In subsequent commits, it will be convenient to have a 'const char **'
of hidden refs (matching `transfer.hiderefs`, `uploadpack.hideRefs`,
etc.), instead of a `string_list`.
Convert spots throughout the tree that store the list of hidden refs
from a `string_list` to a `strvec`.
Note that in `parse_hide_refs_config()` there is an ugly const-cast used
to avoid an extra copy of each value before trimming any trailing slash
characters. This could instead be written as:
ref = xstrdup(value);
len = strlen(ref);
while (len && ref[len - 1] == '/')
ref[--len] = '\0';
strvec_push(hide_refs, ref);
free(ref);
but the double-copy (once when calling `xstrdup()`, and another via
`strvec_push()`) is wasteful.
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The previous commit added low-level tests to ensure that the packed-refs
iterator did not enumerate excluded sections of the refspace.
However, there was no guarantee that these sections weren't being
visited, only that they were being suppressed from the output. To harden
these tests, add a trace2 counter which tracks the number of regions
skipped by the packed-refs iterator, and assert on its value.
Suggested-by: Derrick Stolee <derrickstolee@github.com>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When iterating through the `packed-refs` file in order to answer a query
like:
$ git for-each-ref --exclude=refs/__hidden__
it would be useful to avoid walking over all of the entries in
`refs/__hidden__/*` when possible, since we know that the ref-filter
code is going to throw them away anyways.
In certain circumstances, doing so is possible. The algorithm for doing
so is as follows:
- For each excluded pattern, find the first record that matches it,
and the first record that *doesn't* match it (i.e. the location
you'd next want to consider when excluding that pattern).
- Sort the set of excluded regions from the previous step in ascending
order of the first location within the `packed-refs` file that
matches.
- Clean up the results from the previous step: discard empty regions,
and combine adjacent regions. The set of regions which remains is
referred to as the "jump list", and never contains any references
which should be included in the result set.
Then when iterating through the `packed-refs` file, if `iter->pos` is
ever contained in one of the regions from the previous steps, advance
`iter->pos` past the end of that region, and continue enumeration.
Note that we only perform this optimization when none of the excluded
pattern(s) have special meta-characters in them. For a pattern like
"refs/foo[ac]", the excluded regions ("refs/fooa", "refs/fooc", and
everything underneath them) are not connected. A future implementation
that handles this case may split the character class (pretending as if
two patterns were excluded: "refs/fooa", and "refs/fooc").
There are a few other gotchas worth considering. First, note that the
jump list is sorted, so once we jump past a region, we can avoid
considering it (or any regions preceding it) again. The member
`jump_pos` is used to track the first next-possible region to jump
through.
Second, note that the jump list is best-effort, since we do not handle
loose references, and because of the meta-character issue above. The
jump list may not skip past all references which won't appear in the
results, but will never skip over a reference which does appear in the
result set.
In repositories with a large number of hidden references, the speed-up
can be significant. Tests here are done with a copy of linux.git with a
reference "refs/pull/N" pointing at every commit, as in:
$ git rev-list HEAD | awk '{ print "create refs/pull/" NR " " $0 }' |
git update-ref --stdin
$ git pack-refs --all
, it is significantly faster to have `for-each-ref` jump over the
excluded references, as opposed to filtering them out after the fact:
$ hyperfine \
'git for-each-ref --format="%(objectname) %(refname)" | grep -vE "^[0-9a-f]{40} refs/pull/"' \
'git.prev for-each-ref --format="%(objectname) %(refname)" --exclude="refs/pull"' \
'git.compile for-each-ref --format="%(objectname) %(refname)" --exclude="refs/pull"'
Benchmark 1: git for-each-ref --format="%(objectname) %(refname)" | grep -vE "^[0-9a-f]{40} refs/pull/"
Time (mean ± σ): 798.1 ms ± 3.3 ms [User: 687.6 ms, System: 146.4 ms]
Range (min … max): 794.5 ms … 805.5 ms 10 runs
Benchmark 2: git.prev for-each-ref --format="%(objectname) %(refname)" --exclude="refs/pull"
Time (mean ± σ): 98.9 ms ± 1.4 ms [User: 93.1 ms, System: 5.7 ms]
Range (min … max): 97.0 ms … 104.0 ms 29 runs
Benchmark 3: git.compile for-each-ref --format="%(objectname) %(refname)" --exclude="refs/pull"
Time (mean ± σ): 4.5 ms ± 0.2 ms [User: 0.7 ms, System: 3.8 ms]
Range (min … max): 4.1 ms … 5.8 ms 524 runs
Summary
'git.compile for-each-ref --format="%(objectname) %(refname)" --exclude="refs/pull"' ran
21.87 ± 1.05 times faster than 'git.prev for-each-ref --format="%(objectname) %(refname)" --exclude="refs/pull"'
176.52 ± 8.19 times faster than 'git for-each-ref --format="%(objectname) %(refname)" | grep -vE "^[0-9a-f]{40} refs/pull/"'
(Comparing stock git and this patch isn't quite fair, since an earlier
commit in this series adds a naive implementation of the `--exclude`
option. `git.prev` is built from the previous commit and includes this
naive implementation).
