After running "git pull $there for-linus" to merge a signed tag, the
integrator may need to amend the resulting merge commit to fix typoes
in it. Teach --amend option to read the existing extra headers, and
carry them forward.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Now that we allow pulling a tag from the remote site to validate the
authenticity, we should give the user the final chance to verify and edit
the merge message. The integrator is expected to leave a meaningful merge
commit log in the history. Disallow fast-forwarding in such a case to
ensure that a merge commit is always recorded.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Now MERGE_HEAD records the tag objects without peeling, we could record
the result of manual conflict resolution via "git commit" without losing
the tag information. Introduce a new "mergetag" multi-line header field to
the commit object, and use it to store the entire contents of each signed
tag merged.
A commit header that has a multi-line payload begins with the header tag
(e.g. "mergetag" in this case), SP, the first line of payload, LF, and all
the remaining lines have a SP inserted at the beginning.
In hindsight, it would have been better to make "merge --continue" as the
way to continue from such an interrupted merge, not "commit", but this is
a backward compatibility baggage we would need to carry around for now.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* dm/pack-objects-update:
pack-objects: don't traverse objects unnecessarily
pack-objects: rewrite add_descendants_to_write_order() iteratively
pack-objects: use unsigned int for counter and offset values
pack-objects: mark add_to_write_order() as inline
Otherwise, "git commit" wouldn't have a way to tell that we were in the
middle of merging an annotated or signed tag, not a plain commit, after
"git merge" stops to ask the user to resolve conflicts.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The merge-recursive code uses the commit->util field directly to annotate
the commit objects given from the command line, i.e. the remote heads to
be merged, with a single string to be used to describe it in its trace
messages and conflict markers.
Correct this short-signtedness by redefining the field to be a pointer to
a structure "struct merge_remote_desc" that later enhancements can add
more information. Store the original objects we were told to merge in a
field "obj" in this struct, so that we can recover the tag we were told to
merge.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When a contributor asks the integrator to merge her history, a signed tag
can be a good vehicle to communicate the authenticity of the request while
conveying other information such as the purpose of the topic.
E.g. a signed tag "for-linus" can be created, and the integrator can run:
$ git pull git://example.com/work.git/ for-linus
This would allow the integrator to run "git verify-tag FETCH_HEAD" to
validate the signed tag.
Update fmt-merge-msg so that it pre-fills the merge message template with
the body (but not signature) of the tag object to help the integrator write
a better merge message, in the same spirit as the existing merge.log summary
lines.
The message that comes from GPG signature validation is also included in
the merge message template to help the integrator verify it, but they are
prefixed with "#" to make them comments.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
prune already shows progress meter while pruning. The marking part may
take a few seconds or more, depending on repository size. Show
progress meter during this time too.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2564aa4 started to initialize buf.alloc, but that should actually be one
more byte than the string length due to the trailing \0. Also, do not
modify buf.alloc out of the strbuf code. Use the existing strbuf_attach
instead.
Signed-off-by: Sebastian Schuberth <sschuberth@gmail.com>
Reviewed-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In various places in the codepath, the program tries to return early
assuming there is no more work needed. That is generally untrue when
over time new features are added.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This also updates the autogenerated merge title message from "merge commit X"
to "merge tag X", and its effect can be seen in the changes to the test suite.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Bad copy-paste.
Otherwise "git remote set-branches" without necessary argument
will result in an error message and help for set-url subcommand.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
fsck is usually a long process and it would be nice if it prints
progress from time to time.
Progress meter is not printed when --verbose is given because
--verbose prints a lot, there's no need for "alive" indicator.
Progress meter may provide "% complete" information but it would
be lost anyway in the flood of text.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
During verify_pack() all objects are read for SHA-1 check. Then
fsck_sha1() is called on every object, which read the object again
(fsck_sha1 -> parse_object -> read_sha1_file).
Avoid reading an object twice, do fsck_sha1 while we have an object
uncompressed data in verify_pack.
On git.git, with this patch I got:
$ /usr/bin/time ./git fsck >/dev/null
98.97user 0.90system 1:40.01elapsed 99%CPU (0avgtext+0avgdata 616624maxresident)k
0inputs+0outputs (0major+194186minor)pagefaults 0swaps
Without it:
$ /usr/bin/time ./git fsck >/dev/null
231.23user 2.35system 3:53.82elapsed 99%CPU (0avgtext+0avgdata 636688maxresident)k
0inputs+0outputs (0major+461629minor)pagefaults 0swaps
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Both of these free() calls are freeing a "const unsigned char (*)[20]"
type while free() expects a "void *". This results in the following
warning under clang 2.9:
builtin/diff.c:185:7: warning: passing 'const unsigned char (*)[20]' to parameter of type 'void *' discards qualifiers
free(parent);
^~~~~~
submodule.c:394:7: warning: passing 'const unsigned char (*)[20]' to parameter of type 'void *' discards qualifiers
free(parents);
^~~~~~~
This free()-ing without a cast was added by Jim Meyering to
builtin/diff.c in v1.7.6-rc3~4 and later by Fredrik Gustafsson in
submodule.c in v1.7.7-rc1~25^2.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
According to the C standard size_t is always unsigned, therefore the
comparison "n1 < 0 || n2 < 0" when n1 and n2 are size_t will always be
false.
This was raised by clang 2.9 which throws this warning when compiling
apply.c:
builtin/apply.c:253:9: warning: comparison of unsigned expression < 0 is always false [-Wtautological-compare]
if (n1 < 0 || n2 < 0)
~~ ^ ~
builtin/apply.c:253:19: warning: comparison of unsigned expression < 0 is always false [-Wtautological-compare]
if (n1 < 0 || n2 < 0)
~~ ^ ~
This check was originally added in v1.6.5-rc0~53^2 by Giuseppe Bilotta
while adding an option to git-apply to ignore whitespace differences.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We do not want to record tags as parents of a merge when the user does
"git pull $there tag v1.0" to merge tagged commit, but that is not a good
enough excuse to peel the tag down to commit when storing in FETCH_HEAD.
The caller of underlying "git fetch $there tag v1.0" may have other uses
of information contained in v1.0 tag in mind.
[jc: the test adjustment is to update for the new expectation]
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This mostly moves existing code from builtin/tag.c (for signing)
and builtin/verify-tag.c (for verifying) to a new gpg-interface.c
file to provide a more generic library interface.
- sign_buffer() takes a payload strbuf, a signature strbuf, and a signing
key, runs "gpg" to produce a detached signature for the payload, and
appends it to the signature strbuf. The contents of a signed tag that
concatenates the payload and the detached signature can be produced by
giving the same strbuf as payload and signature strbuf.
- verify_signed_buffer() takes a payload and a detached signature as
<ptr, len> pairs, and runs "gpg --verify" to see if the payload matches
the signature. It can optionally capture the output from GPG to allow
the callers to pretty-print it in a way more suitable for their
contexts.
"verify-tag" (aka "tag -v") used to save the whole tag contents as if it
is a detached signature, and fed gpg the payload part of the tag. It
relied on gpg to fail when the given tag is not signed but just is
annotated. The updated run_gpg_verify() function detects the lack of
detached signature in the input, and errors out without bothering "gpg".
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When pushing to delete a ref, it uses 0{40} as an object name to signal
that the request is a deletion. We shouldn't trigger "deletion of a
corrupt ref" warning in such a case, which was designed to notice that a
ref points at an object that is truly missing from the repository.
Reported-by: Stefan Näwe
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Modify the option parsing heuristic to handle all -m (rename) cases,
including the no-arg case.
Previously, this "fell through" to the (argc <= 2) case and caused
segfault.
Reported-by: Stefan Näwe <stefan.naewe@atlas-elektronik.com>
Signed-off-by: Tay Ray Chuan <rctay89@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* cn/fetch-prune:
fetch: treat --tags like refs/tags/*:refs/tags/* when pruning
fetch: honor the user-provided refspecs when pruning refs
remote: separate out the remote_find_tracking logic into query_refspecs
t5510: add tests for fetch --prune
fetch: free all the additional refspecs
* ss/blame-textconv-fake-working-tree:
(squash) test for previous
blame.c: Properly initialize strbuf after calling, textconv_object()
Conflicts:
t/t8006-blame-textconv.sh
* ef/mingw-upload-archive:
mingw: poll.h is no longer in sys/
upload-archive: use start_command instead of fork
compat/win32/poll.c: upgrade from upstream
mingw: move poll out of sys-folder
* dm/pack-objects-update:
pack-objects: don't traverse objects unnecessarily
pack-objects: rewrite add_descendants_to_write_order() iteratively
pack-objects: use unsigned int for counter and offset values
pack-objects: mark add_to_write_order() as inline
The POSIX-function fork is not supported on Windows. Use our
start_command API instead.
As this is the last call-site that depends on the fork-stub in
compat/mingw.h, remove that as well.
Add an undocumented flag to git-archive that tells it that the
action originated from a remote, so features can be disabled.
Thanks to Jeff King for work on this part.
Remove the NOT_MINGW-prereq for t5000, as git-archive --remote
now works.
Signed-off-by: Erik Faye-Lund <kusmabite@gmail.com>
Helped-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Factor out a small logic out of the private write_pack_file() function
in builtin/pack-objects.c.
This changes the order of finishing multi-pack generation slightly. The
code used to
- adjust shared perm of temporary packfile
- rename temporary packfile to the final name
- update mtime of the packfile under the final name
- adjust shared perm of temporary idxfile
- rename temporary idxfile to the final name
but because the helper does not want to do the mtime thing, the updated
code does that step first and then all the rest.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
For a plain string where only the length is known, strbuf.alloc needs to
be initialized to the length. Otherwise strbuf.alloc is 0 and a later
call to strbuf_setlen() will fail.
This bug surfaced when calling git blame under Windows on a *.doc file.
The *.doc file is converted to plain text by antiword via the textconv
mechanism. However, the plain text returned by antiword contains DOS line
endings instead of Unix line endings which triggered the strbuf_setlen()
which previous to this patch failed.
Signed-off-by: Sebastian Schuberth <sschuberth@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* js/grep-mutex:
builtin/grep: simplify lock_and_read_sha1_file()
builtin/grep: make lock/unlock into static inline functions
git grep: be careful to use mutexes only when they are initialized
Without this patch,
$ git clone foo .
results in this:
Cloning into ....
done.
With it:
Cloning into '.'...
done.
Signed-off-by: Richard Hartmann <richih.mailinglist@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
It is a basic code hygiene to avoid magic constants that are unnamed.
Besides, this helps extending the value later on for "interesting, but
cannot decide if the entry truely matches yet" (ie. prefix matches)
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
tree_entry_len() does not simply take two random arguments and return
a tree length. The two pointers must point to a tree item structure,
or struct name_entry. Passing random pointers will return incorrect
value.
Force callers to pass struct name_entry instead of two pointers (with
hope that they don't manually construct struct name_entry themselves)
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* cn/fetch-prune:
fetch: treat --tags like refs/tags/*:refs/tags/* when pruning
fetch: honor the user-provided refspecs when pruning refs
remote: separate out the remote_find_tracking logic into query_refspecs
t5510: add tests for fetch --prune
fetch: free all the additional refspecs
Conflicts:
remote.c
* jk/argv-array:
run_hook: use argv_array API
checkout: use argv_array API
bisect: use argv_array API
quote: provide sq_dequote_to_argv_array
refactor argv_array into generic code
quote.h: fix bogus comment
add sha1_array API docs
* maint-1.7.6:
notes_merge_commit(): do not pass temporary buffer to other function
gitweb: Fix links to lines in blobs when javascript-actions are enabled
mergetool: no longer need to save standard input
mergetool: Use args as pathspec to unmerged files
t9159-*.sh: skip for mergeinfo test for svn <= 1.4
date.c: Support iso8601 timezone formats
remote: only update remote-tracking branch if updating refspec
remote rename: warn when refspec was not updated
remote: "rename o foo" should not rename ref "origin/bar"
remote: write correct fetch spec when renaming remote 'remote'
* mz/remote-rename:
remote: only update remote-tracking branch if updating refspec
remote rename: warn when refspec was not updated
remote: "rename o foo" should not rename ref "origin/bar"
remote: write correct fetch spec when renaming remote 'remote'
Rather nasty things happen when a mutex is not initialized but locked
nevertheless. Now, when we're not running in a threaded manner, the mutex
is not initialized, which is correct. But then we went and used the mutex
anyway, which -- at least on Windows -- leads to a hard crash (ordinarily
it would be called a segmentation fault, but in Windows speak it is an
access violation).
This problem was identified by our faithful tests when run in the msysGit
environment.
To avoid having to wrap the line due to the 80 column limit, we use
the name "WHEN_THREADED" instead of "IF_USE_THREADS" because it is one
character shorter. Which is all we need in this case.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jk/maint-pack-objects-compete-with-delete:
downgrade "packfile cannot be accessed" errors to warnings
pack-objects: protect against disappearing packs
* nd/maint-autofix-tag-in-head:
Accept tags in HEAD or MERGE_HEAD
merge: remove global variable head[]
merge: use return value of resolve_ref() to determine if HEAD is invalid
merge: keep stash[] a local variable
Conflicts:
builtin/merge.c
This brings back some of the performance lost in optimizing recency
order inside pack objects. We were doing extreme amounts of object
re-traversal: for the 2.14 million objects in the Linux kernel
repository, we were calling add_to_write_order() over 1.03 billion times
(a 0.2% hit rate, making 99.8% of of these calls extraneous).
Two optimizations take place here- we can start our objects array
iteration from a known point where we left off before we started trying
to find our tags, and we don't need to do the deep dives required by
add_family_to_write_order() if the object has already been marked as
filled.
These two optimizations bring some pretty spectacular results via `perf
stat`:
task-clock: 83373 ms --> 43800 ms (50% faster)
cycles: 221,633,461,676 --> 116,307,209,986 (47% fewer)
instructions: 149,299,179,939 --> 122,998,800,184 (18% fewer)
Helped-by: Ramsay Jones (format string fix in "die" message)
Signed-off-by: Dan McGee <dpmcgee@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* js/no-cherry-pick-head-after-punted:
cherry-pick: do not give irrelevant advice when cherry-pick punted
revert.c: defer writing CHERRY_PICK_HEAD till it is safe to do so
This removes the need to call this function recursively, shinking the
code size slightly and netting a small performance increase.
Signed-off-by: Dan McGee <dpmcgee@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This is done in some of the new pack layout code introduced in commit
1b4bb16b9e. This more closely matches the nr_objects global that is
unsigned that these variables are based off of and bounded by.
Signed-off-by: Dan McGee <dpmcgee@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This function is a whole 26 bytes when compiled on x86_64, but is
currently invoked over 1.037 billion times when running pack-objects on
the Linux kernel git repository. This is hitting the point where
micro-optimizations do make a difference, and inlining it only increases
the object file size by 38 bytes.
