The provider chain so far holds the diff-hunks store in front of the
terminal builtin computation. Open it to external processes: a pair on
a path whose driver configures diff.<driver>.process is answered by a
long-running process speaking a pkt-line protocol (following the filter
process protocol), registered at the head of the chain and consulted
before the store and before any blob is loaded.
The protocol starts with the smallest request that can carry an answer:
object names alone. A request is the pathname and the pair's
old-oid/new-oid, with no content. The process answers with hunk lines,
with a zero-hunk success that asserts the blobs equivalent (trailing
newlines included), or with status=need-content, on which the pair
falls through to the builtin answer. This serves the two shapes that
need no content pushed to them: a cache keyed on the blob pair, and a
process that fetches the blobs itself (for example over "git cat-file
--batch"). A pair whose side is not a stored blob carries a NULL id;
the provider sends no request and passes it. Because Git holds no
content for the exchange, the answer is used as sent: hunks are
validated for order, overlap, lockstep alignment, and magnitude, then
replayed without the normalization xdiff applies to diffs it computes
itself. The magnitude bound is the blobs' sizes, read from the object
database without loading content: a blob of N bytes holds at most N
lines.
Because the process's answer is authoritative, it outranks the store,
and its head-of-chain position says so. A pair the process answers
never reaches the store and is never recorded, so nothing it produces
enters the store, which holds the builtin answer only. A request it
does not answer, whether need-content, a missing capability, or a
missing id, passes down the chain to the builtin answer, which is what
the store serves, so the store may serve such a pair and a warming run
may record it. Entries recorded before a process was configured are
not purged; a pair the process answers ignores them, and "git diff-hunks
clear" discards them.
The provider gates itself per request. The driver is looked up by the
old-side path, so a renamed file resolves to the same driver, and by
the repository-relative path, so a diff.relative run from a
subdirectory names the pair the same way. Options the process is never
told about select no process: the whitespace-ignoring options, -I,
--anchored, and an algorithm forced by option or configuration (blame
routes its algorithm through xdl_opts, so --histogram is covered). The
request gains its last field, the path; the consumers change only by
filling it, and neither names the process.
The provider's state is its repository's pool of running processes,
keyed by the configured command, so drivers sharing a command share a
process, a submodule speaks to its own, and releasing the provider
(from repo_clear()) stops them. The pool owns a copy of each command
string, so an entry outlives a config re-read. A command that fails
stays as an entry that is not retried: its request and every later one
pass, so the store may serve the path for the rest of the command.
A protocol error in a response never kills the command. The response
is read through a packet reader gentle about framing, so an error takes
one path: a single warning, the process stopped and marked failed, and
the builtin diff for the rest of the command. That covers garbage
bytes, a truncated response, an empty packet, a bare status, and an
unrecognized status. Semantically invalid coordinates cost only their
pair: the response is drained, the pair is computed, and the process
stays alive. A path the protocol cannot carry (an embedded newline, or
one too long for a packet) falls back per path rather than costing the
command its process. The handshake keeps one fatal check: a process
that announces a capability Git did not request aborts the command, as
the long-running filter protocol does.
Consulting is allowed per command, following the allow_textconv
precedent. "git diff", "git log" and "git show", and "git blame" set
allow_diff_process; the plumbing diff commands and the interactive-patch
machinery never set it, so scripted and staging output stays builtin.
The options adjust the flag:
- --no-ext-diff clears it and --ext-diff sets it;
- --diff-process and --no-diff-process set and clear it alone, leaving
external diff drivers as they were;
- format-patch clears it unconditionally, so a generated patch applies
for recipients without the process;
- range-diff passes --no-ext-diff to the "git log" it compares.
git blame and the summary formats consult the process. For blame, a
pair reported equivalent emits no hunks, so the whole commit passes to
its parent. In the stat formats such a pair sums to a zero-count entry,
which the "nothing changed" rule omits, as under -w. The subprocess is
long-running: one startup cost across a traversal, one round-trip per
consulted pair. Answers travel in struct xdl_hunk, new in
xdiff-interface.h, holding xdiff's 1-based coordinates; nothing feeds
them back to xdiff, since only coordinate consumers consult.
A content-carrying request is the natural extension: it would serve
sides that are not stored blobs and processes that want content pushed
to them, and bring patch output and log -L's range tracking to the same
answer. As it stands, a process's answers show in blame and the summary
formats while patch output stays builtin.
t4080 exercises the protocol, the per-command gate, and the error paths:
- each adversarial response shape warns and falls back to builtin, the
request log proving which failures disable the process and which keep
it alive (a malformed hunk line, coordinates past the blob size, a
count overflowing strtol(), overlapping or misaligned hunks, an
unrecognized status, a bare status, an empty packet, a mid-response
crash, and raw garbage);
- a capability-less process and status=abort degrade without noise, and
a failed start warns once and returns the path to the store;
- a trailing token on a hunk line is ignored, pinning field
appendability;
- positive consults for git diff, git show, and diff-tree under
--ext-diff and --diff-process; textconv output and gitlink sides are
never identified; a diff.relative run consults by the repo-relative
path;
- the equivalence answer is pinned from both consumers, and a warming
run past a deferring process records the pair for a later read.
