Convert the callers to pass struct object_id by changing the function
declaration and definition and applying the following semantic patch:
@@
expression E1, E2;
@@
- sha1_array_append(E1, E2.hash)
+ sha1_array_append(E1, &E2)
@@
expression E1, E2;
@@
- sha1_array_append(E1, E2->hash)
+ sha1_array_append(E1, E2)
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We probe the "17/" loose object directory for auto-gc, and
use a local buffer to format the path. We can just use
git_path() for this. It handles paths of any length
(reducing our error handling). And because we feed the
result straight to a system call, we can just use the static
variant.
Note that git_path also knows the string "objects/" is
special, and will replace it with git_object_directory()
when necessary.
Another alternative would be to use sha1_file_name() for the
pretend object "170000...", but that ends up being more
hassle for no gain, as we have to truncate the final path
component.
Signed-off-by: Jeff King <peff@peff.net>
We'd prefer to avoid unchecked snprintf calls because
truncation can lead to unexpected results.
These are all cases where truncation shouldn't ever happen,
because the input to snprintf is fixed in size. That makes
them candidates for xsnprintf(), but it's simpler still to
just use the heap, and then nobody has to wonder if "100" is
big enough.
We'll use xstrfmt() where possible, and a strbuf when we need
the resulting size or to reuse the same buffer in a loop.
Signed-off-by: Jeff King <peff@peff.net>
After receive-pack reads the pack header from the client, it
feeds the already-read part to index-pack and unpack-objects
via their --pack-header command-line options. To do so, we
format it into a fixed buffer, then duplicate it into the
child's argv_array.
Our buffer is long enough to handle any possible input, so
this isn't wrong. But it's more complicated than it needs to
be; we can just argv_array_pushf() the final value and avoid
the intermediate copy. This drops the magic number and is
more efficient, too.
Note that we need to push to the argv_array in order, which
means we can't do the push until we are in the "unpack-objects
versus index-pack" conditional. Rather than duplicate the
slightly complicated format specifier, I pushed it into a
helper function.
Signed-off-by: Jeff King <peff@peff.net>
When name-rev needs to format an actual name, we do so into
a fixed-size buffer. That includes the actual ref tip, as
well as any traversal information. Since refs can exceed
1024 bytes, this means you can get a bogus result. E.g.,
doing:
git tag $(perl -e 'print join("/", 1..1024)')
git describe --contains HEAD^
results in ".../282/283", when it should be
".../1023/1024~1".
We can solve this by using a heap buffer. We'll use a
strbuf, which lets us write into the same buffer from our
loop without having to reallocate.
Signed-off-by: Jeff King <peff@peff.net>
Many functions which handle refs use a PATH_MAX-sized buffer
to do so. This is mostly reasonable as we have to write
loose refs into the filesystem, and at least on Linux the 4K
PATH_MAX is big enough that nobody would care. But:
1. The static PATH_MAX is not always the filesystem limit.
2. On other platforms, PATH_MAX may be much smaller.
3. As we move to alternate ref storage, we won't be bound
by filesystem limits.
Let's convert these to heap buffers so we don't have to
worry about truncation or size limits.
We may want to eventually constrain ref lengths for sanity
and to prevent malicious names, but we should do so
consistently across all platforms, and in a central place
(like the ref code).
Signed-off-by: Jeff King <peff@peff.net>
Part of the reflog content comes from the environment, which
can be much larger than our fixed buffer. Let's use a heap
buffer so we avoid truncating it.
Signed-off-by: Jeff King <peff@peff.net>
We format the tag header into a fixed 1024-byte buffer. But
since the tag-name and tagger ident can be arbitrarily
large, we may unceremoniously die with "tag header too big".
Let's just use a strbuf instead.
