A handful of code paths had to read the commit object more than
once when showing header fields that are usually not parsed. The
internal data structure to keep track of the contents of the commit
object has been updated to reduce the need for this double-reading,
and to allow the caller find the length of the object.
* jk/commit-buffer-length:
reuse cached commit buffer when parsing signatures
commit: record buffer length in cache
commit: convert commit->buffer to a slab
commit-slab: provide a static initializer
use get_commit_buffer everywhere
convert logmsg_reencode to get_commit_buffer
use get_commit_buffer to avoid duplicate code
use get_cached_commit_buffer where appropriate
provide helpers to access the commit buffer
provide a helper to set the commit buffer
provide a helper to free commit buffer
sequencer: use logmsg_reencode in get_message
logmsg_reencode: return const buffer
do not create "struct commit" with xcalloc
commit: push commit_index update into alloc_commit_node
alloc: include any-object allocations in alloc_report
replace dangerous uses of strbuf_attach
commit_tree: take a pointer/len pair rather than a const strbuf
"git blame" assigned the blame to the copy in the working-tree if
the repository is set to core.autocrlf=input and the file used CRLF
line endings.
* bc/blame-crlf-test:
blame: correctly handle files regardless of autocrlf
Most callsites which use the commit buffer try to use the
cached version attached to the commit, rather than
re-reading from disk. Unfortunately, that interface provides
only a pointer to the NUL-terminated buffer, with no
indication of the original length.
For the most part, this doesn't matter. People do not put
NULs in their commit messages, and the log code is happy to
treat it all as a NUL-terminated string. However, some code
paths do care. For example, when checking signatures, we
want to be very careful that we verify all the bytes to
avoid malicious trickery.
This patch just adds an optional "size" out-pointer to
get_commit_buffer and friends. The existing callers all pass
NULL (there did not seem to be any obvious sites where we
could avoid an immediate strlen() call, though perhaps with
some further refactoring we could).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Like the callsites in the previous commit, logmsg_reencode
already falls back to read_sha1_file when necessary.
However, I split its conversion out into its own commit
because it's a bit more complex.
We return either:
1. The original commit->buffer
2. A newly allocated buffer from read_sha1_file
3. A reencoded buffer (based on either 1 or 2 above).
while trying to do as few extra reads/allocations as
possible. Callers currently free the result with
logmsg_free, but we can simplify this by pointing them
straight to unuse_commit_buffer. This is a slight layering
violation, in that we may be passing a buffer from (3).
However, since the end result is to free() anything except
(1), which is unlikely to change, and because this makes the
interface much simpler, it's a reasonable bending of the
rules.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Right now this is just a one-liner, but abstracting it will
make it easier to change later.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The return value from logmsg_reencode may be either a newly
allocated buffer or a pointer to the existing commit->buffer.
We would not want the caller to accidentally free() or
modify the latter, so let's mark it as const. We can cast
away the constness in logmsg_free, but only once we have
determined that it is a free-able buffer.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In both blame and merge-recursive, we sometimes create a
"fake" commit struct for convenience (e.g., to represent the
HEAD state as if we would commit it). By allocating
ourselves rather than using alloc_commit_node, we do not
properly set the "index" field of the commit. This can
produce subtle bugs if we then use commit-slab on the
resulting commit, as we will share the "0" index with
another commit.
We can fix this by using alloc_commit_node() to allocate.
Note that we cannot free the result, as it is part of our
commit allocator. However, both cases were already leaking
the allocated commit anyway, so there's nothing to fix up.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If a file contained CRLF line endings in a repository with
core.autocrlf=input, then blame always marked lines as "Not
Committed Yet", even if they were unmodified. Don't attempt to
convert the line endings when creating the fake commit so that blame
works correctly regardless of the autocrlf setting.
Reported-by: Ephrim Khong <dr.khong@gmail.com>
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When show date in relative date format for git-blame, the max display
width of datetime is set as the length of the string "Thu Oct 19
16:00:04 2006 -0700" (30 characters long). But actually the max width
for C locale is only 22 (the length of string "x years, xx months ago").
And for other locale, it maybe smaller. E.g. For Chinese locale, only
needs a half (16-character width).