Helped-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Michael Montalbo <mmontalbo@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Before diffing a target blob against a parent, offer the pair's identity
to the hunk provider interface. Blame's requests have gone through
diff_provider_emit_hunks() since the interface arrived, but carried no
identity, so nothing could answer them. Now blame fills in the pair's
blob object ids and its diff options, and the chain serves the pair from
the store, keyed by the ids and the request's xdiff flags, before the
terminal provider falls back to fill-and-compute. Blame diffs at zero
context, which is not part of the key. An answer replays the recorded
hunks through blame_chunk_cb without loading either blob; a request
carrying -I patterns or anchors is outside the key and always computes.
Blame withholds the identity where its diff is not the plain blob-pair
diff the key describes: reverse blame, ignored revisions, textconv
paths, and the working-tree or --contents pseudo-commit, whose blob is
not a stored object. Those requests always compute. Whitespace and
algorithm options such as -w instead change blame's xdl_opts, so the
consult keys a different entry and misses a store warmed without them.
Blame's default xdl_opts now come from DIFF_HUNKS_DEFAULT_XDL_OPTS, new
here, which records the key-relevant defaults a diff_options-based
consumer already carries (today the indent heuristic), so a default
blame run and a default "log --stat" warming run share keys by
construction.
"--show-stats" reports how many pairs the store served and how many
consultations it could not, read from diff_hunks_read_stats(); the store
counts its own consultations, so blame keeps no tally.
Extend t4220 with the blame side:
- parity for plain, --porcelain, and --incremental output, and hit and
miss accounting across warming runs;
- the blame inputs that must bypass or miss the store: -w, indent
heuristics, --reverse, textconv, -M/-C, and the --ignore-rev pass;
- rename and merge handling, and --contents;
- reading a truncated or corrupt store as absent, and a crafted
zero-hunk record as a miss that verify flags.
Add p4218, measuring the cost of a warming run and the read speedups.
Signed-off-by: Michael Montalbo <mmontalbo@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach builtin_diffstat() to consult the hunk provider interface through
diff_provider_consult(), new here: the consult-only entry that answers
without loading content or computing, so it never returns
DIFF_PROVIDER_ERROR. On an answer, the summing callback accumulates the
provided counts directly into the diffstat entry; the blobs were already
loaded for the binary check, so an answer saves the diff run, not the
content load (blame, taught next, skips its loads too). On an
unanswered outcome it computes as before and, with a writer attached,
records what it computed; on unanswered-no-record it computes without
recording.
The provider behind the consult is the diff-hunks store, registered in
front of the terminal builtin computation. Its consult serves a
recorded pair through diff_hunks_replay(), which validates the sequence
before any hunk reaches the callback, so direct accumulation is safe.
The request gains the pair's object ids and the diff options read by the
exclusions below. A side whose bytes are not a stored blob, such as a
working-tree file or a gitlink, has a NULL id; the store passes it by and
the terminal provider computes it. diff_provider_emit_hunks() walks the
same chain, so blame's requests follow these rules the moment blame
supplies identity. The walk also insists, as a BUG check, that a
request's diff options belong to the repository whose chain it walks.
Each exclusion lives with the provider whose key cannot express it. -I
patterns and --anchored shape the diff outside the store key, and break
detection (-B) rescores the pair outside it; the store's consult maps
all three to stop-no-record, so such a request is neither served nor
recorded for any consumer. The consumer-side guard the recording commit
carried for those three comes out here. The compile-time assert on
xpparam_t's layout sits next to that decision, forcing an explicit
keying decision whenever a diff parameter is added. The stat consumer
keeps only the exclusion that is not about the key: --ignore-blank-lines
is part of the key but coalesces hunks differently between the
text-emitting and coordinate-callback paths, so the consumer returns
before consulting. A "log -L" range-scoped stat neither reads nor
records; the line-range filter computes it as before.
"git diff", "git log", "git show", and "git diff-tree" with the --stat,
--numstat, and --shortstat formats consult the interface. Reading is
controlled by core.diffHunks.
An answer is invisible in the output, so the store counts the pairs it
serves and the consultations it cannot, and diff_hunks_read_stats()
reports both; the stat path emits the hits as a trace2 "read-hits" datum
for tests and tuning. The counters live on the store because only the
store knows whether a consultation reached it, and none of its exclusion
legs reaches the replay, so none counts as a miss.
Extend t4220 with the read half:
- output parity with and without the store, at several context lengths
and both directions, and reversed pairs keying apart;
- the consultation made visible through the read-hits datum, and the
trim-divergent pair correct at every context;
- the settings that must bypass the store doing so in both directions
(-I, -B, --anchored, --ignore-blank-lines), asserted through the trace
rather than output parity alone, which a coincidentally equal count
could satisfy;
- a driver-forced algorithm keying apart rather than bypassing: it is
part of the key, so a read under it misses the default entries and a
warm records under its own.
A "log -L" range-scoped stat neither reads nor records.
