With a recent update to refuse access to repositories of other
people by default, "sudo make install" and "sudo git describe"
stopped working. This series intends to loosen it while keeping
the safety.
* cb/path-owner-check-with-sudo:
t0034: add negative tests and allow git init to mostly work under sudo
git-compat-util: avoid failing dir ownership checks if running privileged
t: regression git needs safe.directory when using sudo
"git -c branch.autosetupmerge=simple branch $A $B" will set the $B
as $A's upstream only when $A and $B shares the same name, and "git
-c push.default=simple" on branch $A would push to update the
branch $A at the remote $B came from. Also more places use the
sole remote, if exists, before defaulting to 'origin'.
* tk/simple-autosetupmerge:
push: new config option "push.autoSetupRemote" supports "simple" push
push: default to single remote even when not named origin
branch: new autosetupmerge option 'simple' for matching branches
"git fetch --recurse-submodules" from multiple remotes (either from
a remote group, or "--all") used to make one extra "git fetch" in
the submodules, which has been corrected.
* jc/avoid-redundant-submodule-fetch:
fetch: do not run a redundant fetch from submodule
The "--current" option of "git show-branch" should have been made
incompatible with the "--reflog" mode, but this was not enforced,
which has been corrected.
* jc/show-branch-g-current:
show-branch: -g and --current are incompatible
A bit of test framework fixes with a few fixes to issues found by
valgrind.
* ab/valgrind-fixes:
commit-graph.c: don't assume that stat() succeeds
object-file: fix a unpack_loose_header() regression in 3b6a8db3b0
log test: skip a failing mkstemp() test under valgrind
tests: using custom GIT_EXEC_PATH breaks --valgrind tests
"git show :<path>" learned to work better with the sparse-index
feature.
* ds/sparse-colon-path:
rev-parse: integrate with sparse index
object-name: diagnose trees in index properly
object-name: reject trees found in the index
show: integrate with the sparse index
t1092: add compatibility tests for 'git show'
Teach "git stash" to work better with sparse index entries.
* vd/sparse-stash:
unpack-trees: preserve index sparsity
stash: apply stash using 'merge_ort_nonrecursive()'
read-cache: set sparsity when index is new
sparse-index: expose 'is_sparse_index_allowed()'
stash: integrate with sparse index
stash: expand sparse-checkout compatibility testing
"git bisect" was too silent before it is ready to start computing
the actual bisection, which has been corrected.
* cd/bisect-messages-from-pre-flight-states:
bisect: output bisect setup status in bisect log
bisect: output state before we are ready to compute bisection
"git pull" without "--recurse-submodules=<arg>" made
submodule.recurse take precedence over fetch.recurseSubmodules by
mistake, which has been corrected.
* gc/pull-recurse-submodules:
pull: do not let submodule.recurse override fetch.recurseSubmodules
"git log --since=X" will stop traversal upon seeing a commit that
is older than X, but there may be commits behind it that is younger
than X when the commit was created with a faulty clock. A new
option is added to keep digging without stopping, and instead
filter out commits with timestamp older than X.
* mv/log-since-as-filter:
log: "--since-as-filter" option is a non-terminating "--since" variant
The temporary files fed to external diff command are now generated
inside a new temporary directory under the same basename.
* rs/external-diff-tempfile:
diff: use mks_tempfile_dt()
tempfile: add mks_tempfile_dt()
New tests for the safe.directory mechanism.
* sg/safe-directory-tests-and-docs:
safe.directory: document and check that it's ignored in the environment
t0033-safe-directory: check when 'safe.directory' is ignored
t0033-safe-directory: check the error message without matching the trash dir
When 7dce19d3 (fetch/pull: Add the --recurse-submodules option,
2010-11-12) introduced the "--recurse-submodule" option, the
approach taken was to perform fetches in submodules only once, after
all the main fetching (it may usually be a fetch from a single
remote, but it could be fetching from a group of remotes using
fetch_multiple()) succeeded. Later we added "--all" to fetch from
all defined remotes, which complicated things even more.
If your project has a submodule, and you try to run "git fetch
--recurse-submodule --all", you'd see a fetch for the top-level,
which invokes another fetch for the submodule, followed by another
fetch for the same submodule. All but the last fetch for the
submodule come from a "git fetch --recurse-submodules" subprocess
that is spawned via the fetch_multiple() interface for the remotes,
and the last fetch comes from the code at the end.
