And finally the endgame. Instead of spawning "git mailinfo" via the
run_command() API the same number of times as there are incoming
patches, make direct internal call to the libified mailinfo() from
"git am" to reduce the spawning overhead, which would matter on some
platforms.
mailinfo: handle charset conversion errors in the caller
Instead of dying in convert_to_utf8(), just report an error and let
the callers handle it. Between the two callers:
- decode_header() silently punts when it cannot parse a broken
RFC2047 encoded text (e.g. when it sees anything other than B or
Q after it sees "=?<charset>") by jumping to release_return,
returning the string it successfully parsed out so far, to the
caller. A piece of string that convert_to_utf8() cannot handle
can be treated the same way.
- handle_commit_msg() doesn't cope with a malformed line well, so
die there for now. We'll lift this even higher in later changes
in this series.
When mailinfo() is eventually libified, the calling "git am" still
will have to write out the log message in the "msg" file for hooks
and other users of the information, but it does not have to reopen
and reread what it wrote earlier if the function kept it in a strbuf.
This also removes the need for seeking and truncating the output
file when we see a scissors mark in the input, which in turn allows
us to lose two callsites of die_errno().
mailinfo: handle_commit_msg() shouldn't be called after finding patchbreak
There is a strange "if (!mi->cmitmsg) return 0" at the very beginning
of handle_commit_msg(), but the condition should never trigger, because:
* The only place cmitmsg is set to NULL is after this function sees
a patch break, closes the FILE * to write the commit log message
and returns 1. This function returns non-zero only from that
codepath.
* The caller of this function, upon seeing a non-zero return,
increments filter_stage, starts treating the input as patch text
and will never call handle_commit_msg() again.
Replace it with an assert(!mi->filter_stage) to ensure the above
observation will stay to be true.
mailinfo: move check for metainfo_charset to convert_to_utf8()
All callers of this function refrain from calling it when
mi->metainfo_charset is NULL; move the check to the callee,
as it already has a few conditions at its beginning to turn
it into a no-op.
mailinfo: move filter/header stage to struct mailinfo
Earlier we got rid of two function-scope static variables that kept
track of the states of helper functions by making them extra arguments
that are passed throughout the callchain. Now we have a convenient
place to store and pass them around in the form of "struct mailinfo",
change them into two fields in the struct.
mailinfo: move global "FILE *fin, *fout" to struct mailinfo
This requires us to pass "struct mailinfo" to more functions
throughout the codepath that read input lines. Incidentally,
later steps are helped by this patch passing the struct to
more callchains.
mailinfo: introduce "struct mailinfo" to hold globals
In this first step, move only 'email' and 'name' fields in there and
remove the corresponding globals. In subsequent patches, more
globals will be moved to this and the structure will be passed
around as a new parameter to more functions.
mailinfo: move global "line" into mailinfo() function
With the previous steps, it becomes clear that the mailinfo()
function is the only one that wants the "line" to be directly
touchable. Move it to the function scope of this function.
mailinfo: do not let find_boundary() touch global "line" directly
With the previous two commits, we established that the local
variable "line" in handle_body() and handle_boundary() functions
always refer to the global "line" that is used as the common and
shared "current line from the input". They are the only callers of
the last function that refers to the global line directly, i.e.
find_boundary(). Pass "line" as a parameter to this leaf function
to complete the clean-up. Now the only function that directly refers
to the global "line" is the caller of handle_body() at the very
beginning of this whole callchain.
mailinfo: do not let handle_boundary() touch global "line" directly
This function has a single caller, and called with the global "line"
holding the multi-part boundary line the caller saw while processing
the e-mail body. The function then goes into a loop to process each
line of the input, and fills the same global "line" variable from
the input as it needs to read more lines to process the multi-part
headers.
Let the caller explicitly pass a pointer to this global "line"
variable as an argument, and have the function itself use that
strbuf throughout, instead of referring to the global "line" itself.
There still is a helper function that this function calls that still
touches the global directly; it will be updated as the series progresses.
mailinfo: do not let handle_body() touch global "line" directly
This function has a single caller, and called with the global "line"
holding the first line of the e-mail body after the caller finished
processing the e-mail headers. The function then goes into a loop
to process each line of the input, starting from what was given by
its caller, and fills the same global "line" variable from the input
as it needs to process more lines.
Let the caller explicitly pass a pointer to this global "line"
variable as an argument, and have the function itself use that
strbuf throughout, instead of referring to the global "line" itself.
There are helper functions that this function calls that still touch
the global directly; they will be updated as the series progresses.
Two helper functions use "static int" in their scope to keep track
of the state while repeatedly getting called once for each input
line. Move these state variables to their ultimate caller and pass
down pointers to them along the callchain, as a small step in
preparation for making this entire callchain more reentrant.