Signed-off-by: Michael Montalbo <mmontalbo@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The diff-hunks store has a writer, but nothing fills it. Teach
builtin_diffstat() to do so: on a warming run (a writer is attached), a
modified pair's stat is produced by collecting the pair's hunk
coordinates instead of emitting text, the counts are summed from those
hunks, and the pair is recorded. A run without a writer is unchanged,
and nothing reads the store yet; the read side arrives next.
The store records one context-free entry per pair, and only for a
trim-stable pair: one whose zero-context trimmed diff (what blame will
read) and untrimmed diff (whose counts a nonzero-context stat matches)
are identical. The warming path computes both and hands them to
diff_hunks_writer_record_stable(), new here, which records only when
they agree; a divergent pair is never recorded and every consumer
computes it. The warming run displays the counts it shows a store-less
run: the trimmed ones, since xdi_diff trims at zero context, while the
untrimmed counts serve only the stability comparison.
Not everything the stat path computes may be recorded.
--ignore-blank-lines is part of the key, but it coalesces hunks
differently between the text-emitting and coordinate-callback paths, so
a recorded entry would not match a store-less run's --stat. -I
patterns, --anchored, and break detection (-B) shape the diff outside
the key entirely; the guard for those three sits in this consumer for
now and moves into the store's own provider when it registers, next. A
"log -L" range-scoped stat is not the whole-pair diff the key describes,
so it does not record. Recording also requires both sides to be valid
regular files whose blobs the key can name: a working-tree side,
textconv output, or a gitlink has no usable id.
"git diff", "git log", "git show", and "git diff-tree" with the --stat,
--numstat, and --shortstat formats attach a writer when writing is
enabled and flush it when the traversal finishes, so a warming run such
as
GIT_DIFF_HUNKS_WRITE=1 git log --all --stat >/dev/null
fills the cache as a side effect of the diff work the command already
does. Writing is controlled by diffHunks.write and GIT_DIFF_HUNKS_WRITE.
Add the write half of t4220:
- ordinary commands never create the store, and creation is gated off
by default, the environment overriding the config;
- a warming run builds a store that verifies, and a second refreshes it
in place;
- a warming run displays parity at zero context on a trim-divergent
pair, committed as a fixture (small synthetic pairs cannot diverge:
minimal diffs add and delete equal counts, and trimming preserves
that);
- binary and mode-only pairs do not break the writer;
- a corrupt store is discarded at seed;
- verify and clear run against the files a warming run builds.
Signed-off-by: Michael Montalbo <mmontalbo@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Passing a negative value to -U is silently accepted and produces
corrupt unified diff output with malformed hunk headers:
$ git log -1 -p -U-500 -- GIT-VERSION-GEN | grep '^@@'
@@ -503,999- +503,999- @@
Line 503 of a 106-line file, count "999-" is not a valid integer.
The config variable diff.context already rejects negative values, but
the command line callback diff_opt_unified() uses strtol() with no
range check.
Change the type of diff_options.context and its static default from
int to unsigned int, matching the change to interhunkcontext in the
previous commit. The type change requires reworking the callback and
config parsing to validate in a local variable before assigning to
the now-unsigned field.
Unlike --inter-hunk-context which could be converted to OPT_UNSIGNED,
-U needs OPT_CALLBACK_F for PARSE_OPT_OPTARG (bare -U with no value
enables patch output). Add a range check in the callback instead.
Signed-off-by: Michael Montalbo <mmontalbo@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Negative values for --inter-hunk-context produce structurally invalid
diff output with overlapping hunks:
$ git log -1 -p -U3 --inter-hunk-context=-100 791aeddfa2 \
-- git-compat-util.h | grep '^@@'
@@ -110,6 +110,9 @@
@@ -115,6 +118,9 @@
@@ -116,6 +122,7 @@
Hunk 1 covers lines 110-115, hunk 2 starts at 115 (overlap), hunk 3
starts at 116 (overlaps both). The resulting patch cannot be applied.
The config variable diff.interHunkContext already rejects negative
values, but the command line option does not.
Change the type of diff_options.interhunkcontext and its static
default from int to unsigned int, and switch the option parser from
OPT_INTEGER_F to OPT_UNSIGNED. This rejects negative values at parse
time via git_parse_unsigned() and enforces the correct type at compile
time via BARF_UNLESS_UNSIGNED.
Signed-off-by: Michael Montalbo <mmontalbo@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Halve the memory consumed by artificial filepairs created during
"git diff --find-copioes-harder", also making the operation run
faster.
* rs/diff-index-find-copies-harder-optim:
diff-index: don't queue unchanged filepairs with diff_change()
Both "git apply" and "git diff" learn a new whitespace error class,
"incomplete-line".
* jc/whitespace-incomplete-line:
attr: enable incomplete-line whitespace error for this project
diff: highlight and error out on incomplete lines
apply: check and fix incomplete lines
whitespace: allocate a few more bits and define WS_INCOMPLETE_LINE
apply: revamp the parsing of incomplete lines
diff: update the way rewrite diff handles incomplete lines
diff: call emit_callback ecbdata everywhere
diff: refactor output of incomplete line
diff: keep track of the type of the last line seen
diff: correct suppress_blank_empty hack
diff: emit_line_ws_markup() if/else style fix
whitespace: correct bit assignment comments
diff_cache() queues unchanged filepairs if the flag find_copies_harder
is set, and uses diff_change() for that. This function allocates a
filespec for each side, does a few other things that are unnecessary for
unchanged filepairs and always sets the diff_flag has_changes, which is
simply misleading in this case.