Because recursive fetching from submodules is done in each fetch for
the top-level in fetch_multiple(), the last fetch in the submodule
is redundant. It only matters when fetch_one() interacts with a
single remote at the top-level.
While we are at it, there is one optimization that exists in dealing
with a group of remote, but is missing when "--all" is used. In the
former, when the group turns out to be a group of one, instead of
spawning "git fetch" as a subprocess via the fetch_multiple()
interface, we use the normal fetch_one() code path. Do the same
when handing "--all", if it turns out that we have only one remote
defined.
Reviewed-by: Glen Choo <chooglen@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a support library that provides one function that can be used
to run a "scriplet" of commands through sudo and that helps invoking
sudo in the slightly awkward way that is required to ensure it doesn't
block the call (if shell was allowed as tested in the prerequisite)
and it doesn't run the command through a different shell than the one
we intended.
Add additional negative tests as suggested by Junio and that use a
new workspace that is owned by root.
Document a regression that was introduced by previous commits where
root won't be able anymore to access directories they own unless
SUDO_UID is removed from their environment.
The tests document additional ways that this new restriction could
be worked around and the documentation explains why it might be instead
considered a feature, but a "fix" is planned for a future change.
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Phillip Wood <phillip.wood123@gmail.com>
Signed-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
bdc77d1d68 (Add a function to determine whether a path is owned by the
current user, 2022-03-02) checks for the effective uid of the running
process using geteuid() but didn't account for cases where that user was
root (because git was invoked through sudo or a compatible tool) and the
original uid that repository trusted for its config was no longer known,
therefore failing the following otherwise safe call:
guy@renard ~/Software/uncrustify $ sudo git describe --always --dirty
[sudo] password for guy:
fatal: unsafe repository ('/home/guy/Software/uncrustify' is owned by someone else)
Attempt to detect those cases by using the environment variables that
those tools create to keep track of the original user id, and do the
ownership check using that instead.
This assumes the environment the user is running on after going
privileged can't be tampered with, and also adds code to restrict that
the new behavior only applies if running as root, therefore keeping the
most common case, which runs unprivileged, from changing, but because of
that, it will miss cases where sudo (or an equivalent) was used to change
to another unprivileged user or where the equivalent tool used to raise
privileges didn't track the original id in a sudo compatible way.
Because of compatibility with sudo, the code assumes that uid_t is an
unsigned integer type (which is not required by the standard) but is used
that way in their codebase to generate SUDO_UID. In systems where uid_t
is signed, sudo might be also patched to NOT be unsigned and that might
be able to trigger an edge case and a bug (as described in the code), but
it is considered unlikely to happen and even if it does, the code would
just mostly fail safely, so there was no attempt either to detect it or
prevent it by the code, which is something that might change in the future,
based on expected user feedback.
Reported-by: Guy Maurel <guy.j@maurel.de>
Helped-by: SZEDER Gábor <szeder.dev@gmail.com>
Helped-by: Randall Becker <rsbecker@nexbridge.com>
Helped-by: Phillip Wood <phillip.wood123@gmail.com>
Suggested-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Originally reported after release of v2.35.2 (and other maint branches)
for CVE-2022-24765 and blocking otherwise harmless commands that were
done using sudo in a repository that was owned by the user.
Add a new test script with very basic support to allow running git
commands through sudo, so a reproduction could be implemented and that
uses only `git status` as a proxy of the issue reported.
Note that because of the way sudo interacts with the system, a much
more complete integration with the test framework will require a lot
more work and that was therefore intentionally punted for now.
The current implementation requires the execution of a special cleanup
function which should always be kept as the last "test" or otherwise
the standard cleanup functions will fail because they can't remove
the root owned directories that are used. This also means that if
failures are found while running, the specifics of the failure might
not be kept for further debugging and if the test was interrupted, it
will be necessary to clean the working directory manually before
restarting by running:
$ sudo rm -rf trash\ directory.t0034-root-safe-directory/
The test file also uses at least one initial "setup" test that creates
a parallel execution directory under the "root" sub directory, which
should be used as top level directory for all repositories that are
used in this test file. Unlike all other tests the repository provided
by the test framework should go unused.