As reported by perf, this dropped task-clock from 84183 to 83373 ms, and
total cycles from 223.5 billion to 221.6 billion. Not astronomical, but
worth getting for adding one word.
Signed-off-by: Dan McGee <dpmcgee@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If --tags is specified, add that refspec to the list given to
prune_refs so it knows to treat it as a filter on what refs to
should consider for prunning. This way
git fetch --prune --tags origin
only prunes tags and doesn't delete the branch refs.
Signed-off-by: Carlos Martín Nieto <cmn@elego.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If the user gave us refspecs on the command line, we should use those
when deciding whether to prune a ref instead of relying on the
refspecs in the config.
Previously, running
git fetch --prune origin refs/heads/master:refs/remotes/origin/master
would delete every other ref under the origin namespace because we
were using the refspec to filter the available refs but using the
configured refspec to figure out if a ref had been deleted on the
remote. This is clearly the wrong thing to do.
Change prune_refs and get_stale_heads to simply accept a list of
references and a list of refspecs. The caller of either function needs
to decide what refspecs should be used to decide whether a ref is
stale.
Signed-off-by: Carlos Martín Nieto <cmn@elego.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
These can happen if another process simultaneously prunes a
pack. But that is not usually an error condition, because a
properly-running prune should have repacked the object into
a new pack. So we will notice that the pack has disappeared
unexpectedly, print a message, try other packs (possibly
after re-scanning the list of packs), and find it in the new
pack.
Acked-by: Nicolas Pitre <nico@fluxnic.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
It's possible that while pack-objects is running, a
simultaneously running prune process might delete a pack
that we are interested in. Because we load the pack indices
early on, we know that the pack contains our item, but by
the time we try to open and map it, it is gone.
Since c715f78, we already protect against this in the normal
object access code path, but pack-objects accesses the packs
at a lower level. In the normal access path, we call
find_pack_entry, which will call find_pack_entry_one on each
pack index, which does the actual lookup. If it gets a hit,
we will actually open and verify the validity of the
matching packfile (using c715f78's is_pack_valid). If we
can't open it, we'll issue a warning and pretend that we
didn't find it, causing us to go on to the next pack (or on
to loose objects).
Furthermore, we will cache the descriptor to the opened
packfile. Which means that later, when we actually try to
access the object, we are likely to still have that packfile
opened, and won't care if it has been unlinked from the
filesystem.
Notice the "likely" above. If there is another pack access
in the interim, and we run out of descriptors, we could
close the pack. And then a later attempt to access the
closed pack could fail (we'll try to re-open it, of course,
but it may have been deleted). In practice, this doesn't
happen because we tend to look up items and then access them
immediately.
Pack-objects does not follow this code path. Instead, it
accesses the packs at a much lower level, using
find_pack_entry_one directly. This means we skip the
is_pack_valid check, and may end up with the name of a
packfile, but no open descriptor.
We can add the same is_pack_valid check here. Unfortunately,
the access patterns of pack-objects are not quite as nice
for keeping lookup and object access together. We look up
each object as we find out about it, and the only later when
writing the packfile do we necessarily access it. Which
means that the opened packfile may be closed in the interim.
In practice, however, adding this check still has value, for
three reasons.
1. If you have a reasonable number of packs and/or a
reasonable file descriptor limit, you can keep all of
your packs open simultaneously. If this is the case,
then the race is impossible to trigger.
2. Even if you can't keep all packs open at once, you
may end up keeping the deleted one open (i.e., you may
get lucky).
3. The race window is shortened. You may notice early that
the pack is gone, and not try to access it. Triggering
the problem without this check means deleting the pack
any time after we read the list of index files, but
before we access the looked-up objects. Triggering it
with this check means deleting the pack means deleting
the pack after we do a lookup (and successfully access
the packfile), but before we access the object. Which
is a smaller window.
Acked-by: Nicolas Pitre <nico@fluxnic.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* rs/pending:
commit: factor out clear_commit_marks_for_object_array
checkout: use leak_pending flag
bundle: use leak_pending flag
bisect: use leak_pending flag
revision: add leak_pending flag
checkout: use add_pending_{object,sha1} in orphan check
revision: factor out add_pending_sha1
checkout: check for "Previous HEAD" notice in t2020
Conflicts:
builtin/checkout.c
revision.c
* nd/maint-autofix-tag-in-head:
Accept tags in HEAD or MERGE_HEAD
merge: remove global variable head[]
merge: use return value of resolve_ref() to determine if HEAD is invalid
merge: keep stash[] a local variable
Conflicts:
builtin/merge.c
Implemented internally instead of as "git merge --no-commit && git commit"
so that "merge --edit" is otherwise consistent (hooks, etc) with "merge".
Note: the edit message does not include the status information that one
gets with "commit --status" and it is cleaned up after editing like one
gets with "commit --cleanup=default". A later patch could add the status
information if desired.
Note: previously we were not calling stripspace() after running the
prepare-commit-msg hook. Now we are, stripping comments and
leading/trailing whitespace lines if --edit is given, otherwise only
stripping leading/trailing whitespace lines if not given --edit.
Signed-off-by: Jay Soffian <jaysoffian@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
I noticed this when "git am CORRUPTED" unexpectedly failed with an
odd diagnostic, and even removed one of the files it was supposed
to have patched.
Reproduce with any valid old/new patch from which you have removed
the "+++ b/FILE" line. You'll see a diagnostic like this
fatal: unable to write file '(null)' mode 100644: Bad address
and you'll find that FILE has been removed.
The above is on glibc-based systems. On other systems, rather than
getting "null", you may provoke a segfault as git tries to
dereference the NULL file name.
Signed-off-by: Jim Meyering <meyering@redhat.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* mh/check-ref-format-3: (23 commits)
add_ref(): verify that the refname is formatted correctly
resolve_ref(): expand documentation
resolve_ref(): also treat a too-long SHA1 as invalid
resolve_ref(): emit warnings for improperly-formatted references
resolve_ref(): verify that the input refname has the right format
remote: avoid passing NULL to read_ref()
remote: use xstrdup() instead of strdup()
resolve_ref(): do not follow incorrectly-formatted symbolic refs
resolve_ref(): extract a function get_packed_ref()
resolve_ref(): turn buffer into a proper string as soon as possible
resolve_ref(): only follow a symlink that contains a valid, normalized refname
resolve_ref(): use prefixcmp()
resolve_ref(): explicitly fail if a symlink is not readable
Change check_refname_format() to reject unnormalized refnames
Inline function refname_format_print()
Make collapse_slashes() allocate memory for its result
Do not allow ".lock" at the end of any refname component
Refactor check_refname_format()
Change check_ref_format() to take a flags argument
Change bad_ref_char() to return a boolean value
...
* mz/remote-rename:
remote: only update remote-tracking branch if updating refspec
remote rename: warn when refspec was not updated
remote: "rename o foo" should not rename ref "origin/bar"
remote: write correct fetch spec when renaming remote 'remote'
* cb/common-prefix-unification:
rename pathspec_prefix() to common_prefix() and move to dir.[ch]
consolidate pathspec_prefix and common_prefix
remove prefix argument from pathspec_prefix
The previous logic in show_config was to print the delimiter when the
value was set, but Boolean variables have an implicit value "true" when
they appear with no value in the config file. As a result, we got:
git_Config --get-regexp '.*\.Boolean' #1. Ok: example.boolean
git_Config --bool --get-regexp '.*\.Boolean' #2. NO: example.booleantrue
Fix this by defering the display of the separator until after the value
to display has been computed.
Reported-by: Brian Foster <brian.foster@maxim-ic.com>
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In particular, gcc complains as follows:
CC tree-walk.o
tree-walk.c: In function `traverse_trees':
tree-walk.c:347: warning: 'e' might be used uninitialized in this \
function
CC builtin/revert.o
builtin/revert.c: In function `verify_opt_mutually_compatible':
builtin/revert.c:113: warning: 'opt2' might be used uninitialized in \
this function
Signed-off-by: Ramsay Jones <ramsay@ramsay1.demon.co.uk>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Close FETCH_HEAD and release the string url even if we have to leave the
function store_updated_refs() early.
Reported-by: Chris Wilson <cwilson@vigilantsw.com>
Helped-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Tay Ray Chuan <rctay89@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This teaches "merge --log" and fmt-merge-msg to use branch description
information when merging a local topic branch into the mainline. The
description goes between the branch name label and the list of commit
titles.
The refactoring to share the common configuration parsing between
merge and fmt-merge-msg needs to be made into a separate patch.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* js/maint-no-cherry-pick-head-after-punted:
cherry-pick: do not give irrelevant advice when cherry-pick punted
revert.c: defer writing CHERRY_PICK_HEAD till it is safe to do so
Conflicts:
builtin/revert.c
If a cherry-pick did not even start because the working tree had local
changes that would overlap with the operation, we shouldn't be advising
the users to resolve conflicts nor to conclude it with "git commit".
Signed-off-by: Jay Soffian <jaysoffian@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
do_pick_commit() writes out CHERRY_PICK_HEAD before invoking merge (either
via do_recursive_merge() or try_merge_command()) on the assumption that if
the merge fails it is due to conflict. However, if the tree is dirty, the
merge may not even start, aborting before do_pick_commit() can remove
CHERRY_PICK_HEAD.
Instead, defer writing CHERRY_PICK_HEAD till after merge has returned.
At this point we know the merge has either succeeded or failed due
to conflict. In either case, we want CHERRY_PICK_HEAD to be written
so that it may be picked up by the subsequent invocation of commit.
Note that do_recursive_merge() aborts if the merge cannot start, while
try_merge_command() returns a non-zero value other than 1.
Signed-off-by: Jay Soffian <jaysoffian@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This code calls git_config from a helper function to parse the config entry
it is interested in. Calling git_config in this way may cause a problem if
the helper function can be called after a previous call to git_config by
another function since the second call to git_config may reset some
variable to the value in the config file which was previously overridden.
The above is not a problem in this case since the function passed to
git_config only parses one config entry and the variable it sets is not
assigned outside of the parsing function. But a programmer who desires
all of the standard config options to be parsed may be tempted to modify
git_attr_config() so that it falls back to git_default_config() and then it
_would_ be vulnerable to the above described behavior.
So, move the call to git_config up into the top-level cmd_* function and
move the responsibility for parsing core.attributesfile into the main
config file parser.
Which is only the logical thing to do ;-)
Signed-off-by: Brandon Casey <drafnel@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "it" string would not be free'ed if base_name was non-NULL.
Let's free it.
Signed-off-by: Brandon Casey <drafnel@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "x"-prefixed versions of strdup, malloc, etc. will check whether the
allocation was successful and terminate the process otherwise.
A few uses of malloc were left alone since they already implemented a
graceful path of failure or were in a quasi external library like xdiff.
Additionally, the call to malloc in compat/win32/syslog.c was not modified
since the syslog() implemented there is a die handler and a call to the
x-wrappers within a die handler could result in recursion should memory
allocation fail. This will have to be addressed separately.
Signed-off-by: Brandon Casey <drafnel@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Using branch.$name.description as the configuration key, give users a
place to write about what the purpose of the branch is and things like
that, so that various subsystems, e.g. "push -s", "request-pull", and
"format-patch --cover-letter", can later be taught to use this
information.
The "-m" option similar to "commit/tag" is deliberately omitted, as the
whole point of branch description is about giving descriptive information
(the name of the branch itself is a better place for information that fits
on a single-line).
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Use the description for the branch when preparing the cover letter
when available.
While at it, mark a loosely written codepath that would do a random and
useless thing given an unusual input (e.g. "^master HEAD HEAD^"), which
we may want to fix someday.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since much of the infrastructure does not work correctly with
unnormalized refnames, change check_refname_format() to reject them.
Similarly, change "git check-ref-format" to reject unnormalized
refnames by default. But add an option --normalize, which causes "git
check-ref-format" to normalize the refname before checking its format,
and print the normalized refname. This is exactly the behavior of the
old --print option, which is retained but deprecated.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Soon we will make printing independent of collapsing.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This will make upcoming changes a tiny bit easier.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Change check_ref_format() to take a flags argument that indicates what
is acceptable in the reference name (analogous to "git
check-ref-format"'s "--allow-onelevel" and "--refspec-pattern"). This
is more convenient for callers and also fixes a failure in the test
suite (and likely elsewhere in the code) by enabling "onelevel" and
"refspec-pattern" to be allowed independently of each other.
Also rename check_ref_format() to check_refname_format() to make it
obvious that it deals with refnames rather than references themselves.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Also add tests of the new options. (Actually, one big reason to add
the new options is to make it easy to test check_ref_format(), though
the options should also be useful to other scripts.)
Interpret the result of check_ref_format() based on which types of
refnames are allowed. However, because check_ref_format() can only
return a single value, one test case is still broken. Specifically,
the case "git check-ref-format --onelevel '*'" incorrectly succeeds
because check_ref_format() returns CHECK_REF_FORMAT_ONELEVEL for this
refname even though the refname is also CHECK_REF_FORMAT_WILDCARD.
The type of check that leads to this failure is used elsewhere in
"real" code and could lead to bugs; it will be fixed over the next few
commits.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jk/argv-array:
run_hook: use argv_array API
checkout: use argv_array API
bisect: use argv_array API
quote: provide sq_dequote_to_argv_array
refactor argv_array into generic code
quote.h: fix bogus comment
add sha1_array API docs
* mg/branch-list:
t3200: clean up checks for file existence
branch: -v does not automatically imply --list
branch: allow pattern arguments
branch: introduce --list option
git-branch: introduce missing long forms for the options
git-tag: introduce long forms for the options
t6040: test branch -vv
Conflicts:
Documentation/git-tag.txt
t/t3200-branch.sh
* jk/for-each-ref:
for-each-ref: add split message parts to %(contents:*).
for-each-ref: handle multiline subjects like --pretty
for-each-ref: refactor subject and body placeholder parsing
t6300: add more body-parsing tests
t7004: factor out gpg setup
* jc/receive-verify:
receive-pack: check connectivity before concluding "git push"
check_everything_connected(): libify
check_everything_connected(): refactor to use an iterator
fetch: verify we have everything we need before updating our ref
Conflicts:
builtin/fetch.c
* jc/fetch-verify:
fetch: verify we have everything we need before updating our ref
rev-list --verify-object
list-objects: pass callback data to show_objects()
* rr/revert-cherry-pick-continue:
builtin/revert.c: make commit_list_append() static
revert: Propagate errors upwards from do_pick_commit
revert: Introduce --continue to continue the operation
revert: Don't implicitly stomp pending sequencer operation
revert: Remove sequencer state when no commits are pending
reset: Make reset remove the sequencer state
revert: Introduce --reset to remove sequencer state
revert: Make pick_commits functionally act on a commit list
revert: Save command-line options for continuing operation
revert: Save data for continuing after conflict resolution
revert: Don't create invalid replay_opts in parse_args
revert: Separate cmdline parsing from functional code
revert: Introduce struct to keep command-line options
revert: Eliminate global "commit" variable
revert: Rename no_replay to record_origin
revert: Don't check lone argument in get_encoding
revert: Simplify and inline add_message_to_msg
config: Introduce functions to write non-standard file
advice: Introduce error_resolve_conflict
Make ERR as first packet of remote snapshot reply work like it does in
fetch/push. Lets servers decline remote snapshot with message the same
way as declining fetch/push with a message.
