With rename detection enabled the line-level log is able to trace the
evolution of line ranges across whole-file renames [1]. Alas, to
achieve that it uses the diff machinery very inefficiently, making the
operation very slow [2]. And since rename detection is enabled by
default, the line-level log is very slow by default.
When the line-level log processes a commit with rename detection
enabled, it currently does the following (see queue_diffs()):
1. Computes a full tree diff between the commit and (one of) its
parent(s), i.e. invokes diff_tree_oid() with an empty
'diffopt->pathspec'.
2. Checks whether any paths in the line ranges were modified.
3. Checks whether any modified paths in the line ranges are missing
in the parent commit's tree.
4. If there is such a missing path, then calls diffcore_std() to
figure out whether the path was indeed renamed based on the
previously computed full tree diff.
5. Continues doing stuff that are unrelated to the slowness.
So basically the line-level log computes a full tree diff for each
commit-parent pair in step (1) to be used for rename detection in step
(4) in the off chance that an interesting path is missing from the
parent.
Avoid these expensive and mostly unnecessary full tree diffs by
limiting the diffs to paths in the line ranges. This is much cheaper,
and makes step (2) unnecessary. If it turns out that an interesting
path is missing from the parent, then fall back and compute a full
tree diff, so the rename detection will still work.
Care must be taken when to update the pathspec used to limit the diff
in case of renames. A path might be renamed on one branch and
modified on several parallel running branches, and while processing
commits on these branches the line-level log might have to alternate
between looking at a path's new and old name. However, at any one
time there is only a single 'diffopt->pathspec'.
So add a step (0) to the above to ensure that the paths in the
pathspec match the paths in the line ranges associated with the
currently processed commit, and re-parse the pathspec from the paths
in the line ranges if they differ.
The new test cases include a specially crafted piece of history with
two merged branches and two files, where each branch modifies both
files, renames on of them, and then modifies both again. Then two
separate 'git log -L' invocations check the line-level log of each of
those two files, which ensures that at least one of those invocations
have to do that back-and-forth between the file's old and new name (no
matter which branch is traversed first). 't/t4211-line-log.sh'
already contains two tests involving renames, they don't don't trigger
this back-and-forth.
Avoiding these unnecessary full tree diffs can have huge impact on
performance, especially in big repositories with big trees and mergy
history. Tracing the evolution of a function through the whole
history:
# git.git
$ time git --no-pager log -L:read_alternate_refs:sha1-file.c v2.23.0
Before:
real 0m8.874s
user 0m8.816s
sys 0m0.057s
After:
real 0m2.516s
user 0m2.456s
sys 0m0.060s
# linux.git
$ time ~/src/git/git --no-pager log \
-L:build_restore_work_registers:arch/mips/mm/tlbex.c v5.2
Before:
real 3m50.033s
user 3m48.041s
sys 0m0.300s
After:
real 0m2.599s
user 0m2.466s
sys 0m0.157s
That's just over 88x speedup.
[1] Line-level log's rename following is quite similar to 'git log
--follow path', with the notable differences that it does handle
multiple paths at once as well, and that it doesn't show the
commit performing the rename if it's an exact rename.
[2] This slowness might not have been apparent initially, because back
when the line-level log feature was introduced rename detection
was not yet enabled by default; 12da1d1f6f (Implement line-history
search (git log -L), 2013-03-28) and 5404c116aa (diff: activate
diff.renames by default, 2016-02-25).
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If you use "log -L" with an output format like "--raw" or "--stat",
we'll silently ignore the format and just output the normal patch.
Let's detect and complain about this, which at least tells the user
what's going on.
The tests here aren't exhaustive over the set of all formats, but it
should at least let us know if somebody breaks the format-checking.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When "-L" is in use, we ignore any diff output format that the user
provides to us, and just always print a patch (with extra context lines
covering the whole area of interest). It's not entirely clear what we
should do with all formats (e.g., should "--stat" show just the diffstat
of the touched lines, or the stat for the whole file?).
But "-s" is pretty clear: the user probably wants to see just the
commits that touched those lines, without any diff at all. Let's at
least make that work.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Introduce optname() that does the early half of original opterror() to
come up with the name of the option reported back to the user, and use
it to kill opterror(). The callers of opterror() now directly call
error() using the string returned by opterror() instead.
There are a few issues with opterror()
- it tries to assemble an English sentence from pieces. This is not
great for translators because we give them pieces instead of a full
sentence.
- It's a wrapper around error() and needs some hack to let the
compiler know it always returns -1.
