gitweb.git
git-apply --binary: clean up and prepare for --reverseJunio C Hamano Tue, 15 Aug 2006 09:23:06 +0000 (02:23 -0700)

git-apply --binary: clean up and prepare for --reverse

This cleans up the implementation of "git-apply --binary", and
implements reverse application of binary patches (when git-diff
is converted to emit reversible binary patches).

Earlier, the types of encoding (either deflated literal or
deflated delta) were stored in is_binary field in struct patch,
which meant that we cannot store more than one fragment that
differ in the encoding for a patch. This moves the information
to a field in struct fragment that is otherwise unused for
binary patches, and makes it possible to hang two (or more, but
two is enough) hunks for a binary patch.

The original "binary patch" output from git-diff is internally
parsed into an "is_binary" patch with one fragment. Upcoming
reversible binary patch output will have two fragments, the
first one being the forward patch and the second one the reverse
patch.

Signed-off-by: Junio C Hamano <junkio@cox.net>

Fix detection of ipv6 on SolarisDennis Stosberg Tue, 15 Aug 2006 09:01:27 +0000 (11:01 +0200)

Fix detection of ipv6 on Solaris

The configuration script detects whether linking with -lsocket is
necessary but doesn't add -lsocket to LIBS. This lets the ipv6 test
fail.

Signed-off-by: Dennis Stosberg <dennis@stosberg.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Look for sockaddr_storage in sys/socket.hDennis Stosberg Tue, 15 Aug 2006 09:01:25 +0000 (11:01 +0200)

Look for sockaddr_storage in sys/socket.h

On Solaris and the BSDs the definition of "struct sockaddr_storage"
is not available from "netinet/in.h". On Solaris "sys/socket.h" is
enough, at least OpenBSD needs "sys/types.h", too.

Using "sys/types.h" and "sys/socket.h" seems to be a more portable
way.

Signed-off-by: Dennis Stosberg <dennis@stosberg.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Solaris has strlcpy() at least since version 8Dennis Stosberg Tue, 15 Aug 2006 09:01:20 +0000 (11:01 +0200)

Solaris has strlcpy() at least since version 8

See http://docs.sun.com/app/docs/doc/816-3321/6m9k23sjk?a=view

Signed-off-by: Dennis Stosberg <dennis@stosberg.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Solaris does not support C99 format strings before... Dennis Stosberg Tue, 15 Aug 2006 09:01:22 +0000 (11:01 +0200)

Solaris does not support C99 format strings before version 10

Signed-off-by: Dennis Stosberg <dennis@stosberg.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-apply --reverse: simplify reverse option.Junio C Hamano Tue, 15 Aug 2006 06:26:51 +0000 (23:26 -0700)

git-apply --reverse: simplify reverse option.

Having is_reverse in each patch did not make sense. This will hopefully
simplify the work needed to introduce reversible binary diff format.

Signed-off-by: Junio C Hamano <junkio@cox.net>

t4116 apply --reverse testJunio C Hamano Tue, 15 Aug 2006 06:24:55 +0000 (23:24 -0700)

t4116 apply --reverse test

The binary patch test needs to be made more careful not to have
the postimage blob in the repository in which the patch is applied

Signed-off-by: Junio C Hamano <junkio@cox.net>

Make sha1flush void and remove conditional return.David Rientjes Mon, 14 Aug 2006 20:32:01 +0000 (13:32 -0700)

Make sha1flush void and remove conditional return.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Make upload_pack void and remove conditional return.David Rientjes Mon, 14 Aug 2006 20:40:51 +0000 (13:40 -0700)

Make upload_pack void and remove conditional return.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Make track_tree_refs void.David Rientjes Mon, 14 Aug 2006 20:40:06 +0000 (13:40 -0700)

Make track_tree_refs void.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Make pack_objects void.David Rientjes Mon, 14 Aug 2006 20:38:50 +0000 (13:38 -0700)

Make pack_objects void.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Make fsck_dir void.David Rientjes Mon, 14 Aug 2006 20:36:18 +0000 (13:36 -0700)

Make fsck_dir void.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Make checkout_all void.David Rientjes Mon, 14 Aug 2006 20:20:12 +0000 (13:20 -0700)

Make checkout_all void.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Make show_entry voidDavid Rientjes Mon, 14 Aug 2006 20:39:27 +0000 (13:39 -0700)

Make show_entry void

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Make pprint_tag void and cleans up call in cmd_cat_file.David Rientjes Mon, 14 Aug 2006 20:19:15 +0000 (13:19 -0700)

Make pprint_tag void and cleans up call in cmd_cat_file.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Remove combine-diff.c::uninteresting()Junio C Hamano Tue, 15 Aug 2006 01:36:00 +0000 (18:36 -0700)

Remove combine-diff.c::uninteresting()

A patch from David Rientjes made me realize we do not have to have
this function -- just call diff_unmodified_pair() directly.