Set blame_date_width as the display width of _("4 years, 11 months
ago"), so that translators can make the choice.
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Command `git blame --date relative` aligns the date field with a
fixed-width (defined by blame_date_width), and if time_str is shorter
than that, it adds spaces for padding. But there are two bugs in the
following codes:
time_len = strlen(time_str);
...
memset(time_buf + time_len, ' ', blame_date_width - time_len);
1. The type of blame_date_width is size_t, which is unsigned. If
time_len is greater than blame_date_width, the result of
"blame_date_width - time_len" will never be a negative number, but a
really big positive number, and will cause memory overwrite.
This bug can be triggered if either l10n message for function
show_date_relative() in date.c is longer than 30 characters, then
`git blame --date relative` may exit abnormally.
2. When show blame information with relative time, the UTF-8 characters
in time_str will break the alignment of columns after the date field.
This is because the time_buf padding with spaces should have a
constant display width, not a fixed strlen size. So we should call
utf8_strwidth() instead of strlen() for width calibration.
Helped-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Helped-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Jiang Xin <worldhello.net@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Attempts to show where a single-strand-of-pearls break in "git log"
output.
* nd/log-show-linear-break:
log: add --show-linear-break to help see non-linear history
object.h: centralize object flag allocation
While the field "flags" is mainly used by the revision walker, it is
also used in many other places. Centralize the whole flag allocation to
one place for a better overview (and easier to move flags if we have
too).
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Code clean-up.
* dk/blame-janitorial:
builtin/blame.c::find_copy_in_blob: no need to scan for region end
blame.c: prepare_lines should not call xrealloc for every line
builtin/blame.c::prepare_lines: fix allocation size of sb->lineno
builtin/blame.c: eliminate same_suspect()
builtin/blame.c: struct blame_entry does not need a prev link
Shrink lifetime of variables by moving their definitions to an
inner scope where appropriate.
* ep/varscope:
builtin/gc.c: reduce scope of variables
builtin/fetch.c: reduce scope of variable
builtin/commit.c: reduce scope of variables
builtin/clean.c: reduce scope of variable
builtin/blame.c: reduce scope of variables
builtin/apply.c: reduce scope of variables
bisect.c: reduce scope of variable
Making a single preparation run for counting the lines will avoid memory
fragmentation. Also, fix the allocated memory size which was wrong
when sizeof(int *) != sizeof(int), and would have been too small
for sizeof(int *) < sizeof(int), admittedly unlikely.
Signed-off-by: David Kastrup <dak@gnu.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If we are calling xrealloc on every single line, the least we can do
is get the right allocation size.
Signed-off-by: David Kastrup <dak@gnu.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since the origin pointers are "interned" and reference-counted, comparing
the pointers rather than the content is enough. The only uninterned
origins are cached values kept in commit->util, but same_suspect is not
called on them.
Signed-off-by: David Kastrup <dak@gnu.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
There is no reason to have a hardcoded upper limit of the number of
parents for an octopus merge, created via the graft mechanism.
* js/lift-parent-count-limit:
Remove the line length limit for graft files
Support for grafts predates Git's strbuf, and hence it is understandable
that there was a hard-coded line length limit of 1023 characters (which
was chosen a bit awkwardly, given that it is *exactly* one byte short of
aligning with the 41 bytes occupied by a commit name and the following
space or new-line character).
While regular commit histories hardly win comprehensibility in general
if they merge more than twenty-two branches in one go, it is not Git's
business to limit grafts in such a way.
In this particular developer's case, the use case that requires
substantially longer graft lines to be supported is the visualization of
the commits' order implied by their changes: commits are considered to
have an implicit relationship iff exchanging them in an interactive
rebase would result in merge conflicts.
Thusly implied branches tend to be very shallow in general, and the
resulting thicket of implied branches is usually very wide; It is
actually quite common that *most* of the commits in a topic branch have
not even one implied parent, so that a final merge commit has about as
many implied parents as there are commits in said branch.