This function wants to call find_boundary() and is called only from
one place without any recursing, so it becomes easier to read if it
appears after the called function.
mailinfo: explicitly close file handle to the patch output
This does not make a difference within the context of "git mailinfo"
that runs once and exits, as flushing and closing would happen upon
process termination. It however will matter when we eventually make
it callable as an API function.
Besides, cleaning after yourself once you are done is a good hygiene.
mailinfo: fix an off-by-one error in the boundary stack
We pre-increment the pointer that we will use to store something at,
so the pointer is already beyond the end of the array if it points
at content[MAX_BOUNDARIES].
mailinfo: fold decode_header_bq() into decode_header()
In olden days we might have wanted to behave differently in
decode_header() if the header line was encoded with RFC2047, but we
apparently do not do so, hence this helper function can go, together
with its return value.
mailinfo: remove a no-op call convert_to_utf8(it, "")
The called function checks if the second parameter is either a NULL
or an empty string at the very beginning and returns without doing
anything. Remove the useless call.
merge-file: enforce MAX_XDIFF_SIZE on incoming files
The previous commit enforces MAX_XDIFF_SIZE at the
interfaces to xdiff: xdi_diff (which calls xdl_diff) and
ll_xdl_merge (which calls xdl_merge).
But we have another direct call to xdl_merge in
merge-file.c. If it were written today, this probably would
just use the ll_merge machinery. But it predates that code,
and uses slightly different options to xdl_merge (e.g.,
ZEALOUS_ALNUM).
We could try to abstract out an xdi_merge to match the
existing xdi_diff, but even that is difficult. Rather than
simply report error, we try to treat large files as binary,
and that distinction would happen outside of xdi_merge.
The simplest fix is to just replicate the MAX_XDIFF_SIZE
check in merge-file.c.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
The xdiff code is not prepared to handle extremely large
files. It uses "int" in many places, which can overflow if
we have a very large number of lines or even bytes in our
input files. This can cause us to produce incorrect diffs,
with no indication that the output is wrong. Or worse, we
may even underallocate a buffer whose size is the result of
an overflowing addition.
We're much better off to tell the user that we cannot diff
or merge such a large file. This patch covers both cases,
but in slightly different ways:
1. For merging, we notice the large file and cleanly fall
back to a binary merge (which is effectively "we cannot
merge this").
2. For diffing, we make the binary/text distinction much
earlier, and in many different places. For this case,
we'll use the xdi_diff as our choke point, and reject
any diff there before it hits the xdiff code.
This means in most cases we'll die() immediately after.
That's not ideal, but in practice we shouldn't
generally hit this code path unless the user is trying
to do something tricky. We already consider files
larger than core.bigfilethreshold to be binary, so this
code would only kick in when that is circumvented
(either by bumping that value, or by using a
.gitattribute to mark a file as diffable).
In other words, we can avoid being "nice" here, because
there is already nice code that tries to do the right
thing. We are adding the suspenders to the nice code's
belt, so notice when it has been worked around (both to
protect the user from malicious inputs, and because it
is better to die() than generate bogus output).
The maximum size was chosen after experimenting with feeding
large files to the xdiff code. It's just under a gigabyte,
which leaves room for two obvious cases:
- a diff3 merge conflict result on files of maximum size X
could be 3*X plus the size of the markers, which would
still be only about 3G, which fits in a 32-bit int.
- some of the diff code allocates arrays of one int per
record. Even if each file consists only of blank lines,
then a file smaller than 1G will have fewer than 1G
records, and therefore the int array will fit in 4G.
Since the limit is arbitrary anyway, I chose to go under a
gigabyte, to leave a safety margin (e.g., we would not want
to overflow by allocating "(records + 1) * sizeof(int)" or
similar.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
When we call into xdiff to perform a diff, we generally lose
the return code completely. Typically by ignoring the return
of our xdi_diff wrapper, but sometimes we even propagate
that return value up and then ignore it later. This can
lead to us silently producing incorrect diffs (e.g., "git
log" might produce no output at all, not even a diff header,
for a content-level diff).
In practice this does not happen very often, because the
typical reason for xdiff to report failure is that it
malloc() failed (it uses straight malloc, and not our
xmalloc wrapper). But it could also happen when xdiff
triggers one our callbacks, which returns an error (e.g.,
outf() in builtin/rerere.c tries to report a write failure
in this way). And the next patch also plans to add more
failure modes.
Let's notice an error return from xdiff and react
appropriately. In most of the diff.c code, we can simply
die(), which matches the surrounding code (e.g., that is
what we do if we fail to load a file for diffing in the
first place). This is not that elegant, but we are probably
better off dying to let the user know there was a problem,
rather than simply generating bogus output.
We could also just die() directly in xdi_diff, but the
callers typically have a bit more context, and can provide a
better message (and if we do later decide to pass errors up,
we're one step closer to doing so).