Signed-off-by: Junio C Hamano <junkio@cox.net>

read-cache.c cleanupDavid Rientjes Mon, 14 Aug 2006 20:38:14 +0000 (13:38 -0700)

read-cache.c cleanup

Removes conditional returns.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

http-push.c cleanupDavid Rientjes Mon, 14 Aug 2006 20:37:05 +0000 (13:37 -0700)

http-push.c cleanup

Removes conditional return.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

diff.c cleanupDavid Rientjes Mon, 14 Aug 2006 20:34:16 +0000 (13:34 -0700)

diff.c cleanup

Removes conditional return.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

builtin-push.c cleanupDavid Rientjes Mon, 14 Aug 2006 20:26:58 +0000 (13:26 -0700)

builtin-push.c cleanup

Removes conditional return in builtin-push.c

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

builtin-grep.c cleanupDavid Rientjes Mon, 14 Aug 2006 20:22:15 +0000 (13:22 -0700)

builtin-grep.c cleanup

Removes conditional return.

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

blame.c return cleanupDavid Rientjes Mon, 14 Aug 2006 20:18:11 +0000 (13:18 -0700)

blame.c return cleanup

Removes conditional from return

Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: configurable home link stringYasushi SHOJI Mon, 14 Aug 2006 22:50:49 +0000 (07:50 +0900)

gitweb: configurable home link string

I've always found difficult to figure out git URL for clone from
gitweb URL because git:// and http:// are different on many site
including kernel.org.

I've found this enhancement at http://dev.laptop.org/git when I was on
git channel, and thought that it'd be nice if all public gitweb site
show it's git URL on its page.

This patch allow us to change the home link string. The current
default is "projects" as we all see on gitweb now.

ie. kernel.org might set this variable to "git://git.kernel.org/pub/scm/"

Signed-off-by: Yasushi SHOJI <yashi@atmark-techno.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: Separate printing difftree in git_commit into... Jakub Narebski Mon, 14 Aug 2006 00:18:33 +0000 (02:18 +0200)

gitweb: Separate printing difftree in git_commit into git_difftree_body

Separate printing difftree in git_commit into separate
git_difftree_body subroutine. Add support for "C" (copied) status. For
"M" and "C" add parameter 'fp' (filename parent) to the "diff" link;
currently not supported by git_blobdiff ("blobdiff" action).

Reindented, realigned, added comments.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: True fix: Support for the standard mime.types... Jakub Narebski Mon, 14 Aug 2006 00:16:33 +0000 (02:16 +0200)

gitweb: True fix: Support for the standard mime.types map in gitweb

True fix for error in mimetype_guess, error introduced in original commit
2d00737489b8c61ed616b261c7c9bd314e2b0b41 and later fixed temporarily
by commenting out the line that caused error in commit
57bd4d3523efecf60197040cad34154aff4ddf80.

Gitweb now supports mime.types map $mimetypes_file relative to project.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: Skip comments in mime.types like fileJakub Narebski Mon, 14 Aug 2006 00:15:22 +0000 (02:15 +0200)

gitweb: Skip comments in mime.types like file

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: Change appereance of marker of refs pointing... Jakub Narebski Mon, 14 Aug 2006 00:14:20 +0000 (02:14 +0200)

gitweb: Change appereance of marker of refs pointing to given object

Change git_get_references to include type of ref in the %refs value, which
means putting everything after 'refs/' as a ref name, not only last
part of the name. Instead of separating refs pointing to the same
object by " / " separator, use anonymous array reference to store all
refs pointing to given object.

Use 'git-ls-remote .' if $projectroot/$project/info/refs does not
exist. (Perhaps it should be used always.)

Refs are now in separate span elements. Class is dependent on the ref
type: currently known classes are 'tag', 'head', 'remote', and 'ref'
(last one for HEAD and other refs in the main directory). There is
encompassing span element of class refs, just in case of unknown ref
type.

This might be considered cleaner separating of git_get_references into
filling %refs hash only, and not taking part in formatting ref marker.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: Separate finding project owner into git_get_pro... Jakub Narebski Mon, 14 Aug 2006 00:10:06 +0000 (02:10 +0200)

gitweb: Separate finding project owner into git_get_project_owner

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: Separate main part of git_history into git_hist... Jakub Narebski Mon, 14 Aug 2006 00:09:08 +0000 (02:09 +0200)

gitweb: Separate main part of git_history into git_history_body

Separates main part of git_history into git_history_body subroutine,
and makes output more similar to git_shortlog. Adds "diff to current"
link only for history of regular file (blob).