Signed-off-by: Ilari Liusvaara <ilari.liusvaara@elisanet.fi>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jc/maint-grep-untracked-exclude:
grep: teach --untracked and --exclude-standard options
grep --no-index: don't use git standard exclusions
grep: do not use --index in the short usage output
Conflicts:
Documentation/git-grep.txt
builtin/grep.c
Give each mode of operation (all, from stdin, given commits) its own usage
line to make it easier to see that they are mutually exclusive.
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Factor out the code to clear the commit marks for a whole struct
object_array from builtin/checkout.c into its own exported function
clear_commit_marks_for_object_array and use it in bisect and bundle
as well. It handles tags and commits and ignores objects of any
other type.
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Instead of going through all the references again when we clear the
commit marks, do it like bisect and bundle and gain ownership of the
list of pending objects which we constructed from those references.
We simply copy the struct object_array that points to the list, set
the flag leak_pending and then prepare_revision_walk won't destroy
it and it's ours. We use it to clear the marks and free it at the
end.
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Instead of building a list of textual arguments for setup_revisions, use
add_pending_object and add_pending_sha1 to queue the objects directly.
This is both faster and simpler.
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When a push specifies deletion of non-existent refs, the post post-receive and
post-update hooks receive them as input/arguments.
For instance, for the following push, where refs/heads/nonexistent is a ref
which does not exist on the remote side:
git push origin :refs/heads/nonexistent
the post-receive hook receives from standard input:
<null-sha1> SP <null-sha1> SP refs/heads/nonexistent
and the post-update hook receives as arguments:
refs/heads/nonexistent
which does not make sense since it is a no-op.
Teach receive-pack not to pass non-existent refs to the post-receive and
post-update hooks. If the push only attempts to delete non-existent refs,
these hooks are not even called.
The update and pre-receive hooks are still notified about attempted
deletion of non-existent refs to give them a chance to inspect the
situation and act on it.
[jc: mild fix-ups to avoid introducing an extra list; also added fixes to
some tests]
Signed-off-by: Pang Yan Han <pangyanhan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Checking paths out of a tree is (currently) defined to do:
- Grab the paths from the named tree that match the given pathspec,
and add them to the index;
- Check out the contents from the index for paths that match the
pathspec to the working tree; and while at it
- If the given pathspec did not match anything, suspect a typo from the
command line and error out without updating the index nor the working
tree.
Suppose that the branch you are working on has dir/myfile, and the "other"
branch has dir/other but not dir/myfile. Further imagine that you have
either modified or removed dir/myfile in your working tree, but you have
not run "git add dir/myfile" or "git rm dir/myfile" to tell Git about your
local change. Running
$ git checkout other dir
would add dir/other to the index with the contents taken out of the
"other" branch, and check out the paths from the index that match the
pathspec "dir", namely, "dir/other" and "dir/myfile", overwriting your
local changes to "dir/myfile", even though "other" branch does not even
know about that file.
Fix it by updating the working tree only with the index entries that
was read from the "other" tree.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When running git describe --dirty the index should be refreshed. Previously
the cached index would cause describe to think that the index was dirty when,
in reality, it was just stale.
The issue was exposed by python setuptools which hardlinks files into another
directory when building a distribution.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This option causes check-attr to consider .gitattributes only from
the index, ignoring .gitattributes from the working tree. This allows
the command to be used in situations where a working tree does not exist.
Signed-off-by: Jay Soffian <jaysoffian@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
get_one_patchid() uses a rather dumb heuristic to determine if the
passed buffer is part of the next commit. Whenever the first 40 bytes
are a valid hexadecimal sha1 representation, get_one_patchid() returns
next_sha1.
Once the current line is longer than the fixed buffer, this will break
(provided the additional bytes make a valid hexadecimal sha1). As a result
patch-id returns incorrect results. Instead, use strbuf and read one line
at a time.
Helped-by: Jeff King <peff@peff.net>
Signed-off-by: Michael Schubert <mschub@elegosoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
With the usual "git" transport, a large-ish transfer with "git fetch" and
"git pull" give progress eye-candy to avoid boring users. However, not
when they are reading from a bundle. I.e.
$ git pull ../git-bundle.bndl master
This teaches bundle.c:unbundle() to give "-v" option to index-pack and
tell it to give progress bar when transport decides it is necessary.
The operation in the other direction, "git bundle create", could also
learn to honor --quiet but that is a separate issue.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
HEAD and MERGE_HEAD (among other branch tips) should never hold a
tag. That can only be caused by broken tools and is cumbersome to fix
by an end user with:
$ git update-ref HEAD $(git rev-parse HEAD^{commit})
which may look like a magic to a new person.
Be easy, warn users (so broken tools can be fixed if they bother to
report) and move on.
Be robust, if the given SHA-1 cannot be resolved to a commit object,
die (therefore return value is always valid).
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Also kill head_invalid in favor of "head_commit == NULL".
Local variable "head" in cmd_merge() is renamed to "head_sha1" to make
sure I don't miss any access because this variable should not be used
after head_commit is set (use head_commit->object.sha1 instead).
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
resolve_ref() only updates "head" when it returns non NULL value (it
may update "head" even when returning NULL, but not in all cases).
Because "head" is not initialized before the call, is_null_sha1() is
not enough. Check also resolve_ref() return value.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "git branch" command, while not in listing mode, calls create_branch()
even when the target branch already exists, and it does so even when it is
not interested in updating the value of the branch (i.e. the name of the
commit object that sits at the tip of the existing branch). This happens
when the command is run with "--set-upstream" option.
The earlier safety-measure to prevent "git branch -f $branch $commit" from
updating the currently checked out branch did not take it into account,
and we no longer can update the tracking information of the current branch.
Minimally fix this regression by telling the validation code if it is
called to really update the value of a potentially existing branch, or if
the caller merely is interested in updating auxiliary aspects of a branch.
Reported-and-Tested-by: Jay Soffian
Signed-off-by: Junio C Hamano <gitster@pobox.com>
What should happen if you run this command?
$ git ls-remote -h
It does not give a short-help for the command. Instead because "-h" is a
synonym for "--heads", it runs "git ls-remote --heads", and because there
is no remote specified on the command line, we run it against the default
"origin" remote, hence end up doing the same as
$ git ls-remote --heads origin
Fix this counter-intuitive behaviour by special casing a lone "-h" that
does not have anything else on the command line and calling usage().
Signed-off-by: Junio C Hamano <gitster@pobox.com>
It probably is not such a good idea to use ":/<pattern>" to specify which
commit to merge, as ":/<pattern>" can often hit unexpected commits, but
somebody tried it and got a nonsense error message:
fatal: ':/Foo bar' does not point to a commit
So here is a for-the-sake-of-consistency update that is fairly useless
that allows users to carefully try not shooting in the foot.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We were using a similar ad-hoc rev_list_args structure, but
this saves some code.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Yes, there is a warning that says the function is only used by push in big
red letters in front of this function, but it didn't say a more important
thing it should have said: what the function is for and what it does.
Rename it and document it to avoid future confusion.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Running a hook has to make complex set-up to establish web of
communication between child process and multiplexer, which is common
regardless of what kind of data is fed to the hook. Refactor the parts
that is specific to the data fed to the particular set of hooks from the
part that runs the hook, so that the code can be reused to drive hooks
that take different kind of data.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The message identifies the process as receive-pack when it cannot fork the
sideband demultiplexer. We are actually a send-pack.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Also make common_prefix_len() static as this refactoring makes dir.c
itself the only caller of this helper function.
Signed-off-by: Clemens Buchacher <drizzd@aon.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
commit c9bfb953 (want_color: automatically fallback to color.ui,
2011-08-17) introduced a regression where format-patch produces colorized
patches when color.ui is set to "always".
In f3aafa4 (Disable color detection during format-patch, 2006-07-09),
git_format_config was taught to intercept diff.color to avoid passing it
down to git_log_config and later, git_diff_ui_config.
Teach git_format_config to intercept color.ui in the same way.
Helped-by: Jeff King <peff@peff.net>
Signed-off-by: Pang Yan Han <pangyanhan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
'git remote rename' will only update the remote's fetch refspec if it
looks like a default one. If the remote has no default fetch refspec,
as in
[remote "origin"]
url = git://git.kernel.org/pub/scm/git/git.git
fetch = +refs/heads/*:refs/remotes/upstream/*
we would not update the fetch refspec and even if there is a ref
called "refs/remotes/origin/master", we should not rename it, since it
was not created by fetching from the remote.
Suggested-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Martin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When renaming a remote, we also try to update the fetch refspec
accordingly, but only if it has the default format. For others, such
as refs/heads/master:refs/heads/origin, we are conservative and leave
it untouched. Let's give the user a warning about refspecs that are
not updated, so he can manually update the config if necessary.
Suggested-by: Jeff King <peff@peff.net>
Signed-off-by: Martin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When renaming a remote called 'o' using 'git remote rename o foo', git
should also rename any remote-tracking branches for the remote. This
does happen, but any remote-tracking branches starting with
'refs/remotes/o', such as 'refs/remotes/origin/bar', will also be
renamed (to 'refs/remotes/foorigin/bar' in this case).
Fix it by simply matching one more character, up to the slash
following the remote name.
Signed-off-by: Martin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When renaming a remote whose name is contained in a configured fetch
refspec for that remote, we currently replace the first occurrence of
the remote name in the refspec. This is correct in most cases, but
breaks if the remote name occurs in the fetch refspec before the
expected place. For example, we currently change
[remote "remote"]
url = git://git.kernel.org/pub/scm/git/git.git
fetch = +refs/heads/*:refs/remotes/remote/*
into
[remote "origin"]
url = git://git.kernel.org/pub/scm/git/git.git
fetch = +refs/heads/*:refs/origins/remote/*
Reduce the risk of changing incorrect sections of the refspec by
matching the entire ":refs/remotes/<name>/" instead of just "<name>".
Signed-off-by: Martin von Zweigbergk <martin.von.zweigbergk@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Asking fwrite() to write one item of size bytes results in fwrite()
reporting "I wrote zero item", when size is zero. Instead, we could
ask it to write "size" items of 1 byte and expect it to report that
"I wrote size items" when it succeeds, with any value of size,
including zero.
Noticed and reported by BJ Hargrave.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Extract the helper function and the type definition of the iterator
function it uses out of builtin/fetch.c into a separate source and a
header file.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We will be using the same "rev-list --verify-objects" logic to add a
sanity check to the receiving end of "git push" in the same way, but the
list of commits that are checked come from a structure with a different
shape over there.
Update the function to take an iterator to make it easier to reuse it in
different contexts.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "git fetch" command works in two phases. The remote side tells us what
objects are at the tip of the refs we are fetching from, and transfers the
objects missing from our side. After storing the objects in our repository,
we update our remote tracking branches to point at the updated tips of the
refs.
A broken or malicious remote side could send a perfectly well-formed pack
data during the object transfer phase, but there is no guarantee that the
given data actually fill the gap between the objects we originally had and
the refs we are updating to.
Although this kind of breakage can be caught by running fsck after a
fetch, it is much cheaper to verify that everything that is reachable from
the tips of the refs we fetched are indeed fully connected to the tips of
our current set of refs before we update them.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
It makes no sense to do the - possibly very expensive - call to "rev-list
<new-ref-sha1> --not --all" in check_for_new_submodule_commits() when
there aren't any submodules configured.
Leave check_for_new_submodule_commits() early when no name <-> path
mappings for submodules are found in the configuration. To make that work
reading the configuration had to be moved further up in cmd_fetch(), as
doing that after the actual fetch of the superproject was too late.
Reported-by: Martin Fick <mfick@codeaurora.org>
Signed-off-by: Jens Lehmann <Jens.Lehmann@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"branch -v" without other options or parameters still works in the list
mode, but that is not because there is "-v" but because there is no
parameter nor option.
Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The %(body) placeholder returns the whole body of a tag or
commit, including the signature. However, callers may want
to get just the body without signature, or just the
signature.
Rather than change the meaning of %(body), which might break
some scripts, this patch introduces a new set of
placeholders which break down the %(contents) placeholder
into its constituent parts.
[jk: initial patch by mg, rebased on top of my refactoring
and with tests by me]
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Generally the format of a git tag or commit message is:
subject
body body body
body body body
However, we occasionally see multiline subjects like:
subject
with multiple
lines
body body body
body body body
The rest of git treats these multiline subjects as something
to be concatenated and shown as a single line (e.g., "git
log --pretty=format:%s" will do so since f53bd74). For
consistency, for-each-ref should do the same with its
"%(subject)".
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The find_subpos function was a little hard to use, as well
as to read. It would sometimes write into the subject and
body pointers, and sometimes not. The body pointer sometimes
could be compared to subject, and sometimes not. When
actually duplicating the subject, the caller was forced to
figure out again how long the subject is (which is not too
big a deal when the subject is a single line, but hard to
extend).
The refactoring makes the function more straightforward, both
to read and to use. We will always put something into the
subject and body pointers, and we return explicit lengths
for them, too.
This lays the groundwork both for more complex subject
parsing (e.g., multiline), as well as splitting the body
into subparts (like the text versus the signature).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Passing a prefix to a function that is supposed to find the prefix is
strange. And it's really only used if the pathspec is NULL. Make the
callers handle this case instead.
As we are always returning a fresh copy of a string (or NULL), change the
type of the returned value to non-const "char *".
Signed-off-by: Clemens Buchacher <drizzd@aon.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This reverts commit ffa69e61d3, reversing
changes made to 4a13c4d148.
Adding a new command line option to receive-pack and feed it from
send-pack is not an acceptable way to add features, as there is no
guarantee that your updated send-pack will be talking to updated
receive-pack. New features need to be added via the capability mechanism
negotiated over the protocol.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This single variable can be used to set instead of setting fsckobjects
variable for fetch & receive independently.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This corresponds to receive.fsckobjects configuration variable added (a
lot) earlier in 20dc001 (receive-pack: allow using --strict mode for
unpacking objects, 2008-02-25).