- Since it takes a string instead of printf format, one call site has
to assemble the string manually before passing to it.
Using error() directly solves the second and third problems.
It kind helps the first problem as well because "%s does foo" does
give a translator a full sentence in a sense and let them reorder if
needed. But it has limitations, if the subject part has to change
based on the rest of the sentence, that language is screwed. This is
also why I try to avoid calling optname() when 'flags' is known in
advance.
Mark of these strings for translation as well while at there.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Test fixes.
* sg/test-must-be-empty:
tests: use 'test_must_be_empty' instead of 'test_cmp <empty> <out>'
tests: use 'test_must_be_empty' instead of 'test_cmp /dev/null <out>'
tests: use 'test_must_be_empty' instead of 'test ! -s'
tests: use 'test_must_be_empty' instead of '! test -s'
Using 'test_must_be_empty' is preferable to 'test ! -s', because it
gives a helpful error message if the given file is unexpectedly no
empty, while the latter remains completely silent. Furthermore, it
also catches cases when the given file unexpectedly does not exist at
all.
This patch was created by:
sed -i -e 's/test ! -s/test_must_be_empty/' t[0-9]*.sh
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If the -L option is used to specify a line range in git log, and the end
of the range is past the end of the file, git will fail with a fatal
error. This commit prevents such behaviour - instead we perform the log
for existing lines within the specified range.
This commit also fixes a corner case where -L ,-n:file would be treated
as a log over the whole file. Now we treat this as -L 1,-n:file and
blame the first line of the file instead.
Signed-off-by: Isabella Stephens <istephens@atlassian.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The existing implementation of range_set_union does not correctly
reallocate memory, leading to a heap overflow when it attempts to union
more than 24 separate line ranges.
For struct range_set *out to grow correctly it must have out->nr set to
the current size of the buffer when it is passed to range_set_grow.
However, the existing implementation of range_set_union only updates
out->nr at the end of the function, meaning that it is always zero
before this. This results in range_set_grow never growing the buffer, as
well as some of the union logic itself being incorrect as !out->nr is
always true.
The reason why 24 is the limit is that the first allocation of size 1
ends up allocating a buffer of size 24 (due to the call to alloc_nr in
ALLOC_GROW). This goes some way to explain why this hasn't been
caught before.
Fix the problem by correctly updating out->nr after reallocating the
range_set. As this results in out->nr containing the same value as the
variable o, replace o with out->nr as well.
Finally, add a new test to help prevent the problem reoccurring in the
future. Thanks to Vegard Nossum for writing the test.
Signed-off-by: Allan Xavier <allan.x.xavier@oracle.com>
Reviewed-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The test script t4202-log.sh is already pretty long, and it is a good
idea to test --output with a more obscure option, anyway. So let's
test it in conjunction with line-log.
The most important part of this test, of course, is to ensure that the
file is not closed after writing the diff, but only at the very end
of the log output. That is the entire reason why the test tries to
generate a log that covers more than one commit.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The old message did not mention the :regex:file form.
To avoid overly long lines, split the message into two lines (in case
item->string is long, it will be the only part truncated in a narrow
terminal).
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
line-log tries to access all parents of a commit, but only the first
parent has been loaded if "--first-parent" is specified, resulting
in a crash.
Limit the number of parents to one if "--first-parent" is specified.
Reported-by: Eric N. Vander Weele <ericvw@gmail.com>
Signed-off-by: Tzvetan Mikov <tmikov@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The -L:RE option of blame/log searches from the end of the previous -L
range, if any. Add new notation -L^:RE to override this behavior and
search from start of file.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
For consistency with -L/RE/, teach -L:RE to search relative to the end
of the previous -L range, if any.
The new behavior invalidates one test in t4211 which assumes that -L:RE
begins searching at start of file. This test will be resurrected in a
follow-up patch which teaches -L:RE how to override the default relative
search behavior.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When 12da1d1f added -L support to git-log, a broken bounds check was
copied from git-blame -L which incorrectly allows -LX to extend one line
past end of file without reporting an error. Instead, it generates an
empty range. Fix this bug.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
58960978 and 99780b0a added tests which demonstrated bugs (crashes) in
range-set and line-log when handed empty ranges specified via "log
-LX:file" where X is one greater than the last line of the file. After
these tests were added, it was realized that the ability to specify an
empty range is a loophole due to a bug in -L bounds checking. That bug
is slated to be fixed in a subsequent patch.