Note that it looks at first glance like we can just format
this directly into the "buf" strbuf where it will ultimately
go. But that buffer may already contain the tag message, and
we have no easy way to prepend formatted data to a strbuf
(we can only splice in an already-generated buffer). This
isn't a performance-critical path, so going through an extra
buffer isn't a big deal.
Signed-off-by: Jeff King <peff@peff.net>
"git receive-pack" could have been forced to die by attempting
allocate an unreasonably large amount of memory with a crafted push
certificate; this has been fixed.
* bc/push-cert-receive-fix:
builtin/receive-pack: fix incorrect pointer arithmetic
Some debugging output from "git describe" were marked for l10n,
but some weren't. Mark missing ones for l10n.
* mg/describe-debug-l10n:
l10n: de: translate describe debug terms
describe: localize debug output fully
A few commands that recently learned the "--recurse-submodule"
option misbehaved when started from a subdirectory of the
superproject.
* bw/recurse-submodules-relative-fix:
ls-files: fix bug when recursing with relative pathspec
ls-files: fix typo in variable name
grep: fix bug when recursing with relative pathspec
setup: allow for prefix to be passed to git commands
grep: fix help text typo
"what URL do we want to update this submodule?" and "are we
interested in this submodule?" are split into two distinct
concepts, and then the way used to express the latter got extended,
paving a way to make it easier to manage a project with many
submodules and make it possible to later extend use of multiple
worktrees for a project with submodules.
* bw/submodule-is-active:
submodule add: respect submodule.active and submodule.<name>.active
submodule--helper init: set submodule.<name>.active
clone: teach --recurse-submodules to optionally take a pathspec
submodule init: initialize active submodules
submodule: decouple url and submodule interest
submodule--helper clone: check for configured submodules using helper
submodule sync: use submodule--helper is-active
submodule sync: skip work for inactive submodules
submodule status: use submodule--helper is-active
submodule--helper: add is-active subcommand
Stop supporting "git merge <message> HEAD <commit>" syntax that has
been deprecated since October 2007, and issues a deprecation
warning message since v2.5.0.
* jc/merge-drop-old-syntax:
merge: drop 'git merge <message> HEAD <commit>' syntax
In order to checkout files, difftool reads "diff --raw"
output and feeds the names to checkout_entry(). That
function requires us to have a "struct cache_entry". And
because that struct uses a FLEX_ARRAY for the name field, we
have to actually copy in our new name.
The current code allocates a single re-usable cache_entry
that can hold a name up to PATH_MAX, and then copies
filenames into it using strcpy(). But there's no guarantee
that incoming names are smaller than PATH_MAX. They've come
from "diff --raw" output which might be diffing between two
trees (and hence we'd be subject to the PATH_MAX of some
other system, or even none at all if they were created
directly via "update-index").
We can fix this by using make_cache_entry() to create a
correctly-sized cache_entry for each name. This incurs an
extra allocation per file, but this is negligible compared
to actually writing out the file contents.
To make this simpler, we can push this procedure into a new
helper function. Note that we can also get rid of the "len"
variables for src_path and dst_path (and in fact we must, as
the compiler complains that they are unused).
Signed-off-by: Jeff King <peff@peff.net>
Acked-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
As xgetcwd() returns an allocated buffer, we should free this
buffer when we don't need it any more.
This was found by Coverity.
Reported-by: Jeff King <peff@peff.net>
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git name-rev" assigned a phony "far in the future" date to tips of
refs that are not pointing at tag objects, and favored names based
on a ref with the oldest date. This made it almost impossible for
an unannotated tags and branches to be counted as a viable base,
which was especially problematic when the command is run with the
"--tags" option. If an unannotated tag that points at an ancient
commit and an annotated tag that points at a much newer commit
reaches the commit that is being named, the old unannotated tag was
ignored.
Update the "taggerdate" field of the rev-name structure, which is
initialized from the tip of ref, to have the committer date if the
object at the tip of ref is a commit, not a tag, so that we can
optionally take it into account when doing "is this name better?"
comparison logic.