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "git fetch" command works in two phases. The remote side tells us what
objects are at the tip of the refs we are fetching from, and transfers the
objects missing from our side. After storing the objects in our repository,
we update our remote tracking branches to point at the updated tips of the
refs.
A broken or malicious remote side could send a perfectly well-formed pack
data during the object transfer phase, but there is no guarantee that the
given data actually fill the gap between the objects we originally had and
the refs we are updating to.
Although this kind of breakage can be caught by running fsck after a
fetch, it is much cheaper to verify that everything that is reachable from
the tips of the refs we fetched are indeed fully connected to the tips of
our current set of refs before we update them.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Often we want to verify everything reachable from a given set of commits
are present in our repository and connected without a gap to the tips of
our refs. We used to do this for this purpose:
$ rev-list --objects $commits_to_be_tested --not --all
Even though this is good enough for catching missing commits and trees,
we show the object name but do not verify their existence, let alone their
well-formedness, for the blob objects at the leaf level.
Add a new "--verify-object" option so that we can catch missing and broken
blobs as well.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The traverse_commit_list() API takes two callback functions, one to show
commit objects, and the other to show other kinds of objects. Even though
the former has a callback data parameter, so that the callback does not
have to rely on global state, the latter does not.
Give the show_objects() callback the same callback data parameter.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The current explanation of -e can be misread as allowing the user to say
I know 'git clean -XYZ' (substitute -XYZ with any option and/or
parameter) will remove paths A, B, and C, and I want them all removed
except for paths matching this pattern by adding '-e C' to the same
command line, i.e. 'git clean -e C -XYZ'.
But that is not what this option does. It augments the set of ignore rules
from the command line, just like the same "-e <pattern>" argument does
with the "ls-files" command (the user could probably pass "-e \!C" to tell
the command to clean everything the command would normally remove, except
for C). Also error out when both -x and -e are given with an explanation of
what -e means---it is a symptom of misunderstanding what -e does.
It also fixes small style nit in the parameter to add_exclude() call. The
current code only works because EXC_CMDL happens to be defined as 0.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Allow pattern arguments for the list mode just like for git tag -l.
Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Currently, there is no way to invoke the list mode explicitly, without
giving -v to force verbose output.
Introduce a --list option which invokes the list mode. This will be
beneficial for invoking list mode with pattern matching, which otherwise
would be interpreted as branch creation.
Along with --list, test also combinations of existing options.
Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Long forms are better to memorize and more reliably uniform across
commands.
Names follow precedents, e.g. "git log --remotes".
Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Long forms are better to memorize and more reliably uniform across
commands.
Design notes:
-u,--local-user is named following the analogous gnupg option.
-l,--list is not an argument taking option but a mode switch.
Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* nd/decorate-grafts:
log: Do not decorate replacements with --no-replace-objects
log: decorate "replaced" on to replaced commits
log: decorate grafted commits with "grafted"
Move write_shallow_commits to fetch-pack.c
Add for_each_commit_graft() to iterate all grafts
decoration: do not mis-decorate refs with same prefix
* jk/color-and-pager:
want_color: automatically fallback to color.ui
diff: don't load color config in plumbing
config: refactor get_colorbool function
color: delay auto-color decision until point of use
git_config_colorbool: refactor stdout_is_tty handling
diff: refactor COLOR_DIFF from a flag into an int
setup_pager: set GIT_PAGER_IN_USE
t7006: use test_config helpers
test-lib: add helper functions for config
t7006: modernize calls to unset
Conflicts:
builtin/commit.c
parse-options.c
A stash is created by save_state() and used by restore_state(). Pass
SHA-1 explicitly for clarity and keep stash[] to cmd_merge().
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When asked if "refs///heads/master" is valid, check-ref-format says "Yes,
it is well formed", and when asked to print canonical form, it shows
"refs/heads/master". This is so that it can be tucked after "$GIT_DIR/"
to form a valid pathname for a loose ref, and we normalize a pathname like
"$GIT_DIR/refs///heads/master" to de-dup the slashes in it.
Similarly, when asked if "/refs/heads/master" is valid, check-ref-format
says "Yes, it is Ok", but the leading slash is not removed when printing,
leading to "$GIT_DIR//refs/heads/master".
Fix it to make sure such leading slashes are removed. Add tests that such
refnames are accepted and normalized correctly.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* cb/maint-quiet-push:
receive-pack: do not overstep command line argument array
propagate --quiet to send-pack/receive-pack
Conflicts:
Documentation/git-receive-pack.txt
Documentation/git-send-pack.txt
Cloning from a local repository blindly copies or hardlinks all the files
under objects/ hierarchy. This results in two issues:
- If the repository cloned has an "objects/info/alternates" file, and the
command line of clone specifies --reference, the ones specified on the
command line get overwritten by the copy from the original repository.
- An entry in a "objects/info/alternates" file can specify the object
stores it borrows objects from as a path relative to the "objects/"
directory. When cloning a repository with such an alternates file, if
the new repository is not sitting next to the original repository, such
relative paths needs to be adjusted so that they can be used in the new
repository.
This updates add_to_alternates_file() to take the path to the alternate
object store, including the "/objects" part at the end (earlier, it was
taking the path to $GIT_DIR and was adding "/objects" itself), as it is
technically possible to specify in objects/info/alternates file the path
of a directory whose name does not end with "/objects".
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Also add a test to expose a long-standing bug that is triggered when
cloning with --reference option from a local repository that has its own
alternates. The alternate object stores specified on the command line
are lost, and only alternates copied from the source repository remain.
The bug will be fixed in the next patch.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git branch -M <foo> <current-branch>" allows updating the current branch
which HEAD points, without the necessary house-keeping that git reset
normally does to make this operation sensible. It also leaves the reflog
in a confusing state (you would be warned when trying to read it).
"git checkout -B <current branch> <foo>" is also partly vulnerable to this
bug; due to inconsistent pre-flight checks it would perform half of its
task and then abort just before rewriting the branch. Again this
manifested itself as the index file getting out-of-sync with HEAD.
"git branch -f" already guarded against this problem, and aborts with
a fatal error.
Update "git branch -M", "git checkout -B" and "git branch -f" to share the
same check before allowing a branch to be created. These prevent you from
updating the current branch.
We considered suggesting the use of "git reset" in the failure message
but concluded that it was not possible to discern what the user was
actually trying to do.
Signed-off-by: Conrad Irwin <conrad.irwin@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The function was not gentle at all to the callers and died without giving
them a chance to deal with possible errors. Rename it to read_gitfile(),
and update all the callers.
As no existing caller needs a true "gently" variant, we do not bother
adding one at this point.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
There are two copies of traverse_commit_list callback that show the object
name followed by pathname the object was found, to produce output similar
to "rev-list --objects".
Unify them.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The callback to traverse_commit_list() are to take linked name_path and
a string for the last path component.
If the callee used its parameters, it would have seen duplicated leading
paths. In this particular case, the callee does not use this argument but
that is not a reason to leave the call broken.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When working with submodules it is easy to forget to push a
submodule to the server but pushing a super-project that
contains a commit for that submodule. The result is that the
superproject points at a submodule commit that is not available
on the server.
This adds the option --recurse-submodules=check to push. When
using this option git will check that all submodule commits that
are about to be pushed are present on a remote of the submodule.
To be able to use a combined diff, disabling a diff callback has
been removed from combined-diff.c.
Signed-off-by: Fredrik Gustafsson <iveqy@iveqy.com>
Mentored-by: Jens Lehmann <Jens.Lehmann@web.de>
Mentored-by: Heiko Voigt <hvoigt@hvoigt.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
All of the "do we want color" flags default to -1 to
indicate that we don't have any color configured. This value
is handled in one of two ways:
1. In porcelain, we check early on whether the value is
still -1 after reading the config, and set it to the
value of color.ui (which defaults to 0).
2. In plumbing, it stays untouched as -1, and want_color
defaults it to off.
This works fine, but means that every porcelain has to check
and reassign its color flag. Now that want_color gives us a
place to put this check in a single spot, we can do that,
simplifying the calling code.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
For "git config --get-colorbool color.foo", we use a custom
callback that looks not only for the key that the user gave
us, but also for "diff.color" (for backwards compatibility)
and "color.ui" (as a fallback).
For the former, we use a custom variable to store the
diff.color value. For the latter, though, we store it in the
main "git_use_color_default" variable, turning on color.ui
for any other parts of git that respect this value.
In practice, this doesn't cause any bugs, because git-config
runs without caring about git_use_color_default, and then
exits. But it crosses module boundaries in an unusual and
confusing way, and it makes refactoring color handling
harder than it needs to be.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When we read a color value either from a config file or from
the command line, we use git_config_colorbool to convert it
from the tristate always/never/auto into a single yes/no
boolean value.
This has some timing implications with respect to starting
a pager.
If we start (or decide not to start) the pager before
checking the colorbool, everything is fine. Either isatty(1)
will give us the right information, or we will properly
check for pager_in_use().
However, if we decide to start a pager after we have checked
the colorbool, things are not so simple. If stdout is a tty,
then we will have already decided to use color. However, the
user may also have configured color.pager not to use color
with the pager. In this case, we need to actually turn off
color. Unfortunately, the pager code has no idea which color
variables were turned on (and there are many of them
throughout the code, and they may even have been manipulated
after the colorbool selection by something like "--color" on
the command line).
This bug can be seen any time a pager is started after
config and command line options are checked. This has
affected "git diff" since 89d07f7 (diff: don't run pager if
user asked for a diff style exit code, 2007-08-12). It has
also affect the log family since 1fda91b (Fix 'git log'
early pager startup error case, 2010-08-24).
This patch splits the notion of parsing a colorbool and
actually checking the configuration. The "use_color"
variables now have an additional possible value,
GIT_COLOR_AUTO. Users of the variable should use the new
"want_color()" wrapper, which will lazily determine and
cache the auto-color decision.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The file-scope global variable head_sha1[] was used to hold the object
name of the current HEAD commit (unless we are about to make an initial
commit). Also there is an independent "static int initial_commit".
Fix all the functions on the call-chain that use these two variables to
take a new "(const) struct commit *current_head" argument instead, and
replace their uses, e.g. "if (initial_commit)" becomes "if (!current_head)"
and a reference to "head_sha1" becomes "current_head->object.sha1".
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Usually this function figures out for itself whether stdout
is a tty. However, it has an extra parameter just to allow
git-config to override the auto-detection for its
--get-colorbool option.
Instead of an extra parameter, let's just use a global
variable. This makes calling easier in the common case, and
will make refactoring the colorbool code much simpler.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This lets us store more than just a bit flag for whether we
want color; we can also store whether we want automatic
colors. This can be useful for making the automatic-color
decision closer to the point of use.
This mostly just involves replacing DIFF_OPT_* calls with
manipulations of the flag. The biggest exception is that
calls to DIFF_OPT_TST must check for "o->use_color > 0",
which lets an "unknown" value (i.e., the default) stay at
"no color". In the previous code, a value of "-1" was not
propagated at all.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A malicious server can return ACK with non-existent SHA-1 or not a
commit. lookup_commit() in this case may return NULL. Do not let
fetch-pack crash by accessing NULL address in this case.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This function produces network traffic and should be in fetch-pack. It
has been in commit.c because it needs to iterate (private) graft
list. It can now do so using for_each_commit_graft().
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* mh/check-attr-relative: (29 commits)
test-path-utils: Add subcommand "prefix_path"
test-path-utils: Add subcommand "absolute_path"
git-check-attr: Normalize paths
git-check-attr: Demonstrate problems with relative paths
git-check-attr: Demonstrate problems with unnormalized paths
git-check-attr: test that no output is written to stderr
Rename git_checkattr() to git_check_attr()
git-check-attr: Fix command-line handling to match docs
git-check-attr: Drive two tests using the same raw data
git-check-attr: Add an --all option to show all attributes
git-check-attr: Error out if no pathnames are specified
git-check-attr: Process command-line args more systematically
git-check-attr: Handle each error separately
git-check-attr: Extract a function error_with_usage()
git-check-attr: Introduce a new variable
git-check-attr: Extract a function output_attr()
Allow querying all attributes on a file
Remove redundant check
Remove redundant call to bootstrap_attr_stack()
Extract a function collect_all_attrs()
...
* js/bisect-no-checkout:
bisect: add support for bisecting bare repositories
bisect: further style nitpicks
bisect: replace "; then" with "\n<tab>*then"
bisect: cleanup whitespace errors in git-bisect.sh.
bisect: add documentation for --no-checkout option.
bisect: add tests for the --no-checkout option.
bisect: introduce --no-checkout support into porcelain.
bisect: introduce support for --no-checkout option.
bisect: add tests to document expected behaviour in presence of broken trees.
bisect: use && to connect statements that are deferred with eval.
bisect: move argument parsing before state modification.
* cb/maint-quiet-push:
receive-pack: do not overstep command line argument array
propagate --quiet to send-pack/receive-pack
Conflicts:
Documentation/git-receive-pack.txt
Documentation/git-send-pack.txt
The first paragraph about flag order is no longer true and is
mentioned in git-checkout-index.txt. The rest is also mentioned in
git-checkout-index.txt.
Remove it and keep uptodate document in one place.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* js/ls-tree-error:
Ensure git ls-tree exits with a non-zero exit code if read_tree_recursive fails.
Add a test to check that git ls-tree sets non-zero exit code on error.
* jc/zlib-wrap:
zlib: allow feeding more than 4GB in one go
zlib: zlib can only process 4GB at a time
zlib: wrap deflateBound() too
zlib: wrap deflate side of the API
zlib: wrap inflateInit2 used to accept only for gzip format
zlib: wrap remaining calls to direct inflate/inflateEnd
zlib wrapper: refactor error message formatter
Check if <path> is a valid git-dir or a valid git-file that points
to a valid git-dir.
We want tests to be independent from the fact that a git-dir may
be a git-file. Thus we changed tests to use this feature.
Signed-off-by: Fredrik Gustafsson <iveqy@iveqy.com>
Mentored-by: Jens Lehmann <Jens.Lehmann@web.de>
Mentored-by: Heiko Voigt <hvoigt@hvoigt.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The empty tree passed as common ancestor to merge_trees() when
cherry-picking a parentless commit is allocated on the heap and never
freed. Leaking such a small one-time allocation is not a very big
problem, but now that "git cherry-pick" can cherry-pick multiple
commits it can start to add up.
Avoid the leak by storing the fake tree exactly once in the BSS
section (i.e., use a static). While at it, let's add a test to make
sure cherry-picking multiple parentless commits continues to work.
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The following sequence of commands reveals an issue with error
reporting of relative paths:
$ mkdir sub
$ cd sub
$ git ls-files --error-unmatch ../bbbbb
error: pathspec 'b' did not match any file(s) known to git.
$ git commit --error-unmatch ../bbbbb
error: pathspec 'b' did not match any file(s) known to git.
