_blob($blob, $fh);
+ croak "Failed to read object $blob" if ($size < 0);
$fh->flush == 0 or croak $!;
seek $fh, 0, 0 or croak $!;
my $exp = ::md5sum($fh);
seek $fh, 0, 0 or croak $!;
+ return ($fh, $exp);
+}
+sub chg_file {
+ my ($self, $fbat, $m) = @_;
+ if ($m->{mode_b} =~ /755$/ && $m->{mode_a} !~ /755$/) {
+ $self->change_file_prop($fbat,'svn:executable','*');
+ } elsif ($m->{mode_b} !~ /755$/ && $m->{mode_a} =~ /755$/) {
+ $self->change_file_prop($fbat,'svn:executable',undef);
+ }
+ my ($fh_a, $exp_a) = _chg_file_get_blob $self, $fbat, $m, 'a';
+ my ($fh_b, $exp_b) = _chg_file_get_blob $self, $fbat, $m, 'b';
my $pool = SVN::Pool->new;
- my $atd = $self->apply_textdelta($fbat, undef, $pool);
- my $got = SVN::TxDelta::send_stream($fh, @$atd, $pool);
- die "Checksum mismatch\nexpected: $exp\ngot: $got\n" if ($got ne $exp);
- Git::temp_release($fh, 1);
+ my $atd = $self->apply_textdelta($fbat, $exp_a, $pool);
+ if (-s $fh_a) {
+ my $txstream = SVN::TxDelta::new ($fh_a, $fh_b, $pool);
+ my $res = SVN::TxDelta::send_txstream($txstream, @$atd, $pool);
+ if (defined $res) {
+ die "Unexpected result from send_txstream: $res\n",
+ "(SVN::Core::VERSION: $SVN::Core::VERSION)\n";
+ }
+ } else {
+ my $got = SVN::TxDelta::send_stream($fh_b, @$atd, $pool);
+ die "Checksum mismatch\nexpected: $exp_b\ngot: $got\n"
+ if ($got ne $exp_b);
+ }
+ Git::temp_release($fh_b, 1);
+ Git::temp_release($fh_a, 1);
$pool->clear;
}
}
}
$ra ||= $self;
+ $url_b = escape_url($url_b);
my $reporter = $ra->do_switch($rev_b, '', 1, $url_b, $editor, $pool);
my @lock = $SVN::Core::VERSION ge '1.2.0' ? (undef) : ();
$reporter->set_path('', $rev_a, 0, @lock, $pool);
sub run_pager {
return unless -t *STDOUT && defined $pager;
- pipe my $rfd, my $wfd or return;
+ pipe my ($rfd, $wfd) or return;
defined(my $pid = fork) or ::fatal "Can't fork: $!";
if (!$pid) {
open STDOUT, '>&', $wfd or
if (sizeof(ext) > 1) {
i = strlen(argv[0]) - strlen(ext);
if (i > 0 && !strcmp(argv[0] + i, ext)) {
- char *argv0 = strdup(argv[0]);
+ char *argv0 = xstrdup(argv[0]);
argv[0] = cmd = argv0;
argv0[i] = '\0';
}
cmd, argv[0]);
exit(1);
}
- help_unknown_cmd(cmd);
+ argv[0] = help_unknown_cmd(cmd);
+ handle_internal_command(argc, argv);
+ execv_dashed_external(argv);
}
fprintf(stderr, "Failed to run command '%s': %s\n",
%files cvs
%defattr(-,root,root)
%doc Documentation/*git-cvs*.txt
+%{_bindir}/git-cvsserver
%{_libexecdir}/git-core/*cvs*
%{!?_without_docs: %{_mandir}/man1/*cvs*.1*}
%{!?_without_docs: %doc Documentation/*git-cvs*.html }
# No files for you!
%changelog
+* Fri Sep 12 2008 Quy Tonthat <qtonthat@gmail.com>
+- move git-cvsserver to bindir.
+
* Sun Jun 15 2008 Junio C Hamano <gitster@pobox.com>
- Remove curl from Requires list.
}
proc reset_pending_select {selid} {
- global pending_select mainheadid
+ global pending_select mainheadid selectheadid
if {$selid ne {}} {
set pending_select $selid
+ } elseif {$selectheadid ne {}} {
+ set pending_select $selectheadid
} else {
set pending_select $mainheadid
}
proc readrefs {} {
global tagids idtags headids idheads tagobjid
global otherrefids idotherrefs mainhead mainheadid
+ global selecthead selectheadid
foreach v {tagids idtags headids idheads otherrefids idotherrefs} {
catch {unset $v}
set mainhead [string range $thehead 11 end]
}
}
+ set selectheadid {}
+ if {$selecthead ne {}} {
+ catch {
+ set selectheadid [exec git rev-parse --verify $selecthead]
+ }
+ }
}
# skip over fake commits
-command {flist_hl 1}
$flist_menu add command -label [mc "External diff"] \
-command {external_diff}
+ $flist_menu add command -label [mc "Blame parent commit"] \
+ -command {external_blame 1}
}
# Windows sends all mouse wheel events to the current focused window, not
}
}
+proc external_blame {parent_idx} {
+ global flist_menu_file
+ global nullid nullid2
+ global parentlist selectedline currentid
+
+ if {$parent_idx > 0} {
+ set base_commit [lindex $parentlist $selectedline [expr {$parent_idx-1}]]
+ } else {
+ set base_commit $currentid
+ }
+
+ if {$base_commit eq {} || $base_commit eq $nullid || $base_commit eq $nullid2} {
+ error_popup [mc "No such commit"]
+ return
+ }
+
+ if {[catch {exec git gui blame $base_commit $flist_menu_file &} err]} {
+ error_popup [mc "git gui blame: command failed: $err"]
+ }
+}
+
# delete $dir when we see eof on $f (presumably because the child has exited)
proc delete_at_eof {f dir} {
while {[gets $f line] >= 0} {}
exit 1
}
+set selecthead {}
+set selectheadid {}
+
set revtreeargs {}
set cmdline_files {}
set i 0
set cmdline_files [lrange $argv [expr {$i + 1}] end]
break
}
+ "--select-commit=*" {
+ set selecthead [string range $arg 16 end]
+ }
"--argscmd=*" {
set revtreeargscmd [string range $arg 10 end]
}
incr i
}
+if {$selecthead eq "HEAD"} {
+ set selecthead {}
+}
+
if {$i >= [llength $argv] && $revtreeargs ne {}} {
# no -- on command line, but some arguments (other than --argscmd)
if {[catch {
#include "cache.h"
#include "builtin.h"
#include "exec_cmd.h"
+#include "levenshtein.h"
#include "help.h"
/* most GUI terminals set COLUMNS (although some don't export it) */
cmds->names[cmds->cnt++] = ent;
}
+static void clean_cmdnames(struct cmdnames *cmds)
+{
+ int i;
+ for (i = 0; i < cmds->cnt; ++i)
+ free(cmds->names[i]);
+ free(cmds->names);
+ cmds->cnt = 0;
+ cmds->alloc = 0;
+}
+
static int cmdname_compare(const void *a_, const void *b_)
{
struct cmdname *a = *(struct cmdname **)a_;
return 0;
}
-void help_unknown_cmd(const char *cmd)
+static int autocorrect;
+
+static int git_unknown_cmd_config(const char *var, const char *value, void *cb)
+{
+ if (!strcmp(var, "help.autocorrect"))
+ autocorrect = git_config_int(var,value);
+
+ return git_default_config(var, value, cb);
+}
+
+static int levenshtein_compare(const void *p1, const void *p2)
{
+ const struct cmdname *const *c1 = p1, *const *c2 = p2;
+ const char *s1 = (*c1)->name, *s2 = (*c2)->name;
+ int l1 = (*c1)->len;
+ int l2 = (*c2)->len;
+ return l1 != l2 ? l1 - l2 : strcmp(s1, s2);
+}
+
+const char *help_unknown_cmd(const char *cmd)
+{
+ int i, n, best_similarity = 0;
+ struct cmdnames main_cmds, other_cmds;
+
+ memset(&main_cmds, 0, sizeof(main_cmds));
+ memset(&other_cmds, 0, sizeof(main_cmds));
+
+ git_config(git_unknown_cmd_config, NULL);
+
+ load_command_list("git-", &main_cmds, &other_cmds);
+
+ ALLOC_GROW(main_cmds.names, main_cmds.cnt + other_cmds.cnt,
+ main_cmds.alloc);
+ memcpy(main_cmds.names + main_cmds.cnt, other_cmds.names,
+ other_cmds.cnt * sizeof(other_cmds.names[0]));
+ main_cmds.cnt += other_cmds.cnt;
+ free(other_cmds.names);
+
+ /* This reuses cmdname->len for similarity index */
+ for (i = 0; i < main_cmds.cnt; ++i)
+ main_cmds.names[i]->len =
+ levenshtein(cmd, main_cmds.names[i]->name, 0, 2, 1, 4);
+
+ qsort(main_cmds.names, main_cmds.cnt,
+ sizeof(*main_cmds.names), levenshtein_compare);
+
+ if (!main_cmds.cnt)
+ die ("Uh oh. Your system reports no Git commands at all.");
+
+ best_similarity = main_cmds.names[0]->len;
+ n = 1;
+ while (n < main_cmds.cnt && best_similarity == main_cmds.names[n]->len)
+ ++n;
+ if (autocorrect && n == 1) {
+ const char *assumed = main_cmds.names[0]->name;
+ main_cmds.names[0] = NULL;
+ clean_cmdnames(&main_cmds);
+ fprintf(stderr, "WARNING: You called a Git program named '%s', "
+ "which does not exist.\n"
+ "Continuing under the assumption that you meant '%s'\n",
+ cmd, assumed);
+ if (autocorrect > 0) {
+ fprintf(stderr, "in %0.1f seconds automatically...\n",
+ (float)autocorrect/10.0);
+ poll(NULL, 0, autocorrect * 100);
+ }
+ return assumed;
+ }
+
fprintf(stderr, "git: '%s' is not a git-command. See 'git --help'.\n", cmd);
+
+ if (best_similarity < 6) {
+ fprintf(stderr, "\nDid you mean %s?\n",
+ n < 2 ? "this": "one of these");
+
+ for (i = 0; i < n; i++)
+ fprintf(stderr, "\t%s\n", main_cmds.names[i]->name);
+ }
+
exit(1);
}
int alloc;
int cnt;
struct cmdname {
- size_t len;
+ size_t len; /* also used for similarity index in help.c */
char name[FLEX_ARRAY];
} **names;
};
no_pragma_header = curl_slist_append(no_pragma_header, "Pragma:");
if (remote->url && remote->url[strlen(remote->url)-1] != '/') {
- rewritten_url = malloc(strlen(remote->url)+2);
+ rewritten_url = xmalloc(strlen(remote->url)+2);
strcpy(rewritten_url, remote->url);
strcat(rewritten_url, "/");
remote->url = rewritten_url;
{
CURL* result = curl_easy_init();
- curl_easy_setopt(result, CURLOPT_SSL_VERIFYPEER, curl_ssl_verify);
+ if (!curl_ssl_verify) {
+ curl_easy_setopt(result, CURLOPT_SSL_VERIFYPEER, 0);
+ curl_easy_setopt(result, CURLOPT_SSL_VERIFYHOST, 0);
+ } else {
+ /* Verify authenticity of the peer's certificate */
+ curl_easy_setopt(result, CURLOPT_SSL_VERIFYPEER, 1);
+ /* The name in the cert must match whom we tried to connect */
+ curl_easy_setopt(result, CURLOPT_SSL_VERIFYHOST, 2);
+ }
+
#if LIBCURL_VERSION_NUM >= 0x070907
curl_easy_setopt(result, CURLOPT_NETRC, CURL_NETRC_OPTIONAL);
#endif
void add_fill_function(void *data, int (*fill)(void *))
{
- struct fill_chain *new = malloc(sizeof(*new));
+ struct fill_chain *new = xmalloc(sizeof(*new));
struct fill_chain **linkp = &fill_cfg;
new->data = data;
new->fill = fill;
--- /dev/null
+#include "cache.h"
+#include "levenshtein.h"
+
+int levenshtein(const char *string1, const char *string2,
+ int w, int s, int a, int d)
+{
+ int len1 = strlen(string1), len2 = strlen(string2);
+ int *row0 = xmalloc(sizeof(int) * (len2 + 1));
+ int *row1 = xmalloc(sizeof(int) * (len2 + 1));
+ int *row2 = xmalloc(sizeof(int) * (len2 + 1));
+ int i, j;
+
+ for (j = 0; j <= len2; j++)
+ row1[j] = j * a;
+ for (i = 0; i < len1; i++) {
+ int *dummy;
+
+ row2[0] = (i + 1) * d;
+ for (j = 0; j < len2; j++) {
+ /* substitution */
+ row2[j + 1] = row1[j] + s * (string1[i] != string2[j]);
+ /* swap */
+ if (i > 0 && j > 0 && string1[i - 1] == string2[j] &&
+ string1[i] == string2[j - 1] &&
+ row2[j + 1] > row0[j - 1] + w)
+ row2[j + 1] = row0[j - 1] + w;
+ /* deletion */
+ if (j + 1 < len2 && row2[j + 1] > row1[j + 1] + d)
+ row2[j + 1] = row1[j + 1] + d;
+ /* insertion */
+ if (row2[j + 1] > row2[j] + a)
+ row2[j + 1] = row2[j] + a;
+ }
+
+ dummy = row0;
+ row0 = row1;
+ row1 = row2;
+ row2 = dummy;
+ }
+
+ i = row1[len2];
+ free(row0);
+ free(row1);
+ free(row2);
+
+ return i;
+}
--- /dev/null
+#ifndef LEVENSHTEIN_H
+#define LEVENSHTEIN_H
+
+int levenshtein(const char *string1, const char *string2,
+ int swap_penalty, int substition_penalty,
+ int insertion_penalty, int deletion_penalty);
+
+#endif
extern unsigned int get_max_object_index(void);
extern struct object *get_indexed_object(unsigned int);
-/** Internal only **/
+/*
+ * This can be used to see if we have heard of the object before, but
+ * it can return "yes we have, and here is a half-initialised object"
+ * for an object that we haven't loaded/parsed yet.