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: Refactor printing shortened title in git_shortl... Jakub Narebski Mon, 14 Aug 2006 00:08:27 +0000 (02:08 +0200)

gitweb: Refactor printing shortened title in git_shortlog_body and git_tags_body

Separate printing of perhaps shortened title (subject) in
git_shortlog_body and git_tags_body into format_subject_html.

While at it, remove presentation element <b>...</b> used to format
title (subject) and move formatting to CSS.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: Separate ref parsing in git_get_refs_list into... Jakub Narebski Mon, 14 Aug 2006 00:07:00 +0000 (02:07 +0200)

gitweb: Separate ref parsing in git_get_refs_list into parse_ref

Note that for each ref there are usually two calls to git subroutines:
first to get the type of ref, second to parse ref if ref is of commit
or tag type.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: Great subroutines renamingJakub Narebski Mon, 14 Aug 2006 00:05:47 +0000 (02:05 +0200)

gitweb: Great subroutines renaming

Rename some of subroutines to better reflect what they do.
Some renames were not performed because subroutine name
reflects hash key.

Subroutines name guideline:
* git_ prefix for subroutines related to git commands,
git repository, or to gitweb actions
* git_get_ prefix for inner subroutines calling git command
or reading some file in the repository and returning some output
* parse_ prefix for subroutines parsing some text (or reading and
parsing some text) into hash or list
* format_ prefix for subroutines formatting, post-processing
or generating some HTML/text fragment
* _get_ infix for subroutines which return result
* _print_ infix for subroutines which print fragment of output
* _body suffix for subroutines which outputs main part (body)
of related action (usually table)
* _nav suffix for subroutines related to navigation bars
* _div suffix for subroutines returning or printing div element
* subroutine names should not be based on how the result is obtained,
as this might change easily

Renames performed:
- git_get_referencing => format_ref_marker
- git_get_paging_nav => format_paging_nav
- git_read_head => git_get_head_hash
- git_read_hash => git_get_hash_by_ref
- git_read_description => git_get_project_description
- git_read_projects => git_get_projects_list
- read_info_ref => git_get_references
- git_read_refs => git_get_refs_list
- date_str => parse_date
- git_read_tag => parse_tag
- git_read_commit => parse_commit
- git_blob_plain_mimetype => blob_mimetype
- git_page_nav => git_print_page_nav
- git_header_div => git_print_header_div

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Add a newline before appending "Signed-off-by: " lineFranck Bui-Huu Sun, 13 Aug 2006 18:30:27 +0000 (11:30 -0700)

Add a newline before appending "Signed-off-by: " line

Whef the last line of the commit log message does not end with
"^[-A-Za-z]+: [^@]+@", append a newline after it to separate
the body of the commit log message from the run of sign-off and
ack lines. e.g. "Signed-off-by: A U Thor <au.thor@example.com>" or
"Acked-by: Me <myself@example.org>".

Signed-off-by: Junio C Hamano <junkio@cox.net>

Merge branch 'maint'Junio C Hamano Mon, 14 Aug 2006 05:58:23 +0000 (22:58 -0700)

Merge branch 'maint'

* maint:
sample commit-msg hook: no silent exit on duplicate Signed-off-by lines
Fix regex pattern in commit-msg

Fix type of combine-diff.c::show_patch_diff()Junio C Hamano Mon, 14 Aug 2006 02:19:34 +0000 (19:19 -0700)

Fix type of combine-diff.c::show_patch_diff()

The other function, show_raw_diff() is void and no callers use
return value from neither.

Signed-off-by: Junio C Hamano <junkio@cox.net>

sample commit-msg hook: no silent exit on duplicate... Luben Tuikov Sun, 13 Aug 2006 08:41:22 +0000 (01:41 -0700)

sample commit-msg hook: no silent exit on duplicate Signed-off-by lines

git-commit would silently exit if duplicate Signed-off-by
lines were found. Users of git-commit would not know it,
unless they checked '$?'. This patch makes git-commit
actually print out a message that nothing was commited
since duplicate Signed-off-lines were found.

Signed-off-by: Luben Tuikov <ltuikov@yahoo.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Fix regex pattern in commit-msgLuben Tuikov Sun, 13 Aug 2006 07:34:37 +0000 (00:34 -0700)

Fix regex pattern in commit-msg

Between the count and the line output, some
uniq(1) versions put a TAB character, not a space.
Make sure both are handled.