[jc: squashed in tests by Jonathan]
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jk/robustify-parse-commit:
checkout: do not die when leaving broken detached HEAD
use parse_commit_or_die instead of custom message
use parse_commit_or_die instead of segfaulting
assume parse_commit checks for NULL commit
assume parse_commit checks commit->object.parsed
log_tree_diff: die when we fail to parse a commit
Normally parse_pathspec() is used on command line arguments where it
can do fancy thing like parsing magic on each argument or adding magic
for all pathspecs based on --*-pathspecs options.
There's another use of parse_pathspec(), where pathspec is needed, but
the input is known to be pure paths. In this case we usually don't
want --*-pathspecs to interfere. And we definitely do not want to
parse magic in these paths, regardless of --literal-pathspecs.
Add new flag PATHSPEC_LITERAL_PATH for this purpose. When it's set,
--*-pathspecs are ignored, no magic is parsed. And if the caller
allows PATHSPEC_LITERAL (i.e. the next calls can take literal magic),
then PATHSPEC_LITERAL will be set.
This fixes cases where git chokes when GIT_*_PATHSPECS are set because
parse_pathspec() indicates it won't take any magic. But
GIT_*_PATHSPECS add them anyway. These are
export GIT_LITERAL_PATHSPECS=1
git blame -- something
git log --follow something
git log --merge
"git ls-files --with-tree=path" (aka parse_pathspec() in
overlay_tree_on_cache()) is safe because the input is empty, and
producing one pathspec due to PATHSPEC_PREFER_CWD does not take any
magic into account.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Acked-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The parse_commit function will check the "parsed" flag of
the object and do nothing if it is set. There is no need
for callers to check the flag themselves, and doing so only
clutters the code.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git mv A B" when moving a submodule A does "the right thing",
inclusing relocating its working tree and adjusting the paths in
the .gitmodules file.
* jl/submodule-mv: (53 commits)
rm: delete .gitmodules entry of submodules removed from the work tree
mv: update the path entry in .gitmodules for moved submodules
submodule.c: add .gitmodules staging helper functions
mv: move submodules using a gitfile
mv: move submodules together with their work trees
rm: do not set a variable twice without intermediate reading.
t6131 - skip tests if on case-insensitive file system
parse_pathspec: accept :(icase)path syntax
pathspec: support :(glob) syntax
pathspec: make --literal-pathspecs disable pathspec magic
pathspec: support :(literal) syntax for noglob pathspec
kill limit_pathspec_to_literal() as it's only used by parse_pathspec()
parse_pathspec: preserve prefix length via PATHSPEC_PREFIX_ORIGIN
parse_pathspec: make sure the prefix part is wildcard-free
rename field "raw" to "_raw" in struct pathspec
tree-diff: remove the use of pathspec's raw[] in follow-rename codepath
remove match_pathspec() in favor of match_pathspec_depth()
remove init_pathspec() in favor of parse_pathspec()
remove diff_tree_{setup,release}_paths
convert common_prefix() to use struct pathspec
...
Teaches "git blame" to take more than one -L ranges.
* es/blame-L-twice:
line-range: reject -L line numbers less than 1
t8001/t8002: blame: add tests of -L line numbers less than 1
line-range: teach -L^:RE to search from start of file
line-range: teach -L:RE to search from end of previous -L range
line-range: teach -L^/RE/ to search from start of file
line-range-format.txt: document -L/RE/ relative search
log: teach -L/RE/ to search from end of previous -L range
blame: teach -L/RE/ to search from end of previous -L range
line-range: teach -L/RE/ to search relative to anchor point
blame: document multiple -L support
t8001/t8002: blame: add tests of multiple -L options
blame: accept multiple -L ranges
blame: inline one-line function into its lone caller
range-set: publish API for re-use by git-blame -L
line-range-format.txt: clarify -L:regex usage form
git-log.txt: place each -L option variation on its own line
Range specification -L/RE/ for blame/log unconditionally begins
searching at line one. Mailing list discussion [1] suggests that, in the
presence of multiple -L options, -L/RE/ should search relative to the
endpoint of the previous -L range, if any.
Teach the parsing machinery underlying blame's and log's -L options to
accept a start point for -L/RE/ searches. Follow-up patches will upgrade
blame and log to take advantage of this ability.