When "name-rev" is run without the "--tags" option, the general
expectation is still to name the commit based on a tag if possible,
but use non-tag refs as fallback, and tiebreak among these non-tag
refs by favoring names with shorter hops from the tip. The use of a
phony "far in the future" date in the original code was an effective
way to ensure this expectation is held: a non-tag tip gets the same
"far in the future" timestamp, giving precedence to tags, and among
non-tag tips, names with shorter hops are preferred over longer
hops, without taking the "taggerdate" into account. As we are
taking over the "taggerdate" field to store the committer date for
tips with commits:
(1) keep the original logic when comparing names based on two refs
both of which are from refs/tags/;
(2) favoring a name based on a ref in refs/tags/ hierarchy over
a ref outside the hierarchy;
(3) between two names based on a ref both outside refs/tags/, give
precedence to a name with shorter hops and use "taggerdate"
only to tie-break.
A change to t4202 is a natural consequence. The test creates a
commit on a branch "side" and points at it with an unannotated tag
"refs/tags/side-2". The original code couldn't decide which one to
favor at all, and gave a name based on a branch (simply because
refs/heads/side sorts earlier than refs/tags/side-2). Because the
updated logic is taught to favor refs in refs/tags/ hierarchy, the
the test is updated to expect to see tags/side-2 instead.
[mjg: open-coded the comparisons in is_better_name(), dropping a
helper macro used in the original]
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Michael J Gruber <git@grubix.eu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When we encounter a new ref that could describe the commit we are
looking at, we compare the name that is formed using that ref and
the name we found so far and pick a better one.
Factor the comparison logic out to a separate helper function, while
keeping the current logic the same (i.e. a name that is based on an
older tag is better, and if two tags of the same age can reach the
commit, the one with fewer number of hops to reach the commit is
better).
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Michael J Gruber <git@grubix.eu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The odb_mkstemp() function expects the caller to provide a
fixed buffer to write the resulting tempfile name into. But
it creates the template using snprintf without checking the
return value. This means we could silently truncate the
filename.
In practice, it's unlikely that the truncation would end in
the template-pattern that mkstemp needs to open the file. So
we'd probably end up failing either way, unless the path was
specially crafted.
The simplest fix would be to notice the truncation and die.
However, we can observe that most callers immediately
xstrdup() the result anyway. So instead, let's switch to
using a strbuf, which is easier for them (and isn't a big
deal for the other 2 callers, who can just strbuf_release
when they're done with it).
Note that many of the callers used static buffers, but this
was purely to avoid putting a large buffer on the stack. We
never passed the static buffers out of the function, so
there's no complicated memory handling we need to change.
Signed-off-by: Jeff King <peff@peff.net>
The odb_mkstemp function does not return an error; it dies
on failure instead. But many of its callers compare the
resulting descriptor against -1 and die themselves.
Mostly this is just pointless, but it does raise a question
when looking at the callers: if they show the results of the
"template" buffer after a failure, what's in it? The answer
is: it doesn't matter, because it cannot happen.
So let's make that clear by removing the bogus error checks.
In bitmap_writer_finish(), we can drop the error-handling
code entirely. In the other two cases, it's shared with the
open() in another code path; we can just move the
error-check next to that open() call.
And while we're at it, let's flesh out the function's
docstring a bit to make the error behavior clear.
Signed-off-by: Jeff King <peff@peff.net>
Code clean-up.
* jk/fast-import-cleanup:
pack.h: define largest possible encoded object size
encode_in_pack_object_header: respect output buffer length
fast-import: use xsnprintf for formatting headers
fast-import: use xsnprintf for writing sha1s
"git checkout" is taught the "--recurse-submodules" option.