Signed-off-by: Luben Tuikov <ltuikov@yahoo.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Merge branch 'jc/upload-pack'Junio C Hamano Sun, 13 Aug 2006 05:16:51 +0000 (22:16 -0700)

Merge branch 'jc/upload-pack'

Merge branch 'jc/racy-delay'Junio C Hamano Sun, 13 Aug 2006 02:41:44 +0000 (19:41 -0700)

Merge branch 'jc/racy-delay'

Merge branch 'th/diff-extra'Junio C Hamano Sun, 13 Aug 2006 02:34:41 +0000 (19:34 -0700)

Merge branch 'th/diff-extra'

Merge branch 'jc/pack-objects'Junio C Hamano Sun, 13 Aug 2006 02:33:16 +0000 (19:33 -0700)

Merge branch 'jc/pack-objects'

Merge branch 'js/read-tree'Junio C Hamano Sun, 13 Aug 2006 02:29:11 +0000 (19:29 -0700)

Merge branch 'js/read-tree'

Merge branch 'js/http-mb'Junio C Hamano Sun, 13 Aug 2006 02:24:51 +0000 (19:24 -0700)

Merge branch 'js/http-mb'

Merge branch 'js/color-diff'Junio C Hamano Sun, 13 Aug 2006 02:24:47 +0000 (19:24 -0700)

Merge branch 'js/color-diff'

Merge branch 'jn/web'Junio C Hamano Sun, 13 Aug 2006 02:24:15 +0000 (19:24 -0700)

Merge branch 'jn/web'

Merge branch 'lt/web'Junio C Hamano Sun, 13 Aug 2006 02:24:09 +0000 (19:24 -0700)

Merge branch 'lt/web'

Merge branch 'jn/conf'Junio C Hamano Sun, 13 Aug 2006 02:23:09 +0000 (19:23 -0700)

Merge branch 'jn/conf'

Merge branch 'jc/grep'Junio C Hamano Sun, 13 Aug 2006 02:16:33 +0000 (19:16 -0700)

Merge branch 'jc/grep'

Merge branch 'mk/rename'Junio C Hamano Sun, 13 Aug 2006 02:13:31 +0000 (19:13 -0700)

Merge branch 'mk/rename'

Merge branch 'ml/pager'Junio C Hamano Sun, 13 Aug 2006 02:13:25 +0000 (19:13 -0700)

Merge branch 'ml/pager'

Merge branch 'master' into js/c-merge-recursiveJunio C Hamano Sun, 13 Aug 2006 01:34:36 +0000 (18:34 -0700)

Merge branch 'master' into js/c-merge-recursive

Adjust to hold_lock_file_for_update() change on the master.

GIT 1.4.2 v1.4.2Junio C Hamano Sun, 13 Aug 2006 01:32:31 +0000 (18:32 -0700)

GIT 1.4.2

Merge git://git.kernel.org/pub/scm/gitk/gitkJunio C Hamano Sun, 13 Aug 2006 01:32:17 +0000 (18:32 -0700)

Merge git://git.kernel.org/pub/scm/gitk/gitk

* git://git.kernel.org/pub/scm/gitk/gitk:
gitk: Show the currently checked-out head in bold font
gitk: Allow the user to set some colors

t/t4013: fix futzing with the version string.Junio C Hamano Sun, 13 Aug 2006 01:04:07 +0000 (18:04 -0700)

t/t4013: fix futzing with the version string.

Signed-off-by: Junio C Hamano <junkio@cox.net>

Better error message when we are unable to lock the... Junio C Hamano Sat, 12 Aug 2006 08:03:47 +0000 (01:03 -0700)

Better error message when we are unable to lock the index file

Most of the callers except the one in refs.c use the function to
update the index file. Among the index writers, everybody
except write-tree dies if they cannot open it for writing.

This gives the function an extra argument, to tell it to die
when it cannot create a new file as the lockfile.

The only caller that does not have to die is write-tree, because
updating the index for the cache-tree part is optional and not
being able to do so does not affect the correctness. I think we
do not have to be so careful and make the failure into die() the
same way as other callers, but that would be a different patch.

Signed-off-by: Junio C Hamano <junkio@cox.net>

git-am: give better diagnostics when the patch does... Junio C Hamano Sat, 12 Aug 2006 23:16:47 +0000 (16:16 -0700)

git-am: give better diagnostics when the patch does not apply during --3way

If the user tries to apply a patch that was hand-edited in such
a way that it does not apply to the original file recorded on
its "index" line anymore, we did detect the situation but did
not issue an error message that is specific enough.