Unfortunately, the closure of this loophole makes it impossible to
continue checking range-set and line-log behavior with regard to empty
ranges since there is no other way to specify empty ranges via the
command-line. APIs of both facilities are private (file static) so
there likewise is no way to test their behaviors programmatically.
Consequently, retire these two tests.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A bounds checking bug allows the X in -LX to extend one line past the
end of file. For example, given a file with 5 lines, -L6 is accepted as
valid. Demonstrate this problem.
While here, also add tests to check that the remaining cases of X and Y
in -LX,Y are handled correctly at and in the vicinity of end-of-file.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Wnen I rewrote "cat b.c | wc -l" into "wc -l <b.c" to squash in a
suggestion on the list to this series, I screwed up subsequent
rebase. Fix it up.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
range-set invariants are: ranges must be (1) non-empty, (2) disjoint,
(3) sorted in ascending order.
line_log_data_insert() breaks the non-empty invariant under the
following conditions: the incoming range is empty and the pathname
attached to the range has not yet been encountered. In this case,
line_log_data_insert() assigns the empty range to a new line_log_data
record without taking any action to ensure that the empty range is
eventually folded out. Subsequent range-set functions crash or throw an
assertion failure upon encountering such an anomaly. Fix this bug.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Acked-by: Thomas Rast <trast@inf.ethz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
range-set invariants are: ranges must be (1) non-empty, (2) disjoint,
(3) sorted in ascending order.
During processing, various range-set utility functions break the
invariants (for instance, by adding empty ranges), with the
expectation that a finalizing sort_and_merge_range_set() will restore
sanity.
sort_and_merge_range_set(), however, neglects to fold out empty
ranges, thus it fails to satisfy the non-empty constraint. Subsequent
range-set functions crash or throw an assertion failure upon
encountering such an anomaly. Rectify the situation by having
sort_and_merge_range_set() fold out empty ranges.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Acked-by: Thomas Rast <trast@inf.ethz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Acked-by: Thomas Rast <trast@inf.ethz.ch>
Helped-by: SZEDER Gábor <szeder@ira.uka.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Acked-by: Thomas Rast <trast@inf.ethz.ch>
Helped-by: SZEDER Gábor <szeder@ira.uka.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When coalescing ranges, sort_and_merge_range_set() unconditionally
assumes that the end of a range being folded into a preceding range
should become the end of the coalesced range. This assumption, however,
is invalid when one range is a subset of another. For example, given
ranges 1-5 and 2-3 added via range_set_append_unsafe(),
sort_and_merge_range_set() incorrectly coalesces them to range 1-3
rather than the correct union range 1-5. Fix this bug.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
t4211 attempts to test multiple git-log -L ranges where one range is a
superset of the other, and falsely succeeds because its "expected"
output is incorrect.
Overlapping -L ranges handed to git-log are coalesced by
line-log.c:sort_and_merge_range_set() into a set of non-overlapping,
disjoint ranges. When one range is a subset of another,
sort_and_merge_range_set() should coalesce both ranges to the superset
range, but instead the coalesced range often is incorrectly truncated to
the end of the subset range. For example, ranges 2-8 and 3-4 are
coalesced incorrectly to 2-4.
One can observe this incorrect behavior with git-log -L using the test
repository created by t4211. The superset/subset ranges t4211 employs
are 4-$ and 8-12 (where $ represents end-of-file). The coalesced range
should be 4-$. Manually invoking git-log with the same ranges the test
employs, we see:
% git log -L 4:a.c simple |
awk '/^commit [0-9a-f]{40}/ { print substr($2,1,7) }'
4659538
100b61a
39b6eb2
a6eb826
f04fb20
de4c48a
% git log -L 8,12:a.c simple | awk ...
f04fb20
de4c48a
% git log -L 4:a.c -L 8,12:a.c simple | awk ...
a6eb826
f04fb20
de4c48a
This last output is incorrect. 8-12 is a subset of 4-$, hence the output
of the coalesced range should be the same as the 4-$ output shown first.
In fact, the above incorrect output is the truncated bogus range 4-12:
% git log -L 4,12:a.c simple | awk ...
a6eb826
f04fb20
de4c48a
Fix the test to correctly fail in the presence of the
sort_and_merge_range_set() coalescing bug. Do so by changing the
"expected" output to the commits mentioned in the 4-$ output above.
Signed-off-by: Eric Sunshine <sunshine@sunshineco.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
line_log_data has held a diff_filespec* since the very early versions
of the code. However, the only place in the code where we actually
need the full filespec is parse_range_arg(); in all other cases, we
are only interested in the path, so there is hardly a reason to store
a filespec. Even worse, it causes a lot of redundant ->spec->path
pointer dereferencing.