This bug is visible only if the normalized path (i.e., the relative
path from the repository root) is longer than the prefix.
Otherwise, the code skips over the normalized path and reads from
an unused memory location which still contains a leftover of the
original command line argument.
So instead, use the existing facilities to deal with relative paths
correctly.
Also fix inconsistency between "checkout" and "commit", e.g.
$ cd Documentation
$ git checkout nosuch.txt
error: pathspec 'Documentation/nosuch.txt' did not match...
$ git commit nosuch.txt
error: pathspec 'nosuch.txt' did not match...
by propagating the prefix down the codepath that reports the error.
Signed-off-by: Clemens Buchacher <drizzd@aon.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When running git describe --dirty the index should be refreshed. Previously
the cached index would cause describe to think that the index was dirty when,
in reality, it was just stale.
The issue was exposed by python setuptools which hardlinks files into another
directory when building a distribution.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* js/ls-tree-error:
Ensure git ls-tree exits with a non-zero exit code if read_tree_recursive fails.
Add a test to check that git ls-tree sets non-zero exit code on error.
Currently, revert_or_cherry_pick can fail in two ways. If it
encounters a conflict, it returns a positive number indicating the
intended exit status for the git wrapper to pass on; for all other
errors, it calls die(). The latter behavior is inconsiderate towards
callers, as it denies them the opportunity to recover from errors and
do other things.
After this patch, revert_or_cherry_pick will still return a positive
return value to indicate an exit status for conflicts as before, while
for some other errors, it will print an error message and return -1
instead of die()-ing. The cmd_revert and cmd_cherry_pick are adjusted
to handle the fatal errors by die()-ing themselves.
While the full benefits of this patch will only be seen once all the
"die" calls are replaced with calls to "error", its immediate impact
is to change some "fatal:" messages to say "error:" and to add a new
"fatal: cherry-pick failed" message at the end when the operation
fails.
Inspired-by: Christian Couder <chriscool@tuxfamily.org>
Mentored-by: Jonathan Nieder <jrnieder@gmail.com>
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Introduce a new "git cherry-pick --continue" command which uses the
information in ".git/sequencer" to continue a cherry-pick that stopped
because of a conflict or other error. It works by dropping the first
instruction from .git/sequencer/todo and performing the remaining
cherry-picks listed there, with options (think "-s" and "-X") from the
initial command listed in ".git/sequencer/opts".
So now you can do:
$ git cherry-pick -Xpatience foo..bar
... description conflict in commit moo ...
$ git cherry-pick --continue
error: 'cherry-pick' is not possible because you have unmerged files.
fatal: failed to resume cherry-pick
$ echo resolved >conflictingfile
$ git add conflictingfile && git commit
$ git cherry-pick --continue; # resumes with the commit after "moo"
During the "git commit" stage, CHERRY_PICK_HEAD will aid by providing
the commit message from the conflicting "moo" commit. Note that the
cherry-pick mechanism has no control at this stage, so the user is
free to violate anything that was specified during the first
cherry-pick invocation. For example, if "-x" was specified during the
first cherry-pick invocation, the user is free to edit out the message
during commit time. Note that the "--signoff" option specified at
cherry-pick invocation time is not reflected in the commit message
provided by CHERRY_PICK_HEAD; the user must take care to add
"--signoff" during the "git commit" invocation.
Helped-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Protect the user from forgetting about a pending sequencer operation
by immediately erroring out when an existing cherry-pick or revert
operation is in progress like:
$ git cherry-pick foo
... conflict ...
$ git cherry-pick moo
error: .git/sequencer already exists
hint: A cherry-pick or revert is in progress
hint: Use --reset to forget about it
fatal: cherry-pick failed
A naive version of this would break the following established ways of
working:
$ git cherry-pick foo
... conflict ...
$ git reset --hard # I actually meant "moo" when I said "foo"
$ git cherry-pick moo
$ git cherry-pick foo
... conflict ...
$ git commit # commit the resolution
$ git cherry-pick moo # New operation
However, the previous patches "reset: Make reset remove the sequencer
state" and "revert: Remove sequencer state when no commits are
pending" make sure that this does not happen.
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When cherry-pick or revert is called on a list of commits, and a
conflict encountered somewhere in the middle, the data in
".git/sequencer" is required to continue the operation. However, when
a conflict is encountered in the very last commit, the user will have
to "continue" after resolving the conflict and committing just so that
the sequencer state is removed. This is how the current "rebase -i"
script works as well.
$ git cherry-pick foo..bar
... conflict encountered while picking "bar" ...
$ echo "resolved" >problematicfile
$ git add problematicfile
$ git commit
$ git cherry-pick --continue # This would be a no-op
Change this so that the sequencer state is cleared when a conflict is
encountered in the last commit. Incidentally, this patch makes sure
that some existing tests don't break when features like "--reset" and
"--continue" are implemented later in the series.
A better way to implement this feature is to get the last "git commit"
to remove the sequencer state. However, that requires tighter
coupling between "git commit" and the sequencer, a goal that can be
pursued once the sequencer is made more general.
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Acked-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Many pathnames in a fast-import stream need to be quoted. In
particular:
1. Pathnames at the end of an "M" or "D" line need quoting
if they contain a LF or start with double-quote.
2. Pathnames on a "C" or "R" line need quoting as above,
but also if they contain spaces.
For (1), we weren't quoting at all. For (2), we put
double-quotes around the paths to handle spaces, but ignored
the possibility that they would need further quoting.
This patch checks whether each pathname needs c-style
quoting, and uses it. This is slightly overkill for (1),
which doesn't actually need to quote many characters that
vanilla c-style quoting does. However, it shouldn't hurt, as
any implementation needs to be ready to handle quoted
strings anyway.
In addition to adding a test, we have to tweak a test which
blindly assumed that case (2) would always use
double-quotes, whether it needed to or not.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Normalize the path arguments (relative to the working tree root, if
applicable) before looking up their attributes. This requires passing
the prefix down the call chain.
This fixes two test cases for different reasons:
* "unnormalized paths" is fixed because the .gitattribute-file-seeking
code is not confused into reading the top-level file twice.
* "relative paths" is fixed because the canonical pathnames are passed
to get_check_attr() or get_all_attrs(), allowing them to match the
pathname patterns as expected.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Suggested by: Junio Hamano <gitster@pobox.com>
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
According to the git-check-attr synopsis, if the '--stdin' option is
used then no pathnames are expected on the command line. Change the
behavior to match this description; namely, if '--stdin' is used but
not '--', then treat all command-line arguments as attribute names.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add new usage patterns
git check-attr [-a | --all] [--] pathname...
git check-attr --stdin [-a | --all] < <list-of-paths>
which display all attributes associated with the specified file(s).
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If no pathnames are passed as command-line arguments and the --stdin
option is not specified, fail with the error message "No file
specified". Add tests of this behavior.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Avoid reusing variable "doubledash" to mean something other than the
expected "position of a double-dash, if any".
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
To explicitly remove the sequencer state for a fresh cherry-pick or
revert invocation, introduce a new subcommand called "--reset" to
remove the sequencer state.
Take the opportunity to publicly expose the sequencer paths, and a
generic function called "remove_sequencer_state" that various git
programs can use to remove the sequencer state in a uniform manner;
"git reset" uses it later in this series. Introducing this public API
is also in line with our long-term goal of eventually factoring out
functions from revert.c into a generic commit sequencer.
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Apart from its central objective of calling into the picking
mechanism, pick_commits creates a sequencer directory, prepares a todo
list, and even acts upon the "--reset" subcommand. This makes for a
bad API since the central worry of callers is to figure out whether or
not any conflicts were encountered during the cherry picking. The
current API is like:
if (pick_commits(opts) < 0)
print "Something failed, we're not sure what"
So, change pick_commits so that it's only responsible for picking
commits in a loop and reporting any errors, leaving the rest to a new
function called pick_revisions. Consequently, the API of pick_commits
becomes much clearer:
act_on_subcommand(opts->subcommand);
todo_list = prepare_todo_list();
if (pick_commits(todo_list, opts) < 0)
print "Error encountered while picking commits"
Now, callers can easily call-in to the cherry-picking machinery by
constructing an arbitrary todo list along with some options.
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In the same spirit as ".git/sequencer/head" and ".git/sequencer/todo",
introduce ".git/sequencer/opts" to persist the replay_opts structure
for continuing after a conflict resolution. Use the gitconfig format
for this file so that it looks like:
[options]
signoff = true
record-origin = true
mainline = 1
strategy = recursive
strategy-option = patience
strategy-option = ours
Helped-by: Jonathan Nieder <jrnieder@gmail.com>
Helped-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Ever since v1.7.2-rc1~4^2~7 (revert: allow cherry-picking more than
one commit, 2010-06-02), a single invocation of "git cherry-pick" or
"git revert" can perform picks of several individual commits. To
implement features like "--continue" to continue the whole operation,
we will need to store some information about the state and the plan at
the beginning. Introduce a ".git/sequencer/head" file to store this
state, and ".git/sequencer/todo" file to store the plan. The head
file contains the SHA-1 of the HEAD before the start of the operation,
and the todo file contains an instruction sheet whose format is
inspired by the format of the "rebase -i" instruction sheet. As a
result, a typical todo file looks like:
pick 8537f0e submodule add: test failure when url is not configured
pick 4d68932 submodule add: allow relative repository path
pick f22a17e submodule add: clean up duplicated code
pick 59a5775 make copy_ref globally available
Since SHA-1 hex is abbreviated using an find_unique_abbrev(), it is
unambiguous. This does not guarantee that there will be no ambiguity
when more objects are added to the repository.
These two files alone are not enough to implement a "--continue" that
remembers the command-line options specified; later patches in the
series save them too.
These new files are unrelated to the existing .git/CHERRY_PICK_HEAD,
which will still be useful while committing after a conflict
resolution.
Inspired-by: Christian Couder <chriscool@tuxfamily.org>
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "--ff" command-line option cannot be used with some other
command-line options. However, parse_args still parses these
incompatible options into a replay_opts structure for use by the rest
of the program. Although pick_commits, the current gatekeeper to the
cherry-pick machinery, checks the validity of the replay_opts
structure before before starting its operation, there will be multiple
entry points to the cherry-pick machinery in future. To futureproof
the code and catch these errors in one place, make sure that an
invalid replay_opts structure is not created by parse_args in the
first place. We still check the replay_opts structure for validity in
pick_commits, but this is an assert() now to emphasize that it's the
caller's responsibility to get it right.
Inspired-by: Christian Couder <chriscool@tuxfamily.org>
Mentored-by: Jonathan Nieder <jrnieder@gmail.com>
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Currently, revert_or_cherry_pick sets up a default git config, parses
command-line arguments, before preparing to pick commits. This makes
for a bad API as the central worry of callers is to assert whether or
not a conflict occured while cherry picking. The current API is like:
if (revert_or_cherry_pick(argc, argv, opts) < 0)
print "Something failed, we're not sure what"
Simplify and rename revert_or_cherry_pick to pick_commits so that it
only has the responsibility of setting up the revision walker and
picking commits in a loop. Transfer the remaining work to its
callers. Now, the API is simplified as:
if (parse_args(argc, argv, opts) < 0)
print "Can't parse arguments"
if (pick_commits(opts) < 0)
print "Error encountered in picking machinery"
Later in the series, pick_commits will also serve as the starting
point for continuing a cherry-pick or revert.
Inspired-by: Christian Couder <chriscool@tuxfamily.org>
Helped-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The current code uses a set of file-scope static variables to tell the
cherry-pick/ revert machinery how to replay the changes, and
initializes them by parsing the command-line arguments. In later
steps in this series, we would like to introduce an API function that
calls into this machinery directly and have a way to tell it what to
do. Hence, introduce a structure to group these variables, so that
the API can take them as a single replay_options parameter. The only
exception is the variable "me" -- remove it since it not an
independent option, and can be inferred from the action.
Unfortunately, this patch introduces a minor regression. Parsing
strategy-option violates a C89 rule: Initializers cannot refer to
variables whose address is not known at compile time. Currently, this
rule is violated by some other parts of Git as well, and it is
possible to get GCC to report these instances using the "-std=c89
-pedantic" option.
Inspired-by: Christian Couder <chriscool@tuxfamily.org>
Mentored-by: Jonathan Nieder <jrnieder@gmail.com>
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Functions which act on commits currently rely on a file-scope static
variable to be set before they're called. Consequently, the API and
corresponding callsites are ugly and unclear. Remove this variable
and change their API to accept the commit to act on as additional
argument so that the callsites change from looking like
commit = prepare_a_commit();
act_on_commit();
to looking like
commit = prepare_a_commit();
act_on_commit(commit);
This change is also in line with our long-term goal of exposing some
of these functions through a public API.
Inspired-by: Christian Couder <chriscool@tuxfamily.org>
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "-x" command-line option is used to record the name of the
original commits being picked in the commit message. The variable
corresponding to this option is named "no_replay" for historical
reasons; the name is especially confusing because the term "replay" is
used to describe what cherry-pick does (for example, in the
documentation of the "--mainline" option). So, give the variable
corresponding to the "-x" command-line option a better name:
"record_origin".
Mentored-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The only place get_encoding uses the global "commit" variable is when
writing an error message explaining that its lone argument was NULL.
Since the function's only caller ensures that a NULL argument isn't
passed, we can remove this check with two beneficial consequences:
1. Since the function doesn't use the global "commit" variable any
more, it won't need to change when we eliminate the global variable
later in the series.
2. Translators no longer need to localize an error message that will
never be shown.
Suggested-by: Junio C Hamano <gitster@pobox.com>
Mentored-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The add_message_to_msg function has some dead code, an unclear API,
only one callsite. While it originally intended fill up an empty
commit message with the commit object name while picking, it really
doesn't do this -- a bug introduced in v1.5.1-rc1~65^2~2 (Make
git-revert & git-cherry-pick a builtin, 2007-03-01). Today, tests in
t3505-cherry-pick-empty.sh indicate that not filling up an empty
commit message is the desired behavior. Re-implement and inline the
function accordingly, with a beneficial side-effect: don't dereference
a NULL pointer when the commit doesn't have a delimeter after the
header.
Helped-by: Junio C Hamano <gitster@pobox.com>
Mentored-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Enable future callers to report a conflict and not die immediately by
introducing a new function called error_resolve_conflict.
Re-implement die_resolve_conflict as a call to error_resolve_conflict
followed by a call to die. Consequently, the message printed by
die_resolve_conflict changes from
fatal: 'commit' is not possible because you have unmerged files.
Please, fix them up in the work tree ...
...
to
error: 'commit' is not possible because you have unmerged files.
hint: Fix them up in the work tree ...
hint: ...
fatal: Exiting because of an unresolved conflict.