Signed-off-by: Junio C Hamano <junkio@cox.net>

Merge branch 'rj/header'Junio C Hamano Sat, 12 Aug 2006 23:09:19 +0000 (16:09 -0700)

Merge branch 'rj/header'

* rj/header:
Fix header breakage with _XOPEN_SOURCE.

git-svn: split the path from the url correctly with... Eric Wong Sat, 12 Aug 2006 06:21:41 +0000 (23:21 -0700)

git-svn: split the path from the url correctly with limited perms

This version of the splitter (that only affects SVN:: library
users) works when one only has limited read-permissions to
the repository they're fetching from.

Updated from the original patch to workaround some SVN bug
somewhere, which only seems to happen against file://
repositories... Here's the diff against the original patch I
submitted:

@@ -1159,8 +1159,8 @@ sub repo_path_split {
}

if ($_use_lib) {
- $SVN = libsvn_connect($full_url);
- my $url = $SVN->get_repos_root;
+ my $tmp = libsvn_connect($full_url);
+ my $url = $tmp->get_repos_root;
$full_url =~ s#^\Q$url\E/*##;
push @repo_path_split_cache, qr/^(\Q$url\E)/;
return ($url, $full_url);

Somehow connecting to a repository with the full url makes the
returned SVN::Ra object act strangely and break things, so now
we just drop the SVN::Ra object that we made our initial
connection with.

Signed-off-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-grep: show pathnames relative to the current directoryJunio C Hamano Fri, 11 Aug 2006 07:44:42 +0000 (00:44 -0700)

git-grep: show pathnames relative to the current directory

By default, the command shows pathnames relative to the current
directory. Use --full-name (the same flag to do so in ls-files)
if you want to see the full pathname relative to the project root.

This makes it very pleasant to run in Emacs compilation (or
"grep-find") buffer.

Signed-off-by: Junio C Hamano <junkio@cox.net>

git-sh-setup: do not use repo-config to test the git... Junio C Hamano Sat, 12 Aug 2006 01:47:50 +0000 (18:47 -0700)

git-sh-setup: do not use repo-config to test the git directory

Since repo-config does not fail in non-git directory, it is not
a good command to use to test the git-ness nor validate the
repository revision of $GIT_DIR.

Original patch by Robert Shearman but with minor fixes.

Signed-off-by: Junio C Hamano <junkio@cox.net>

git-svn: bugfix: allow SVN:: lib users to track the... Eric Wong Fri, 11 Aug 2006 18:11:29 +0000 (11:11 -0700)

git-svn: bugfix: allow SVN:: lib users to track the root of the repository

I'm not sure if anybody has hit this (besides me), but this
fixes the problem where I ran into while attempting to import a
small repo at the root level: I ended up with all the commits, but
with no file/tree changes at all throughout the entire history.

Also, fix a warning if the commit message is not defined for revision 0.

Signed-off-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-svn: correctly kill keyword expansion without mungi... Eric Wong Fri, 11 Aug 2006 11:34:07 +0000 (04:34 -0700)

git-svn: correctly kill keyword expansion without munging EOLs

This bugfix applies to users of the svn command-line client only.

We no longer muck with newlines when killing keyword expansion.
This tended to generate unintended diffs in commits because svn
revert -R would destroy the manual EOL changes we were doing. Of
course, we didn't need the EOL munging in the first place, as
svn seems to do it for us even in the text-base files.

Now we set the mtime and atime the files changed by keyword
expansion killing to avoid triggering a change on svn revert,
which svn still seems to want to do.

Thanks to Seth Falcon for reporting this bug.

Signed-off-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>

drop length argument of has_extensionRene Scharfe Fri, 11 Aug 2006 12:01:45 +0000 (14:01 +0200)

drop length argument of has_extension

As Fredrik points out the current interface of has_extension() is
potentially confusing. Its parameters include both a nul-terminated
string and a length-limited string.

This patch drops the length argument, requiring two nul-terminated
strings; all callsites are updated. I checked that all of them indeed
provide nul-terminated strings. Filenames need to be nul-terminated
anyway if they are to be passed to open() etc. The performance penalty
of the additional strlen() is negligible compared to the system calls
which inevitably surround has_extension() calls.