+ *
+ * When parsing a commit to create an in-core commit object, its
+ * parents list holds commit objects that represent its parents, but
+ * they are expected to be lazily initialized and do not know what
+ * their trees or parents are yet. When this function returns such a
+ * half-initialised objects, the caller is expected to initialize them
+ * by calling parse_object() on them.
+ */
struct object *lookup_object(const unsigned char *sha1);
extern void *create_object(const unsigned char *sha1, int type, void *obj);
command_bidi_pipe command_close_bidi_pipe
version exec_path hash_object git_cmd_try
remote_refs
- temp_acquire temp_release temp_reset);
+ temp_acquire temp_release temp_reset temp_path);
=head1 DESCRIPTION
{ # %TEMP_* Lexical Context
-my (%TEMP_LOCKS, %TEMP_FILES);
+my (%TEMP_FILEMAP, %TEMP_FILES);
=item temp_acquire ( NAME )
my $temp_fd = _temp_cache($name);
- $TEMP_LOCKS{$temp_fd} = 1;
+ $TEMP_FILES{$temp_fd}{locked} = 1;
$temp_fd;
}
sub temp_release {
my ($self, $temp_fd, $trunc) = _maybe_self(@_);
- if (ref($temp_fd) ne 'File::Temp') {
+ if (exists $TEMP_FILEMAP{$temp_fd}) {
$temp_fd = $TEMP_FILES{$temp_fd};
}
- unless ($TEMP_LOCKS{$temp_fd}) {
+ unless ($TEMP_FILES{$temp_fd}{locked}) {
carp "Attempt to release temp file '",
$temp_fd, "' that has not been locked";
}
temp_reset($temp_fd) if $trunc and $temp_fd->opened;
- $TEMP_LOCKS{$temp_fd} = 0;
+ $TEMP_FILES{$temp_fd}{locked} = 0;
undef;
}
_verify_require();
- my $temp_fd = \$TEMP_FILES{$name};
+ my $temp_fd = \$TEMP_FILEMAP{$name};
if (defined $$temp_fd and $$temp_fd->opened) {
- if ($TEMP_LOCKS{$$temp_fd}) {
+ if ($TEMP_FILES{$$temp_fd}{locked}) {
throw Error::Simple("Temp file with moniker '",
$name, "' already in use");
}
carp "Temp file '", $name,
"' was closed. Opening replacement.";
}
- $$temp_fd = File::Temp->new(
- TEMPLATE => 'Git_XXXXXX',
- DIR => File::Spec->tmpdir
+ my $fname;
+ ($$temp_fd, $fname) = File::Temp->tempfile(
+ 'Git_XXXXXX', UNLINK => 1
) or throw Error::Simple("couldn't open new temp file");
$$temp_fd->autoflush;
binmode $$temp_fd;
+ $TEMP_FILES{$$temp_fd}{fname} = $fname;
}
$$temp_fd;
}
or throw Error::Simple("expected file position to be reset");
}
+=item temp_path ( NAME )
+
+=item temp_path ( FILEHANDLE )
+
+Returns the filename associated with the given tempfile.
+
+=cut
+
+sub temp_path {
+ my ($self, $temp_fd) = _maybe_self(@_);
+
+ if (exists $TEMP_FILEMAP{$temp_fd}) {
+ $temp_fd = $TEMP_FILEMAP{$temp_fd};
+ }
+ $TEMP_FILES{$temp_fd}{fname};
+}
+
sub END {
- unlink values %TEMP_FILES if %TEMP_FILES;
+ unlink values %TEMP_FILEMAP if %TEMP_FILEMAP;
}
} # %TEMP_* Lexical Context
istate->cache_nr = 0;
istate->cache_changed = 0;
istate->timestamp = 0;
+ istate->name_hash_initialized = 0;
free_hash(&istate->name_hash);
cache_tree_free(&(istate->cache_tree));
free(istate->alloc);
return retval;
}
+/*
+ * If the "reading" argument is set, this function finds out what _object_
+ * the ref points at by "reading" the ref. The ref, if it is not symbolic,
+ * has to exist, and if it is symbolic, it has to point at an existing ref,
+ * because the "read" goes through the symref to the ref it points at.
+ *
+ * The access that is not "reading" may often be "writing", but does not
+ * have to; it can be merely checking _where it leads to_. If it is a
+ * prelude to "writing" to the ref, a write to a symref that points at
+ * yet-to-be-born ref will create the real ref pointed by the symref.
+ * reading=0 allows the caller to check where such a symref leads to.
+ */
const char *resolve_ref(const char *ref, unsigned char *sha1, int reading, int *flag)
{
int depth = MAXDEPTH;
if (--depth < 0)
return NULL;
- /* Special case: non-existing file.
- * Not having the refs/heads/new-branch is OK
- * if we are writing into it, so is .git/HEAD
- * that points at refs/heads/master still to be
- * born. It is NOT OK if we are resolving for
- * reading.
- */
+ /* Special case: non-existing file. */
if (lstat(path, &st) < 0) {
struct ref_list *list = get_packed_refs();
while (list) {
if (!longest)
return url;
- ret = malloc(rewrite[longest_i]->baselen +
+ ret = xmalloc(rewrite[longest_i]->baselen +
(strlen(url) - longest->len) + 1);
strcpy(ret, rewrite[longest_i]->base);
strcpy(ret + rewrite[longest_i]->baselen, url + longest->len);
ret->name = xstrndup(name, len);
else
ret->name = xstrdup(name);
- refname = malloc(strlen(name) + strlen("refs/heads/") + 1);
+ refname = xmalloc(strlen(name) + strlen("refs/heads/") + 1);
strcpy(refname, "refs/heads/");
strcpy(refname + strlen("refs/heads/"), ret->name);
ret->refname = refname;
. ./test-lib.sh
+remotes_git_svn=remotes/git""-svn
+git_svn_id=git""-svn-id
+
if test -n "$NO_SVN_TESTS"
then
- test_expect_success 'skipping git-svn tests, NO_SVN_TESTS defined' :
+ test_expect_success 'skipping git svn tests, NO_SVN_TESTS defined' :
test_done
exit
fi
svn >/dev/null 2>&1
if test $? -ne 1
then
- test_expect_success 'skipping git-svn tests, svn not found' :
+ test_expect_success 'skipping git svn tests, svn not found' :
test_done
exit
fi
mkdir "$GIT_DIR"/logs
cat > "$GIT_DIR/httpd.conf" <<EOF
-ServerName "git-svn test"
+ServerName "git svn test"
ServerRoot "$GIT_DIR"
DocumentRoot "$GIT_DIR"
PidFile "$GIT_DIR/httpd.pid"
test_must_fail git diff --no-index Beer.java Beer-correct.java
'
+test_expect_success 'alternation in pattern' '
+ git config diff.java.funcname "^[ ]*\\(\\(public\\|static\\).*\\)$"
+ git diff --no-index Beer.java Beer-correct.java |
+ grep "^@@.*@@ public static void main("
+'
+
test_done
'
+test_expect_success 'do not color trailing cr in context' '
+ git config --unset core.whitespace
+ rm -f .gitattributes &&
+ echo AAAQ | tr Q "\015" >G &&
+ git add G &&
+ echo BBBQ | tr Q "\015" >>G
+ git diff --color G | tr "\015" Q >output &&
+ grep "BBB.*${blue_grep}Q" output &&
+ grep "AAA.*\[mQ" output
+
+'
+
test_done
'
-sed -e 's/deerit./&\n\n\n\n/' -e "s/locavit,/locavit;/" < new6.txt > new8.txt
-sed -e 's/deerit./&\n\n\n\n/' -e "s/locavit,/locavit --/" < new7.txt > new9.txt
+sed -e 's/deerit./&%%%%/' -e "s/locavit,/locavit;/"< new6.txt | tr '%' '\012' > new8.txt
+sed -e 's/deerit./&%%%%/' -e "s/locavit,/locavit --/" < new7.txt | tr '%' '\012' > new9.txt
test_expect_success 'ZEALOUS_ALNUM' '
'checkout with --track, but without -b, fails with too short tracked name' '
test_must_fail git checkout --track renamer'
+test_expect_success 'checkout an unmerged path should fail' '
+ rm -f .git/index &&
+ O=$(echo original | git hash-object -w --stdin) &&
+ A=$(echo ourside | git hash-object -w --stdin) &&
+ B=$(echo theirside | git hash-object -w --stdin) &&
+ (
+ echo "100644 $A 0 fild" &&
+ echo "100644 $O 1 file" &&
+ echo "100644 $A 2 file" &&
+ echo "100644 $B 3 file" &&
+ echo "100644 $A 0 filf"
+ ) | git update-index --index-info &&
+ echo "none of the above" >sample &&
+ cat sample >fild &&
+ cat sample >file &&
+ cat sample >filf &&
+ test_must_fail git checkout fild file filf &&
+ test_cmp sample fild &&
+ test_cmp sample filf &&
+ test_cmp sample file
+'
+
test_done
test_expect_success \
'validate git rev-list output.' \
- 'diff current expected'
+ 'test_cmp expected current'
test_expect_success 'partial commit that involves removal (1)' '
git commit -m "Partial: add elif" elif &&
git diff-tree --name-status HEAD^ HEAD >current &&
echo "A elif" >expected &&
- diff expected current
+ test_cmp expected current
'
git commit -m "Partial: remove file" file &&
git diff-tree --name-status HEAD^ HEAD >current &&
echo "D file" >expected &&
- diff expected current
+ test_cmp expected current
'
git commit -m "Partial: modify elif" elif &&
git diff-tree --name-status HEAD^ HEAD >current &&
echo "M elif" >expected &&
- diff expected current
+ test_cmp expected current
'
expected &&
git commit --amend --author="$author" &&
git cat-file -p HEAD > current &&
- diff expected current
+ test_cmp expected current
'
expected &&
git commit --amend --author="$author" &&
git cat-file -p HEAD > current &&
- diff expected current
+ test_cmp expected current
'
#!/bin/sh
-test_description='git-merge
+test_description='git merge
Testing a custom strategy.'