Add a new streamlined function for queuing unchanged filepairs and
use it in show_modified(), which is called by diff_cache() via
oneway_diff() and do_oneway_diff(). It allocates only a single filespec
for each filepair and uses it twice with reference counting. This has a
measurable effect if there are a lot of them, like in the Linux repo:
Benchmark 1: ./git_v2.52.0 -C ../linux diff --cached --find-copies-harder
Time (mean ± σ): 31.8 ms ± 0.2 ms [User: 24.2 ms, System: 6.3 ms]
Range (min … max): 31.5 ms … 32.3 ms 85 runs
Benchmark 2: ./git -C ../linux diff --cached --find-copies-harder
Time (mean ± σ): 23.9 ms ± 0.2 ms [User: 18.1 ms, System: 4.6 ms]
Range (min … max): 23.5 ms … 24.4 ms 111 runs
Summary
./git -C ../linux diff --cached --find-copies-harder ran
1.33 ± 0.01 times faster than ./git_v2.52.0 -C ../linux diff --cached --find-copies-harder
Signed-off-by: René Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Reserve a few more bits in the diff flags word to be used for future
whitespace rules. Add WS_INCOMPLETE_LINE without implementing the
behaviour (yet).
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A comment in diff.c claimed that bits up to 12th (counting from 0th)
are whitespace rules, and 13th thru 15th are for new/old/context,
but it turns out it was miscounting. Correct them, and clarify
where the whitespace rule bits come from in the comment. Extend bit
assignment comments to cover bits used for color-moved, which
weren't described.
Also update the way these bit constants are defined to use (1 << N)
notation, instead of octal constants, as it tends to make it easier
to notice a breakage like this.
Sprinkle a few blank lines between logically distinct groups of CPP
macro definitions to make them easier to read.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We introduced a dry_run flag to diff_options in b55e6d36eb (diff: ensure
consistent diff behavior with ignore options, 2025-08-08), with the idea
that the lower-level diff code could skip output when it is set.
As we saw with the bugs fixed by 3ed5d8bd73 (diff: stop output garbled
message in dry run mode, 2025-10-20), it is easy to miss spots. In the
end, we located all of them by checking where diff_options.file is used.
That suggests another possible approach: we can replace the dry_run
boolean with a NULL pointer for "file", as we know that using "file" in
dry_run mode would always be an error. This turns any missed spots from
producing extra output[1] into a segfault. Which is less forgiving, but
that is the point: this is indicative of a programming error, and
complaining loudly and immediately is good.
[1] We protect ourselves against garbled output as a separate step,
courtesy of 623f7af284 (diff: restore redirection to /dev/null for
diff_from_contents, 2025-10-17). So in that sense this patch can
only introduce user-visible errors (since any "bugs" were going to
/dev/null before), but the idea is to catch them rather than quietly
send garbage to /dev/null.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git last-modified" operating in non-recursive mode used to trigger
a BUG(), which has been corrected.
* tc/last-modified-recursive-fix:
last-modified: fix bug when some paths remain unhandled
Some places in the code confused a variable that is *not* a boolean
to enable color but is an enum that records what the user requested
to do about color. A couple of bugs of this sort have been fixed,
while the code has been cleaned up to prevent similar bugs in the
future.
* jk/color-variable-fixes:
config: store want_color() result in a separate bool
add-interactive: retain colorbool values longer
color: return bool from want_color()
color: use git_colorbool enum type to store colorbools
pretty: use format_commit_context.auto_color as colorbool
diff: stop passing ecbdata->use_color as boolean
diff: pass o->use_color directly to fill_metainfo()
diff: don't use diff_options.use_color as a strict bool
diff: simplify color_moved check when flushing
grep: don't treat grep_opt.color as a strict bool
color: return enum from git_config_colorbool()
color: use GIT_COLOR_* instead of numeric constants
The recently introduced new subcommand git-last-modified(1) runs into an
error in some scenarios. It then would exit with the message:
BUG: paths remaining beyond boundary in last-modified
This seems to happens for example when criss-cross merges are involved.
In that scenario, the function diff_tree_combined() gets called.
The function diff_tree_combined() copies the `struct diff_options` from
the input `struct rev_info` to override some flags. One flag is
`recursive`, which is always set to 1. This has been the case since the
inception of this function in af3feefa1d (diff-tree -c: show a merge
commit a bit more sensibly., 2006-01-24).
This behavior is incompatible with git-last-modified(1), when called
non-recursive (which is the default).
The last-modified machinery uses a hashmap for all the paths it wants to
get the last-modified commit for. Through log_tree_commit() the callback
mark_path() is called. The diff machinery uses diff_tree_combined()
internally, and due to it's recursive behavior the callback receives
entries inside subtrees, but not the subtree entries themselves. So a
directory is never expelled from the hashmap, and the BUG() statement
gets hit.