Hints are printed using the same advise function introduced in
v1.7.3-rc0~26^2~3 (Introduce advise() to print hints, 2010-08-11).
Inspired-by: Christian Couder <chistian.couder@gmail.com>
Helped-by: Jonathan Nieder <jrnieder@gmail.com>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Ramkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If --no-checkout is specified, then the bisection process uses:
git update-ref --no-deref HEAD <trial>
at each trial instead of:
git checkout <trial>
Improved-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Jon Seymour <jon.seymour@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In order to do partial commits, git-commit overlays a tree on the
cache and checks pathspecs against the result. Currently, the
overlaying is done using "prefix" which prevents relative pathspecs
with ".." and absolute pathspec from matching when they refer to
files not under "prefix" and absent from the index, but still in
the tree (i.e. files staged for removal).
The point of providing a prefix at all is performance optimization.
If we say there is no common prefix for the files of interest, then
we have to read the entire tree into the index.
But even if we cannot use the working directory as a prefix, we can
still figure out if there is a common prefix for all given paths,
and use that instead. The pathspec_prefix() routine from ls-files.c
does exactly that.
Any use of global variables is removed from pathspec_prefix() so
that it can be called from commit.c.
Reported-by: Reuben Thomas <rrt@sc3d.org>
Analyzed-by: Michael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: Clemens Buchacher <drizzd@aon.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Take long option names for -A (--after-context), -B (--before-context)
and -C (--context) from GNU grep and add a similar long option name
for -W (--function-context).
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a new option, -W, to show the whole surrounding function of a match.
It uses the same regular expressions as -p and diff to find the beginning
of sections.
Currently it will not display comments in front of a function, but those
that are following one. Despite this shortcoming it is already useful,
e.g. to simply see a more complete applicable context or to extract whole
functions.
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jk/clone-detached:
clone: always fetch remote HEAD
make copy_ref globally available
consider only branches in guess_remote_head
t: add tests for cloning remotes with detached HEAD
* sr/transport-helper-fix: (21 commits)
transport-helper: die early on encountering deleted refs
transport-helper: implement marks location as capability
transport-helper: Use capname for refspec capability too
transport-helper: change import semantics
transport-helper: update ref status after push with export
transport-helper: use the new done feature where possible
transport-helper: check status code of finish_command
transport-helper: factor out push_update_refs_status
fast-export: support done feature
fast-import: introduce 'done' command
git-remote-testgit: fix error handling
git-remote-testgit: only push for non-local repositories
remote-curl: accept empty line as terminator
remote-helpers: export GIT_DIR variable to helpers
git_remote_helpers: push all refs during a non-local export
transport-helper: don't feed bogus refs to export push
git-remote-testgit: import non-HEAD refs
t5800: document some non-functional parts of remote helpers
t5800: use skip_all instead of prereq
t5800: factor out some ref tests
...
The reflog manpage says:
git reflog [show] [log-options] [<ref>]
the subcommand 'show' is the default "in the absence of any
subcommands". Currently this is only true if the user provided either
at least one option or no additional argument at all. For example:
git reflog master
won't work. Change this by actually calling cmd_log_reflog in
absence of any subcommand.
Signed-off-by: Michael Schubert <mschub@elegosoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Currently, git push --quiet produces some non-error output, e.g.:
$ git push --quiet
Unpacking objects: 100% (3/3), done.
Add the --quiet option to send-pack/receive-pack and pass it to
unpack-objects in the receive-pack codepath and to receive-pack in
the push codepath.
This fixes a bug reported for the fedora git package:
https://bugzilla.redhat.com/show_bug.cgi?id=725593
Reported-by: Jesse Keating <jkeating@redhat.com>
Cc: Todd Zullinger <tmz@pobox.com>
Signed-off-by: Clemens Buchacher <drizzd@aon.at>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In the case of a corrupt repository, git ls-tree may report an error but
presently it exits with a code of 0.
This change uses the return code of read_tree_recursive instead.
Improved-by: Jens Lehmann <Jens.Lehmann@web.de>
Signed-off-by: Jon Seymour <jon.seymour@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jk/tag-contains-ab:
Revert clock-skew based attempt to optimize tag --contains traversal
git skew: a tool to find how big a clock skew exists in the history
default core.clockskew variable to one day
limit "contains" traversals based on commit timestamp
tag: speed up --contains calculation
The reset command creates its reflog entry from argv.
However, it does so after having run parse_options, which
means the only thing left in argv is any non-option
arguments. Thus you would end up with confusing reflog
entries like:
$ git reset --hard HEAD^
$ git reset --soft HEAD@{1}
$ git log -2 -g --oneline
8e46cad HEAD@{0}: HEAD@{1}: updating HEAD
1eb9486 HEAD@{1}: HEAD^: updating HEAD
However, we must also consider that some scripts may set
GIT_REFLOG_ACTION before calling reset, and we need to show
their reflog action (with our text appended). For example:
rebase -i (squash): updating HEAD
On top of that, we also set the ORIG_HEAD reflog action
(even though it doesn't generally exist). In that case, the
reset argument is somewhat meaningless, as it has nothing to
do with what's in ORIG_HEAD.
This patch changes the reset reflog code to show:
$GIT_REFLOG_ACTION: updating {HEAD,ORIG_HEAD}
as before, but only if GIT_REFLOG_ACTION is set. Otherwise,
show:
reset: moving to $rev
for HEAD, and:
reset: updating ORIG_HEAD
for ORIG_HEAD (this is still somewhat superfluous, since we
are in the ORIG_HEAD reflog, obviously, but at least we now
mention which command was used to update it).
While we're at it, we can clean up the code a bit:
- Use strbufs to make the message.
- Use the "rev" parameter instead of showing all options.
This makes more sense, since it is the only thing
impacting the writing of the ref.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If fast-export is being used to generate a fast-import stream that
will be used afterwards it is desirable to indicate the end of the
stream with the new 'done' command.
Add a flag that causes fast-export to end with 'done'.
Signed-off-by: Sverre Rabbelier <srabbelier@gmail.com>
Acked-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jc/index-pack:
verify-pack: use index-pack --verify
index-pack: show histogram when emulating "verify-pack -v"
index-pack: start learning to emulate "verify-pack -v"
index-pack: a miniscule refactor
index-pack --verify: read anomalous offsets from v2 idx file
write_idx_file: need_large_offset() helper function
index-pack: --verify
write_idx_file: introduce a struct to hold idx customization options
index-pack: group the delta-base array entries also by type
Conflicts:
builtin/verify-pack.c
cache.h
sha1_file.c
* jk/clone-cmdline-config:
clone: accept config options on the command line
config: make git_config_parse_parameter a public function
remote: use new OPT_STRING_LIST
parse-options: add OPT_STRING_LIST helper
* jc/zlib-wrap:
zlib: allow feeding more than 4GB in one go
zlib: zlib can only process 4GB at a time
zlib: wrap deflateBound() too
zlib: wrap deflate side of the API
zlib: wrap inflateInit2 used to accept only for gzip format
zlib: wrap remaining calls to direct inflate/inflateEnd
zlib wrapper: refactor error message formatter
Conflicts:
sha1_file.c
Because "diff --cached HEAD" showed an incorrect blob object name on the
LHS of the diff, we ended up updating the index entry with bogus value,
not what we read from the tree.
Noticed by John Nowak.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Change upload-pack and receive-pack to use the namespace-prefixed refs
when working with the repository, and use the unprefixed refs when
talking to the client, maintaining the masquerade. This allows
clone, pull, fetch, and push to work with a suitably configured
GIT_NAMESPACE.
receive-pack advertises refs outside the current namespace as .have refs
(as it currently does for refs in alternates), so that the client can
use them to minimize data transfer but will otherwise ignore them.
With appropriate configuration, this also allows http-backend to expose
namespaces as multiple repositories with different paths. This only
requires setting GIT_NAMESPACE, which http-backend passes through to
upload-pack and receive-pack.
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Jamey Sharp <jamey@minilop.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This optimizes the "recency order" (see pack-heuristics.txt in
Documentation/technical/ directory) used to order objects within a
packfile in three ways:
- Commits at the tip of tags are written together, in the hope that
revision traversal done in incremental fetch (which starts by
putting them in a revision queue marked as UNINTERESTING) will see a
better locality of these objects;
- In the original recency order, trees and blobs are intermixed. Write
trees together before blobs, in the hope that this will improve
locality when running pathspec-limited revision traversal, i.e.
"git log paths...";
- When writing blob objects out, write the whole family of blobs that use
the same delta base object together, by starting from the root of the
delta chain, and writing its immediate children in a width-first
manner, in the hope that this will again improve locality when reading
blobs that belong to the same path, which are likely to be deltified
against each other.
I tried various workloads in the Linux kernel repositories (HEAD at
v3.0-rc6-71-g4dd1b49) packed with v1.7.6 and with this patch, counting how
large seeks are needed between adjacent accesses to objects in the pack,
and the result looks promising. The history has 2072052 objects, weighing
some 490MiB.
* Simple commit-only log.
$ git log >/dev/null
There are 254656 commits in total.
v1.7.6 with patch
Total number of access : 258,031 258,032
0.0% percentile : 12 12
10.0% percentile : 259 259
20.0% percentile : 294 294
30.0% percentile : 326 326
40.0% percentile : 363 363
50.0% percentile : 415 415
60.0% percentile : 513 513
70.0% percentile : 857 858
80.0% percentile : 10,434 10,441
90.0% percentile : 91,985 91,996
95.0% percentile : 260,852 260,885
99.0% percentile : 1,150,680 1,152,811
99.9% percentile : 3,148,435 3,148,435
Less than 2MiB seek: 99.70% 99.69%
95% of the pack accesses look at data that is no further than 260kB
from the previous location we accessed. The patch does not change the
order of commit objects very much, and the result is very similar.
* Pathspec-limited log.
$ git log drivers/net >/dev/null
The path is touched by 26551 commits and merges (among 254656 total).
v1.7.6 with patch
Total number of access : 559,511 558,663
0.0% percentile : 0 0
10.0% percentile : 182 167
20.0% percentile : 259 233
30.0% percentile : 357 304
40.0% percentile : 714 485
50.0% percentile : 5,046 3,976
60.0% percentile : 688,671 443,578
70.0% percentile : 319,574,732 110,370,100
80.0% percentile : 361,647,599 123,707,229
90.0% percentile : 393,195,669 128,947,636
95.0% percentile : 405,496,875 131,609,321
99.0% percentile : 412,942,470 133,078,115
99.5% percentile : 413,172,266 133,163,349
99.9% percentile : 413,354,356 133,240,445
Less than 2MiB seek: 61.71% 62.87%
With the current pack heuristics, more than 30% of accesses have to
seek further than 300MB; the updated pack heuristics ensures that less
than 0.1% of accesses have to seek further than 135MB. This is largely
due to the fact that the updated heuristics does not mix blobs and
trees together.
* Blame.
$ git blame drivers/net/ne.c >/dev/null
The path is touched by 34 commits and merges.
v1.7.6 with patch
Total number of access : 178,147 178,166
0.0% percentile : 0 0
10.0% percentile : 142 139
20.0% percentile : 222 194
30.0% percentile : 373 300
40.0% percentile : 1,168 837
50.0% percentile : 11,248 7,334
60.0% percentile : 305,121,284 106,850,130
70.0% percentile : 361,427,854 123,709,715
80.0% percentile : 388,127,343 128,171,047
90.0% percentile : 399,987,762 130,200,707
95.0% percentile : 408,230,673 132,174,308
99.0% percentile : 412,947,017 133,181,160
99.5% percentile : 413,312,798 133,220,425
99.9% percentile : 413,352,366 133,269,051
Less than 2MiB seek: 56.47% 56.83%
The result is very similar to the pathspec-limited log above, which
only looks at the tree objects.
* Packing recent history.
$ (git for-each-ref --format='^%(refname)' refs/tags; echo HEAD) |
git pack-objects --revs --stdout >/dev/null
This should pack data worth 71 commits.
v1.7.6 with patch
Total number of access : 11,511 11,514
0.0% percentile : 0 0
10.0% percentile : 48 47
20.0% percentile : 134 98
30.0% percentile : 332 178
40.0% percentile : 1,386 293
50.0% percentile : 8,030 478
60.0% percentile : 33,676 1,195
70.0% percentile : 147,268 26,216
80.0% percentile : 9,178,662 464,598
90.0% percentile : 67,922,665 965,782
95.0% percentile : 87,773,251 1,226,102
99.0% percentile : 98,011,763 1,932,377
99.5% percentile : 100,074,427 33,642,128
99.9% percentile : 105,336,398 275,772,650
Less than 2MiB seek: 77.09% 99.04%
The long-tail part of the result looks worse with the patch, but
the change helps majority of the access. 99.04% of the accesses
need less than 2MiB of seeking, compared to 77.09% with the current
packing heuristics.
* Index pack.
$ git index-pack -v .git/objects/pack/pack*.pack
v1.7.6 with patch
Total number of access : 2,791,228 2,788,802
0.0% percentile : 9 9
10.0% percentile : 140 89
20.0% percentile : 233 167
30.0% percentile : 322 235
40.0% percentile : 464 310
50.0% percentile : 862 423
60.0% percentile : 2,566 686
70.0% percentile : 25,827 1,498
80.0% percentile : 1,317,862 4,971
90.0% percentile : 11,926,385 119,398
95.0% percentile : 41,304,149 952,519
99.0% percentile : 227,613,070 6,709,650
99.5% percentile : 321,265,121 11,734,871
99.9% percentile : 382,919,785 33,155,191
Less than 2MiB seek: 81.73% 96.92%
As the index-pack command already walks objects in the delta chain
order, writing the blobs out in the delta chain order seems to
drastically improve the locality of access.
Note that a half-a-gigabyte packfile comfortably fits in the buffer cache,
and you would unlikely to see much performance difference on a modern and
reasonably beefy machine with enough memory and local disks. Benchmarking
with cold cache (or over NFS) would be interesting.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* commit 'v1.7.6': (3211 commits)
Git 1.7.6
completion: replace core.abbrevguard to core.abbrev
Git 1.7.6-rc3
Documentation: git diff --check respects core.whitespace
gitweb: 'pickaxe' and 'grep' features requires 'search' to be enabled
t7810: avoid unportable use of "echo"
plug a few coverity-spotted leaks
builtin/gc.c: add missing newline in message
tests: link shell libraries into valgrind directory
t/Makefile: pass test opts to valgrind target properly
sh-i18n--envsubst.c: do not #include getopt.h
Fix typo: existant->existent
Git 1.7.6-rc2
gitweb: do not misparse nonnumeric content tag files that contain a digit
Git 1.7.6-rc1
fetch: do not leak a refspec
t3703: skip more tests using colons in file names on Windows
gitweb: Fix usability of $prevent_xss
gitweb: Move "Requirements" up in gitweb/INSTALL
gitweb: Describe CSSMIN and JSMIN in gitweb/INSTALL
...