# Copyright (c) 2006 Eric Wong
#
-test_description='git-svn basic tests'
+test_description='git svn basic tests'
GIT_SVN_LC_ALL=${LC_ALL:-$LANG}
case "$GIT_SVN_LC_ALL" in
. ./lib-git-svn.sh
-say 'define NO_SVN_TESTS to skip git-svn tests'
+say 'define NO_SVN_TESTS to skip git svn tests'
test_expect_success \
- 'initialize git-svn' '
+ 'initialize git svn' '
mkdir import &&
cd import &&
echo foo > foo &&
echo "zzz" > bar/zzz &&
echo "#!/bin/sh" > exec.sh &&
chmod +x exec.sh &&
- svn import -m "import for git-svn" . "$svnrepo" >/dev/null &&
+ svn import -m "import for git svn" . "$svnrepo" >/dev/null &&
cd .. &&
rm -rf import &&
- git-svn init "$svnrepo"'
+ git svn init "$svnrepo"'
test_expect_success \
'import an SVN revision into git' \
- 'git-svn fetch'
+ 'git svn fetch'
test_expect_success "checkout from svn" 'svn co "$svnrepo" "$SVN_TREE"'
name='try a deep --rmdir with a commit'
test_expect_success "$name" '
- git checkout -f -b mybranch remotes/git-svn &&
+ git checkout -f -b mybranch ${remotes_git_svn} &&
mv dir/a/b/c/d/e/file dir/file &&
cp dir/file file &&
git update-index --add --remove dir/a/b/c/d/e/file dir/file file &&
git commit -m "$name" &&
- git-svn set-tree --find-copies-harder --rmdir \
- remotes/git-svn..mybranch &&
+ git svn set-tree --find-copies-harder --rmdir \
+ ${remotes_git_svn}..mybranch &&
svn up "$SVN_TREE" &&
test -d "$SVN_TREE"/dir && test ! -d "$SVN_TREE"/dir/a'
git update-index --remove dir/file &&
git update-index --add dir/file/file &&
git commit -m '$name' &&
- test_must_fail git-svn set-tree --find-copies-harder --rmdir \
- remotes/git-svn..mybranch" || true
+ test_must_fail git svn set-tree --find-copies-harder --rmdir \
+ ${remotes_git_svn}..mybranch" || true
name='detect node change from directory to file #1'
test_expect_success "$name" '
rm -rf dir "$GIT_DIR"/index &&
- git checkout -f -b mybranch2 remotes/git-svn &&
+ git checkout -f -b mybranch2 ${remotes_git_svn} &&
mv bar/zzz zzz &&
rm -rf bar &&
mv zzz bar &&
git update-index --remove -- bar/zzz &&
git update-index --add -- bar &&
git commit -m "$name" &&
- test_must_fail git-svn set-tree --find-copies-harder --rmdir \
- remotes/git-svn..mybranch2' || true
+ test_must_fail git svn set-tree --find-copies-harder --rmdir \
+ ${remotes_git_svn}..mybranch2' || true
name='detect node change from file to directory #2'
test_expect_success "$name" '
rm -f "$GIT_DIR"/index &&
- git checkout -f -b mybranch3 remotes/git-svn &&
+ git checkout -f -b mybranch3 ${remotes_git_svn} &&
rm bar/zzz &&
git update-index --remove bar/zzz &&
mkdir bar/zzz &&
echo yyy > bar/zzz/yyy &&
git update-index --add bar/zzz/yyy &&
git commit -m "$name" &&
- test_must_fail git-svn set-tree --find-copies-harder --rmdir \
- remotes/git-svn..mybranch3' || true
+ test_must_fail git svn set-tree --find-copies-harder --rmdir \
+ ${remotes_git_svn}..mybranch3' || true
name='detect node change from directory to file #2'
test_expect_success "$name" '
rm -f "$GIT_DIR"/index &&
- git checkout -f -b mybranch4 remotes/git-svn &&
+ git checkout -f -b mybranch4 ${remotes_git_svn} &&
rm -rf dir &&
git update-index --remove -- dir/file &&
touch dir &&
echo asdf > dir &&
git update-index --add -- dir &&
git commit -m "$name" &&
- test_must_fail git-svn set-tree --find-copies-harder --rmdir \
- remotes/git-svn..mybranch4' || true
+ test_must_fail git svn set-tree --find-copies-harder --rmdir \
+ ${remotes_git_svn}..mybranch4' || true
name='remove executable bit from a file'
test_expect_success "$name" '
rm -f "$GIT_DIR"/index &&
- git checkout -f -b mybranch5 remotes/git-svn &&
+ git checkout -f -b mybranch5 ${remotes_git_svn} &&
chmod -x exec.sh &&
git update-index exec.sh &&
git commit -m "$name" &&
- git-svn set-tree --find-copies-harder --rmdir \
- remotes/git-svn..mybranch5 &&
+ git svn set-tree --find-copies-harder --rmdir \
+ ${remotes_git_svn}..mybranch5 &&
svn up "$SVN_TREE" &&
test ! -x "$SVN_TREE"/exec.sh'
chmod +x exec.sh &&
git update-index exec.sh &&
git commit -m "$name" &&
- git-svn set-tree --find-copies-harder --rmdir \
- remotes/git-svn..mybranch5 &&
+ git svn set-tree --find-copies-harder --rmdir \
+ ${remotes_git_svn}..mybranch5 &&
svn up "$SVN_TREE" &&
test -x "$SVN_TREE"/exec.sh'
ln -s bar/zzz exec.sh &&
git update-index exec.sh &&
git commit -m "$name" &&
- git-svn set-tree --find-copies-harder --rmdir \
- remotes/git-svn..mybranch5 &&
+ git svn set-tree --find-copies-harder --rmdir \
+ ${remotes_git_svn}..mybranch5 &&
svn up "$SVN_TREE" &&
test -L "$SVN_TREE"/exec.sh'
ln -s bar/zzz exec-2.sh &&
git update-index --add bar/zzz exec-2.sh &&
git commit -m "$name" &&
- git-svn set-tree --find-copies-harder --rmdir \
- remotes/git-svn..mybranch5 &&
+ git svn set-tree --find-copies-harder --rmdir \
+ ${remotes_git_svn}..mybranch5 &&
svn up "$SVN_TREE" &&
test -x "$SVN_TREE"/bar/zzz &&
test -L "$SVN_TREE"/exec-2.sh'
cp help exec-2.sh &&
git update-index exec-2.sh &&
git commit -m "$name" &&
- git-svn set-tree --find-copies-harder --rmdir \
- remotes/git-svn..mybranch5 &&
+ git svn set-tree --find-copies-harder --rmdir \
+ ${remotes_git_svn}..mybranch5 &&
svn up "$SVN_TREE" &&
test -f "$SVN_TREE"/exec-2.sh &&
test ! -L "$SVN_TREE"/exec-2.sh &&
echo '# hello' >> exec-2.sh &&
git update-index exec-2.sh &&
git commit -m 'éï∏' &&
- git-svn set-tree HEAD"
+ git svn set-tree HEAD"
unset LC_ALL
else
say "UTF-8 locale not set, test skipped ($GIT_SVN_LC_ALL)"
GIT_SVN_ID=alt
export GIT_SVN_ID
test_expect_success "$name" \
- 'git-svn init "$svnrepo" && git-svn fetch &&
- git rev-list --pretty=raw remotes/git-svn | grep ^tree | uniq > a &&
+ 'git svn init "$svnrepo" && git svn fetch &&
+ git rev-list --pretty=raw ${remotes_git_svn} | grep ^tree | uniq > a &&
git rev-list --pretty=raw remotes/alt | grep ^tree | uniq > b &&
test_cmp a b'
test_expect_success 'exit if remote refs are ambigious' "
git config --add svn-remote.svn.fetch \
- bar:refs/remotes/git-svn &&
- test_must_fail git-svn migrate
+ bar:refs/${remotes_git_svn} &&
+ test_must_fail git svn migrate
"
test_expect_success 'exit if init-ing a would clobber a URL' '
svnadmin create "${PWD}/svnrepo2" &&
svn mkdir -m "mkdir bar" "${svnrepo}2/bar" &&
git config --unset svn-remote.svn.fetch \
- "^bar:refs/remotes/git-svn$" &&
- test_must_fail git-svn init "${svnrepo}2/bar"
+ "^bar:refs/${remotes_git_svn}$" &&
+ test_must_fail git svn init "${svnrepo}2/bar"
'
test_expect_success \
'init allows us to connect to another directory in the same repo' '
- git-svn init --minimize-url -i bar "$svnrepo/bar" &&
+ git svn init --minimize-url -i bar "$svnrepo/bar" &&
git config --get svn-remote.svn.fetch \
"^bar:refs/remotes/bar$" &&
git config --get svn-remote.svn.fetch \
- "^:refs/remotes/git-svn$"
+ "^:refs/${remotes_git_svn}$"
'
test_expect_success 'able to dcommit to a subdirectory' "
- git-svn fetch -i bar &&
+ git svn fetch -i bar &&
git checkout -b my-bar refs/remotes/bar &&
echo abc > d &&
git update-index --add d &&
git commit -m '/bar/d should be in the log' &&
- git-svn dcommit -i bar &&
+ git svn dcommit -i bar &&
test -z \"\`git diff refs/heads/my-bar refs/remotes/bar\`\" &&
mkdir newdir &&
echo new > newdir/dir &&
git update-index --add newdir/dir &&
git commit -m 'add a new directory' &&
- git-svn dcommit -i bar &&
+ git svn dcommit -i bar &&
test -z \"\`git diff refs/heads/my-bar refs/remotes/bar\`\" &&
echo foo >> newdir/dir &&
git update-index newdir/dir &&
git commit -m 'modify a file in new directory' &&
- git-svn dcommit -i bar &&
+ git svn dcommit -i bar &&
test -z \"\`git diff refs/heads/my-bar refs/remotes/bar\`\"
"
echo cba > d &&
git update-index d &&
git commit -m 'update /bar/d' &&
- git-svn set-tree -i bar HEAD &&
+ git svn set-tree -i bar HEAD &&
test -z \"\`git diff refs/heads/my-bar refs/remotes/bar\`\"
"
# Copyright (c) 2006 Eric Wong
#
-test_description='git-svn property tests'
+test_description='git svn property tests'
. ./lib-git-svn.sh
mkdir import
EOF
printf "Hello\r\nWorld\r\n" > crlf
- a_crlf=`git-hash-object -w crlf`
+ a_crlf=`git hash-object -w crlf`
printf "Hello\rWorld\r" > cr
- a_cr=`git-hash-object -w cr`
+ a_cr=`git hash-object -w cr`
printf "Hello\nWorld\n" > lf
- a_lf=`git-hash-object -w lf`
+ a_lf=`git hash-object -w lf`
printf "Hello\r\nWorld" > ne_crlf
- a_ne_crlf=`git-hash-object -w ne_crlf`
+ a_ne_crlf=`git hash-object -w ne_crlf`
printf "Hello\nWorld" > ne_lf
- a_ne_lf=`git-hash-object -w ne_lf`
+ a_ne_lf=`git hash-object -w ne_lf`
printf "Hello\rWorld" > ne_cr
- a_ne_cr=`git-hash-object -w ne_cr`
+ a_ne_cr=`git hash-object -w ne_cr`
touch empty
- a_empty=`git-hash-object -w empty`
+ a_empty=`git hash-object -w empty`
printf "\n" > empty_lf
- a_empty_lf=`git-hash-object -w empty_lf`
+ a_empty_lf=`git hash-object -w empty_lf`
printf "\r" > empty_cr
- a_empty_cr=`git-hash-object -w empty_cr`
+ a_empty_cr=`git hash-object -w empty_cr`
printf "\r\n" > empty_crlf
- a_empty_crlf=`git-hash-object -w empty_crlf`
+ a_empty_crlf=`git hash-object -w empty_crlf`
- svn import --no-auto-props -m 'import for git-svn' . "$svnrepo" >/dev/null
+ svn import --no-auto-props -m 'import for git svn' . "$svnrepo" >/dev/null
cd ..
rm -rf import
svn commit -m "Propset Id" &&
cd ..'
-test_expect_success 'initialize git-svn' 'git-svn init "$svnrepo"'
-test_expect_success 'fetch revisions from svn' 'git-svn fetch'
+test_expect_success 'initialize git svn' 'git svn init "$svnrepo"'
+test_expect_success 'fetch revisions from svn' 'git svn fetch'
name='test svn:keywords ignoring'
test_expect_success "$name" \
- 'git checkout -b mybranch remotes/git-svn &&
+ 'git checkout -b mybranch ${remotes_git_svn} &&
echo Hi again >> kw.c &&
git commit -a -m "test keywords ignoring" &&
- git-svn set-tree remotes/git-svn..mybranch &&
- git pull . remotes/git-svn'
+ git svn set-tree ${remotes_git_svn}..mybranch &&
+ git pull . ${remotes_git_svn}'
expect='/* $Id$ */'
got="`sed -ne 2p kw.c`"
cd ..'
test_expect_success 'fetch and pull latest from svn and checkout a new wc' \
- 'git-svn fetch &&
- git pull . remotes/git-svn &&
+ 'git svn fetch &&
+ git pull . ${remotes_git_svn} &&
svn co "$svnrepo" new_wc'
for i in crlf ne_crlf lf ne_lf cr ne_cr empty_cr empty_lf empty empty_crlf
cd test_wc
printf '$Id$\rHello\rWorld\r' > cr
printf '$Id$\rHello\rWorld' > ne_cr
- a_cr=`printf '$Id$\r\nHello\r\nWorld\r\n' | git-hash-object --stdin`
- a_ne_cr=`printf '$Id$\r\nHello\r\nWorld' | git-hash-object --stdin`
+ a_cr=`printf '$Id$\r\nHello\r\nWorld\r\n' | git hash-object --stdin`
+ a_ne_cr=`printf '$Id$\r\nHello\r\nWorld' | git hash-object --stdin`
test_expect_success 'Set CRLF on cr files' \
'svn propset svn:eol-style CRLF cr &&
svn propset svn:eol-style CRLF ne_cr &&
svn commit -m "propset CRLF on cr files"'
cd ..
test_expect_success 'fetch and pull latest from svn' \
- 'git-svn fetch && git pull . remotes/git-svn'
+ 'git svn fetch && git pull . ${remotes_git_svn}'
-b_cr="`git-hash-object cr`"
-b_ne_cr="`git-hash-object ne_cr`"
+b_cr="`git hash-object cr`"
+b_ne_cr="`git hash-object ne_cr`"
test_expect_success 'CRLF + $Id$' "test '$a_cr' = '$b_cr'"
test_expect_success 'CRLF + $Id$ (no newline)' "test '$a_ne_cr' = '$b_ne_cr'"
svn propset -R svn:ignore 'no-such-file*' .
svn commit -m 'propset svn:ignore'
cd .. &&
- git-svn show-ignore > show-ignore.got &&
+ git svn show-ignore > show-ignore.got &&
cmp show-ignore.expect show-ignore.got
"
EOF
test_expect_success 'test create-ignore' "
- git-svn fetch && git pull . remotes/git-svn &&
- git-svn create-ignore &&
+ git svn fetch && git pull . ${remotes_git_svn} &&
+ git svn create-ignore &&
cmp ./.gitignore create-ignore.expect &&
cmp ./deeply/.gitignore create-ignore.expect &&
cmp ./deeply/nested/.gitignore create-ignore.expect &&
# pattern, it can pass even though the propget did not execute on the
# right directory.
test_expect_success 'test propget' "
- git-svn propget svn:ignore . | cmp - prop.expect &&
+ git svn propget svn:ignore . | cmp - prop.expect &&
cd deeply &&
- git-svn propget svn:ignore . | cmp - ../prop.expect &&
- git-svn propget svn:entry:committed-rev nested/directory/.keep \
+ git svn propget svn:ignore . | cmp - ../prop.expect &&
+ git svn propget svn:entry:committed-rev nested/directory/.keep \
| cmp - ../prop2.expect &&
- git-svn propget svn:ignore .. | cmp - ../prop.expect &&
- git-svn propget svn:ignore nested/ | cmp - ../prop.expect &&
- git-svn propget svn:ignore ./nested | cmp - ../prop.expect &&
- git-svn propget svn:ignore .././deeply/nested | cmp - ../prop.expect
+ git svn propget svn:ignore .. | cmp - ../prop.expect &&
+ git svn propget svn:ignore nested/ | cmp - ../prop.expect &&
+ git svn propget svn:ignore ./nested | cmp - ../prop.expect &&
+ git svn propget svn:ignore .././deeply/nested | cmp - ../prop.expect
"
cat >prop.expect <<\EOF
EOF
test_expect_success 'test proplist' "
- git-svn proplist . | cmp - prop.expect &&
- git-svn proplist nested/directory/.keep | cmp - prop2.expect
+ git svn proplist . | cmp - prop.expect &&
+ git svn proplist nested/directory/.keep | cmp - prop2.expect
"
test_done
#!/bin/sh
-test_description='git-svn rmdir'
+test_description='git svn rmdir'
. ./lib-git-svn.sh
test_expect_success 'initialize repo' '
mkdir -p deeply/nested/directory/number/2 &&
echo foo > deeply/nested/directory/number/1/file &&
echo foo > deeply/nested/directory/number/2/another &&
- svn import -m "import for git-svn" . "$svnrepo" &&
+ svn import -m "import for git svn" . "$svnrepo" &&
cd ..