Because there are many callers calling into diff_tree_combined(), both
directly and indirectly, we cannot simply change it's behavior.
Instead, add a flag `no_recursive_diff_tree_combined` which supresses
the behavior of diff_tree_combined() to override `recursive` and set
this flag in builtin/last-modified.c.
Signed-off-by: Toon Claes <toon@iotcl.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We traditionally used "int" to store and pass around the values defined
by "enum git_colorbool" (which were originally just #define macros).
Using an int doesn't produce incorrect results, but using the actual
enum makes the intent of the code more clear.
It would be nice if the compiler could catch cases where we used the
enum and an int interchangeably, since it's very easy to accidentally
check the boolean true/false of a colorbool like:
if (branch_use_color)
This is wrong because GIT_COLOR_UNKNOWN and GIT_COLOR_AUTO evaluate to
true in C, even though we may ultimately decide not to use color. But C
is pretty happy to convert between ints and enums (even with various
-Wenum-* warnings). So this sadly doesn't protect us from such mistakes,
but it hopefully does make the code easier to read.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In git-diff, options like `-w` and `-I<regex>`, two files are considered
equivalent under the specified "ignore" rules, even when they are not
bit-for-bit identical. For options like `--raw`, `--name-status`,
and `--name-only`, git-diff deliberately compares only the SHA values
to determine whether two files are equivalent, for performance reasons.
As a result, a file shown in `git diff --name-status` may not appear
in `git diff --patch`.
To quickly determine whether two files are equivalent, add a helper
function diff_flush_patch_quietly() in diff.c. Add `.dry_run` field in
`struct diff_options`. When `.dry_run` is true, builtin_diff() returns
immediately upon finding any change. Call diff_flush_patch_quietly()
to determine if we should flush `--raw`, `--name-only` or `--name-status`
output.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Lidong Yan <yldhome2d2@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When you are doing a tree-diff, there are basically two options: do not
recurse into subtrees at all, or recurse indefinitely. While most
callers would want to always recurse and see full pathnames, some may
want the efficiency of looking only at a particular level of the tree.
This is currently easy to do for the top-level (just turn off
recursion), but you cannot say "show me what changed in subdir/, but do
not recurse".
This patch adds a max-depth parameter which is measured from the closest
pathspec match, so that you can do:
git log --raw --max-depth=1 -- a/b/c
and see the raw output for a/b/c/, but not those of a/b/c/d/
(instead of the raw output you would see for a/b/c/d).
Co-authored-by: Toon Claes <toon@iotcl.com>
Signed-off-by: Toon Claes <toon@iotcl.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The object layer has been updated to take an explicit repository
instance as a parameter in more code paths.
* ps/object-wo-the-repository:
hash: stop depending on `the_repository` in `null_oid()`
hash: fix "-Wsign-compare" warnings
object-file: split out logic regarding hash algorithms
delta-islands: stop depending on `the_repository`
object-file-convert: stop depending on `the_repository`
pack-bitmap-write: stop depending on `the_repository`
pack-revindex: stop depending on `the_repository`
pack-check: stop depending on `the_repository`
environment: move access to "core.bigFileThreshold" into repo settings
pack-write: stop depending on `the_repository` and `the_hash_algo`
object: stop depending on `the_repository`
csum-file: stop depending on `the_repository`
The `null_oid()` function returns the object ID that only consists of
zeroes. Naturally, this ID also depends on the hash algorithm used, as
the number of zeroes is different between SHA1 and SHA256. Consequently,
the function returns the hash-algorithm-specific null object ID.
This is currently done by depending on `the_hash_algo`, which implicitly
makes us depend on `the_repository`. Refactor the function to instead
pass in the hash algorithm for which we want to retrieve the null object
ID. Adapt callsites accordingly by passing in `the_repository`, thus
bubbling up the dependency on that global variable by one layer.
There are a couple of trivial exceptions for subsystems that already got
rid of `the_repository`. These subsystems instead use the repository
that is available via the calling context:
- "builtin/grep.c"
- "grep.c"
- "refs/debug.c"
There are also two non-trivial exceptions:
- "diff-no-index.c": Here we know that we may not have a repository
initialized at all, so we cannot rely on `the_repository`. Instead,
we adapt `diff_no_index()` to get a `struct git_hash_algo` as
parameter. The only caller is located in "builtin/diff.c", where we
know to call `repo_set_hash_algo()` in case we're running outside of
a Git repository. Consequently, it is fine to continue passing
`the_repository->hash_algo` even in this case.
- "builtin/ls-files.c": There is an in-flight patch series that drops
`USE_THE_REPOSITORY_VARIABLE` in this file, which causes a semantic
conflict because we use `null_oid()` in `show_submodule()`. The
value is passed to `repo_submodule_init()`, which may use the object
ID to resolve a tree-ish in the superproject from which we want to
read the submodule config. As such, the object ID should refer to an
object in the superproject, and consequently we need to use its hash
algorithm.
This means that we could in theory just not bother about this edge
case at all and just use `the_repository` in "diff-no-index.c". But
doing so would feel misdesigned.
Remove the `USE_THE_REPOSITORY_VARIABLE` preprocessor define in
"hash.c".