* sb/checkout-recurse-submodules:
builtin/read-tree: add --recurse-submodules switch
builtin/checkout: add --recurse-submodules switch
entry.c: create submodules when interesting
unpack-trees: check if we can perform the operation for submodules
unpack-trees: pass old oid to verify_clean_submodule
update submodules: add submodule_move_head
submodule.c: get_super_prefix_or_empty
update submodules: move up prepare_submodule_repo_env
submodules: introduce check to see whether to touch a submodule
update submodules: add a config option to determine if submodules are updated
update submodules: add submodule config parsing
make is_submodule_populated gently
lib-submodule-update.sh: define tests for recursing into submodules
lib-submodule-update.sh: replace sha1 by hash
lib-submodule-update: teach test_submodule_content the -C <dir> flag
lib-submodule-update.sh: do not use ./. as submodule remote
lib-submodule-update.sh: reorder create_lib_submodule_repo
submodule--helper.c: remove duplicate code
connect_work_tree_and_git_dir: safely create leading directories
Code clean-up.
* jk/pack-name-cleanups:
index-pack: make pointer-alias fallbacks safer
replace snprintf with odb_pack_name()
odb_pack_keep(): stop generating keepfile name
sha1_file.c: make pack-name helper globally accessible
move odb_* declarations out of git-compat-util.h
"git fetch" that requests a commit by object name, when the other
side does not allow such an request, failed without much
explanation.
* mm/fetch-show-error-message-on-unadvertised-object:
fetch-pack: add specific error for fetching an unadvertised object
fetch_refs_via_pack: call report_unmatched_refs
fetch-pack: move code to report unmatched refs to a function
"git branch @" created refs/heads/@ as a branch, and in general the
code that handled @{-1} and @{upstream} was a bit too loose in
disambiguating.
* jk/interpret-branch-name:
checkout: restrict @-expansions when finding branch
strbuf_check_ref_format(): expand only local branches
branch: restrict @-expansions when deleting
t3204: test git-branch @-expansion corner cases
interpret_branch_name: allow callers to restrict expansions
strbuf_branchname: add docstring
strbuf_branchname: drop return value
interpret_branch_name: move docstring to header file
interpret_branch_name(): handle auto-namelen for @{-1}
"git repack --depth=<n>" for a long time busted the specified depth
when reusing delta from existing packs. This has been corrected.
* jk/delta-chain-limit:
pack-objects: convert recursion to iteration in break_delta_chain()
pack-objects: enforce --depth limit in reused deltas
Make the internal storage for struct sha1_array use an array of struct
object_id internally. Update the users of this struct which inspect its
internals.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert virtually all uses of unsigned char [20] to struct object_id.
Leave all the arguments that come from struct sha1_array, as these will
be converted in a later patch.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
All of the callers of this function have been converted, so convert this
function and update the callers. This function also calls
sha1_array_append, which we'll convert shortly.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert some hardcoded constants into uses of parse_oid_hex.
Additionally, convert all uses of struct command, and miscellaneous
other functions necessary for that. This work is necessary to be able
to convert sha1_array_append later on.
To avoid needing to specify a constant, reject shallow lines with the
wrong length instead of simply ignoring them.
Note that in queue_command we are guaranteed to have a NUL-terminated
buffer or at least one byte of overflow that we can safely read, so the
linelen check can be elided. We would die in such a case, but not read
invalid memory.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If we had already processed the last newline in a push certificate, we
would end up subtracting NULL from the end-of-certificate pointer when
computing the length of the line. This would have resulted in an
absurdly large length, and possibly a buffer overflow. Instead,
subtract the beginning-of-certificate pointer from the
end-of-certificate pointer, which is what's expected.
Note that this situation should never occur, since not only do we
require the certificate to be newline terminated, but the signature will
only be read from the beginning of a line. Nevertheless, it seems
prudent to correct it.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
git describe --debug localizes all debug messages but not the terms
head, lightweight, annotated that it outputs for the candidates.
Localize them, too.