'
-test_expect_success 'mirror via git-svn' '
- git-svn init "$svnrepo" &&
- git-svn fetch &&
- git checkout -f -b test-rmdir remotes/git-svn
+test_expect_success 'mirror via git svn' '
+ git svn init "$svnrepo" &&
+ git svn fetch &&
+ git checkout -f -b test-rmdir ${remotes_git_svn}
'
test_expect_success 'Try a commit on rmdir' '
git rm -f deeply/nested/directory/number/2/another &&
git commit -a -m "remove another" &&
- git-svn set-tree --rmdir HEAD &&
+ git svn set-tree --rmdir HEAD &&
svn ls -R "$svnrepo" | grep ^deeply/nested/directory/number/1
'
# Copyright (c) 2007 Eric Wong
#
-test_description='git-svn tracking removed top-level path'
+test_description='git svn tracking removed top-level path'
. ./lib-git-svn.sh
test_expect_success 'make history for tracking' '
# Copyright (c) 2006 Eric Wong
#
-test_description='git-svn fetching'
+test_description='git svn fetching'
. ./lib-git-svn.sh
test_expect_success 'initialize repo' '
'
test_expect_success 'init and fetch a moved directory' '
- git-svn init --minimize-url -i thunk "$svnrepo"/thunk &&
- git-svn fetch -i thunk &&
+ git svn init --minimize-url -i thunk "$svnrepo"/thunk &&
+ git svn fetch -i thunk &&
test "`git rev-parse --verify refs/remotes/thunk@2`" \
= "`git rev-parse --verify refs/remotes/thunk~1`" &&
test "`git cat-file blob refs/remotes/thunk:readme |\
trunk:refs/remotes/svn/trunk &&
git config --add svn-remote.svn.fetch \
thunk:refs/remotes/svn/thunk &&
- git-svn fetch -i svn/thunk &&
+ git svn fetch -i svn/thunk &&
test "`git rev-parse --verify refs/remotes/svn/trunk`" \
= "`git rev-parse --verify refs/remotes/svn/thunk~1`" &&
test "`git cat-file blob refs/remotes/svn/thunk:readme |\
-r2 "$svnrepo"/trunk "$svnrepo"/junk) &&
git config --add svn-remote.svn.fetch \
junk:refs/remotes/svn/junk &&
- git-svn fetch -i svn/thunk &&
- git-svn fetch -i svn/junk &&
+ git svn fetch -i svn/thunk &&
+ git svn fetch -i svn/junk &&
test -z "`git diff svn/junk svn/trunk`" &&
test "`git merge-base svn/junk svn/trunk`" \
= "`git rev-parse svn/trunk`"
echo hi > import/trunk/thunk/bump/thud/file &&
svn import -m "import a larger parent" import "$svnrepo"/larger-parent &&
svn cp -m "hi" "$svnrepo"/larger-parent "$svnrepo"/another-larger &&
- git-svn init --minimize-url -i larger \
+ git svn init --minimize-url -i larger \
"$svnrepo"/another-larger/trunk/thunk/bump/thud &&
- git-svn fetch -i larger &&
+ git svn fetch -i larger &&
git rev-parse --verify refs/remotes/larger &&
git rev-parse --verify \
refs/remotes/larger-parent/trunk/thunk/bump/thud &&
cd ..
svn mkdir -m "new glob at top level" "$svnrepo"/glob &&
svn mv -m "move blob down a level" "$svnrepo"/blob "$svnrepo"/glob/blob &&
- git-svn init --minimize-url -i blob "$svnrepo"/glob/blob &&
- git-svn fetch -i blob
+ git svn init --minimize-url -i blob "$svnrepo"/glob/blob &&
+ git svn fetch -i blob
'
test_expect_success 'follow deleted directory' '
svn mv -m "bye!" "$svnrepo"/glob/blob/hi "$svnrepo"/glob/blob/bye &&
svn rm -m "remove glob" "$svnrepo"/glob &&
- git-svn init --minimize-url -i glob "$svnrepo"/glob &&
- git-svn fetch -i glob &&
+ git svn init --minimize-url -i glob "$svnrepo"/glob &&
+ git svn fetch -i glob &&
test "`git cat-file blob refs/remotes/glob:blob/bye`" = hi &&
test "`git ls-tree refs/remotes/glob | wc -l `" -eq 1
'
poke native/t/c.t &&
svn commit -m "reorg test" &&
cd .. &&
- git-svn init --minimize-url -i r9270-t \
+ git svn init --minimize-url -i r9270-t \
"$svnrepo"/r9270/trunk/subversion/bindings/swig/perl/native/t &&
- git-svn fetch -i r9270-t &&
+ git svn fetch -i r9270-t &&
test `git rev-list r9270-t | wc -l` -eq 2 &&
test "`git ls-tree --name-only r9270-t~1`" = \
"`git ls-tree --name-only r9270-t`"
test_expect_success "track initial change if it was only made to parent" '
svn cp -m "wheee!" "$svnrepo"/r9270/trunk "$svnrepo"/r9270/drunk &&
- git-svn init --minimize-url -i r9270-d \
+ git svn init --minimize-url -i r9270-d \
"$svnrepo"/r9270/drunk/subversion/bindings/swig/perl/native/t &&
- git-svn fetch -i r9270-d &&
+ git svn fetch -i r9270-d &&
test `git rev-list r9270-d | wc -l` -eq 3 &&
test "`git ls-tree --name-only r9270-t`" = \
"`git ls-tree --name-only r9270-d`" &&
test_expect_success "track multi-parent paths" '
svn cp -m "resurrect /glob" "$svnrepo"/r9270 "$svnrepo"/glob &&
- git-svn multi-fetch &&
+ git svn multi-fetch &&
test `git cat-file commit refs/remotes/glob | \
grep "^parent " | wc -l` -eq 2
'
test_expect_success "multi-fetch continues to work" "
- git-svn multi-fetch
+ git svn multi-fetch
"
test_expect_success "multi-fetch works off a 'clean' repository" '
rm -r "$GIT_DIR/svn" "$GIT_DIR/refs/remotes" "$GIT_DIR/logs" &&
mkdir "$GIT_DIR/svn" &&
- git-svn multi-fetch
+ git svn multi-fetch
'
test_debug 'gitk --all &'
#!/bin/sh
#
# Copyright (c) 2006 Eric Wong
-test_description='git-svn commit-diff'
+test_description='git svn commit-diff'
. ./lib-git-svn.sh
test_expect_success 'initialize repo' '
test_expect_success 'test the commit-diff command' '
test -n "$prev" && test -n "$head" &&
- git-svn commit-diff -r1 "$prev" "$head" "$svnrepo" &&
+ git svn commit-diff -r1 "$prev" "$head" "$svnrepo" &&
svn co "$svnrepo" wc &&
cmp readme wc/readme
'
-test_expect_success 'commit-diff to a sub-directory (with git-svn config)' '
+test_expect_success 'commit-diff to a sub-directory (with git svn config)' '
svn import -m "sub-directory" import "$svnrepo"/subdir &&
- git-svn init --minimize-url "$svnrepo"/subdir &&
- git-svn fetch &&
- git-svn commit-diff -r3 "$prev" "$head" &&
+ git svn init --minimize-url "$svnrepo"/subdir &&
+ git svn fetch &&
+ git svn commit-diff -r3 "$prev" "$head" &&
svn cat "$svnrepo"/subdir/readme > readme.2 &&
cmp readme readme.2
'
#!/bin/sh
#
# Copyright (c) 2006 Eric Wong
-test_description='git-svn commit-diff clobber'
+test_description='git svn commit-diff clobber'
. ./lib-git-svn.sh
test_expect_success 'initialize repo' '
test_expect_success 'commit conflicting change from git' '
echo second line from git >> file &&
git commit -a -m "second line from git" &&
- test_must_fail git-svn commit-diff -r1 HEAD~1 HEAD "$svnrepo"
+ test_must_fail git svn commit-diff -r1 HEAD~1 HEAD "$svnrepo"
'
test_expect_success 'commit complementing change from git' '
git commit -a -m "second line from svn" &&
echo third line from git >> file &&
git commit -a -m "third line from git" &&
- git-svn commit-diff -r2 HEAD~1 HEAD "$svnrepo"
+ git svn commit-diff -r2 HEAD~1 HEAD "$svnrepo"
'
test_expect_success 'dcommit fails to commit because of conflict' '
- git-svn init "$svnrepo" &&
- git-svn fetch &&
- git reset --hard refs/remotes/git-svn &&
+ git svn init "$svnrepo" &&
+ git svn fetch &&
+ git reset --hard refs/${remotes_git_svn} &&
svn co "$svnrepo" t.svn &&
cd t.svn &&
echo fourth line from svn >> file &&
rm -rf t.svn &&
echo "fourth line from git" >> file &&
git commit -a -m "fourth line from git" &&
- test_must_fail git-svn dcommit
+ test_must_fail git svn dcommit
'
test_expect_success 'dcommit does the svn equivalent of an index merge' "
- git reset --hard refs/remotes/git-svn &&
+ git reset --hard refs/${remotes_git_svn} &&
echo 'index merge' > file2 &&
git update-index --add file2 &&
git commit -a -m 'index merge' &&
echo 'more changes' >> file2 &&
git update-index file2 &&
git commit -a -m 'more changes' &&
- git-svn dcommit
+ git svn dcommit
"
test_expect_success 'commit another change from svn side' '
rm -rf t.svn
'
-test_expect_success 'multiple dcommit from git-svn will not clobber svn' "
- git reset --hard refs/remotes/git-svn &&
+test_expect_success 'multiple dcommit from git svn will not clobber svn' "
+ git reset --hard refs/${remotes_git_svn} &&
echo new file >> new-file &&
git update-index --add new-file &&
git commit -a -m 'new file' &&
#!/bin/sh
#
# Copyright (c) 2007 Eric Wong
-test_description='git-svn dcommit clobber series'
+test_description='git svn dcommit clobber series'
. ./lib-git-svn.sh
test_expect_success 'initialize repo' '
#!/bin/sh
# Copyright (c) 2006 Eric Wong
-test_description='git-svn metadata migrations from previous versions'
+test_description='git svn metadata migrations from previous versions'
. ./lib-git-svn.sh
test_expect_success 'setup old-looking metadata' '
done && \
svn import -m test . "$svnrepo"
cd .. &&
- git-svn init "$svnrepo" &&
- git-svn fetch &&
+ git svn init "$svnrepo" &&
+ git svn fetch &&
mv "$GIT_DIR"/svn/* "$GIT_DIR"/ &&
mv "$GIT_DIR"/svn/.metadata "$GIT_DIR"/ &&
rmdir "$GIT_DIR"/svn &&
- git update-ref refs/heads/git-svn-HEAD refs/remotes/git-svn &&
- git update-ref refs/heads/svn-HEAD refs/remotes/git-svn &&
- git update-ref -d refs/remotes/git-svn refs/remotes/git-svn
+ git update-ref refs/heads/git-svn-HEAD refs/${remotes_git_svn} &&
+ git update-ref refs/heads/svn-HEAD refs/${remotes_git_svn} &&
+ git update-ref -d refs/${remotes_git_svn} refs/${remotes_git_svn}
'
head=`git rev-parse --verify refs/heads/git-svn-HEAD^0`
test_expect_success 'git-svn-HEAD is a real HEAD' "test -n '$head'"
-test_expect_success 'initialize old-style (v0) git-svn layout' '
+test_expect_success 'initialize old-style (v0) git svn layout' '
mkdir -p "$GIT_DIR"/git-svn/info "$GIT_DIR"/svn/info &&
echo "$svnrepo" > "$GIT_DIR"/git-svn/info/url &&
echo "$svnrepo" > "$GIT_DIR"/svn/info/url &&
- git-svn migrate &&
- ! test -d "$GIT_DIR"/git-svn &&
- git rev-parse --verify refs/remotes/git-svn^0 &&
+ git svn migrate &&
+ ! test -d "$GIT_DIR"/git svn &&