Signed-off-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
By default, `diffcore_std()` resolves the statuses for queued diff file
pairs by calling `diff_resolve_rename_copy()`. If status information is
already manually set, invoking `diffcore_std()` may change the status
value.
Introduce the `skip_resolving_statuses` diff option that prevents
`diffcore_std()` from resolving file pair statuses when enabled.
Signed-off-by: Justin Tobler <jltobler@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The `diff_addremove()` and `diff_change()` functions set up and queue
diffs, but do not return the `diff_filepair` added to the queue. In a
subsequent commit, modifications to `diff_filepair` need to occur in
certain cases after being queued.
Since the existing `diff_addremove()` and `diff_change()` are also used
for callbacks in `diff_options` as types `add_remove_fn_t` and
`change_fn_t`, modifying the existing function signatures requires
further changes. The diff options for pruning use `file_add_remove()`
and `file_change()` where file pairs do not even get queued. Thus,
separate functions are implemented instead.
Split out the queuing operations into `diff_queue_addremove()` and
`diff_queue_change()` which also return a handle to the queued
`diff_filepair`. Both `diff_addremove()` and `diff_change()` are
reimplemented as thin wrappers around the new functions.
Signed-off-by: Justin Tobler <jltobler@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Further code clean-up on the use of hash functions. Now the
context object knows what hash function it is working with.
* ps/hash-cleanup:
global: adapt callers to use generic hash context helpers
hash: provide generic wrappers to update hash contexts
hash: stop typedeffing the hash context
hash: convert hashing context to a structure
We generally avoid using `typedef` in the Git codebase. One exception
though is the `git_hash_ctx`, likely because it used to be a union
rather than a struct until the preceding commit refactored it. But now
that it is a normal `struct` there isn't really a need for a typedef
anymore.
Drop the typedef and adapt all callers accordingly.
Signed-off-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The internals of the path diffing code, including ll_diff_tree_paths(),
all take an extra combine_diff_path parameter which they use as the tail
of a list of results, appending any new entries to it.
The public-facing diff_tree_paths() takes the same argument, but it just
makes the callers more awkward. They always start with a clean list, and
have to set up a fake head struct to pass in.
Let's keep the public API clean by always returning a new list. That
keeps the fake struct as an implementation detail of tree-diff.c.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We want callers to use combine_diff_path_new() to allocate structs,
rather than using combine_diff_path_size() and xmalloc(). That gives us
more consistency over the initialization of the fields.
Now that the final external user of combine_diff_path_size() is gone, we
can stop declaring it publicly. And since our constructor is the only
caller, we can just inline it there.
Breaking the size computation into two parts also lets us reuse the
intermediate multiplication result of the parent length, since we need
to know it to perform our memset(). The result is a little easier to
read.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We only fill in the per-parent "path" field when it differs from what's
in combine_diff_path.path (and even then only when the option is
appropriate). Let's document that.
Suggested-by: Wink Saville <wink@saville.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Commit d76ce4f734 (log,diff-tree: add --combined-all-paths option,
2019-02-07) added a "path" field to each combine_diff_parent struct.
It's defined as a strbuf, but this is overkill. We never manipulate the
buffer beyond inserting a single string into it.
And in fact there's a small bug: we zero the parent structs, including
the path strbufs. For the 0th parent, we strbuf_init() the strbuf before
adding to it. But for subsequent parents, we never do the init. This is
technically violating the strbuf API, though the code there is resilient
enough to handle this zero'd state.
This patch switches us to just store an allocated string pointer.
Zeroing it is enough to properly initialize it there (modulo the usual
assumption we make that a NULL pointer is all-zeroes).
And as a bonus, we can just check for a non-NULL value to see if it is
present, rather than repeating the combined_all_paths logic at each
site.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The combine_diff_path struct has variable size, since it embeds both the
memory allocation for the path field as well as a variable-sized parent
array. This makes allocating one a bit tricky.
We have a helper to compute the required size, but it's up to individual
sites to actually initialize all of the fields. Let's provide a
constructor function to make that a little nicer. Besides being shorter,
it also hides away tricky bits like the computation of the "path"
pointer (which is right after the "parent" flex array).
As a bonus, using the same constructor everywhere means that we'll
consistently initialize all parts of the struct. A few code paths left
the parent array unitialized. This didn't cause any bugs, but we'll be
able to simplify some code in the next few patches knowing that the
parent fields have all been zero'd.
This also gets rid of some questionable uses of "int" to store buffer
lengths. Though we do use them to allocate, I don't think there are any
integer overflow vulnerabilities here (the allocation helper promotes
them to size_t and checks arithmetic for overflow, and the actual memcpy
of the bytes is done using the possibly-truncated "int" value).
Sadly we can't use the FLEX_* macros to simplify the allocation here,
because there are two variable-sized parts to the struct (and those
macros only handle one).
Nor can we get stop publicly declaring combine_diff_path_size(). This
patch does not touch the code in path_appendnew() at all, which is not
ready to be moved to our new constructor for a few reasons:
- path_appendnew() has a memory-reuse optimization where it tries to
reuse combine_diff_path structs rather than freeing and
reallocating.