Signed-off-by: Michael J Gruber <git@grubix.eu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git describe --dirty" dies when it cannot be determined if the
state in the working tree matches that of HEAD (e.g. broken
repository or broken submodule). The command learned a new option
"git describe --broken" to give "$name-broken" (where $name is the
description of HEAD) in such a case.
* sb/describe-broken:
builtin/describe: introduce --broken flag
Recently we started passing the "--push-options" through the
external remote helper interface; now the "smart HTTP" remote
helper understands what to do with the passed information.
* sb/push-options-via-transport:
remote-curl: allow push options
send-pack: send push options correctly in stateless-rpc case
Code clean-up with minor bugfixes.
* jk/prefix-filename:
bundle: use prefix_filename with bundle path
prefix_filename: simplify windows #ifdef
prefix_filename: return newly allocated string
prefix_filename: drop length parameter
prefix_filename: move docstring to header file
hash-object: fix buffer reuse with --path in a subdirectory
Convert the caller of sha1_array_append to struct object_id.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since we will likely be introducing a new hash function at some point,
and that hash function might be longer than 20 bytes, use the constant
GIT_MAX_RAWSZ, which is designed to be suitable for allocations, instead
of GIT_SHA1_RAWSZ. This will ease the transition down the line by
distinguishing between places where we need to allocate memory suitable
for the largest hash from those where we need to handle the current
hash.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since we will likely be introducing a new hash function at some point,
and that hash function might be longer than 40 hex characters, use the
constant GIT_MAX_HEXSZ, which is designed to be suitable for
allocations, instead of GIT_SHA1_HEXSZ. This will ease the transition
down the line by distinguishing between places where we need to allocate
memory suitable for the largest hash from those where we need to handle
the current hash.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The code to parse the command line "git grep <patterns>... <rev>
[[--] <pathspec>...]" has been cleaned up, and a handful of bugs
have been fixed (e.g. we used to check "--" if it is a rev).
* jk/grep-no-index-fix:
grep: treat revs the same for --untracked as for --no-index
grep: do not diagnose misspelt revs with --no-index
grep: avoid resolving revision names in --no-index case
grep: fix "--" rev/pathspec disambiguation
grep: re-order rev-parsing loop
grep: do not unnecessarily query repo for "--"
grep: move thread initialization a little lower
"git push" had a handful of codepaths that could lead to a deadlock
when unexpected error happened, which has been fixed.
* jk/push-deadlock-regression-fix:
send-pack: report signal death of pack-objects
send-pack: read "unpack" status even on pack-objects failure
send-pack: improve unpack-status error messages
send-pack: use skip_prefix for parsing unpack status
send-pack: extract parsing of "unpack" response
receive-pack: fix deadlock when we cannot create tmpdir
Several callers use fixed buffers for storing the pack
object header, and they've picked 10 as a magic number. This
is reasonable, since it handles objects up to 2^67. But
let's give them a constant so it's clear that the number
isn't pulled out of thin air.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The encode_in_pack_object_header() writes a variable-length
header to an output buffer, but it doesn't actually know
long the buffer is. At first glance, this looks like it
might be possible to overflow.
In practice, this is probably impossible. The smallest
buffer we use is 10 bytes, which would hold the header for
an object up to 2^67 bytes. Obviously we're not likely to
see such an object, but we might worry that an object could
lie about its size (causing us to overflow before we realize
it does not actually have that many bytes). But the argument
is passed as a uintmax_t. Even on systems that have __int128
available, uintmax_t is typically restricted to 64-bit by
the ABI.
So it's unlikely that a system exists where this could be
exploited. Still, it's easy enough to use a normal out/len
pair and make sure we don't write too far. That protects the
hypothetical 128-bit system, makes it harder for callers to
accidentally specify a too-small buffer, and makes the
resulting code easier to audit.
Note that the one caller in fast-import tried to catch such
a case, but did so _after_ the call (at which point we'd
have already overflowed!). This check can now go away.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Change the tag, branch & for-each-ref commands to have a --no-contains
option in addition to their longstanding --contains options.