Using the jump list is fairly straightforward (see the changes to
`refs/packed-backend.c::next_record()`), but constructing the list is
not. To ensure that the construction is correct, add a new suite of
tests in t1419 covering various corner cases (overlapping regions,
partially overlapping regions, adjacent regions, etc.).
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The function `find_reference_location()` is used to perform a
binary search-like function over the contents of a repository's
`$GIT_DIR/packed-refs` file.
The search it implements is unlike a standard binary search in that the
records it searches over are not of a fixed width, so the comparison
must locate the end of a record before comparing it.
Extract the core routine of `find_reference_location()` in order to
implement a function in the following patch which will find the first
location in the `packed-refs` file that *doesn't* match the given
pattern.
The behavior of `find_reference_location()` is unchanged.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The subsequent patch will want to access an optional `excluded_patterns`
array within `refs/packed-backend.c` that will cull out certain
references matching any of the given patterns on a best-effort basis.
To do so, the refs subsystem needs to be updated to pass this value
across a number of different locations.
Prepare for a future patch by introducing this plumbing now, passing
NULLs at top-level APIs in order to make that patch less noisy and more
easily readable.
Signed-off-by: Taylor Blau <me@ttaylorr.co>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When using `for-each-ref`, it is sometimes convenient for the caller to
be able to exclude certain parts of the references.
For example, if there are many `refs/__hidden__/*` references, the
caller may want to emit all references *except* the hidden ones.
Currently, the only way to do this is to post-process the output, like:
$ git for-each-ref --format='%(refname)' | grep -v '^refs/hidden/'
Which is do-able, but requires processing a potentially large quantity
of references.
Teach `git for-each-ref` a new `--exclude=<pattern>` option, which
excludes references from the results if they match one or more excluded
patterns.
This patch provides a naive implementation where the `ref_filter` still
sees all references (including ones that it will discard) and is left to
check whether each reference matches any excluded pattern(s) before
emitting them.
By culling out references we know the caller doesn't care about, we can
avoid allocating memory for their storage, as well as spending time
sorting the output (among other things). Even the naive implementation
provides a significant speed-up on a modified copy of linux.git (that
has a hidden ref pointing at each commit):
$ hyperfine \
'git.compile for-each-ref --format="%(objectname) %(refname)" | grep -vE "[0-9a-f]{40} refs/pull/"' \
'git.compile for-each-ref --format="%(objectname) %(refname)" --exclude refs/pull/'
Benchmark 1: git.compile for-each-ref --format="%(objectname) %(refname)" | grep -vE "[0-9a-f]{40} refs/pull/"
Time (mean ± σ): 820.1 ms ± 2.0 ms [User: 703.7 ms, System: 152.0 ms]
Range (min … max): 817.7 ms … 823.3 ms 10 runs
Benchmark 2: git.compile for-each-ref --format="%(objectname) %(refname)" --exclude refs/pull/
Time (mean ± σ): 106.6 ms ± 1.1 ms [User: 99.4 ms, System: 7.1 ms]
Range (min … max): 104.7 ms … 109.1 ms 27 runs
Summary
'git.compile for-each-ref --format="%(objectname) %(refname)" --exclude refs/pull/' ran
7.69 ± 0.08 times faster than 'git.compile for-each-ref --format="%(objectname) %(refname)" | grep -vE "[0-9a-f]{40} refs/pull/"'
Subsequent patches will improve on this by avoiding visiting excluded
sections of the `packed-refs` file in certain cases.
Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
`match_pattern()` and `match_name_as_path()` both take a `struct
ref_filter *`, and then store a stack variable `patterns` pointing at
`filter->patterns`.
The subsequent patch will add a new array of patterns to match over (the
excluded patterns, via a new `git for-each-ref --exclude` option),
treating the return value of these functions differently depending on
which patterns are being used to match.
Tweak `match_pattern()` and `match_name_as_path()` to take an array of
patterns to prepare for passing either in.
Once we start passing either in, `match_pattern()` will have little to
do with a particular `struct ref_filter *` instance. To clarify this,
drop it from the argument list, and replace it with the only bit of the
`ref_filter` that we care about (`filter->ignore_case`).
Co-authored-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We did not bother to clean up at all in `git branch` or `git tag`, and
`git for-each-ref` only cleans up a couple of members.
Add and call `ref_filter_clear()` when cleaning up a `struct
ref_filter`. Running this patch (without any test changes) indicates a
couple of now leak-free tests. This was found by running:
$ make SANITIZE=leak
$ make -C t GIT_TEST_PASSING_SANITIZE_LEAK=check GIT_TEST_OPTS=--immediate
(Note that the `reachable_from` and `unreachable_from` lists should be
cleaned as they are used. So this is just covering any case where we
might bail before running the reachability check.)
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In `reach_filter()`, we pop all commits from the reachable lists,
leaving them empty. But because we're operating on a list pointer that
was passed by value, the original `filter.reachable_from` and
`filter.unreachable_from` pointers are left dangling.
As is the case with the previous commit, nobody touches either of these
fields after calling `reach_filter()`, so leaving them dangling is OK.
But this future proofs against dangerous situations.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Provide a sane initialization value for `struct ref_filter`, which in a
subsequent patch will be used to initialize a new field.
In the meantime, ensure that the `ref_filter` struct used in the
test-helper's `cmd__reach()` is zero-initialized. The lack of
initialization is OK, since `commit_contains()` only looks at the single
`with_commit_tag_algo` field that *is* initialized directly above.
So this does not fix a bug, but rather prevents one from biting us in
the future.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The refs machinery has its own implementation of a `ref_filter` (used by
`for-each-ref`), which is distinct from the `ref-filter.h` API (also
used by `for-each-ref`, among other things).
Rename the one within refs.c to more clearly indicate its purpose.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* 'master' of github.com:git/git:
A few more topics after 2.41-rc1
Git 2.41-rc1
t/lib-httpd: make CGIPassAuth support conditional
t9001: mark the script as no longer leak checker clean
send-email: clear the $message_id after validation
upload-pack: advertise capabilities when cloning empty repos
A bit more before -rc1
imap-send: include strbuf.h
run-command.c: fix missing include under `NO_PTHREADS`
test: do not negate test_path_is_* to assert absense
t2021: do not negate test_path_is_dir
tests: do not negate test_path_exists
doc/git-config: add unit for http.lowSpeedLimit
rebase -r: fix the total number shown in the progress
rebase --update-refs: fix loops
attr: teach "--attr-source=<tree>" global option to "git"
A few bugs in the sequencer machinery that results in miscounting
the steps have been corrected.
* js/rebase-count-fixes:
rebase -r: fix the total number shown in the progress
rebase --update-refs: fix loops
Small fixes.
* jc/do-not-negate-test-helpers:
test: do not negate test_path_is_* to assert absense
t2021: do not negate test_path_is_dir
tests: do not negate test_path_exists
We started unconditionally testing with CGIPassAuth directive but
it is unavailable in older Apache that ships with CentOS 7 that has
about a year of shelf-life still left. The test has conditionally
been disabled when running with an ancient Apache. This was a fix
for a recent regression caught before the release, so no need to
mention it in the release notes.
* jk/http-test-cgipassauth-unavailable-in-older-apache:
t/lib-httpd: make CGIPassAuth support conditional
The server side of "git clone" now advertises the necessary hints
to clients to help them to clone from an empty repository and learn
object hash algorithm and the (unborn) branch pointed at by HEAD,
even over the older v0/v1 protocol.