- path_appendnew() does not create the struct from a single path
string, but rather allocates and copies into the buffer from
multiple sources.
These can be addressed by some refactoring, but let's leave it as-is for
now.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The `struct diff_flags` structure is essentially an array of flags, all
of which have the same type. We can thus use `sizeof()` to iterate
through all of the flags, which we do in `diff_flags_or()`. But while
the statement returns an unsigned integer, we used a signed integer to
iterate through the flags, which generates a warning.
Fix this by using `size_t` for the index instead.
Signed-off-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Code clean-up.
* jk/output-prefix-cleanup:
diff: store graph prefix buf in git_graph struct
diff: return line_prefix directly when possible
diff: return const char from output_prefix callback
diff: drop line_prefix_length field
line-log: use diff_line_prefix() instead of custom helper
The diff_options structure has an output_prefix callback for returning a
prefix string, but it does so by returning a pointer to a strbuf.
This makes the interface awkward. There's no reason the callback should
need to use a strbuf, and it creates questions about whether the
ownership of the resulting buffer should be transferred to the caller
(it should not be, but a recent attempt to clean up this code led to a
double-free in some cases).
The one advantage we get is that the strbuf contains a ptr/len pair, so
we could in theory have a prefix with embedded NULs. But we can observe
that none of the existing callbacks would ever produce such a NUL (they
are usually just indentation or graph symbols, and even the
"--line-prefix" option takes a NUL-terminated string).
And anyway, only one caller (the one in log_tree_diff_flush) actually
looks at the strbuf length. In every other case we use a helper function
which discards the length and just returns the NUL-terminated string.
So let's just have the callback return a "const char *" pointer. It's up
to the callbacks themselves if they want to use a strbuf under the hood.
And now the caller in log_tree_diff_flush() can just use the helper
function along with everybody else. That lets us even simplify out the
function pointer check, since the helper returns an empty string
(technically this does mean we'll sometimes issue an empty fputs() call,
but I don't think this code path is hot enough to care about that).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The diff_options structure holds a line_prefix string and an associated
length. But the length is always just the strlen() of the NUL-terminated
string. Let's simplify the code by just storing the string pointer and
assuming it is NUL-terminated when we use it.
This will cause us to compute the string length in a few extra spots,
but I don't think any of these are particularly hot code paths.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The `orderfile` diff option is being assigned via `OPT_FILENAME()`,
which assigns an allocated string to the variable. We never free it
though, causing a memory leak.
Change the type of the string to `char *` and free it to plug the leak.
This also requires us to use `xstrdup()` to assign the global config to
it in case it is set.
This leak is being hit in t7621, but plugging it alone does not make the
test suite pass.
Signed-off-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Code clean-up.
* jc/range-diff-lazy-setup:
remerge-diff: clean up temporary objdir at a central place
remerge-diff: lazily prepare temporary objdir on demand
After running a diff between two things, or a series of diffs while
walking the history, the diff computation is concluded by a call to
diff_result_code() to extract the exit status of the diff machinery.
The function can work on "struct diffopt", but all the callers
historically and currently pass "struct diffopt" that is embedded in
the "struct rev_info" that is used to hold the remerge_diff bit and
the remerge_objdir variable that points at the temporary object
directory in use.
Redefine diff_result_code() to take the whole "struct rev_info" to
give it an access to these members related to remerge-diff, so that
it can get rid of the temporary object directory for any and all
callers that used the feature. We can lose the equivalent code to
do so from the code paths for individual commands, diff-tree, diff,
and log.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "hash-ll.h" header was introduced via d1cbe1e6d8 (hash-ll.h: split
out of hash.h to remove dependency on repository.h, 2023-04-22) to make
explicit the split between hash-related functions that rely on the
global `the_repository`, and those that don't. This split is no longer
necessary now that we we have removed the reliance on `the_repository`.
Merge "hash-ll.h" back into "hash.h". This causes some code units to not
include "repository.h" anymore, which requires us to add some forward
declarations.
Signed-off-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Extract the commonly used initialization of the --stat-width=<width>,
--stat-name-width=<width> and --stat-graph-with=<width> parameters to their
internal default values into a helper function, to avoid repeating the same
initialization code in a few places.
Add a couple of tests to additionally cover existing configuration options
diff.statNameWidth=<width> and diff.statGraphWidth=<width> when used by
git-merge to generate --stat outputs. This closes the gap that existed
previously in the --stat tests, and reduces the chances for having any
regressions introduced by this commit.
While there, perform a small bunch of minor wording tweaks in the improved
unit test, to improve its test-level consistency a bit.
Signed-off-by: Dragan Simic <dsimic@manjaro.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Many programs use diff_result_code() to get a user-visible program exit
code from a diff result (e.g., checking opts.found_changes if
--exit-code was requested).
This function also takes a "status" parameter, which seems at first
glance that it could be used to propagate an error encountered when
computing the diff. But it doesn't work that way:
- negative values are passed through as-is, but are not appropriate as
program exit codes
- when --exit-code or --check is in effect, we _ignore_ the passed-in
status completely. So a failed diff which did not have a chance to
set opts.found_changes would erroneously report "success, no
changes" instead of propagating the error.