Special care should be taken when invoking commands through sudo, since
the environment is otherwise independent from what the test framework
setup and might have changed the values for HOME, SHELL and dropped
several relevant environment variables for your test. Indeed `git status`
was used as a proxy because it doesn't even require commits in the
repository to work and usually doesn't require much from the environment
to run, but a future patch will add calls to `git init` and that will
fail to honor the default branch name, unless that setting is NOT
provided through an environment variable (which means even a CI run
could fail that test if enabled incorrectly).
A new SUDO prerequisite is provided that does some sanity checking
to make sure the sudo command that will be used allows for passwordless
execution as root without restrictions and doesn't mess with git's
execution path. This matches what is provided by the macOS agents that
are used as part of GitHub actions and probably nowhere else.
Most of those characteristics make this test mostly only suitable for
CI, but it might be executed locally if special care is taken to provide
for all of them in the local configuration and maybe making use of the
sudo credential cache by first invoking sudo, entering your password if
needed, and then invoking the test with:
$ GIT_TEST_ALLOW_SUDO=YES ./t0034-root-safe-directory.sh
If it fails to run, then it means your local setup wouldn't work for the
test because of the configuration sudo has or other system settings, and
things that might help are to comment out sudo's secure_path config, and
make sure that the account you are using has no restrictions on the
commands it can run through sudo, just like is provided for the user in
the CI.
For example (assuming a username of marta for you) something probably
similar to the following entry in your /etc/sudoers (or equivalent) file:
marta ALL=(ALL:ALL) NOPASSWD: ALL
Reported-by: SZEDER Gábor <szeder.dev@gmail.com>
Helped-by: Phillip Wood <phillip.wood123@gmail.com>
Signed-off-by: Carlo Marcelo Arenas Belón <carenas@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since 4c28e4ada0 (commit: die before asking to edit the log
message, 2010-12-20), we have been "leaking" the "author_ident" when
prepare_to_commit() fails. Instead of returning from right there,
introduce an exit status variable and jump to the clean-up label
at the end.
Instead of explicitly releasing the resource with strbuf_release(),
mark the variable with UNLEAK() at the end, together with two other
variables that are already marked as such. If this were in a
utility function that is called number of times, but these are
different, we should explicitly release resources that grow
proportionally to the size of the problem being solved, but
cmd_commit() is like main() and there is no point in spending extra
cycles to release individual pieces of resource at the end, just
before process exit will clean everything for us for free anyway.
This fixes a leak demonstrated by e.g. "t3505-cherry-pick-empty.sh",
but unfortunately we cannot mark it or other affected tests as passing
now with "TEST_PASSES_SANITIZE_LEAK=true" as we'll need to fix many
other memory leaks before doing so.
Incidentally there are two tests that always passes the leak checker
with or without this change. Mark them as such.
This is based on an earlier patch by Ævar, but takes a different
approach that is more maintainable.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Fix a regression in my 3b6a8db3b0 (object-file.c: use "enum" return
type for unpack_loose_header(), 2021-10-01) revealed both by running
the test suite with --valgrind, and with the amended "git fsck" test.
In practice this regression in v2.34.0 caused us to claim that we
couldn't parse the header, as opposed to not being able to unpack
it. Before the change in the C code the test_cmp added here would emit:
-error: unable to unpack header of ./objects/e6/9de29bb2d1d6434b8b29ae775ad8c2e48c5391
+error: unable to parse header of ./objects/e6/9de29bb2d1d6434b8b29ae775ad8c2e48c5391
I.e. we'd proceed to call parse_loose_header() on the uninitialized
"hdr" value, and it would have been very unlikely for that
uninitialized memory to be a valid git object.
The other callers of unpack_loose_header() were already checking the
enum values exhaustively. See 3b6a8db3b0 and
5848fb11ac (object-file.c: return ULHR_TOO_LONG on "header too long",
2021-10-01).