[1]: http://thread.gmane.org/gmane.comp.version-control.git/229755/focus=229966
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
git-blame accepts only a single -L option or none. Clients requiring
blame information for multiple disjoint ranges are therefore forced
either to invoke git-blame multiple times, once for each range, or only
once with no -L option to cover the entire file, both of which can be
costly. Teach git-blame to accept multiple -L ranges. Overlapping and
out-of-order ranges are accepted.
In this patch, the X in -LX,Y is absolute (for instance, /RE/ patterns
search from line 1), and Y is relative to X. Follow-up patches provide
more flexibility over how X is anchored.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
As of 25ed3412 (Refactor parse_loc; 2013-03-28),
blame.c:prepare_blame_range() became effectively a one-line function
which merely passes its arguments along to another function. This
indirection does not bring clarity to the code. Simplify by inlining
prepare_blame_range() into its lone caller.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since inception, -LX,Y has correctly reported an out-of-range error when
Y is beyond end of file, however, X was not checked, and an out-of-range
X would cause a crash. 92f9e273 (blame: prevent a segv when -L given
start > EOF; 2010-02-08) attempted to rectify this shortcoming but has
its own off-by-one error which allows X to extend one line past end of
file. For example, given a file with 5 lines:
git blame -L5 foo # OK, blames line 5
git blame -L6 foo # accepted, no error, no output, huh?
git blame -L7 foo # error "fatal: file foo has only 5 lines"
Fix this bug.
In order to avoid regressing "blame foo" when foo is an empty file, the
fix is slightly more complicated than changing '<' to '<='.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This task emerged from b04ba2bb (parse-options: deprecate OPT_BOOLEAN,
2011-09-27). All occurrences of the respective variables have
been reviewed and none of them relied on the counting up mechanism,
but all of them were using the variable as a true boolean.
This patch does not change semantics of any command intentionally.
Signed-off-by: Stefan Beller <stefanbeller@googlemail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
While at there, move free_pathspec() to pathspec.c
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* tr/line-log:
git-log(1): remove --full-line-diff description
line-log: fix documentation formatting
log -L: improve comments in process_all_files()
log -L: store the path instead of a diff_filespec
log -L: test merge of parallel modify/rename
t4211: pass -M to 'git log -M -L...' test
log -L: fix overlapping input ranges
log -L: check range set invariants when we look it up
Speed up log -L... -M
log -L: :pattern:file syntax to find by funcname
Implement line-history search (git log -L)
Export rewrite_parents() for 'log -L'
Refactor parse_loc
pretty-printing body of the commit that is stored in non UTF-8
encoding did not work well. The early part of this series fixes
it. And then it adds %C(auto) specifier that turns the coloring on
when we are emitting to the terminal, and adds column-aligning
format directives.
* nd/pretty-formats:
pretty: support %>> that steal trailing spaces
pretty: support truncating in %>, %< and %><
pretty: support padding placeholders, %< %> and %><
pretty: add %C(auto) for auto-coloring
pretty: split color parsing into a separate function
pretty: two phase conversion for non utf-8 commits
utf8.c: add reencode_string_len() that can handle NULs in string
utf8.c: add utf8_strnwidth() with the ability to skip ansi sequences
utf8.c: move display_mode_esc_sequence_len() for use by other functions
pretty: share code between format_decoration and show_decorations
pretty-formats.txt: wrap long lines
pretty: get the correct encoding for --pretty:format=%e
pretty: save commit encoding from logmsg_reencode if the caller needs it
The commit encoding is parsed by logmsg_reencode, there's no need for
the caller to re-parse it again. The reencoded message now has the new
encoding, not the original one. The caller would need to read commit
object again before parsing.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
split_ident_line() can leave us with the pointers date_begin, date_end,
tz_begin and tz_end all set to NULL. Check them before use and supply
the same fallback values as in the case of a negative return code from
split_ident_line().
The "(unknown)" is not actually shown in the output, though, because it
will be converted to a number (zero) eventually.
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This new syntax finds a funcname matching /pattern/, and then takes from there
up to (but not including) the next funcname. So you can say
git log -L:main:main.c
and it will dig up the main() function and show its line-log, provided
there are no other funcnames matching 'main'.