+ git rev-parse --verify refs/${remotes_git_svn}^0 &&
git rev-parse --verify refs/remotes/svn^0 &&
test "$(git config --get svn-remote.svn.url)" = "$svnrepo" &&
test `git config --get svn-remote.svn.fetch` = \
- ":refs/remotes/git-svn"
+ ":refs/${remotes_git_svn}"
'
test_expect_success 'initialize a multi-repository repo' '
- git-svn init "$svnrepo" -T trunk -t tags -b branches &&
+ git svn init "$svnrepo" -T trunk -t tags -b branches &&
git config --get-all svn-remote.svn.fetch > fetch.out &&
grep "^trunk:refs/remotes/trunk$" fetch.out &&
test -n "`git config --get svn-remote.svn.branches \
# refs should all be different, but the trees should all be the same:
test_expect_success 'multi-fetch works on partial urls + paths' "
- git-svn multi-fetch &&
+ git svn multi-fetch &&
for i in trunk a b tags/0.1 tags/0.2 tags/0.3; do
git rev-parse --verify refs/remotes/\$i^0 >> refs.out || exit 1;
done &&
( mkdir -p "$GIT_DIR"/svn/$ref/info/ &&
echo "$svnrepo"$path > "$GIT_DIR"/svn/$ref/info/url ) || exit 1;
done &&
- git-svn migrate --minimize &&
+ git svn migrate --minimize &&
test -z "`git config -l |grep -v "^svn-remote\.git-svn\."`" &&
git config --get-all svn-remote.svn.fetch > fetch.out &&
grep "^trunk:refs/remotes/trunk$" fetch.out &&
grep "^tags/0\.1:refs/remotes/tags/0\.1$" fetch.out &&
grep "^tags/0\.2:refs/remotes/tags/0\.2$" fetch.out &&
grep "^tags/0\.3:refs/remotes/tags/0\.3$" fetch.out
- grep "^:refs/remotes/git-svn" fetch.out
+ grep "^:refs/${remotes_git_svn}" fetch.out
'
test_expect_success ".rev_db auto-converted to .rev_map.UUID" '
- git-svn fetch -i trunk &&
+ git svn fetch -i trunk &&
test -z "$(ls "$GIT_DIR"/svn/trunk/.rev_db.* 2>/dev/null)" &&
expect="$(ls "$GIT_DIR"/svn/trunk/.rev_map.*)" &&
test -n "$expect" &&
convert_to_rev_db "$expect" "$rev_db" &&
rm -f "$expect" &&
test -f "$rev_db" &&
- git-svn fetch -i trunk &&
+ git svn fetch -i trunk &&
test -z "$(ls "$GIT_DIR"/svn/trunk/.rev_db.* 2>/dev/null)" &&
test ! -e "$GIT_DIR"/svn/trunk/.rev_db &&
test -f "$expect"
#!/bin/sh
# Copyright (c) 2007 Eric Wong
-test_description='git-svn globbing refspecs'
+test_description='git svn globbing refspecs'
. ./lib-git-svn.sh
cat > expect.end <<EOF
"branches/*/src/a:refs/remotes/branches/*" &&
git config --add svn-remote.svn.tags\
"tags/*/src/a:refs/remotes/tags/*" &&
- git-svn multi-fetch &&
+ git svn multi-fetch &&
git log --pretty=oneline refs/remotes/tags/end | \
sed -e "s/^.\{41\}//" > output.end &&
test_cmp expect.end output.end &&
poke tags/end/src/b/readme &&
svn commit -m "try to try"
) &&
- git-svn fetch two &&
+ git svn fetch two &&
test `git rev-list refs/remotes/two/tags/end | wc -l` -eq 6 &&
test `git rev-list refs/remotes/two/branches/start | wc -l` -eq 3 &&
test `git rev-parse refs/remotes/two/branches/start~2` = \
poke tags/end/src/b/readme &&
svn commit -m "try to try"
) &&
- test_must_fail git-svn fetch three 2> stderr.three &&
+ test_must_fail git svn fetch three 2> stderr.three &&
test_cmp expect.three stderr.three
'
#!/bin/sh
# Copyright (c) 2007 Eric Wong
-test_description='git-svn globbing refspecs'
+test_description='git svn globbing refspecs'
. ./lib-git-svn.sh
cat > expect.end <<EOF
"branches/*/*/src/a:refs/remotes/branches/*/*" &&
git config --add svn-remote.svn.tags\
"tags/*/src/a:refs/remotes/tags/*" &&
- git-svn multi-fetch &&
+ git svn multi-fetch &&
git log --pretty=oneline refs/remotes/tags/end | \
sed -e "s/^.\{41\}//" > output.end &&
test_cmp expect.end output.end &&
poke tags/end/src/b/readme &&
svn commit -m "try to try"
) &&
- git-svn fetch two &&
+ git svn fetch two &&
test `git rev-list refs/remotes/two/tags/end | wc -l` -eq 6 &&
test `git rev-list refs/remotes/two/branches/v1/start | wc -l` -eq 3 &&
test `git rev-parse refs/remotes/two/branches/v1/start~2` = \
"branches/*/*:refs/remotes/four/branches/*/*" &&
git config --add svn-remote.four.tags \
"tags/*:refs/remotes/four/tags/*" &&
- git-svn fetch four &&
+ git svn fetch four &&
test `git rev-list refs/remotes/four/tags/next | wc -l` -eq 5 &&
test `git rev-list refs/remotes/four/branches/v2/start | wc -l` -eq 3 &&
test `git rev-parse refs/remotes/four/branches/v2/start~2` = \
poke tags/end/src/b/readme &&
svn commit -m "try to try"
) &&
- test_must_fail git-svn fetch three 2> stderr.three &&
+ test_must_fail git svn fetch three 2> stderr.three &&
test_cmp expect.three stderr.three
'
# Copyright (c) 2007 Eric Wong
#
-test_description='git-svn useSvmProps test'
+test_description='git svn useSvmProps test'
. ./lib-git-svn.sh
test_expect_success 'load svm repo' '
svnadmin load -q "$rawsvnrepo" < "$TEST_DIRECTORY"/t9110/svm.dump &&
- git-svn init --minimize-url -R arr -i bar "$svnrepo"/mirror/arr &&
- git-svn init --minimize-url -R argh -i dir "$svnrepo"/mirror/argh &&
- git-svn init --minimize-url -R argh -i e \
+ git svn init --minimize-url -R arr -i bar "$svnrepo"/mirror/arr &&
+ git svn init --minimize-url -R argh -i dir "$svnrepo"/mirror/argh &&
+ git svn init --minimize-url -R argh -i e \
"$svnrepo"/mirror/argh/a/b/c/d/e &&
git config svn.useSvmProps true &&
- git-svn fetch --all
+ git svn fetch --all
'
uuid=161ce429-a9dd-4828-af4a-52023f968c89
bar_url=http://mayonaise/svnrepo/bar
test_expect_success 'verify metadata for /bar' "
git cat-file commit refs/remotes/bar | \
- grep '^git-svn-id: $bar_url@12 $uuid$' &&
+ grep '^${git_svn_id}: $bar_url@12 $uuid$' &&
git cat-file commit refs/remotes/bar~1 | \
- grep '^git-svn-id: $bar_url@11 $uuid$' &&
+ grep '^${git_svn_id}: $bar_url@11 $uuid$' &&
git cat-file commit refs/remotes/bar~2 | \
- grep '^git-svn-id: $bar_url@10 $uuid$' &&
+ grep '^${git_svn_id}: $bar_url@10 $uuid$' &&
git cat-file commit refs/remotes/bar~3 | \
- grep '^git-svn-id: $bar_url@9 $uuid$' &&
+ grep '^${git_svn_id}: $bar_url@9 $uuid$' &&
git cat-file commit refs/remotes/bar~4 | \
- grep '^git-svn-id: $bar_url@6 $uuid$' &&
+ grep '^${git_svn_id}: $bar_url@6 $uuid$' &&
git cat-file commit refs/remotes/bar~5 | \
- grep '^git-svn-id: $bar_url@1 $uuid$'
+ grep '^${git_svn_id}: $bar_url@1 $uuid$'
"
e_url=http://mayonaise/svnrepo/dir/a/b/c/d/e
test_expect_success 'verify metadata for /dir/a/b/c/d/e' "
git cat-file commit refs/remotes/e | \
- grep '^git-svn-id: $e_url@1 $uuid$'
+ grep '^${git_svn_id}: $e_url@1 $uuid$'
"
dir_url=http://mayonaise/svnrepo/dir
test_expect_success 'verify metadata for /dir' "
git cat-file commit refs/remotes/dir | \
- grep '^git-svn-id: $dir_url@2 $uuid$' &&
+ grep '^${git_svn_id}: $dir_url@2 $uuid$' &&
git cat-file commit refs/remotes/dir~1 | \
- grep '^git-svn-id: $dir_url@1 $uuid$'
+ grep '^${git_svn_id}: $dir_url@1 $uuid$'
"
test_expect_success 'find commit based on SVN revision number' "
- git-svn find-rev r12 |
+ git svn find-rev r12 |
grep `git rev-parse HEAD`
"
test_expect_success 'empty rebase' "
- git-svn rebase
+ git svn rebase
"
test_done
# Copyright (c) 2007 Eric Wong
#
-test_description='git-svn useSvnsyncProps test'
+test_description='git svn useSvnsyncProps test'
. ./lib-git-svn.sh
test_expect_success 'load svnsync repo' '
svnadmin load -q "$rawsvnrepo" < "$TEST_DIRECTORY"/t9111/svnsync.dump &&
- git-svn init --minimize-url -R arr -i bar "$svnrepo"/bar &&
- git-svn init --minimize-url -R argh -i dir "$svnrepo"/dir &&
- git-svn init --minimize-url -R argh -i e "$svnrepo"/dir/a/b/c/d/e &&
+ git svn init --minimize-url -R arr -i bar "$svnrepo"/bar &&
+ git svn init --minimize-url -R argh -i dir "$svnrepo"/dir &&
+ git svn init --minimize-url -R argh -i e "$svnrepo"/dir/a/b/c/d/e &&
git config svn.useSvnsyncProps true &&
- git-svn fetch --all
+ git svn fetch --all
'
uuid=161ce429-a9dd-4828-af4a-52023f968c89
bar_url=http://mayonaise/svnrepo/bar
test_expect_success 'verify metadata for /bar' "
git cat-file commit refs/remotes/bar | \
- grep '^git-svn-id: $bar_url@12 $uuid$' &&
+ grep '^${git_svn_id}: $bar_url@12 $uuid$' &&
git cat-file commit refs/remotes/bar~1 | \
- grep '^git-svn-id: $bar_url@11 $uuid$' &&
+ grep '^${git_svn_id}: $bar_url@11 $uuid$' &&
git cat-file commit refs/remotes/bar~2 | \
- grep '^git-svn-id: $bar_url@10 $uuid$' &&
+ grep '^${git_svn_id}: $bar_url@10 $uuid$' &&
git cat-file commit refs/remotes/bar~3 | \
- grep '^git-svn-id: $bar_url@9 $uuid$' &&
+ grep '^${git_svn_id}: $bar_url@9 $uuid$' &&
git cat-file commit refs/remotes/bar~4 | \
- grep '^git-svn-id: $bar_url@6 $uuid$' &&
+ grep '^${git_svn_id}: $bar_url@6 $uuid$' &&
git cat-file commit refs/remotes/bar~5 | \
- grep '^git-svn-id: $bar_url@1 $uuid$'
+ grep '^${git_svn_id}: $bar_url@1 $uuid$'
"
e_url=http://mayonaise/svnrepo/dir/a/b/c/d/e
test_expect_success 'verify metadata for /dir/a/b/c/d/e' "
git cat-file commit refs/remotes/e | \
- grep '^git-svn-id: $e_url@1 $uuid$'
+ grep '^${git_svn_id}: $e_url@1 $uuid$'
"
dir_url=http://mayonaise/svnrepo/dir
test_expect_success 'verify metadata for /dir' "
git cat-file commit refs/remotes/dir | \
- grep '^git-svn-id: $dir_url@2 $uuid$' &&
+ grep '^${git_svn_id}: $dir_url@2 $uuid$' &&
git cat-file commit refs/remotes/dir~1 | \
- grep '^git-svn-id: $dir_url@1 $uuid$'
+ grep '^${git_svn_id}: $dir_url@1 $uuid$'
"
test_done
test_expect_success 'load svn dumpfile' 'svnadmin load "$rawsvnrepo" < dumpfile.svn'
-test_expect_success 'initialize git-svn' 'git-svn init "$svnrepo"'
-test_expect_success 'fetch revisions from svn' 'git-svn fetch'
+test_expect_success 'initialize git svn' 'git svn init "$svnrepo"'
+test_expect_success 'fetch revisions from svn' 'git svn fetch'