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Skip a test added in f1e3df3169 (t: increase test coverage of
signature verification output, 2020-03-04) when running under
valgrind. Due to valgrind's interception of mkstemp() this test will
fail with:
+ pwd
+ TMPDIR=[...]/t/trash directory.t4202-log/bogus git log --show-signature -n1 plain-fail
==7696== VG_(mkstemp): failed to create temp file: [...]/t/trash directory.t4202-log/bogus/valgrind_proc_7696_cmdline_d545ddcf
[... 10 more similar lines omitted ..]
valgrind: Startup or configuration error:
valgrind: Can't create client cmdline file in [...]/t/trash directory.t4202-log/bogus/valgrind_proc_7696_cmdline_6e542d1d
valgrind: Unable to start up properly. Giving up.
error: last command exited with $?=1
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Fix a regression in b7d11a0f5d (tests: exercise the RUNTIME_PREFIX
feature, 2021-07-24) where tests that want to set up and test a "git"
wrapper in $PATH conflicted with the t/bin/valgrind wrapper(s) doing
the same.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Fix a bug in "git pull" where `submodule.recurse` is preferred over
`fetch.recurseSubmodules` when performing a fetch
(Documentation/config/fetch.txt says that `fetch.recurseSubmodules`
should be preferred.). Do this by passing the value of the
"--recurse-submodules" CLI option to the underlying fetch, instead of
passing a value that combines the CLI option and config variables.
In other words, this bug occurred because builtin/pull.c is conflating
two similar-sounding, but different concepts:
- Whether "git pull" itself should care about submodules e.g. whether it
should update the submodule worktrees after performing a merge.
- The value of "--recurse-submodules" to pass to the underlying "git
fetch".
Thus, when `submodule.recurse` is set, the underlying "git fetch" gets
invoked with "--recurse-submodules[=value]", overriding the value of
`fetch.recurseSubmodules`.
An alternative (and more obvious) approach to fix the bug would be to
teach "git pull" to understand `fetch.recurseSubmodules`, but the
proposed solution works better because:
- We don't maintain two identical config-parsing implementions in "git
pull" and "git fetch".
- It works better with other commands invoked by "git pull" e.g. "git
merge" won't accidentally respect `fetch.recurseSubmodules`.
Reported-by: Huang Zou <huang.zou@schrodinger.com>
Helped-by: Philippe Blain <levraiphilippeblain@gmail.com>
Signed-off-by: Glen Choo <chooglen@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The commit summary shown after making a commit is matched to what
is given in "git status" not to use the break-rewrite heuristics.
* rs/commit-summary-wo-break-rewrite:
commit, sequencer: turn off break_opt for commit summary
Avoid problems from interaction between malloc_check and address
sanitizer.
* pw/test-malloc-with-sanitize-address:
tests: make SANITIZE=address imply TEST_NO_MALLOC_CHECK
"git rebase --keep-base <upstream> <branch-to-rebase>" computed the
commit to rebase onto incorrectly, which has been corrected.
* ah/rebase-keep-base-fix:
rebase: use correct base for --keep-base when a branch is given
This allows seeing the current intermediate status without adding a new
good or bad commit:
$ git bisect log | tail -1
# status: waiting for bad commit, 1 good commit known
Signed-off-by: Chris Down <chris@chrisdown.name>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Commit 73c6de06af ("bisect: don't use invalid oid as rev when
starting") changes the behaviour of `git bisect` to consider invalid
oids as pathspecs again, as in the old shell implementation.
While that behaviour may be desirable, it can also cause confusion. For
example, while bisecting in a particular repo I encountered this:
$ git bisect start d93ff48803f0 v6.3
$
...which led to me sitting for a few moments, wondering why there's no
printout stating the first rev to check.
It turns out that the tag was actually "6.3", not "v6.3", and thus the
bisect was still silently started with only a bad rev, because
d93ff48803f0 was a valid oid and "v6.3" was silently considered to be a
pathspec.
While this behaviour may be desirable, it can be confusing, especially
with different repo conventions either using or not using "v" before
release names, or when a branch name or tag is simply misspelled on the
command line.
In order to avoid situations like this, make it more clear what we're
waiting for:
$ git bisect start d93ff48803f0 v6.3
status: waiting for good commit(s), bad commit known
We already have good output once the bisect process has begun in
earnest, so we don't need to do anything more there.
Signed-off-by: Chris Down <chris@chrisdown.name>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The progress meter of "git blame" was showing incorrect numbers
when processing only parts of the file.