* bc/clone-empty-repo-via-protocol-v0:
upload-pack: advertise capabilities when cloning empty repos
When "git send-email" that uses the validate hook is fed a message
without and then with Message-ID, it failed to auto-assign a unique
Message-ID to the former and instead reused the Message-ID from the
latter, which has been corrected. This was a fix for a recent
regression caught before the release, so no need to mention it in
the release notes.
* jc/send-email-pre-process-fix:
t9001: mark the script as no longer leak checker clean
send-email: clear the $message_id after validation
Fix the build problem with NO_PTHREADS defined, a fallout from
recent header file shuffling.
* tb/run-command-needs-alloc-h:
run-command.c: fix missing include under `NO_PTHREADS`
Commit 988aad99b4 (t5563: add tests for basic and anoymous HTTP access,
2023-02-27) added tests that require Apache to support the CGIPassAuth
directive, which was added in Apache 2.4.13. This is fairly old (~8
years), but recent enough that we still encounter it in the wild (e.g.,
RHEL/CentOS 7, which is not EOL until June 2024).
We can live with skipping the new tests on such a platform. But
unfortunately, since the directive is used unconditionally in our
apache.conf, it means the web server fails to start entirely, and we
cannot run other HTTP tests at all (e.g., the basic ones in t5551).
We can fix that by making the config conditional, and only triggering it
for t5563. That solves the problem for t5551 (which then ignores the
directive entirely). For t5563, we'd see apache complain in start_httpd;
with the default setting of GIT_TEST_HTTPD, we'd then skip the whole
script.
But that leaves one small problem: people may set GIT_TEST_HTTPD=1
explicitly, which instructs the tests to fail (rather than skip) when we
can't start the webserver (to avoid accidentally missing some tests).
This could be worked around by having the user manually set
GIT_SKIP_TESTS on a platform with an older Apache. But we can be a bit
friendlier by doing the version check ourselves and setting an
appropriate prereq. We'll use the (lack of) prereq to then skip the rest
of t5563. In theory we could use the prereq to skip individual tests, but
in practice this whole script depends on it.
Reported-by: Todd Zullinger <tmz@pobox.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The test uses "format-patch --thread" which is known to leak the
generated message ID list.
Plugging these leaks involves straightening out the memory ownership
rules around rev_info.message_id and rev_info.ref_message_ids, and
is beyond the scope of send-email fix, so for now mark the test as
leaky to unblock the topic before the release.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Recently git-send-email started parsing the same message twice, once
to validate _all_ the message before sending even the first one, and
then after the validation hook is happy and each message gets sent,
to read the contents to find out where to send to etc.
Unfortunately, the effect of reading the messages for validation
lingered even after the validation is done. Namely $message_id gets
assigned if exists in the input files but the variable is global,
and it is not cleared before pre_process_file runs. This causes
reading a message without a message-id followed by reading a message
with a message-id to misbehave---the sub reports as if the message
had the same id as the previously written one.
Clear the variable before starting to read the headers in
pre_process_file.
Tested-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When cloning an empty repository, protocol versions 0 and 1 currently
offer nothing but the header and flush packets for the /info/refs
endpoint. This means that no capabilities are provided, so the client
side doesn't know what capabilities are present.
However, this does pose a problem when working with SHA-256
repositories, since we use the capabilities to know the remote side's
object format (hash algorithm). As of 8b214c2e9d ("clone: propagate
object-format when cloning from void", 2023-04-05), this has been fixed
for protocol v2, since there we always read the hash algorithm from the
remote.
Fortunately, the push version of the protocol already indicates a clue
for how to solve this. When the /info/refs endpoint is accessed for a
push and the remote is empty, we include a dummy "capabilities^{}" ref
pointing to the all-zeros object ID. The protocol documentation already
indicates this should _always_ be sent, even for fetches and clones, so
let's just do that, which means we'll properly announce the hash
algorithm as part of the capabilities. This just works with the
existing code because we share the same ref code for fetches and clones,
and libgit2, JGit, and dulwich do as well.
There is one minor issue to fix, though. If we called send_ref with
namespaces, we would return NULL with the capabilities entry, which
would cause a crash. Instead, let's refactor out a function to print
just the ref itself without stripping the namespace and use it for our
special capabilities entry.
Add several sets of tests for HTTP as well as for local clones. The
behavior can be slightly different for HTTP versus a local or SSH clone
because of the stateless-rpc functionality, so it's worth testing both.
Signed-off-by: brian m. carlson <bk2204@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