Use the value from 'core.abbrev' configuration variable unless user
specifies the length on command line when showing commit object name
in "branch -v" output.
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
> As you probably guessed from the specificity of the number, I wrote a
> short program to actually traverse and find the worst skew. It takes
> about 5 seconds to run (unsurprisingly, since it is doing the same full
> traversal that we end up doing in the above numbers). So we could
> "autoskew" by setting up the configuration on clone, and then
> periodically updating it as part of "git gc".
This patch doesn't implement auto-detection of skew, but is the program
I used to calculate, and would provide the basis for such
auto-detection. It would be interesting to see average skew numbers for
popular repositories. You can run it as "git skew --all".
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jc/streaming:
sha1_file: use the correct type (ssize_t, not size_t) for read-style function
streaming: read loose objects incrementally
sha1_file.c: expose helpers to read loose objects
streaming: read non-delta incrementally from a pack
streaming_write_entry(): support files with holes
convert: CRLF_INPUT is a no-op in the output codepath
streaming_write_entry(): use streaming API in write_entry()
streaming: a new API to read from the object store
write_entry(): separate two helper functions out
unpack_object_header(): make it public
sha1_object_info_extended(): hint about objects in delta-base cache
sha1_object_info_extended(): expose a bit more info
packed_object_info_detail(): do not return a string
Since 03feddd (git-check-ref-format: reject funny ref names, 2005-10-13),
"git branch -d" can take more than one branch names to remove.
The documentation was correct, but the usage string was not.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Clone does all of init, "remote add", fetch, and checkout
without giving the user a chance to intervene and set any
configuration. This patch allows you to set config options
in the newly created repository after the clone, but before
we do any other operations.
In many cases, this is a minor convenience over something
like:
git clone git://...
git config core.whatever true
But in some cases, it can bring extra efficiency by changing
how the fetch or checkout work. For example, setting
line-ending config before the checkout avoids having to
re-checkout all of the contents with the correct line
endings.
It also provides a mechanism for passing information to remote
helpers during a clone; the helpers may read the git config
to influence how they operate.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Some tar filters may be very expensive to run, so sites do
not want to expose them via upload-archive. This patch lets
users configure tar.<filter>.remote to turn them off.
By default, gzip filters are left on, as they are about as
expensive as creating zip archives.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The process for guessing an archive output format based on
the filename is something like this:
a. parse --output in cmd_archive; check the filename
against a static set of mapping heuristics (right now
it just matches ".zip" for zip files).
b. if found, stick a fake "--format=zip" at the beginning
of the arguments list (if the user did specify a
--format manually, the later option will override our
fake one)
c. if it's a remote call, ship the arguments to the remote
(including the fake), which will call write_archive on
their end
d. if it's local, ship the arguments to write_archive
locally
There are two problems:
1. The set of mappings is static and at too high a level.
The write_archive level is going to check config for
user-defined formats, some of which will specify
extensions. We need to delay lookup until those are
parsed, so we can match against them.
2. For a remote archive call, our set of mappings (or
formats) may not match the remote side's. This is OK in
practice right now, because all versions of git
understand "zip" and "tar". But as new formats are
added, there is going to be a mismatch between what the
client can do and what the remote server can do.
To fix (1), this patch refactors the location guessing to
happen at the write_archive level, instead of the
cmd_archive level. So instead of sticking a fake --format
field in the argv list, we actually pass a "name hint" down
the callchain; this hint is used at the appropriate time to
guess the format (if one hasn't been given already).
This patch leaves (2) unfixed. The name_hint is converted to
a "--format" option as before, and passed to the remote.
This means the local side's idea of how extensions map to
formats will take precedence.
Another option would be to pass the name hint to the remote
side and let the remote choose. This isn't a good idea for
two reasons:
1. There's no room in the protocol for passing that
information. We can pass a new argument, but older
versions of git on the server will choke on it.
2. Letting the remote side decide creates a silent
inconsistency in user experience. Consider the case
that the locally installed git knows about the "tar.gz"
format, but a remote server doesn't.
Running "git archive -o foo.tar.gz" will use the tar.gz
format. If we use --remote, and the local side chooses
the format, then we send "--format=tar.gz" to the
remote, which will complain about the unknown format.
But if we let the remote side choose the format, then
it will realize that it doesn't know about "tar.gz" and
output uncompressed tar without even issuing a warning.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Until now, "git tag -l foo* bar*" would silently ignore the
second argument, showing only refs starting with "foo". It's
not just unfriendly not to take a second pattern; we
actually generated subtly wrong results (from the user's
perspective) because some of the requested tags were
omitted.
This patch allows an arbitrary number of patterns on the
command line; if any of them matches, the ref is shown.
While we're tweaking the documentation, let's also make it
clear that the pattern is fnmatch.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This is the slop value used by name-rev, so presumably is a
reasonable default.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When looking for commits that contain other commits (e.g.,
via "git tag --contains"), we can end up traversing useless
portions of the graph. For example, if I am looking for a
tag that contains a commit made last week, there is not much
point in traversing portions of the history graph made five
years ago.
This optimization can provide massive speedups. For example,
doing "git tag --contains HEAD~200" in the linux-2.6
repository goes from:
real 0m5.302s
user 0m5.116s
sys 0m0.184s
to:
real 0m0.030s
user 0m0.020s
sys 0m0.008s
The downside is that we will no longer find some answers in
the face of extreme clock skew, as we will stop the
traversal early when seeing commits skewed too far into the
past.
Name-rev already implements a similar optimization, using a
"slop" of one day to allow for a certain amount of clock
skew in commit timestamps. This patch introduces a
"core.clockskew" variable, which allows specifying the
allowable amount of clock skew in seconds. For safety, it
defaults to "none", causing a full traversal (i.e., no
change in behavior from previous versions).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When we want to know if commit A contains commit B (or any
one of a set of commits, B through Z), we generally
calculate the merge bases and see if B is a merge base of A
(or for a set, if any of the commits B through Z have that
property).
When we are going to check a series of commits A1 through An
to see whether each contains B (e.g., because we are
deciding which tags to show with "git tag --contains"), we
do a series of merge base calculations. This can be very
expensive, as we repeat a lot of traversal work.
Instead, let's leverage the fact that we are going to use
the same --contains list for each tag, and mark areas of the
commit graph is definitely containing those commits, or
definitely not containing those commits. Later tags can then
stop traversing as soon as they see a previously calculated
answer.
This sped up "git tag --contains HEAD~200" in the linux-2.6
repository from:
real 0m15.417s
user 0m15.197s
sys 0m0.220s
to:
real 0m5.329s
user 0m5.144s
sys 0m0.184s
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The size of objects we read from the repository and data we try to put
into the repository are represented in "unsigned long", so that on larger
architectures we can handle objects that weigh more than 4GB.
But the interface defined in zlib.h to communicate with inflate/deflate
limits avail_in (how many bytes of input are we calling zlib with) and
avail_out (how many bytes of output from zlib are we ready to accept)
fields effectively to 4GB by defining their type to be uInt.
In many places in our code, we allocate a large buffer (e.g. mmap'ing a
large loose object file) and tell zlib its size by assigning the size to
avail_in field of the stream, but that will truncate the high octets of
the real size. The worst part of this story is that we often pass around
z_stream (the state object used by zlib) to keep track of the number of
used bytes in input/output buffer by inspecting these two fields, which
practically limits our callchain to the same 4GB limit.
Wrap z_stream in another structure git_zstream that can express avail_in
and avail_out in unsigned long. For now, just die() when the caller gives
a size that cannot be given to a single zlib call. In later patches in the
series, we would make git_inflate() and git_deflate() internally loop to
give callers an illusion that our "improved" version of zlib interface can
operate on a buffer larger than 4GB in one go.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Wrap deflateInit, deflate, and deflateEnd for everybody, and the sole use
of deflateInit2 in remote-curl.c to tell the library to use gzip header
and trailer in git_deflate_init_gzip().
There is only one caller that cares about the status from deflateEnd().
Introduce git_deflate_end_gently() to let that sole caller retrieve the
status and act on it (i.e. die) for now, but we would probably want to
make inflate_end/deflate_end die when they ran out of memory and get
rid of the _gently() kind.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In most cases, fetching the remote HEAD explicitly is
unnecessary. It's just a symref pointing to a branch which
we are already fetching, so we will already ask for its sha1.
However, if the remote has a detached HEAD, things are less
certain. We do not ask for HEAD's sha1, but we do try to
write it into a local detached HEAD. In most cases this is
fine, as the remote HEAD is pointing to some part of the
history graph that we will fetch via the refs.
But if the remote HEAD points to an "orphan" commit (one
which was is not an ancestor of any refs), then we will not
have the object, and update_ref will complain when we try to
write the detached HEAD, aborting the whole clone.
This patch makes clone always explicitly ask the remote for
the sha1 of its HEAD commit. In the non-detached case, this
is a no-op, as we were going to ask for that sha1 anyway. In
the regular detached case, this will add an extra "want" to
the protocol negotiation, but will not change the history
that gets sent. And in the detached orphan case, we will
fetch the orphaned history so that we can write it into our
local detached HEAD.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* bc/maint-status-z-to-use-porcelain:
builtin/commit.c: set status_format _after_ option parsing
t7508: demonstrate status's failure to use --porcelain format with -z
Conflicts:
builtin/commit.c
This finally gets rid of the inefficient verify-pack implementation that
walks objects in the packfile in their object name order and replaces it
with a call to index-pack --verify. As a side effect, it also removes
packed_object_info_detail() API which is rather expensive.
As this changes the way errors are reported (verify-pack used to rely on
the usual runtime error detection routine unpack_entry() to diagnose the
CRC errors in an entry in the *.idx file; index-pack --verify checks the
whole *.idx file in one go), update a test that expected the string "CRC"
to appear in the error message.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The histogram produced by "verify-pack -v" always had an artificial
limit of 50, but index-pack knows what the maximum delta depth is, so
we do not have to limit it.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "index-pack" machinery already has almost enough knowledge to produce
the same output as "verify-pack -v". Fill small gaps in its bookkeeping,
and teach it to show what it knows.
Add a few more command line options that do not have to be advertised to
the end users. They will be used internally when verify-pack calls this.
The eventual goal is to remove verify-pack implementation and redo it as a
thin wrapper around the index-pack, so that we can remove the rather
expensive packed_object_info_detail() API.
This still does not do the delta-chain-depth histogram yet but that part
is easy.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Introduce a helper function that takes the type of an object and
tell if it is a delta, as we seem to use this check in many places.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When create a new branch, we fed "refs/heads/<proposed name>" as a string
to get_sha1() and expected it to fail when a branch already exists.
The right way to check if a ref exists is to check with resolve_ref().
A naïve solution that might appear attractive but does not work is to
forbid slashes in get_describe_name() but that will not work. A describe
name is is in the form of "ANYTHING-g<short sha1>", and that ANYTHING part
comes from a original tag name used in the repository the user ran the
describe command. A sick user could have a confusing hierarchical tag
whose name is "refs/heads/foobar" (stored as refs/tags/refs/heads/foobar")
to generate a describe name "refs/heads/foobar-6-g02ac983", and we should
be able to use that name to refer to the object whose name is 02ac983.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
With --heading, the filename is printed once before matches from that
file instead of at the start of each line, giving more screen space to
the actual search results.
This option is taken from ack (http://betterthangrep.com/). And now
git grep can dress up like it:
$ git config alias.ack "grep --break --heading --line-number"
$ git ack -e --heading
Documentation/git-grep.txt
154:--heading::
t/t7810-grep.sh
785:test_expect_success 'grep --heading' '
786: git grep --heading -e char -e lo_w hello.c hello_world >actual &&
808: git grep --break --heading -n --color \
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
With --break, an empty line is printed between matches from different
files, increasing readability. This option is taken from ack
(http://betterthangrep.com/).
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Commit 431d6e7b (grep: enable threading for context line printing)
split the printing of the "--\n" mark between results from different
files out into two places: show_line() in grep.c for the non-threaded
case and work_done() in builtin/grep.c for the threaded case. Commit
55f638bd (grep: Colorize filename, line number, and separator) updated
the former, but not the latter, so the separators between files are
not colored if threads are used.
This patch merges the two. In the threaded case, hunk marks are now
printed by show_line() for every file, including the first one, and the
very first mark is simply skipped in work_done(). This ensures that the
output is properly colored and works just as well.
Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jk/format-patch-am:
format-patch: preserve subject newlines with -k
clean up calling conventions for pretty.c functions
pretty: add pp_commit_easy function for simple callers
mailinfo: always clean up rfc822 header folding
t: test subject handling in format-patch / am pipeline
Conflicts:
builtin/branch.c
builtin/log.c
commit.h
* ab/i18n-fixup: (24 commits)
i18n: use test_i18n{cmp,grep} in t7600, t7607, t7611 and t7811
i18n: use test_i18n{grep,cmp} in t7508
i18n: use test_i18ngrep in t7506
i18n: use test_i18ngrep and test_i18ncmp in t7502
i18n: use test_i18ngrep in t7501
i18n: use test_i18ncmp in t7500
i18n: use test_i18ngrep in t7201
i18n: use test_i18ncmp and test_i18ngrep in t7102 and t7110
i18n: use test_i18ncmp and test_i18ngrep in t5541, t6040, t6120, t7004, t7012 and t7060
i18n: use test_i18ncmp and test_i18ngrep in t3700, t4001 and t4014
i18n: use test_i18ncmp and test_i18ngrep in t3203, t3501 and t3507
i18n: use test_i18ngrep in t2020, t2204, t3030, and t3200
i18n: use test_i18ngrep in lib-httpd and t2019
i18n: do not overuse C_LOCALE_OUTPUT (grep)
i18n: use test_i18ncmp in t1200 and t2200
i18n: .git file is not a human readable message (t5601)
i18n: do not overuse C_LOCALE_OUTPUT
i18n: mark init-db messages for translation
i18n: mark checkout plural warning for translation
i18n: mark checkout --detach messages for translation
...