* ea/progress-partial-blame:
blame: report correct number of lines in progress when using ranges
Reimplement "vimdiff[123]" mergetool drivers with a more generic
layout mechanism.
* fr/vimdiff-layout:
mergetools: add description to all diff/merge tools
vimdiff: add tool documentation
vimdiff: integrate layout tests in the unit tests framework ('t' folder)
vimdiff: new implementation with layout support
The performance of the "untracked cache" feature has been improved
when "--untracked-files=<mode>" and "status.showUntrackedFiles"
are combined.
* tk/untracked-cache-with-uall:
untracked-cache: support '--untracked-files=all' if configured
untracked-cache: test untracked-cache-bypassing behavior with -uall
When unpacking trees, set the default sparsity of the resultant index based
on repo settings and 'is_sparse_index_allowed()'.
Normally, when executing 'unpack_trees', the output index is marked sparse
when (and only when) it unpacks a sparse directory. However, an index may be
"sparse" even if it contains no sparse directories - when all files fall
inside the sparse-checkout definition or otherwise have SKIP_WORKTREE
disabled. Therefore, the output index may be marked "full" even when it is
"sparse", resulting in unnecessary 'ensure_full_index' calls when writing to
disk. Avoid this by setting the "default" index sparsity to match what is
expected for the repository.
As a consequence of this fix, the (non-merge) 'read-tree' performed when
applying a stash with untracked files no longer expands the index. Update
the corresponding test in 't1092'.
Signed-off-by: Victoria Dye <vdye@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Update 'stash' to use 'merge_ort_nonrecursive()' to apply a stash to the
current working tree. When 'git stash apply' was converted from its shell
script implementation to a builtin in 8a0fc8d19d (stash: convert apply to
builtin, 2019-02-25), 'merge_recursive_generic()' was used to merge a stash
into the working tree as part of 'git stash (apply|pop)'. However, with the
single merge base used in 'do_apply_stash()', the commit wrapping done by
'merge_recursive_generic()' is not only unnecessary, but misleading (the
*real* merge base is labeled "constructed merge base"). Therefore, a
non-recursive merge of the working tree, stashed tree, and stash base tree
is more appropriate.
There are two options for a non-recursive merge-then-update-worktree
function: 'merge_trees()' and 'merge_ort_nonrecursive()'. Use
'merge_ort_nonrecursive()' to align with the default merge strategy used by
'git merge' (6a5fb96672 (Change default merge backend from recursive to ort,
2021-08-04)) and, because merge-ort does not operate in-place on the index,
avoid unnecessary index expansion. Update tests in 't1092' verifying index
expansion for 'git stash' accordingly.
Signed-off-by: Victoria Dye <vdye@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When the index read in 'do_read_index()' does not exist on-disk, mark the
index "sparse" if the executing command does not require a full index and
sparse index is otherwise enabled.
Some commands (such as 'git stash -u') implicitly create a new index (when
the 'GIT_INDEX_FILE' variable points to a non-existent file) and perform
some operation on it. However, when this index is created, it isn't created
with the same sparsity settings as the repo index. As a result, while these
indexes may be sparse during the operation, they are always expanded before
being written to disk. We can avoid that expansion by defaulting the index
to "sparse", in which case it will only be expanded if the full index is
needed.
Note that the function 'set_new_index_sparsity()' is created despite having
only a single caller because additional callers will be added in a
subsequent patch.
Signed-off-by: Victoria Dye <vdye@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Enable sparse index in 'git stash' by disabling
'command_requires_full_index'.
With sparse index enabled, some subcommands of 'stash' work without
expanding the index, e.g., 'git stash', 'git stash list', 'git stash drop',
etc. Others ensure the index is expanded either directly (as in the case of
'git stash [pop|apply]', where the call to 'merge_recursive_generic()' in
'do_apply_stash()' triggers the expansion), or in a command called
internally by stash (e.g., 'git update-index' in 'git stash -u'). So, in
addition to enabling sparse index, add tests to 't1092' demonstrating which
variants of 'git stash' expand the index, and which do not.
Finally, add the option to skip writing 'untracked.txt' in
'ensure_not_expanded', and use that option to successfully apply stashed
untracked files without a conflict in 'untracked.txt'.