Additionally, change has_extension() to use size_t inside instead of
int, as that is the exact type strlen() returns and memcmp() expects.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Merge branch 'rn/push-dav'Junio C Hamano Fri, 11 Aug 2006 06:13:50 +0000 (23:13 -0700)

Merge branch 'rn/push-dav'

* rn/push-dav:
http-push: Make WebDAV work with (broken?) default apache2 WebDAV module

Add the --color-words option to the diff options familyJohannes Schindelin Fri, 28 Jul 2006 21:56:15 +0000 (23:56 +0200)

Add the --color-words option to the diff options family

With this option, the changed words are shown inline. For example,
if a file containing "This is foo" is changed to "This is bar", the diff
will now show "This is " in plain text, "foo" in red, and "bar" in green.

Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Add Documentation/howto/setup-git-server-over-http.txtRutger Nijlunsing Thu, 10 Aug 2006 20:00:26 +0000 (22:00 +0200)

Add Documentation/howto/setup-git-server-over-http.txt

A small howto on how to setup GIT over HTTP transport protocol by
setting up WebDAV access on apache2.

[jc: minimum ispell fixes applied]

Signed-off-by: Rutger Nijlunsing <git@tux.tmfweb.nl>
Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

gitweb: Whitespace cleanup - tabs are for indent, space... Jakub Narebski Thu, 10 Aug 2006 10:38:47 +0000 (12:38 +0200)

gitweb: Whitespace cleanup - tabs are for indent, spaces are for align

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

discard_cache(): discard index, even if no file was... Johannes Schindelin Thu, 10 Aug 2006 14:47:21 +0000 (16:47 +0200)

discard_cache(): discard index, even if no file was mmap()ed

Since add_cacheinfo() can be called without a mapped index file,
discard_cache() _has_ to discard the entries, even when
cache_mmap == NULL.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-verify-pack: make builtinRene Scharfe Thu, 10 Aug 2006 15:02:38 +0000 (17:02 +0200)

git-verify-pack: make builtin

Convert git-verify-pack to a builtin command. Also rename ac to argc
and av to argv for consistancy.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Merge branch 'master' into mk/renameJunio C Hamano Thu, 10 Aug 2006 21:17:49 +0000 (14:17 -0700)

Merge branch 'master' into mk/rename

* master:
git-verify-pack: no need to count errors
git-verify-pack: buffer overrun paranoia
git-verify-pack: free pack after use and a cleanup
git-verify-pack: get rid of while loop
git-verify-pack: insist on .idx extension
git-verify-pack: more careful path handling
git-verify-pack: show usage when no pack was specified
Add has_extension()
builtin-apply: remove unused increment
Fix git-diff A...B
combine-diff: use color
git-apply: applying a patch to make a symlink shorter.
allow diff.renamelimit to be set regardless of -M/-C
make --find-copies-harder imply -C
find_unique_abbrev() with len=0 should not abbreviate
check return value from diff_setup_done()
Fix tutorial-2.html
Documentation: git-status takes the same options as git-commit
Update git-init-db(1) and documentation of core.sharedRepository

git-verify-pack: no need to count errorsRene Scharfe Thu, 10 Aug 2006 15:02:37 +0000 (17:02 +0200)

git-verify-pack: no need to count errors

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-verify-pack: buffer overrun paranoiaRene Scharfe Thu, 10 Aug 2006 15:02:36 +0000 (17:02 +0200)

git-verify-pack: buffer overrun paranoia

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-verify-pack: free pack after use and a cleanupRene Scharfe Thu, 10 Aug 2006 15:02:35 +0000 (17:02 +0200)

git-verify-pack: free pack after use and a cleanup

Plug memory leak in verify_one_pack() by freeing the struct packed_git
we got from add_packed_git(). Also rename g to pack and pull an
assignment out of an if statement while we're at it.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-verify-pack: get rid of while loopRene Scharfe Thu, 10 Aug 2006 15:02:34 +0000 (17:02 +0200)

git-verify-pack: get rid of while loop

Get rid of that while loop which was apparently used as a way to avoid
goto's (why?). It's easy now because there is only one break left at
the end of it. Also make the comment clearer.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-verify-pack: insist on .idx extensionRene Scharfe Thu, 10 Aug 2006 15:02:33 +0000 (17:02 +0200)

git-verify-pack: insist on .idx extension

git-verify-pack can be called with a filename without .idx extension.
add_packed_git() on the other hand depends on its presence. So
instead of trying to call it with whatever the user gave us check for
that extension and add it if it's missing.