* jc/rename-degrade-cc-to-c:
diffcore-rename: fall back to -C when -C -C busts the rename limit
diffcore-rename: record filepair for rename src
diffcore-rename: refactor "too many candidates" logic
builtin/diff.c: remove duplicated call to diff_result_code()
* mk/grep-pcre:
git-grep: Fix problems with recently added tests
git-grep: Update tests (mainly for -P)
Makefile: Pass USE_LIBPCRE down in GIT-BUILD-OPTIONS
git-grep: update tests now regexp type is "last one wins"
git-grep: do not die upon -F/-P when grep.extendedRegexp is set.
git-grep: Bail out when -P is used with -F or -E
grep: Add basic tests
configure: Check for libpcre
git-grep: Learn PCRE
grep: Extract compile_regexp_failed() from compile_regexp()
grep: Fix a typo in a comment
grep: Put calls to fixmatch() and regmatch() into patmatch()
contrib/completion: --line-number to git grep
Documentation: Add --line-number to git-grep synopsis
* jc/notes-batch-removal:
show: --ignore-missing
notes remove: --stdin reads from the standard input
notes remove: --ignore-missing
notes remove: allow removing more than one
* jk/haves-from-alternate-odb:
receive-pack: eliminate duplicate .have refs
bisect: refactor sha1_array into a generic sha1 list
refactor refs_from_alternate_cb to allow passing extra data
The previous commit simply hijacked --quiet and essentially made it into a
no-op. Instead, take it as a cue that the end user wants to omit the patch
output from commands that default to show patches, e.g. "show".
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In older versions of git, we used rfc822 header folding to
indicate that the original subject line had multiple lines
in it. But since a1f6baa (format-patch: wrap long header
lines, 2011-02-23), we now use header folding whenever there
is a long line.
This means that "git am" cannot trust header folding as a
sign from format-patch that newlines should be preserved.
Instead, format-patch needs to signal more explicitly that
the newlines are significant. This patch does so by
rfc2047-encoding the newlines in the subject line. No
changes are needed on the "git am" end; it already decodes
the newlines properly.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We have a pretty_print_context representing the parameters
for a pretty-print session, but we did not use it uniformly.
As a result, functions kept growing more and more arguments.
Let's clean this up in a few ways:
1. All pretty-print pp_* functions now take a context.
This lets us reduce the number of arguments to these
functions, since we were just passing around the
context values separately.
2. The context argument now has a cmit_fmt field, which
was passed around separately. That's one less argument
per function.
3. The context argument always comes first, which makes
calling a little more uniform.
This drops lines from some callers, and adds lines in a few
places (because we need an extra line to set the context's
fmt field). Overall, we don't save many lines, but the lines
that are there are a lot simpler and more readable.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Many callers don't actually care about the pretty print
context at all; let's just give them a simple way of
pretty-printing a commit without having to create a context
struct.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Without the "-k" option, mailinfo will convert a folded
subject header like:
Subject: this is a
subject that doesn't
fit on one line
into a single line. With "-k", however, we assumed that
these newlines were significant and represented something
that the sending side would want us to preserve.
For messages created by format-patch, this assumption was
broken by a1f6baa (format-patch: wrap long header lines,
2011-02-23). For messages sent by arbitrary MUAs, this was
probably never a good assumption to make, as they may have
been folding subjects in accordance with rfc822's line
length recommendations all along.
This patch now joins folded lines with a single whitespace
character. This treats header folding purely as a syntactic
feature of the transport mechanism, not as something that
format-patch is trying to tell us about the original
subject.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Commit 13fc2c1 (remote: disallow some nonsensical option
combinations, 2011-03-30) made it impossible to use "remote
add -t foo --mirror". The argument was that specifying
specific branches is useless because:
1. Push mirrors do not want a refspec at all.
2. The point of fetch mirroring is to use a broad refspec
like "refs/*", but using "-t" overrides that.
Point (1) is valid; "-t" with push mirrors is useless. But
point (2) ignored another side effect of using --mirror: it
fetches the refs directly into the refs/ namespace as they
are found upstream, instead of placing them in a
separate-remote layout.
So 13fc2c1 was overly constrictive, and disallowed
reasonable specific-branch mirroring, like:
git remote add -t heads/foo -t heads/bar --mirror=fetch
which makes the local "foo" and "bar" branches direct
mirrors of the remote, but does not fetch anything else.
This patch restores the original behavior, but only for
fetch mirrors.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jk/git-connection-deadlock-fix:
test core.gitproxy configuration
send-pack: avoid deadlock on git:// push with failed pack-objects
connect: let callers know if connection is a socket
connect: treat generic proxy processes like ssh processes
Conflicts:
connect.c
* js/maint-send-pack-stateless-rpc-deadlock-fix:
sideband_demux(): fix decl-after-stmt
send-pack: unbreak push over stateless rpc
send-pack: avoid deadlock when pack-object dies early
* jk/git-connection-deadlock-fix:
test core.gitproxy configuration
send-pack: avoid deadlock on git:// push with failed pack-objects
connect: let callers know if connection is a socket
connect: treat generic proxy processes like ssh processes
Conflicts:
connect.c
* js/maint-send-pack-stateless-rpc-deadlock-fix:
sideband_demux(): fix decl-after-stmt
send-pack: unbreak push over stateless rpc
send-pack: avoid deadlock when pack-object dies early
* jc/bigfile:
Bigfile: teach "git add" to send a large file straight to a pack
index_fd(): split into two helper functions
index_fd(): turn write_object and format_check arguments into one flag
The option can be used to check if read-tree with the same set of other
options like "-m" and "-u" would succeed without actually changing either
the index or the working tree.
The relevant tests in the t10?? range were extended to do a read-tree -n
before the real read-tree to make sure neither the index nor any local
files were changed with -n and the same exit code as without -n is
returned. The helper functions added for that purpose reside in the new
t/lib-read-tree.sh file.
The only exception is #13 in t1004 ("unlinking an un-unlink-able
symlink"). As this is an issue of wrong directory permissions it is not
detected with -n.
Signed-off-by: Jens Lehmann <Jens.Lehmann@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Ever since the merge command was made multi-strategy aware, we said
Merge made by octopus.
at the end of a session. Reword it to
Merge made by the 'octopus' strategy.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When switching away from a detached HEAD with "git checkout", we give a
listing of the commits about to be lost, and then tell how to resurrect
them since 8e2dc6a (commit: give final warning when reattaching HEAD to
leave commits behind, 2011-02-18).
This is a good safety measure for people who are not comfortable with the
detached HEAD state, but the advice on how to keep the state you just left
was given even to those who set advice.detachedHead to false.
Keep the warning and informational commit listing, but honor the setting
of advice.detachedHead to squelch the advice.
Helped-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jc/magic-pathspec:
setup.c: Fix some "symbol not declared" sparse warnings
t3703: Skip tests using directory name ":" on Windows
revision.c: leave a note for "a lone :" enhancement
t3703, t4208: add test cases for magic pathspec
rev/path disambiguation: further restrict "misspelled index entry" diag
fix overslow :/no-such-string-ever-existed diagnostics
fix overstrict :<path> diagnosis
grep: use get_pathspec() correctly
pathspec: drop "lone : means no pathspec" from get_pathspec()
Revert "magic pathspec: add ":(icase)path" to match case insensitively"
magic pathspec: add ":(icase)path" to match case insensitively
magic pathspec: futureproof shorthand form
magic pathspec: add tentative ":/path/from/top/level" pathspec support
* jc/replacing:
read_sha1_file(): allow selective bypassing of replacement mechanism
inline lookup_replace_object() calls
read_sha1_file(): get rid of read_sha1_file_repl() madness
t6050: make sure we test not just commit replacement
Declare lookup_replace_object() in cache.h, not in commit.h
Conflicts:
environment.c
* jk/git-connection-deadlock-fix:
test core.gitproxy configuration
send-pack: avoid deadlock on git:// push with failed pack-objects
connect: let callers know if connection is a socket
connect: treat generic proxy processes like ssh processes
Conflicts:
connect.c
When receiving a push, we advertise ref tips from any
alternate repositories, in case that helps the client send a
smaller pack. Since these refs don't actually exist in the
destination repository, we don't transmit the real ref
names, but instead use the pseudo-ref ".have".
If your alternate has a large number of duplicate refs (for
example, because it is aggregating objects from many related
repositories, some of which will have the same tags and
branch tips), then we will send each ".have $sha1" line
multiple times. This is a pointless waste of bandwidth, as
we are simply repeating the same fact to the client over and
over.
This patch eliminates duplicate .have refs early on. It does
so efficiently by sorting the complete list and skipping
duplicates. This has the side effect of re-ordering the
.have lines by ascending sha1; this isn't a problem, though,
as the original order was meaningless.
There is a similar .have system in fetch-pack, but it
does not suffer from the same problem. For each alternate
ref we consider in fetch-pack, we actually open the object
and mark it with the SEEN flag, so duplicates are
automatically culled.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The foreach_alt_odb function triggers a callback for each
alternate object db we have, with room for a single void
pointer as data. Currently, we always call refs_from_alternate_cb
as the callback function, and then pass another callback (to
receive each ref individually) as the void pointer.
This has two problems:
1. C technically forbids stuffing a function pointer into
a "void *". In practice, this probably doesn't matter
on any architectures git runs on, but it never hurts to
follow the letter of the law.
2. There is no room for an extra data pointer. Indeed, the
alternate_ref_fn that refs_from_alternate_cb calls
takes a void* for data, but we always pass it NULL.
Instead, let's properly stuff our function pointer into a
data struct, which also leaves room for an extra
caller-supplied data pointer. And to keep things simple for
existing callers, let's make a for_each_alternate_ref
function that takes care of creating the extra struct.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We add every local ref to a list so that we can mark them
and all of their ancestors back to a certain cutoff point.
However, if some refs point to the same commit, we will end
up adding them to the list many times.
Furthermore, since commit_lists are stored as linked lists,
we must do an O(n) traversal of the list in order to find
the right place to insert each commit. This makes building
the list O(n^2) in the number of refs.
For normal repositories, this isn't a big deal. We have a
few hundreds refs at most, and most of them are unique. But
consider an "alternates" repo that serves as an object
database for many other similar repos. For reachability, it
needs to keep a copy of the refs in each child repo. This
means it may have a large number of refs, many of which
point to the same commits.
By noting commits we have already added to the list, we can
shrink the size of "n" in such a repo to the number of
unique commits, which is on the order of what a normal repo
would contain (it's actually more than a normal repo, since child repos
may have branches at different states, but in practice it tends
to be much smaller than the list with duplicates).
Here are the results on one particular giant repo
(containing objects for all Rails forks on GitHub):
$ git for-each-ref | wc -l
112514
[before]
$ git fetch --no-tags ../remote.git
63.52user 0.12system 1:03.68elapsed 99%CPU (0avgtext+0avgdata 137648maxresident)k
1856inputs+48outputs (11major+19603minor)pagefaults 0swaps
$ git fetch --no-tags ../remote.git
6.15user 0.08system 0:06.25elapsed 99%CPU (0avgtext+0avgdata 123856maxresident)k
0inputs+40outputs (0major+18872minor)pagefaults 0swaps
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Instead of barfing, simply ignore bad object names seen in the
input. This is useful when reading from "git notes list" output
that may refer to objects that have already been garbage collected.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach the command to read object names to remove from the standard
input, in addition to the object names given from the command line.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Depending on the application, it is not necessarily an error for an object
to lack a note, especially if the only thing the caller wants to make sure
is that notes are cleared for an object. By passing this option from the
command line, the "git notes remove" command considers it a success if the
object did not have any note to begin with.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
While "xargs -n1 git notes rm" is certainly a possible way to remove notes
from many objects, this would create one notes "commit" per removal, which
is not quite suitable for seasonal housekeeping.
Allow taking more than one on the command line, and record their removal
as a single atomic event if everthing goes well.
Even though the old code insisted that "git notes rm" must be given only
one object (or zero, in which case it would default to HEAD), this
condition was not tested. Add tests to handle the new case where we feed
multiple objects, and also make sure if there is a bad input, no change
is recorded.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "git ls-remote" uses its exit status to indicate if it successfully
talked with the remote repository. A new option "--exit-code" makes the
command exit with status "2" when there is no refs to be listed, even when
the command successfully talked with the remote repository.
This way, the caller can tell if we failed to contact the remote, or the
remote did not have what we wanted to see. Of course, you can inspect the
output from the command, which has been and will continue to be a valid
way to check the same thing.
Signed-off-by: Michael Schubert <mschub@elegosoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add log.abbrevCommit config variable as a convenience for users who
often use --abbrev-commit with git log and friends. Allow the option
to be overridden with --no-abbrev-commit. Per 635530a2fc and 4f62c2bc57,
the config variable is ignored when log is given "--pretty=raw".
(Also, a drive-by spelling correction in git log's short help.)
Signed-off-by: Jay Soffian <jaysoffian@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When trying to set a multivar with "git config var value", "git config"
issues
warning: remote.repoor.push has multiple values
leaving the user under the impression that the operation succeeded,
unless one checks the return value.
Instead, make it
warning: remote.repoor.push has multiple values
error: cannot overwrite multiple values with a single value
Use a regexp, --add or --set-all to change remote.repoor.push.
to be clear and helpful.
Note: The "warning" is raised through other code paths also so that it
needs to remain a warning for these (which do not raise the error). Only
the caller can determine how to go on from that.
Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Instead return an integer that can be given to typename() if
the caller wants a string, just like everybody else does.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* ci/commit--interactive-atomic:
Test atomic git-commit --interactive
Add commit to list of config.singlekey commands
Add support for -p/--patch to git-commit
Allow git commit --interactive with paths
t7501.8: feed a meaningful command
Use a temporary index for git commit --interactive
* mg/merge-ff-config:
tests: check git does not barf on merge.ff values for future versions of git
merge: introduce merge.ff configuration variable
Conflicts:
t/t7600-merge.sh
* jc/maint-add-p-overlapping-hunks:
t3701: add-p-fix makes the last test to pass
"add -p": work-around an old laziness that does not coalesce hunks
add--interactive.perl: factor out repeated --recount option
t3701: Editing a split hunk in an "add -p" session
add -p: 'q' should really quit
Although it is probably an uncommon operation, there is no
reason to disallow it, as it works just fine. It is the
reverse of a cherry-pick of a root commit, which is already
allowed.
We do have to tweak one check on whether we have a merge
commit, which assumed we had at least one parent.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The merge-recursive strategy already handles root commits;
it cherry-picks the difference between the empty tree and
the root commit's tree.
However, for external strategies, we dereference NULL and
segfault while building the argument list. Instead, let's
handle this by passing the empty tree sha1 to the merge
script.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jc/convert:
convert: make it harder to screw up adding a conversion attribute
convert: make it safer to add conversion attributes
convert: give saner names to crlf/eol variables, types and functions
convert: rename the "eol" global variable to "core_eol"
* jc/bigfile:
Bigfile: teach "git add" to send a large file straight to a pack
index_fd(): split into two helper functions
index_fd(): turn write_object and format_check arguments into one flag
* jc/replacing:
read_sha1_file(): allow selective bypassing of replacement mechanism
inline lookup_replace_object() calls
read_sha1_file(): get rid of read_sha1_file_repl() madness
t6050: make sure we test not just commit replacement
Declare lookup_replace_object() in cache.h, not in commit.h