There is one interesting case, which is in diff_grep(). Here
if we cannot generate the diff, there is nothing to match,
and we silently return "no hits". This is actually what the
existing code does already, but we make it a little more
explicit.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
By default, libcurl will follow circular http redirects
forever. Let's put a cap on this so that somebody who can
trigger an automated fetch of an arbitrary repository (e.g.,
for CI) cannot convince git to loop infinitely.
The value chosen is 20, which is the same default that
Firefox uses.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Previously, libcurl would follow redirection to any protocol
it was compiled for support with. This is desirable to allow
redirection from HTTP to HTTPS. However, it would even
successfully allow redirection from HTTP to SFTP, a protocol
that git does not otherwise support at all. Furthermore
git's new protocol-whitelisting could be bypassed by
following a redirect within the remote helper, as it was
only enforced at transport selection time.
This patch limits redirects within libcurl to HTTP, HTTPS,
FTP and FTPS. If there is a protocol-whitelist present, this
list is limited to those also allowed by the whitelist. As
redirection happens from within libcurl, it is impossible
for an HTTP redirect to a protocol implemented within
another remote helper.
When the curl version git was compiled with is too old to
support restrictions on protocol redirection, we warn the
user if GIT_ALLOW_PROTOCOL restrictions were requested. This
is a little inaccurate, as even without that variable in the
environment, we would still restrict SFTP, etc, and we do
not warn in that case. But anything else means we would
literally warn every time git accesses an http remote.
This commit includes a test, but it is not as robust as we
would hope. It redirects an http request to ftp, and checks
that curl complained about the protocol, which means that we
are relying on curl's specific error message to know what
happened. Ideally we would redirect to a working ftp server
and confirm that we can clone without protocol restrictions,
and not with them. But we do not have a portable way of
providing an ftp server, nor any other protocol that curl
supports (https is the closest, but we would have to deal
with certificates).
[jk: added test and version warning]
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
submodule: allow only certain protocols for submodule fetches
Some protocols (like git-remote-ext) can execute arbitrary
code found in the URL. The URLs that submodules use may come
from arbitrary sources (e.g., .gitmodules files in a remote
repository). Let's restrict submodules to fetching from a
known-good subset of protocols.
Note that we apply this restriction to all submodule
commands, whether the URL comes from .gitmodules or not.
This is more restrictive than we need to be; for example, in
the tests we run:
git submodule add ext::...
which should be trusted, as the URL comes directly from the
command line provided by the user. But doing it this way is
simpler, and makes it much less likely that we would miss a
case. And since such protocols should be an exception
(especially because nobody who clones from them will be able
to update the submodules!), it's not likely to inconvenience
anyone in practice.
Reported-by: Blake Burkhart <bburky@bburky.com> Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
transport: add a protocol-whitelist environment variable
If we are cloning an untrusted remote repository into a
sandbox, we may also want to fetch remote submodules in
order to get the complete view as intended by the other
side. However, that opens us up to attacks where a malicious
user gets us to clone something they would not otherwise
have access to (this is not necessarily a problem by itself,
but we may then act on the cloned contents in a way that
exposes them to the attacker).
Ideally such a setup would sandbox git entirely away from
high-value items, but this is not always practical or easy
to set up (e.g., OS network controls may block multiple
protocols, and we would want to enable some but not others).
We can help this case by providing a way to restrict
particular protocols. We use a whitelist in the environment.
This is more annoying to set up than a blacklist, but
defaults to safety if the set of protocols git supports
grows). If no whitelist is specified, we continue to default
to allowing all protocols (this is an "unsafe" default, but
since the minority of users will want this sandboxing
effect, it is the only sensible one).
A note on the tests: ideally these would all be in a single
test file, but the git-daemon and httpd test infrastructure
is an all-or-nothing proposition rather than a test-by-test
prerequisite. By putting them all together, we would be
unable to test the file-local code on machines without
apache.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
send-email: fix uninitialized var warning for $smtp_auth
On the latest version of git-send-email, I see this error just before
running SMTP auth (I didn't provide any --smtp-auth= parameter):
Use of uninitialized value $smtp_auth in pattern match (m//) at \
/home/briannorris/git/git/git-send-email.perl line 1139.
Signed-off-by: Brian Norris <computersforpeace@gmail.com> Reviewed-by: Eric Sunshine <sunshine@sunshineco.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Introduce three i18n improvements from the following commits:
* tag, update-ref: improve description of option "create-reflog"
* pull: don't mark values for option "rebase" for translation
* show-ref: place angle brackets around variables in usage string
The experimental untracked-cache feature were buggy when paths with
a few levels of subdirectories are involved.