Signed-off-by: Victoria Dye <vdye@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add tests verifying expected 'git stash' behavior in
't1092-sparse-checkout-compatibility'. These cases establish the expected
behavior of 'git stash' in a sparse-checkout and verify consistency both
with and without a sparse index. Although no sparse index compatibility has
been integrated into 'git stash' yet, the tests are all 'expect_success' -
we don't want the cone-mode sparse-checkout behavior to change depending on
whether it is using a sparse index or not. Therefore, we expect these tests
to continue passing once sparse index is integrated with 'git stash'.
Additionally, add performance test cases for 'git stash' both with and
without untracked files. Note that, unlike the other tests in
'p2000-sparse-operations.sh', the tests added for 'stash' are combination
operations. This is done to ensure the stash/unstash is not blocked by the
modification of '$SPARSE_CONE/a' performed as part of 'test_perf_on_all'.
Signed-off-by: Victoria Dye <vdye@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git format-patch <args> -- <pathspec>" lost the pathspec when
showing the second and subsequent commits, which has been
corrected.
source: <c36896a1-6247-123b-4fa3-b7eb24af1897@web.de>
* rs/format-patch-pathspec-fix:
2.36 format-patch regression fix
"git fast-export -- <pathspec>" lost the pathspec when showing the
second and subsequent commits, which has been corrected.
source: <2c988c7b-0efe-4222-4a43-8124fe1a9da6@web.de>
* rs/fast-export-pathspec-fix:
2.36 fast-export regression fix
"git show <commit1> <commit2>... -- <pathspec>" lost the pathspec
when showing the second and subsequent commits, which has been
corrected.
source: <xmqqo80j87g0.fsf_-_@gitster.g>
* jc/show-pathspec-fix:
2.36 show regression fix
"diff-tree --stdin" has been broken for about a year, but 2.36
release broke it even worse by breaking running the command with
<pathspec>, which in turn broke "gitk" and got noticed. This has
been corrected by aligning its behaviour to that of "log".
This fixes a regression in 2.36 and is slate to go to 2.36.1
source: <xmqq7d7bsu2n.fsf@gitster.g>
* jc/diff-tree-stdin-fix:
2.36 gitk/diff-tree --stdin regression fix
"git submodule update" without pathspec should silently skip an
uninitialized submodule, but it started to become noisy by mistake.
This fixes a regression in 2.36 and is slate to go to 2.36.1
source: <pull.1258.v2.git.git.1650890741430.gitgitgadget@gmail.com>
* gc/submodule-update-part2:
submodule--helper: fix initialization of warn_if_uninitialized
git-p4 is designed to run correctly under python2.7 and python3, but
its functional behavior wrt importing user-entered text differs across
these environments:
Under python2, git-p4 "naively" writes the Perforce bytestream into git
metadata (and does not set an "encoding" header on the commits); this
means that any non-utf-8 byte sequences end up creating invalidly-encoded
commit metadata in git.
Under python3, git-p4 attempts to decode the Perforce bytestream as utf-8
data, and fails badly (with an unhelpful error) when non-utf-8 data is
encountered.
Perforce clients (especially p4v) encourage user entry of changelist
descriptions (and user full names) in OS-local encoding, and store the
resulting bytestream to the server unmodified - such that different
clients can end up creating mutually-unintelligible messages. The most
common inconsistency, in many Perforce environments, is likely to be utf-8
(typical in linux) vs cp-1252 (typical in windows).
Make the changelist-description- and user-fullname-handling code
python-runtime-agnostic, introducing three "strategies" selectable via
config:
- 'passthrough', behaving as previously under python2,
- 'strict', behaving as previously under python3, and
- 'fallback', favoring utf-8 but supporting a secondary encoding when
utf-8 decoding fails, and finally escaping high-range bytes if the
decoding with the secondary encoding also fails.
Keep the python2 default behavior as-is ('legacy' strategy), but switch
the python3 default strategy to 'fallback' with default fallback encoding
'cp1252'.
Also include tests exercising these encoding strategies, documentation for
the new config, and improve the user-facing error messages when decoding
does fail.
Signed-off-by: Tao Klerks <tao@klerks.biz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