test_done
# daemon running on a users system if the test fails.
# Not all git users will need to interact with SVN.
-test_description='git-svn dcommit new files over svn:// test'
+test_description='git svn dcommit new files over svn:// test'
. ./lib-git-svn.sh
# Copyright (c) 2007 Eric Wong
# Based on a script by Joakim Tjernlund <joakim.tjernlund@transmode.se>
-test_description='git-svn dcommit handles merges'
+test_description='git svn dcommit handles merges'
. ./lib-git-svn.sh
# Copyright (c) 2007 Eric Wong
-test_description='git-svn dcommit can commit renames of files with ugly names'
+test_description='git svn dcommit can commit renames of files with ugly names'
. ./lib-git-svn.sh
git svn dcommit
'
-test_expect_success 'git-svn rebase works inside a fresh-cloned repository' '
+test_expect_success 'git svn rebase works inside a fresh-cloned repository' '
cd test-rebase &&
git svn rebase &&
test -e test-rebase-main &&
# Copyright (c) 2007 Eric Wong
#
-test_description='git-svn log tests'
+test_description='git svn log tests'
. ./lib-git-svn.sh
test_expect_success 'setup repository and import' '
done && \
svn import -m test . "$svnrepo"
cd .. &&
- git-svn init "$svnrepo" -T trunk -b branches -t tags &&
- git-svn fetch &&
+ git svn init "$svnrepo" -T trunk -b branches -t tags &&
+ git svn fetch &&
git reset --hard trunk &&
echo bye >> README &&
git commit -a -m bye &&
# Copyright (c) 2007 Eric Wong
#
-test_description='git-svn init/clone tests'
+test_description='git svn init/clone tests'
. ./lib-git-svn.sh
# Copyright (c) 2007 Eric Wong
#
-test_description='git-svn funky branch names'
+test_description='git svn funky branch names'
. ./lib-git-svn.sh
+# Abo-Uebernahme (Bug #994)
+scary_uri='Abo-Uebernahme%20%28Bug%20%23994%29'
+scary_ref='Abo-Uebernahme%20(Bug%20#994)'
+
test_expect_success 'setup svnrepo' '
mkdir project project/trunk project/branches project/tags &&
echo foo > project/trunk/foo &&
"$svnrepo/pr ject/branches/fun plugin" &&
svn cp -m "more fun!" "$svnrepo/pr ject/branches/fun plugin" \
"$svnrepo/pr ject/branches/more fun plugin!" &&
+ svn cp -m "scary" "$svnrepo/pr ject/branches/fun plugin" \
+ "$svnrepo/pr ject/branches/$scary_uri" &&
start_httpd
'
cd project &&
git rev-parse "refs/remotes/fun%20plugin" &&
git rev-parse "refs/remotes/more%20fun%20plugin!" &&
+ git rev-parse "refs/remotes/$scary_ref" &&
cd ..
'
cd ..
"
+test_expect_success 'test dcommit to scary branch' '
+ cd project &&
+ git reset --hard "refs/remotes/$scary_ref" &&
+ echo urls are scary >> foo &&
+ git commit -m "eep" -- foo &&
+ git svn dcommit &&
+ cd ..
+ '
+
stop_httpd
test_done
#
# Copyright (c) 2007 David D. Kilzer
-test_description='git-svn info'
+test_description='git svn info'
. ./lib-git-svn.sh
-set -e
-
# Tested with: svn, version 1.4.4 (r25188)
-v=`svn --version | sed -n -e 's/^svn, version \(1\.4\.[0-9]\).*$/\1/p'`
+v=`svn --version | sed -n -e 's/^svn, version \(1\.[0-9]*\.[0-9]*\).*$/\1/p'`
case $v in
-1.4.*)
+1.[45].*)
;;
*)
say "skipping svn-info test (SVN version: $v not supported)"
' "`svn info $2 | grep '^Text Last Updated:'`" "$1"
}
+quoted_svnrepo="$(echo $svnrepo | sed 's/ /%20/')"
+
test_expect_success 'setup repository and import' '
mkdir info &&
cd info &&
ln -s directory symlink-directory &&
svn import -m "initial" . "$svnrepo" &&
cd .. &&
+ svn co "$svnrepo" svnwc &&
+ cd svnwc &&
+ echo foo > foo &&
+ svn add foo &&
+ svn commit -m "change outside directory" &&
+ svn update &&
+ cd .. &&
mkdir gitwc &&
cd gitwc &&
- git-svn init "$svnrepo" &&
- git-svn fetch &&
+ git svn init "$svnrepo" &&
+ git svn fetch &&
cd .. &&
- svn co "$svnrepo" svnwc &&
- ptouch svnwc/file gitwc/file &&
- ptouch svnwc/directory gitwc/directory &&
- ptouch svnwc/symlink-file gitwc/symlink-file &&
- ptouch svnwc/symlink-directory gitwc/symlink-directory
+ ptouch gitwc/file svnwc/file &&
+ ptouch gitwc/directory svnwc/directory &&
+ ptouch gitwc/symlink-file svnwc/symlink-file &&
+ ptouch gitwc/symlink-directory svnwc/symlink-directory
'
test_expect_success 'info' "
(cd svnwc; svn info) > expected.info &&
- (cd gitwc; git-svn info) > actual.info &&
- git-diff expected.info actual.info
+ (cd gitwc; git svn info) > actual.info &&
+ test_cmp expected.info actual.info
"
test_expect_success 'info --url' '
- test "$(cd gitwc; git-svn info --url)" = "$svnrepo"
+ test "$(cd gitwc; git svn info --url)" = "$quoted_svnrepo"
'
test_expect_success 'info .' "
(cd svnwc; svn info .) > expected.info-dot &&
- (cd gitwc; git-svn info .) > actual.info-dot &&
- git-diff expected.info-dot actual.info-dot
+ (cd gitwc; git svn info .) > actual.info-dot &&
+ test_cmp expected.info-dot actual.info-dot
"
test_expect_success 'info --url .' '
- test "$(cd gitwc; git-svn info --url .)" = "$svnrepo"
+ test "$(cd gitwc; git svn info --url .)" = "$quoted_svnrepo"
'
test_expect_success 'info file' "
(cd svnwc; svn info file) > expected.info-file &&
- (cd gitwc; git-svn info file) > actual.info-file &&
- git-diff expected.info-file actual.info-file
+ (cd gitwc; git svn info file) > actual.info-file &&
+ test_cmp expected.info-file actual.info-file
"
test_expect_success 'info --url file' '
- test "$(cd gitwc; git-svn info --url file)" = "$svnrepo/file"
+ test "$(cd gitwc; git svn info --url file)" = "$quoted_svnrepo/file"
'
test_expect_success 'info directory' "
(cd svnwc; svn info directory) > expected.info-directory &&
- (cd gitwc; git-svn info directory) > actual.info-directory &&
- git-diff expected.info-directory actual.info-directory
+ (cd gitwc; git svn info directory) > actual.info-directory &&
+ test_cmp expected.info-directory actual.info-directory
+ "
+
+test_expect_success 'info inside directory' "
+ (cd svnwc/directory; svn info) > expected.info-inside-directory &&
+ (cd gitwc/directory; git svn info) > actual.info-inside-directory &&
+ test_cmp expected.info-inside-directory actual.info-inside-directory
"
test_expect_success 'info --url directory' '
- test "$(cd gitwc; git-svn info --url directory)" = "$svnrepo/directory"
+ test "$(cd gitwc; git svn info --url directory)" = "$quoted_svnrepo/directory"
'
test_expect_success 'info symlink-file' "
(cd svnwc; svn info symlink-file) > expected.info-symlink-file &&
- (cd gitwc; git-svn info symlink-file) > actual.info-symlink-file &&
- git-diff expected.info-symlink-file actual.info-symlink-file
+ (cd gitwc; git svn info symlink-file) > actual.info-symlink-file &&
+ test_cmp expected.info-symlink-file actual.info-symlink-file
"
test_expect_success 'info --url symlink-file' '
- test "$(cd gitwc; git-svn info --url symlink-file)" \
- = "$svnrepo/symlink-file"
+ test "$(cd gitwc; git svn info --url symlink-file)" \
+ = "$quoted_svnrepo/symlink-file"
'
test_expect_success 'info symlink-directory' "
(cd svnwc; svn info symlink-directory) \
> expected.info-symlink-directory &&
- (cd gitwc; git-svn info symlink-directory) \
+ (cd gitwc; git svn info symlink-directory) \
> actual.info-symlink-directory &&
- git-diff expected.info-symlink-directory actual.info-symlink-directory
+ test_cmp expected.info-symlink-directory actual.info-symlink-directory
"
test_expect_success 'info --url symlink-directory' '
- test "$(cd gitwc; git-svn info --url symlink-directory)" \
- = "$svnrepo/symlink-directory"
+ test "$(cd gitwc; git svn info --url symlink-directory)" \
+ = "$quoted_svnrepo/symlink-directory"
'
test_expect_success 'info added-file' "
svn add added-file > /dev/null &&
cd .. &&
(cd svnwc; svn info added-file) > expected.info-added-file &&
- (cd gitwc; git-svn info added-file) > actual.info-added-file &&
- git-diff expected.info-added-file actual.info-added-file
+ (cd gitwc; git svn info added-file) > actual.info-added-file &&
+ test_cmp expected.info-added-file actual.info-added-file
"
test_expect_success 'info --url added-file' '
- test "$(cd gitwc; git-svn info --url added-file)" \
- = "$svnrepo/added-file"
+ test "$(cd gitwc; git svn info --url added-file)" \
+ = "$quoted_svnrepo/added-file"
'
test_expect_success 'info added-directory' "
cd .. &&
(cd svnwc; svn info added-directory) \
> expected.info-added-directory &&
- (cd gitwc; git-svn info added-directory) \
+ (cd gitwc; git svn info added-directory) \
> actual.info-added-directory &&
- git-diff expected.info-added-directory actual.info-added-directory
+ test_cmp expected.info-added-directory actual.info-added-directory
"
test_expect_success 'info --url added-directory' '
- test "$(cd gitwc; git-svn info --url added-directory)" \
- = "$svnrepo/added-directory"
+ test "$(cd gitwc; git svn info --url added-directory)" \
+ = "$quoted_svnrepo/added-directory"
'
test_expect_success 'info added-symlink-file' "
ptouch gitwc/added-symlink-file svnwc/added-symlink-file &&
(cd svnwc; svn info added-symlink-file) \
> expected.info-added-symlink-file &&
- (cd gitwc; git-svn info added-symlink-file) \
+ (cd gitwc; git svn info added-symlink-file) \
> actual.info-added-symlink-file &&
- git-diff expected.info-added-symlink-file \
+ test_cmp expected.info-added-symlink-file \
actual.info-added-symlink-file
"
test_expect_success 'info --url added-symlink-file' '
- test "$(cd gitwc; git-svn info --url added-symlink-file)" \
- = "$svnrepo/added-symlink-file"
+ test "$(cd gitwc; git svn info --url added-symlink-file)" \
+ = "$quoted_svnrepo/added-symlink-file"
'
test_expect_success 'info added-symlink-directory' "
ptouch gitwc/added-symlink-directory svnwc/added-symlink-directory &&
(cd svnwc; svn info added-symlink-directory) \
> expected.info-added-symlink-directory &&
- (cd gitwc; git-svn info added-symlink-directory) \
+ (cd gitwc; git svn info added-symlink-directory) \
> actual.info-added-symlink-directory &&
- git-diff expected.info-added-symlink-directory \
+ test_cmp expected.info-added-symlink-directory \
actual.info-added-symlink-directory
"
test_expect_success 'info --url added-symlink-directory' '
- test "$(cd gitwc; git-svn info --url added-symlink-directory)" \
- = "$svnrepo/added-symlink-directory"
+ test "$(cd gitwc; git svn info --url added-symlink-directory)" \
+ = "$quoted_svnrepo/added-symlink-directory"
'
# The next few tests replace the "Text Last Updated" value with a
(cd svnwc; svn info file) |
sed -e 's/^\(Text Last Updated:\).*/\1 TEXT-LAST-UPDATED-STRING/' \
> expected.info-deleted-file &&
- (cd gitwc; git-svn info file) |
+ (cd gitwc; git svn info file) |
sed -e 's/^\(Text Last Updated:\).*/\1 TEXT-LAST-UPDATED-STRING/' \
> actual.info-deleted-file &&
- git-diff expected.info-deleted-file actual.info-deleted-file
+ test_cmp expected.info-deleted-file actual.info-deleted-file
"
test_expect_success 'info --url file (deleted)' '
- test "$(cd gitwc; git-svn info --url file)" \
- = "$svnrepo/file"
+ test "$(cd gitwc; git svn info --url file)" \
+ = "$quoted_svnrepo/file"
'
test_expect_success 'info deleted-directory' "
(cd svnwc; svn info directory) |
sed -e 's/^\(Text Last Updated:\).*/\1 TEXT-LAST-UPDATED-STRING/' \
> expected.info-deleted-directory &&
- (cd gitwc; git-svn info directory) |
+ (cd gitwc; git svn info directory) |
sed -e 's/^\(Text Last Updated:\).*/\1 TEXT-LAST-UPDATED-STRING/' \
> actual.info-deleted-directory &&
- git-diff expected.info-deleted-directory actual.info-deleted-directory
+ test_cmp expected.info-deleted-directory actual.info-deleted-directory
"
test_expect_success 'info --url directory (deleted)' '
- test "$(cd gitwc; git-svn info --url directory)" \
- = "$svnrepo/directory"
+ test "$(cd gitwc; git svn info --url directory)" \