This allows for finding the last-good rollout tag given a known-bad
<commit>. Given a hypothetically bad commit cf5c7253e0, the git
version to revert to can be found with this hacky two-liner:
(git tag -l 'v[0-9]*'; git tag -l --contains cf5c7253e0 'v[0-9]*') |
sort | uniq -c | grep -E '^ *1 ' | awk '{print $2}' | tail -n 10
With this new --no-contains option the same can be achieved with:
git tag -l --no-contains cf5c7253e0 'v[0-9]*' | sort | tail -n 10
As the filtering machinery is shared between the tag, branch &
for-each-ref commands, implement this for those commands too. A
practical use for this with "branch" is e.g. finding branches which
were branched off between v2.8.0 and v2.10.0:
git branch --contains v2.8.0 --no-contains v2.10.0
The "describe" command also has a --contains option, but its semantics
are unrelated to what tag/branch/for-each-ref use --contains for. A
--no-contains option for "describe" wouldn't make any sense, other
than being exactly equivalent to not supplying --contains at all,
which would be confusing at best.
Add a --without option to "tag" as an alias for --no-contains, for
consistency with --with and --contains. The --with option is
undocumented, and possibly the only user of it is
Junio (<xmqqefy71iej.fsf@gitster.mtv.corp.google.com>). But it's
trivial to support, so let's do that.
The additions to the the test suite are inverse copies of the
corresponding --contains tests. With this change --no-contains for
tag, branch & for-each-ref is just as well tested as the existing
--contains option.
In addition to those tests, add a test for "tag" which asserts that
--no-contains won't find tree/blob tags, which is slightly
unintuitive, but consistent with how --contains works & is documented.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Change the --points-at option to default to HEAD for consistency with
its siblings --contains, --merged etc. which default to
HEAD. Previously we'd get:
$ git tag --points-at 2>&1 | head -n 1
error: option `points-at' requires a value
This changes behavior added in commit ae7706b9ac (tag: add --points-at
list option, 2012-02-08).
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Change the "tag" command to implicitly turn on its --list mode when
provided with a list-like option such as --contains, --points-at etc.
This is for consistency with how "branch" works. When "branch" is
given a list-like option, such as --contains, it implicitly provides
--list. Before this change "tag" would error out on those sorts of
invocations. I.e. while both of these worked for "branch":
git branch --contains v2.8.0 <pattern>
git branch --list --contains v2.8.0 <pattern>
Only the latter form worked for "tag":
git tag --contains v2.8.0 '*rc*'
git tag --list --contains v2.8.0 '*rc*'
Now "tag", like "branch", will implicitly supply --list when a
list-like option is provided, and no other conflicting non-list
options (such as -d) are present on the command-line.
Spelunking through the history via:
git log --reverse -p -G'only allowed with' -- '*builtin*tag*c'
Reveals that there was no good reason for not allowing this in the
first place. The --contains option added in 32c35cfb1e ("git-tag: Add
--contains option", 2009-01-26) made this an error. All the other
subsequent list-like options that were added copied its pattern of
making this usage an error.
The only tests that break as a result of this change are tests that
were explicitly checking that this "branch-like" usage wasn't
permitted. Change those failing tests to check that this invocation
mode is permitted, add extra tests for the list-like options we
weren't testing, and tests to ensure that e.g. we don't toggle the
list mode in the presence of other conflicting non-list options.
With this change errors messages such as "--contains option is only
allowed with -l" don't make sense anymore, since options like
--contain turn on -l. Instead we error out when list-like options such
as --contain are used in conjunction with conflicting options such as
-d or -v.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Change mentions of <object> to <commit> in the help output of
for-each-ref as appropriate.
Both --[no-]merged and --contains only take commits, but --points-at
can take any object, such as a tag pointing to a tree or blob.