That means that you can't name your index file "blah" and your pack
file ".pack" anymore ("git-verify-pack blah" currently works in that
case). I think this regression is a good change. ;-)

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-verify-pack: more careful path handlingRene Scharfe Thu, 10 Aug 2006 15:02:32 +0000 (17:02 +0200)

git-verify-pack: more careful path handling

Use strlcpy() to copy the filename into a buffer and complain if it
doesn't fit. Also move the path buffer into verify_one_pack(); it is
used only there. Now we can const'ify the first argument of this
function.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-verify-pack: show usage when no pack was specifiedRene Scharfe Thu, 10 Aug 2006 15:02:31 +0000 (17:02 +0200)

git-verify-pack: show usage when no pack was specified

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Add has_extension()Rene Scharfe Thu, 10 Aug 2006 15:02:30 +0000 (17:02 +0200)

Add has_extension()

The little helper has_extension() documents through its name what we are
trying to do and makes sure we don't forget the underrun check.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

builtin-apply: remove unused incrementJunio C Hamano Thu, 10 Aug 2006 07:56:40 +0000 (00:56 -0700)

builtin-apply: remove unused increment

We do not use desc.alloc after assigning desc.buffer to patch->result;
do not bother to increment it.

Signed-off-by: Junio C Hamano <junkio@cox.net>

Fix git-diff A...BJunio C Hamano Thu, 10 Aug 2006 07:50:15 +0000 (00:50 -0700)

Fix git-diff A...B

Commit 9919f41 meant to make git-diff A...B to (usually) mean
"git-diff `git-merge-base A B` B", but it got the parameters wrong
and ended up showing "git-diff `git-merge-base A B` A" by mistake.

Signed-off-by: Junio C Hamano <junkio@cox.net>

combine-diff: use colorJunio C Hamano Thu, 10 Aug 2006 07:30:33 +0000 (00:30 -0700)

combine-diff: use color

Using the same mechanism as the regular diffs, color combined diff
output.

Signed-off-by: Junio C Hamano <junkio@cox.net>

http-push: Make WebDAV work with (broken?) default... Rutger Nijlunsing Wed, 9 Aug 2006 18:54:23 +0000 (20:54 +0200)

http-push: Make WebDAV work with (broken?) default apache2 WebDAV module

WebDAV on Debian unstable cannot handle renames on WebDAV from
file.ext to newfile (without ext) when newfile* already
exists. Normally, git creates a file like 'objects/xx/sha1.token',
which is renamed to 'objects/xx/sha1' when transferred completely.

Just use '_' instead of '.' so WebDAV doesn't see it as an extension
change.

Signed-off-by: Rutger Nijlunsing <git@tux.tmfweb.nl>
Acked-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-apply: applying a patch to make a symlink shorter.Junio C Hamano Thu, 10 Aug 2006 05:47:25 +0000 (22:47 -0700)

git-apply: applying a patch to make a symlink shorter.

The internal representation of the result is counted string
(i.e. char *buf and ulong size), which is fine for writing out
to regular file, but throwing the buf at symlink(2) was a
no-no.

Reported by Willy Tarreau.

Signed-off-by: Junio C Hamano <junkio@cox.net>

merge-recur: do not die unnecessarilyJohannes Schindelin Wed, 9 Aug 2006 20:31:49 +0000 (22:31 +0200)

merge-recur: do not die unnecessarily

When the cache is dirty, and we switch the index file from temporary
to final, we want to discard the cache without complaint.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

merge-recur: try to merge older merge bases firstJohannes Schindelin Wed, 9 Aug 2006 20:30:58 +0000 (22:30 +0200)

merge-recur: try to merge older merge bases first

It seems to be the only sane way to do it: when a two-head merge is
done, and the merge-base and one of the two branches agree, the
merge assumes that the other branch has something new.

If we start creating virtual commits from newer merge-bases, and go
back to older merge-bases, and then merge with newer commits again,
chances are that a patch is lost, _because_ the merge-base and the
head agree on it. Unlikely, yes, but it happened to me.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

merge-recur: if there is no common ancestor, fake empty oneJohannes Schindelin Wed, 9 Aug 2006 16:43:03 +0000 (18:43 +0200)

merge-recur: if there is no common ancestor, fake empty one

This fixes the coolest merge ever.

[jc: with two "Oops that's not it" fixes from Johannes and Alex,
and an obvious type mismatch fix.]