+ = "$quoted_svnrepo/directory"
'
test_expect_success 'info deleted-symlink-file' "
(cd svnwc; svn info symlink-file) |
sed -e 's/^\(Text Last Updated:\).*/\1 TEXT-LAST-UPDATED-STRING/' \
> expected.info-deleted-symlink-file &&
- (cd gitwc; git-svn info symlink-file) |
+ (cd gitwc; git svn info symlink-file) |
sed -e 's/^\(Text Last Updated:\).*/\1 TEXT-LAST-UPDATED-STRING/' \
> actual.info-deleted-symlink-file &&
- git-diff expected.info-deleted-symlink-file \
+ test_cmp expected.info-deleted-symlink-file \
actual.info-deleted-symlink-file
"
test_expect_success 'info --url symlink-file (deleted)' '
- test "$(cd gitwc; git-svn info --url symlink-file)" \
- = "$svnrepo/symlink-file"
+ test "$(cd gitwc; git svn info --url symlink-file)" \
+ = "$quoted_svnrepo/symlink-file"
'
test_expect_success 'info deleted-symlink-directory' "
(cd svnwc; svn info symlink-directory) |
sed -e 's/^\(Text Last Updated:\).*/\1 TEXT-LAST-UPDATED-STRING/' \
> expected.info-deleted-symlink-directory &&
- (cd gitwc; git-svn info symlink-directory) |
+ (cd gitwc; git svn info symlink-directory) |
sed -e 's/^\(Text Last Updated:\).*/\1 TEXT-LAST-UPDATED-STRING/' \
> actual.info-deleted-symlink-directory &&
- git-diff expected.info-deleted-symlink-directory \
+ test_cmp expected.info-deleted-symlink-directory \
actual.info-deleted-symlink-directory
"
test_expect_success 'info --url symlink-directory (deleted)' '
- test "$(cd gitwc; git-svn info --url symlink-directory)" \
- = "$svnrepo/symlink-directory"
+ test "$(cd gitwc; git svn info --url symlink-directory)" \
+ = "$quoted_svnrepo/symlink-directory"
'
# NOTE: git does not have the concept of replaced objects,
test_expect_success 'info unknown-file' "
echo two > gitwc/unknown-file &&
- cp gitwc/unknown-file svnwc/unknown-file &&
- ptouch gitwc/unknown-file svnwc/unknown-file &&
- (cd svnwc; svn info unknown-file) 2> expected.info-unknown-file &&
- (cd gitwc; git-svn info unknown-file) 2> actual.info-unknown-file &&
- git-diff expected.info-unknown-file actual.info-unknown-file
+ (cd gitwc; test_must_fail git svn info unknown-file) \
+ 2> actual.info-unknown-file &&
+ grep unknown-file actual.info-unknown-file
"
test_expect_success 'info --url unknown-file' '
- test -z "$(cd gitwc; git-svn info --url unknown-file \
- 2> ../actual.info--url-unknown-file)" &&
- git-diff expected.info-unknown-file actual.info--url-unknown-file
+ echo two > gitwc/unknown-file &&
+ (cd gitwc; test_must_fail git svn info --url unknown-file) \
+ 2> actual.info-url-unknown-file &&
+ grep unknown-file actual.info-url-unknown-file
'
test_expect_success 'info unknown-directory' "
mkdir gitwc/unknown-directory svnwc/unknown-directory &&
- ptouch gitwc/unknown-directory svnwc/unknown-directory &&
- touch gitwc/unknown-directory/.placeholder &&
- (cd svnwc; svn info unknown-directory) \
- 2> expected.info-unknown-directory &&
- (cd gitwc; git-svn info unknown-directory) \
- 2> actual.info-unknown-directory &&
- git-diff expected.info-unknown-directory actual.info-unknown-directory
+ (cd gitwc; test_must_fail git svn info unknown-directory) \
+ 2> actual.info-unknown-directory &&
+ grep unknown-directory actual.info-unknown-directory
"
test_expect_success 'info --url unknown-directory' '
- test -z "$(cd gitwc; git-svn info --url unknown-directory \
- 2> ../actual.info--url-unknown-directory)" &&
- git-diff expected.info-unknown-directory \
- actual.info--url-unknown-directory
+ (cd gitwc; test_must_fail git svn info --url unknown-directory) \
+ 2> actual.info-url-unknown-directory &&
+ grep unknown-directory actual.info-url-unknown-directory
'
test_expect_success 'info unknown-symlink-file' "
cd gitwc &&
ln -s unknown-file unknown-symlink-file &&
cd .. &&
- cd svnwc &&
- ln -s unknown-file unknown-symlink-file &&
- cd .. &&
- ptouch gitwc/unknown-symlink-file svnwc/unknown-symlink-file &&
- (cd svnwc; svn info unknown-symlink-file) \
- 2> expected.info-unknown-symlink-file &&
- (cd gitwc; git-svn info unknown-symlink-file) \
- 2> actual.info-unknown-symlink-file &&
- git-diff expected.info-unknown-symlink-file \
- actual.info-unknown-symlink-file
+ (cd gitwc; test_must_fail git svn info unknown-symlink-file) \
+ 2> actual.info-unknown-symlink-file &&
+ grep unknown-symlink-file actual.info-unknown-symlink-file
"
test_expect_success 'info --url unknown-symlink-file' '
- test -z "$(cd gitwc; git-svn info --url unknown-symlink-file \
- 2> ../actual.info--url-unknown-symlink-file)" &&
- git-diff expected.info-unknown-symlink-file \
- actual.info--url-unknown-symlink-file
+ (cd gitwc; test_must_fail git svn info --url unknown-symlink-file) \
+ 2> actual.info-url-unknown-symlink-file &&
+ grep unknown-symlink-file actual.info-url-unknown-symlink-file
'
test_expect_success 'info unknown-symlink-directory' "
cd gitwc &&
ln -s unknown-directory unknown-symlink-directory &&
cd .. &&
- cd svnwc &&
- ln -s unknown-directory unknown-symlink-directory &&
- cd .. &&
- ptouch gitwc/unknown-symlink-directory \
- svnwc/unknown-symlink-directory &&
- (cd svnwc; svn info unknown-symlink-directory) \
- 2> expected.info-unknown-symlink-directory &&
- (cd gitwc; git-svn info unknown-symlink-directory) \
- 2> actual.info-unknown-symlink-directory &&
- git-diff expected.info-unknown-symlink-directory \
- actual.info-unknown-symlink-directory
+ (cd gitwc; test_must_fail git svn info unknown-symlink-directory) \
+ 2> actual.info-unknown-symlink-directory &&
+ grep unknown-symlink-directory actual.info-unknown-symlink-directory
"
test_expect_success 'info --url unknown-symlink-directory' '
- test -z "$(cd gitwc; git-svn info --url unknown-symlink-directory \
- 2> ../actual.info--url-unknown-symlink-directory)" &&
- git-diff expected.info-unknown-symlink-directory \
- actual.info--url-unknown-symlink-directory
+ (cd gitwc; test_must_fail git svn info --url unknown-symlink-directory) \
+ 2> actual.info-url-unknown-symlink-directory &&
+ grep unknown-symlink-directory actual.info-url-unknown-symlink-directory
'
test_done
# Copyright (c) 2008 Kevin Ballard
#
-test_description='git-svn clone with percent escapes'
+test_description='git svn clone with percent escapes'
. ./lib-git-svn.sh
test_expect_success 'setup svnrepo' '
test_expect_success 'test clone with percent escapes' '
git svn clone "$svnrepo/pr%20ject" clone &&
cd clone &&
- git rev-parse refs/remotes/git-svn &&
+ git rev-parse refs/${remotes_git_svn} &&
cd ..
'
fi
# Copyright (c) 2008 Santhosh Kumar Mani
-test_description='git-svn can fetch renamed directories'
+test_description='git svn can fetch renamed directories'
. ./lib-git-svn.sh
)
'
-test_expect_success 'interact with it via git-svn' '
+test_expect_success 'interact with it via git svn' '
mkdir work.git &&
(
cd work.git &&
# Copyright (c) 2008 Jan Krüger
#
-test_description='git-svn respects rewriteRoot during rebuild'
+test_description='git svn respects rewriteRoot during rebuild'
. ./lib-git-svn.sh
mkdir import
cd import
touch foo
- svn import -m 'import for git-svn' . "$svnrepo" >/dev/null
+ svn import -m 'import for git svn' . "$svnrepo" >/dev/null
cd ..
rm -rf import
test_expect_success 'init, fetch and checkout repository' '
git svn init --rewrite-root=http://invalid.invalid/ "$svnrepo" &&
git svn fetch
- git checkout -b mybranch remotes/git-svn
+ git checkout -b mybranch ${remotes_git_svn}
'
test_expect_success 'remove rev_map' '
#
# Copyright (c) 2008 Brad King
-test_description='git-svn dcommit honors auto-props'
+test_description='git svn dcommit honors auto-props'
. ./lib-git-svn.sh
EOF
}
-test_expect_success 'initialize git-svn' '
+test_expect_success 'initialize git svn' '
mkdir import &&
(
cd import &&
echo foo >foo &&
- svn import -m "import for git-svn" . "$svnrepo"
+ svn import -m "import for git svn" . "$svnrepo"
) &&
rm -rf import &&
- git-svn init "$svnrepo"
- git-svn fetch
+ git svn init "$svnrepo"
+ git svn fetch
'
test_expect_success 'enable auto-props config' '
- cd "$gittestrepo" &&
mkdir user &&
generate_auto_props yes >user/config
'
test_expect_success 'add files matching auto-props' '
- cd "$gittestrepo" &&
echo "#!$SHELL_PATH" >exec1.sh &&
chmod +x exec1.sh &&
echo "hello" >hello.txt &&
'
test_expect_success 'disable auto-props config' '
- cd "$gittestrepo" &&
generate_auto_props no >user/config
'
test_expect_success 'add files matching disabled auto-props' '
- cd "$gittestrepo" &&
echo "#$SHELL_PATH" >exec2.sh &&
chmod +x exec2.sh &&
echo "world" >world.txt &&
'
test_expect_success 'check resulting svn repository' '
+(
mkdir work &&
cd work &&
svn co "$svnrepo" &&
test "x$(svn propget svn:mime-type world.txt)" = "x" &&
test "x$(svn propget svn:eol-style world.txt)" = "x" &&
test "x$(svn propget svn:mime-type zot)" = "x"
+)
+'
+
+test_expect_success 'check renamed file' '
+ test -d user &&
+ generate_auto_props yes > user/config &&
+ git mv foo foo.sh &&
+ git commit -m "foo => foo.sh" &&
+ git svn dcommit --config-dir=user &&
+ (
+ cd work/svnrepo &&
+ svn up &&
+ test ! -e foo &&
+ test -e foo.sh &&
+ test "x$(svn propget svn:mime-type foo.sh)" = \
+ "xapplication/x-shellscript" &&
+ test "x$(svn propget svn:eol-style foo.sh)" = "xLF"
+ )
'
test_done
#!/bin/sh
# Copyright (c) 2008 Marcus Griep
-test_description='git-svn multi-glob branch names'
+test_description='git svn multi-glob branch names'
. ./lib-git-svn.sh
test_expect_success 'setup svnrepo' '
cvs >/dev/null 2>&1
if test $? -ne 1
then
- test_expect_success 'skipping git-cvsexportcommit tests, cvs not found' :
+ test_expect_success 'skipping git cvsexportcommit tests, cvs not found' :
test_done
exit
fi
diff F/newfile6.png ../F/newfile6.png
)'
-# Should fail (but only on the git-cvsexportcommit stage)
+# Should fail (but only on the git cvsexportcommit stage)
test_expect_success \
'Fail to change binary more than one generation old' \
'cat F/newfile6.png >>D/newfile4.png &&
git commit -a -m "With spaces" &&
id=$(git rev-list --max-count=1 HEAD) &&
(cd "$CVSWORK" &&
- git-cvsexportcommit -c $id &&
+ git cvsexportcommit -c $id &&
check_entries "G g" "with spaces.png/1.1/-kb|with spaces.txt/1.1/"
)'
git commit -a -m "Update with spaces" &&
id=$(git rev-list --max-count=1 HEAD) &&
(cd "$CVSWORK" &&
- git-cvsexportcommit -c $id
+ git cvsexportcommit -c $id
check_entries "G g" "with spaces.png/1.2/-kb|with spaces.txt/1.2/"
)'
git commit -a -m "Går det så går det" && \
id=$(git rev-list --max-count=1 HEAD) &&
(cd "$CVSWORK" &&
- git-cvsexportcommit -v -c $id &&
+ git cvsexportcommit -v -c $id &&
check_entries \
"Å/goo/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/å/ä/ö" \
"gårdetsågårdet.png/1.1/-kb|gårdetsågårdet.txt/1.1/"
git commit -a -m "Update two" &&
id=$(git rev-list --max-count=1 HEAD) &&
(cd "$CVSWORK" &&
- test_must_fail git-cvsexportcommit -c $id
+ test_must_fail git cvsexportcommit -c $id
)'
case "$(git config --bool core.filemode)" in
git add G/off &&
git commit -a -m "Execute test" &&
(cd "$CVSWORK" &&
- git-cvsexportcommit -c HEAD
+ git cvsexportcommit -c HEAD
test -x G/on &&
! test -x G/off
)'
# Copyright (c) 2007 Shawn Pearce
#
-test_description='test git-fast-import utility'
+test_description='test git fast-import utility'
. ./test-lib.sh
. "$TEST_DIRECTORY"/diff-lib.sh ;# test-lib chdir's into trash
INPUT_END
test_expect_success \
'A: create pack from stdin' \
- 'git-fast-import --export-marks=marks.out <input &&
+ 'git fast-import --export-marks=marks.out <input &&
git whatchanged master'
test_expect_success \
'A: verify pack' \
test_expect_success \