* dt/untracked-subdir:
untracked cache: fix entry invalidation
untracked-cache: fix subdirectory handling
t7063: use --force-untracked-cache to speed up a bit
untracked-cache: support sparse checkout
My 'demon' email address is no longer functional since, after 16+
years with demon, I have had to change my ISP. :(
Also, take the opportunity to remove my middle name, which I only
use on official documents (or in the GECOS field when creating a
user account on unix).
Signed-off-by: Ramsay Jones <ramsay@ramsayjones.plus.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Recent versions of scripted "git am" has a performance regression in
"git am --skip" codepath, which no longer exists in the built-in
version on the 'master' front. Fix the regression in the last
scripted version that appear in 2.5.x maintenance track and older.
* js/maint-am-skip-performance-regression:
am --skip/--abort: merge HEAD/ORIG_HEAD tree into index
Both "git show-ref -h" and "git show-ref --help" illustrated that the
"--exclude-existing" option makes the command read list of refs
from its standard input. Change only the "show-ref -h" output to
have a pair of "<>" around the placeholder that designate an input
file, i.e. "git show-ref --exclude-existing < <ref-list>".
* ah/show-ref-usage-string:
show-ref: place angle brackets around variables in usage string
Ensure that when passing a pipe, the gnulib poll replacement will not
return 0 before the timeout has passed.
Not obeying the timeout (and merely returning 0) causes pathological
behavior when preparing a packfile for a repository and taking a
long time to do so. If poll were to return 0 immediately, this would
cause keep-alives to get sent as quickly as possible until the packfile
was created. Such deviance from the standard would cause megabytes (or
more) of keep-alive packets to be sent.
GetTickCount is used as it is efficient, stable and monotonically
increasing. (Neither GetSystemTime nor QueryPerformanceCounter have
all three of these properties.)
Signed-off-by: Edward Thomson <ethomson@microsoft.com> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
The branch description will be included in 'git format-patch
--cover-letter' and in 'git pull-request' emails. It can also
be used in the automatic merge message. Tell the reader.
While here, clarify that the description may be a multi-line
explanation of the purpose of the branch's patch series.
Signed-off-by: Philip Oakley <philipoakley@iee.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
* git://ozlabs.org/~paulus/gitk:
gitk: Accelerators for the main menu
gitk: Adjust the menu line numbers to compensate for the new entry
gitk: Add a "Copy commit summary" command
gitk: Update Bulgarian translation (307t)
gitk: Update .po files
gitk: Update Bulgarian translation (304t)
gitk: Use translated version of "Command line" in getcommitlines
gitk: Make it easier to go quickly to a specific commit
gitk: Show the current view's name in the window title
gitk: Add mouse right-click options to copy path and branch name
gitk: Remove mc parameter from proc show_error
gitk: Fix error when changing colors after closing "List references" window
gitk: Replace catch {unset foo} with unset -nocomplain foo
gitk: Rearrange window title to be more conventional
gitk: sv.po: Update Swedish translation (305t0f0u)
gitk: Fix bad English grammar "Matches none Commit Info"
gitk: Adjust the menu line numbers to compensate for the new entry
Commit d835dbb9 ("gitk: Add a "Copy commit summary" command",
2015-08-13) in the upstream gitk repo added a new context menu entry.
Therefore, the line numbers of the entries below the new one need to be
adjusted when their text or state is changed.
Signed-off-by: Beat Bolli <dev+git@drbeat.li> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
tag, update-ref: improve description of option "create-reflog"
The description of option "create-reflog" is "create_reflog", which
is neither a good description, nor a sensible string to translate.
Change it to a more meaningful message.
Signed-off-by: Ralf Thielow <ralf.thielow@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Makefile: use SHELL_PATH when running generate-cmdlist.sh
Non-POSIX shells, such as /bin/sh on SunOS, do not support $((...))
arithmetic expansion or $(...) command substitution needed by
generate-cmdlist.sh. Make sure that we use a POSIX compliant shell
$(SHELL_PATH) when running generate-cmdlist.sh.
Signed-off-by: Alejandro R. Sedeño <asedeno@mit.edu> Acked-by: Eric Sunshine <sunshine@sunshineco.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Currently, git-svn parses an authors file using the perl regex
/^(.+?|\(no author\))\s*=\s*(.+?)\s*<(.+)>\s*$/
in order to extract svn user name, real name and e-mail.
This does not match an empty e-mail field like "<>". On the other hand,
the output of an authors-prog is parsed with the perl regex
as the authors prog gives different results compared to specifying
/tmp/authors as the authors file directly.
Instead, make git svn uses the perl regex
/^(.+?|\(no author\))\s*=\s*(.+?)\s*<(.*)>\s*$/
for parsing the authors file so that the same (slightly more lenient)
regex is used in both cases.
Reported-by: Till Schäfer <till2.schaefer@tu-dortmund.de> Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net> Signed-off-by: Eric Wong <normalperson@yhbt.net>