And *even* worse, it caused the following bug. If you merge a rename
with a modification to the old filename, like so:
* Merge
| \
| * Modify foo
| |
* | Rename foo->bar
| /
* Create foo
we internally -- in process_ranges_merge_commit() -- scan all parents.
We are mainly looking for one that doesn't have any modifications, so
that we can assign all the blame to it and simplify away the merge.
In doing so, we run the normal machinery on all parents in a loop.
For each parent, we prepare a "working set" line_log_data by making a
copy with line_log_data_copy(), which does *not* make a copy of the
spec.
Now suppose the rename is the first parent. The diff machinery tells
us that the filepair is ('foo', 'bar'). We duly update the path we
are interested in:
rg->spec->path = xstrdup(pair->one->path);
But that 'struct spec' is shared between the output line_log_data and
the original input line_log_data. So we just wrecked the state of
process_ranges_merge_commit(). When we get around to the second
parent, the ranges tell us we are interested in a file 'foo' while the
commits touch 'bar'.
So most of this patch is just s/->spec->path/->path/ and associated
management changes. This implicitly fixes the bug because we removed
the shared parts between input and output of line_log_data_copy(); it
is now safe to overwrite the path in the copy.
There's one only somewhat related change: the comment in
process_all_files() explains the reasoning behind using 'range' there.
That bit of half-correct code had me sidetracked for a while.
Signed-off-by: Thomas Rast <trast@inf.ethz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This tests a toy example of a history like
* Merge
| \
| * Modify foo
| |
* | Rename foo->bar
| /
* Create foo
Current log -L fails on this; we'll fix it in the next commit.
Signed-off-by: Thomas Rast <trast@inf.ethz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Embarrassingly, the -M test did not actually invoke -M, and thus not
really test the feature.
Signed-off-by: Thomas Rast <trast@inf.ethz.ch>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The existing code was too defensive, and would trigger the assert in
range_set_append() if the user gave overlapping ranges.
The intent was always to define overlapping ranges as just the union
of all of them, as evidenced by the call to sort_and_merge_range_set().
(Which was already used, unlike what the comment said.)
Fix by splitting out the meat of range_set_append() to a new _unsafe()
function that lacks the paranoia. sort_and_merge_range_set will fix
up the ranges, so we don't need the checks there.
Signed-off-by: Thomas Rast <trast@inf.ethz.ch>
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>
This is a rewrite of much of Bo's work, mainly in an effort to split
it into smaller, easier to understand routines.
The algorithm is built around the struct range_set, which encodes a
series of line ranges as intervals [a,b). This is used in two
contexts:
* A set of lines we are tracking (which will change as we dig through
history).
* To encode diffs, as pairs of ranges.
The main routine is range_set_map_across_diff(). It processes the
diff between a commit C and some parent P. It determines which diff
hunks are relevant to the ranges tracked in C, and computes the new
ranges for P.
The algorithm is then simply to process history in topological order
from newest to oldest, computing ranges and (partial) diffs. At
branch points, we need to merge the ranges we are watching. We will
find that many commits do not affect the chosen ranges, and mark them
TREESAME (in addition to those already filtered by pathspec limiting).
Another pass of history simplification then gets rid of such commits.
This is wired as an extra filtering pass in the log machinery. This
currently only reduces code duplication, but should allow for other
simplifications and options to be used.
Finally, we hook a diff printer into the output chain. Ideally we
would wire directly into the diff logic, to optionally use features
like word diff. However, that will require some major reworking of
the diff chain, so we completely replace the output with our own diff
for now.
As this was a GSoC project, and has quite some history by now, many
people have helped. In no particular order, thanks go to
Jakub Narebski <jnareb@gmail.com>
Jens Lehmann <Jens.Lehmann@web.de>
Jonathan Nieder <jrnieder@gmail.com>
Junio C Hamano <gitster@pobox.com>
Ramsay Jones <ramsay@ramsay1.demon.co.uk>
Will Palmer <wmpalmer@gmail.com>
Apologies to everyone I forgot.
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>
SZEDER Gábor
Jeff King
Nguyễn Thái Ngọc Duy
Junio C Hamano
Isabella Stephens
Allan Xavier
Johannes Schindelin
Matthieu Moy
Tzvetan Mikov
Eric Sunshine
Thomas Rast
Thomas Rast