'A: verify marks import' \
- 'git-fast-import \
+ 'git fast-import \
--import-marks=marks.out \
--export-marks=marks.new \
</dev/null &&
INPUT_END
test_expect_success \
'A: verify marks import does not crash' \
- 'git-fast-import --import-marks=marks.out <input &&
+ 'git fast-import --import-marks=marks.out <input &&
git whatchanged verify--import-marks'
test_expect_success \
'A: verify pack' \
INPUT_END
test_expect_success 'B: fail on invalid blob sha1' '
- test_must_fail git-fast-import <input
+ test_must_fail git fast-import <input
'
rm -f .git/objects/pack_* .git/objects/index_*
INPUT_END
test_expect_success 'B: fail on invalid branch name ".badbranchname"' '
- test_must_fail git-fast-import <input
+ test_must_fail git fast-import <input
'
rm -f .git/objects/pack_* .git/objects/index_*
INPUT_END
test_expect_success 'B: fail on invalid branch name "bad[branch]name"' '
- test_must_fail git-fast-import <input
+ test_must_fail git fast-import <input
'
rm -f .git/objects/pack_* .git/objects/index_*
INPUT_END
test_expect_success \
'B: accept branch name "TEMP_TAG"' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
test -f .git/TEMP_TAG &&
test `git rev-parse master` = `git rev-parse TEMP_TAG^`'
rm -f .git/TEMP_TAG
### series C
###
-newf=`echo hi newf | git-hash-object -w --stdin`
+newf=`echo hi newf | git hash-object -w --stdin`
oldf=`git rev-parse --verify master:file2`
test_tick
cat >input <<INPUT_END
INPUT_END
test_expect_success \
'C: incremental import create pack from stdin' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
git whatchanged branch'
test_expect_success \
'C: verify pack' \
INPUT_END
test_expect_success \
'D: inline data in commit' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
git whatchanged branch'
test_expect_success \
'D: verify pack' \
INPUT_END
test_expect_success 'E: rfc2822 date, --date-format=raw' '
- test_must_fail git-fast-import --date-format=raw <input
+ test_must_fail git fast-import --date-format=raw <input
'
test_expect_success \
'E: rfc2822 date, --date-format=rfc2822' \
- 'git-fast-import --date-format=rfc2822 <input'
+ 'git fast-import --date-format=rfc2822 <input'
test_expect_success \
'E: verify pack' \
'for p in .git/objects/pack/*.pack;do git verify-pack $p||exit;done'
INPUT_END
test_expect_success \
'F: non-fast-forward update skips' \
- 'if git-fast-import <input
+ 'if git fast-import <input
then
echo BAD gfi did not fail
return 1
INPUT_END
test_expect_success \
'G: non-fast-forward update forced' \
- 'git-fast-import --force <input'
+ 'git fast-import --force <input'
test_expect_success \
'G: verify pack' \
'for p in .git/objects/pack/*.pack;do git verify-pack $p||exit;done'
INPUT_END
test_expect_success \
'H: deletall, add 1' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
git whatchanged H'
test_expect_success \
'H: verify pack' \
INPUT_END
test_expect_success \
'I: export-pack-edges' \
- 'git-fast-import --export-pack-edges=edges.list <input'
+ 'git fast-import --export-pack-edges=edges.list <input'
cat >expect <<EOF
.git/objects/pack/pack-.pack: `git rev-parse --verify export-boundary`
INPUT_END
test_expect_success \
'J: reset existing branch creates empty commit' \
- 'git-fast-import <input'
+ 'git fast-import <input'
test_expect_success \
'J: branch has 1 commit, empty tree' \
'test 1 = `git rev-list J | wc -l` &&
INPUT_END
test_expect_success \
'K: reinit branch with from' \
- 'git-fast-import <input'
+ 'git fast-import <input'
test_expect_success \
'K: verify K^1 = branch^1' \
'test `git rev-parse --verify branch^1` \
test_expect_success \
'L: verify internal tree sorting' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
git diff-tree --abbrev --raw L^ L >output &&
test_cmp expect output'
EOF
test_expect_success \
'M: rename file in same subdirectory' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
git diff-tree -M -r M1^ M1 >actual &&
compare_diff_raw expect actual'
EOF
test_expect_success \
'M: rename file to new subdirectory' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
git diff-tree -M -r M2^ M2 >actual &&
compare_diff_raw expect actual'
EOF
test_expect_success \
'M: rename subdirectory to new subdirectory' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
git diff-tree -M -r M3^ M3 >actual &&
compare_diff_raw expect actual'
EOF
test_expect_success \
'N: copy file in same subdirectory' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
git diff-tree -C --find-copies-harder -r N1^ N1 >actual &&
compare_diff_raw expect actual'
EOF
test_expect_success \
'N: copy then modify subdirectory' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
git diff-tree -C --find-copies-harder -r N2^^ N2 >actual &&
compare_diff_raw expect actual'
test_expect_success \
'N: copy dirty subdirectory' \
- 'git-fast-import <input &&
- test `git-rev-parse N2^{tree}` = `git-rev-parse N3^{tree}`'
+ 'git fast-import <input &&
+ test `git rev-parse N2^{tree}` = `git rev-parse N3^{tree}`'
###
### series O
test_expect_success \
'O: comments are all skipped' \
- 'git-fast-import <input &&
- test `git-rev-parse N3` = `git-rev-parse O1`'
+ 'git fast-import <input &&
+ test `git rev-parse N3` = `git rev-parse O1`'
cat >input <<INPUT_END
commit refs/heads/O2
test_expect_success \
'O: blank lines not necessary after data commands' \
- 'git-fast-import <input &&
- test `git-rev-parse N3` = `git-rev-parse O2`'
+ 'git fast-import <input &&
+ test `git rev-parse N3` = `git rev-parse O2`'
test_expect_success \
'O: repack before next test' \
INPUT_END
test_expect_success \
'O: blank lines not necessary after other commands' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
test 8 = `find .git/objects/pack -type f | wc -l` &&
test `git rev-parse refs/tags/O3-2nd` = `git rev-parse O3^` &&
git log --reverse --pretty=oneline O3 | sed s/^.*z// >actual &&
INPUT_END
test_expect_success \
'O: progress outputs as requested by input' \
- 'git-fast-import <input >actual &&
+ 'git fast-import <input >actual &&
grep "progress " <input >expect &&
test_cmp expect actual'
test_expect_success \
'P: supermodule & submodule mix' \
- 'git-fast-import <input &&
+ 'git fast-import <input &&
git checkout subuse1 &&
rm -rf sub && mkdir sub && cd sub &&
git init &&
git submodule init &&
git submodule update'
-SUBLAST=$(git-rev-parse --verify sub)
-SUBPREV=$(git-rev-parse --verify sub^)
+SUBLAST=$(git rev-parse --verify sub)
+SUBPREV=$(git rev-parse --verify sub^)
cat >input <<INPUT_END
blob
'P: verbatim SHA gitlinks' \
'git branch -D sub &&
git gc && git prune &&
- git-fast-import <input &&
- test $(git-rev-parse --verify subuse2) = $(git-rev-parse --verify subuse1)'
+ git fast-import <input &&
+ test $(git rev-parse --verify subuse2) = $(git rev-parse --verify subuse1)'
test_tick
cat >input <<INPUT_END
INPUT_END
test_expect_success 'P: fail on inline gitlink' '
- test_must_fail git-fast-import <input'
+ test_must_fail git fast-import <input'
test_tick
cat >input <<INPUT_END
INPUT_END
test_expect_success 'P: fail on blob mark in gitlink' '
- test_must_fail git-fast-import <input'
+ test_must_fail git fast-import <input'
test_done
# Copyright (c) 2007 Johannes E. Schindelin
#
-test_description='git-fast-export'
+test_description='git fast-export'
. ./test-lib.sh
test_expect_success 'setup' '
! grep -v "^master[ ]\+master$" < out
'
+#------------
+# CVS ANNOTATE
+#------------
+
+cd "$WORKDIR"
+test_expect_success 'cvs annotate' '
+ cd cvswork &&
+ GIT_CONFIG="$git_config" cvs annotate merge >../out &&
+ sed -e "s/ .*//" ../out >../actual &&
+ for i in 3 1 1 1 1 1 1 1 2 4; do echo 1.$i; done >../expect &&
+ test_cmp ../expect ../actual
+'
+
test_done
#!/bin/sh
-test_description='git-cvsimport basic tests'
+test_description='git cvsimport basic tests'
. ./test-lib.sh
CVSROOT=$(pwd)/cvsroot
echo "changed file 1" > file1 &&
git commit -a -m "second commit" &&
- git-config --add color.test.slot1 green &&
- git-config --add test.string value &&
- git-config --add test.dupstring value1 &&
- git-config --add test.dupstring value2 &&
- git-config --add test.booltrue true &&
- git-config --add test.boolfalse no &&
- git-config --add test.boolother other &&
- git-config --add test.int 2k
+ git config --add color.test.slot1 green &&
+ git config --add test.string value &&
+ git config --add test.dupstring value1 &&
+ git config --add test.dupstring value2 &&
+ git config --add test.booltrue true &&
+ git config --add test.boolfalse no &&
+ git config --add test.boolother other &&
+ git config --add test.int 2k
'
test_external_without_stderr \
use Cwd;
use File::Basename;
-use File::Temp;
BEGIN { use_ok('Git') }
# Failure cases for config:
# Save and restore STDERR; we will probably extract this into a
# "dies_ok" method and possibly move the STDERR handling to Git.pm.
-open our $tmpstderr, ">&", STDERR or die "cannot save STDERR"; close STDERR;
+open our $tmpstderr, ">&STDERR" or die "cannot save STDERR"; close STDERR;
eval { $r->config("test.dupstring") };
ok($@, "config: duplicate entry in scalar context fails");
eval { $r->config_bool("test.boolother") };
# objects and hashes
ok(our $file1hash = $r->command_oneline('rev-parse', "HEAD:file1"), "(get file hash)");
-our $tmpfile = File::Temp->new;
-is($r->cat_blob($file1hash, $tmpfile), 15, "cat_blob: size");
+my $tmpfile = "file.tmp";
+open TEMPFILE, "+>$tmpfile" or die "Can't open $tmpfile: $!";
+is($r->cat_blob($file1hash, \*TEMPFILE), 15, "cat_blob: size");
our $blobcontents;
-{ local $/; seek $tmpfile, 0, 0; $blobcontents = <$tmpfile>; }
+{ local $/; seek TEMPFILE, 0, 0; $blobcontents = <TEMPFILE>; }
is($blobcontents, "changed file 1\n", "cat_blob: data");
-seek $tmpfile, 0, 0;
+close TEMPFILE or die "Failed writing to $tmpfile: $!";
is(Git::hash_object("blob", $tmpfile), $file1hash, "hash_object: roundtrip");
-$tmpfile = File::Temp->new();
-print $tmpfile my $test_text = "test blob, to be inserted\n";
+open TEMPFILE, ">$tmpfile" or die "Can't open $tmpfile: $!";
+print TEMPFILE my $test_text = "test blob, to be inserted\n";
+close TEMPFILE or die "Failed writing to $tmpfile: $!";
like(our $newhash = $r->hash_and_insert_object($tmpfile), qr/[0-9a-fA-F]{40}/,
"hash_and_insert_object: returns hash");
-$tmpfile = File::Temp->new;
-is($r->cat_blob($newhash, $tmpfile), length $test_text, "cat_blob: roundtrip size");
-{ local $/; seek $tmpfile, 0, 0; $blobcontents = <$tmpfile>; }
+open TEMPFILE, "+>$tmpfile" or die "Can't open $tmpfile: $!";
+is($r->cat_blob($newhash, \*TEMPFILE), length $test_text, "cat_blob: roundtrip size");
+{ local $/; seek TEMPFILE, 0, 0; $blobcontents = <TEMPFILE>; }
is($blobcontents, $test_text, "cat_blob: roundtrip data");
+close TEMPFILE;
+unlink $tmpfile;
# paths
is($r->repo_path, "./.git", "repo_path");
dir=`expr "$$dst" : '\(.*\)/'` && \
mkdir -p blt/$$dir && \
case "$$boilerplate" in \
- *--) ;; \
- *) cp -p $$boilerplate blt/$$dst ;; \
- esac || exit; \
+ *--) continue;; \
+ esac && \
+ cp $$boilerplate blt/$$dst && \
+ if test -x "blt/$$dst"; then rx=rx; else rx=r; fi && \
+ chmod a+$$rx "blt/$$dst" || exit; \
done && \
date >$@
return -1;
}
}
- else if (newtree)
+ else if (newtree) {
+ if (oldtree && !o->initial_checkout) {
+ /*
+ * deletion of the path was staged;
+ */
+ if (same(oldtree, newtree))
+ return 1;
+ return reject_merge(oldtree, o);
+ }
return merged_entry(newtree, current, o);
+ }
return deleted_entry(oldtree, current, o);
}
verbose_update:1,
aggressive:1,
skip_unmerged:1,
+ initial_checkout:1,
gently:1;
const char *prefix;
int pos;