Signed-off-by: Thomas Rast <trast@student.ethz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We want to use the same style of -L n,m argument for 'git log -L' as
for git-blame. Refactor the argument parsing of the range arguments
from builtin/blame.c to the (new) file that will hold the 'git log -L'
logic.
To accommodate different data structures in blame and log -L, the file
contents are abstracted away; parse_range_arg takes a callback that it
uses to get the contents of a line of the (notional) file.
The new test is for a case that made me pause during debugging: the
'blame -L with invalid end' test was the only one that noticed an
outright failure to parse the end *at all*. So make a more explicit
test for that.
Signed-off-by: Bo Yang <struggleyb.nku@gmail.com>
Signed-off-by: Thomas Rast <trast@student.ethz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Usually a commit that makes it to logmsg_reencode will have
been parsed, and the commit->buffer struct member will be
valid. However, some code paths will free commit buffers
after having used them (for example, the log traversal
machinery will do so to keep memory usage down).
Most of the time this is fine; log should only show a commit
once, and then exits. However, there are some code paths
where this does not work. At least two are known:
1. A commit may be shown as part of a regular ref, and
then it may be shown again as part of a submodule diff
(e.g., if a repo contains refs to both the superproject
and subproject).
2. A notes-cache commit may be shown during "log --all",
and then later used to access a textconv cache during a
diff.
Lazily loading in logmsg_reencode does not necessarily catch
all such cases, but it should catch most of them. Users of
the commit buffer tend to be either parsing for structure
(in which they will call parse_commit, and either we will
already have parsed, or we will load commit->buffer lazily
there), or outputting (either to the user, or fetching a
part of the commit message via format_commit_message). In
the latter case, we should always be using logmsg_reencode
anyway (and typically we do so via the pretty-print
machinery).
If there are any cases that this misses, we can fix them up
to use logmsg_reencode (or handle them on a case-by-case
basis if that is inappropriate).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The logmsg_reencode function will return the reencoded
commit buffer, or NULL if reencoding failed or no reencoding
was necessary. Since every caller then ends up checking for NULL
and just using the commit's original buffer, anyway, we can
be a bit more helpful and just return that buffer when we
would have returned NULL.
Since the resulting string may or may not need to be freed,
we introduce a logmsg_free, which checks whether the buffer
came from the commit object or not (callers either
implemented the same check already, or kept two separate
pointers, one to mark the buffer to be used, and one for the
to-be-freed string).
Pushing this logic into logmsg_* simplifies the callers, and
will let future patches lazily load the commit buffer in a
single place.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach commands in the "log" family to optionally pay attention to
the mailmap.
* ap/log-mailmap:
log --use-mailmap: optimize for cases without --author/--committer search
log: add log.mailmap configuration option
log: grep author/committer using mailmap
test: add test for --use-mailmap option
log: add --use-mailmap option
pretty: use mailmap to display username and email
mailmap: add mailmap structure to rev_info and pp
mailmap: simplify map_user() interface
mailmap: remove email copy and length limitation
Use split_ident_line to parse author and committer
string-list: allow case-insensitive string list
Simplify map_user(), mostly to avoid copies of string buffers. It
also simplifies caller functions.
map_user() directly receive pointers and length from the commit buffer
as mail and name. If mapping of the user and mail can be done, the
pointer is updated to a new location. Lengths are also updated if
necessary.
The caller of map_user() can then copy the new email and name if
necessary.
Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Currently blame.c::get_acline(), pretty.c::pp_user_info() and
shortlog.c::insert_one_record() are parsing author name, email, time
and tz themselves.
Use ident.c::split_ident_line() for better code reuse.
Signed-off-by: Antoine Pelisse <apelisse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This function has only two callsites, and is a thin wrapper whose
usefulness is dubious. When the caller needs to learn the log
output encoding, it should be able to do so by directly calling
get_log_output_encoding() and calling the underlying
logmsg_reencode() with it.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If you know your history did not have renames, or if you care only
about the history after a large rename that happened some time ago,
"git blame --no-follow $path" is a way to tell the command not to
bother about renames.
When you use -C, the lines that came from the renamed file will
still be found without the whole-file rename detection, so it is not
all that interesting either way, though.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