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

merge-recur: do not setenv("GIT_INDEX_FILE")Johannes Schindelin Wed, 9 Aug 2006 13:07:31 +0000 (15:07 +0200)

merge-recur: do not setenv("GIT_INDEX_FILE")

Since there are no external calls left in merge-recur, we do not need
to set the environment variable GIT_INDEX_FILE all the time.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

merge-recur: do not call git-write-treeJohannes Schindelin Wed, 9 Aug 2006 13:04:16 +0000 (15:04 +0200)

merge-recur: do not call git-write-tree

Since merge-recur is in C, and uses libgit, it can call the relevant
functions directly, without writing the index file.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

allow diff.renamelimit to be set regardless of -M/-C v1.4.2-rc4Junio C Hamano Wed, 9 Aug 2006 21:05:23 +0000 (14:05 -0700)

allow diff.renamelimit to be set regardless of -M/-C

Signed-off-by: Junio C Hamano <junkio@cox.net>

make --find-copies-harder imply -CJunio C Hamano Wed, 9 Aug 2006 20:17:19 +0000 (13:17 -0700)

make --find-copies-harder imply -C

Signed-off-by: Junio C Hamano <junkio@cox.net>

find_unique_abbrev() with len=0 should not abbreviateJunio C Hamano Wed, 9 Aug 2006 20:17:04 +0000 (13:17 -0700)

find_unique_abbrev() with len=0 should not abbreviate

Signed-off-by: Junio C Hamano <junkio@cox.net>

check return value from diff_setup_done()Junio C Hamano Wed, 9 Aug 2006 19:45:27 +0000 (12:45 -0700)

check return value from diff_setup_done()

Signed-off-by: Junio C Hamano <junkio@cox.net>

Fix tutorial-2.htmlJunio C Hamano Wed, 9 Aug 2006 07:57:01 +0000 (00:57 -0700)

Fix tutorial-2.html

Honza Pazdziora noticed that one example did not match reality.

Signed-off-by: Junio C Hamano <junkio@cox.net>

autoconf: Add support for setting CURLDIR, OPENSSLDIR... Jakub Narebski Wed, 9 Aug 2006 00:15:10 +0000 (02:15 +0200)

autoconf: Add support for setting CURLDIR, OPENSSLDIR, EXPATDIR

Add support for --with-openssl=PATH and --without-openssl,
--with-curl=PATH and --without-curl, --with-expat=PATH and
--without-expat ./configure options, each setting or unsetting
appropriate NO_PACKAGE and if called with argument also PACKAGEDIR
(of which only CURLDIR is documented in Makefile)

All above options are supported as override to autodetection; more to
come in the same style (override to autodetection), so moved the bulk
of comments for site configuration down.

Needs review by someone well versed in autoconf and m4.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Documentation: git-status takes the same options as... Junio C Hamano Tue, 8 Aug 2006 05:41:32 +0000 (22:41 -0700)

Documentation: git-status takes the same options as git-commit

Signed-off-by: Junio C Hamano <junkio@cox.net>

autoconf: It is --without-python, not --no-pythonJakub Narebski Wed, 9 Aug 2006 00:19:22 +0000 (02:19 +0200)

autoconf: It is --without-python, not --no-python

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Update git-init-db(1) and documentation of core.sharedR... Jonas Fonseca Wed, 9 Aug 2006 00:26:23 +0000 (02:26 +0200)

Update git-init-db(1) and documentation of core.sharedRepository

Combine option descriptions in git-init-db(1). Reflect the changes to
additionally allow all users to read the created git repository.

Signed-off-by: Jonas Fonseca <fonseca@diku.dk>
Signed-off-by: Junio C Hamano <junkio@cox.net>

autoconf: clean temporary file mak.appendJunio C Hamano Tue, 8 Aug 2006 20:42:35 +0000 (13:42 -0700)

autoconf: clean temporary file mak.append

When configure is interrupted in the middle it leaves
config.mak.append behind. Add it to .gitignore and make sure
$(MAKE) clean removes it.

Also earlier .gitignore listed config.mak.in which is a tracked
file. Fix it.

Signed-off-by: Junio C Hamano <junkio@cox.net>

autoconf: Move variables which we always set to config... Jakub Narebski Tue, 8 Aug 2006 16:39:06 +0000 (18:39 +0200)

autoconf: Move variables which we always set to config.mak.in

Move detected NO_STH and NEED_STH variables, which we always output,
either setting or unsetting (setting to empty string) to config.mak.in
and use setting appropriately named variables and doing AC_SUBST
instead of adding them via GIT_CONF_APPEND_LINE macro and
config.mak.append temporary file.

Variables which might and might not be set are still added via
config.mak.append; this include all STH_PATH variables.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

read-cache: tweak racy-git delay logicJunio C Hamano Tue, 8 Aug 2006 21:47:32 +0000 (14:47 -0700)

read-cache: tweak racy-git delay logic

Instead of looping over the entries and writing out, use a
separate loop after all entries have been written out to check
how many entries are racily clean. Make sure that the newly
created index file gets the right timestamp when we check by
flushing the buffered data by ce_write().

Signed-off-by: Junio C Hamano <junkio@cox.net>