After recent cleanups, neither of these bugs is possible to trigger, as
every caller just passes in "0". So rather than fixing them, we can
simply drop the useless parameter instead.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Neither of these functions ever returns a value other than zero.
Instead, they expect unrecoverable errors to exit immediately, and
things like "--exit-code" are stored inside the diff_options struct to
be handled later via diff_result_code().
Some callers do check the return values, but many don't bother. Let's
drop the useless return values, which are misleading callers about how
the functions work. This could be seen as a step in the wrong direction,
as we might want to eventually "lib-ify" these to more cleanly return
errors up the stack, in which case we'd have to add the return values
back in. But there are some benefits to doing this now:
1. In the current code, somebody could accidentally add a "return -1"
to one of the functions, which would be erroneously ignored by many
callers. By removing the return code, the compiler can notice the
mismatch and force the developer to decide what to do.
Obviously the other option here is that we could start consistently
checking the error code in every caller. But it would be dead code,
and we wouldn't get any compile-time help in catching new cases.
2. It communicates the situation to callers, who may want to choose a
different function. These functions are really thin wrappers for
doing git-diff-files and git-diff-index within the process. But
callers who care about recovering from an error here are probably
better off using the underlying library functions, many of
which do return errors.
If somebody eventually wants to teach these functions to propagate
errors, they'll have to switch back to returning a value, effectively
reverting this patch. But at least then they will be starting with a
level playing field: they know that they will need to inspect each
caller to see how it should handle the error.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Reduce reliance on a global state in the config reading API.
* gc/config-context:
config: pass source to config_parser_event_fn_t
config: add kvi.path, use it to evaluate includes
config.c: remove config_reader from configsets
config: pass kvi to die_bad_number()
trace2: plumb config kvi
config.c: pass ctx with CLI config
config: pass ctx with config files
config.c: pass ctx in configsets
config: add ctx arg to config_fn_t
urlmatch.h: use config_fn_t type
config: inline git_color_default_config
Add a new "const struct config_context *ctx" arg to config_fn_t to hold
additional information about the config iteration operation.
config_context has a "struct key_value_info kvi" member that holds
metadata about the config source being read (e.g. what kind of config
source it is, the filename, etc). In this series, we're only interested
in .kvi, so we could have just used "struct key_value_info" as an arg,
but config_context makes it possible to add/adjust members in the future
without changing the config_fn_t signature. We could also consider other
ways of organizing the args (e.g. moving the config name and value into
config_context or key_value_info), but in my experiments, the
incremental benefit doesn't justify the added complexity (e.g. a
config_fn_t will sometimes invoke another config_fn_t but with a
different config value).
In subsequent commits, the .kvi member will replace the global "struct
config_reader" in config.c, making config iteration a global-free
operation. It requires much more work for the machinery to provide
meaningful values of .kvi, so for now, merely change the signature and
call sites, pass NULL as a placeholder value, and don't rely on the arg
in any meaningful way.
Most of the changes are performed by
contrib/coccinelle/config_fn_ctx.pending.cocci, which, for every
config_fn_t:
- Modifies the signature to accept "const struct config_context *ctx"
- Passes "ctx" to any inner config_fn_t, if needed
- Adds UNUSED attributes to "ctx", if needed
Most config_fn_t instances are easily identified by seeing if they are
called by the various config functions. Most of the remaining ones are
manually named in the .cocci patch. Manual cleanups are still needed,
but the majority of it is trivial; it's either adjusting config_fn_t
that the .cocci patch didn't catch, or adding forward declarations of
"struct config_context ctx" to make the signatures make sense.
The non-trivial changes are in cases where we are invoking a config_fn_t
outside of config machinery, and we now need to decide what value of
"ctx" to pass. These cases are:
- trace2/tr2_cfg.c:tr2_cfg_set_fl()
This is indirectly called by git_config_set() so that the trace2
machinery can notice the new config values and update its settings
using the tr2 config parsing function, i.e. tr2_cfg_cb().
- builtin/checkout.c:checkout_main()
This calls git_xmerge_config() as a shorthand for parsing a CLI arg.
This might be worth refactoring away in the future, since
git_xmerge_config() can call git_default_config(), which can do much
more than just parsing.
Handle them by creating a KVI_INIT macro that initializes "struct
key_value_info" to a reasonable default, and use that to construct the
"ctx" arg.
Signed-off-by: Glen Choo <chooglen@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This also made it clear that several .c files depended upon various
things that oidset included, but had omitted the direct #include for
those headers. Add those now.
Signed-off-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git [-c log.follow=true] log [--follow] ':(glob)f**'" used to barf.
* jk/log-follow-with-non-literal-pathspec:
diff: detect pathspec magic not supported by --follow
diff: factor out --follow pathspec check
pathspec: factor out magic-to-name function
In --follow mode, we require exactly one pathspec. We check this
condition in two places:
- in diff_setup_done(), we complain if --follow is used with an
inapropriate pathspec
- in git-log's revision "tweak" function, we enable log.follow only if
the pathspec allows it
The duplication isn't a big deal right now, since the logic is so
simple. But in preparation for it becoming more complex, let's pull it
into a shared function.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Michael Montalbo