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach remote-curl to understand push options and to be able to convey
them across HTTP.
Signed-off-by: Brandon Williams <bmwill@google.com>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
git-describe tells you the version number you're at, or errors out, e.g.
when you run it outside of a repository, which may happen when downloading
a tar ball instead of using git to obtain the source code.
To keep this property of only erroring out, when not in a repository,
severe (submodule) errors must be downgraded to reporting them gently
instead of having git-describe error out completely.
To achieve that a flag '--broken' is introduced, which is in the same
vein as '--dirty' but uses an actual child process to check for dirtiness.
When that child dies unexpectedly, we'll append '-broken' instead of
'-dirty'.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Code clean-up.
* jk/pack-name-cleanups:
index-pack: make pointer-alias fallbacks safer
replace snprintf with odb_pack_name()
odb_pack_keep(): stop generating keepfile name
sha1_file.c: make pack-name helper globally accessible
move odb_* declarations out of git-compat-util.h
"git show-branch" expected there were only very short branch names
in the repository and used a fixed-length buffer to hold them
without checking for overflow.
* jk/show-branch-lift-name-len-limit:
show-branch: use skip_prefix to drop magic numbers
show-branch: store resolved head in heap buffer
show-branch: drop head_len variable
"git remote rm X", when a branch has remote X configured as the
value of its branch.*.remote, tried to remove branch.*.remote and
branch.*.merge and failed if either is unset.
* rl/remote-allow-missing-branch-name-merge:
remote: ignore failure to remove missing branch.<name>.merge
A "gc.log" file left by a backgrounded "gc --auto" disables further
automatic gc; it has been taught to run at least once a day (by
default) by ignoring a stale "gc.log" file that is too old.
* dt/gc-ignore-old-gc-logs:
gc: ignore old gc.log files
We may take the path to a bundle file as an argument, and
need to adjust the filename based on the prefix we
discovered while setting up the git directory. We do so
manually into a fixed-size buffer, but using
prefix_filename() is the normal way.
Besides being more concise, there are two subtle
improvements:
1. The original inserted a "/" between the two paths, even
though the "prefix" argument always has the "/"
appended. That means that:
cd subdir && git bundle verify ../foo.bundle
was looking at (and reporting) subdir//../foo.bundle.
Harmless, but ugly. Using prefix_filename() gets this
right.
2. The original checked for an absolute path by looking
for a leading '/'. It should have been using
is_absolute_path(), which also covers more cases on
Windows (backslashes and dos drive prefixes).
But it's easier still to just pass the name to
prefix_filename(), which handles this case
automatically.
Note that we'll just leak the resulting buffer in the name
of simplicity, since it needs to last through the duration
of the program anyway.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The prefix_filename() function returns a pointer to static
storage, which makes it easy to use dangerously. We already
fixed one buggy caller in hash-object recently, and the
calls in apply.c are suspicious (I didn't dig in enough to
confirm that there is a bug, but we call the function once
in apply_all_patches() and then again indirectly from
parse_chunk()).
Let's make it harder to get wrong by allocating the return
value. For simplicity, we'll do this even when the prefix is
empty (and we could just return the original file pointer).
That will cause us to allocate sometimes when we wouldn't
otherwise need to, but this function isn't called in
performance critical code-paths (and it already _might_
allocate on any given call, so a caller that cares about
performance is questionable anyway).
The downside is that the callers need to remember to free()
the result to avoid leaking. Most of them already used
xstrdup() on the result, so we know they are OK. The
remainder have been converted to use free() as appropriate.
I considered retaining a prefix_filename_unsafe() for cases
where we know the static lifetime is OK (and handling the
cleanup is awkward). This is only a handful of cases,
though, and it's not worth the mental energy in worrying
about whether the "unsafe" variant is OK to use in any
situation.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This function takes the prefix as a ptr/len pair, but in
every caller the length is exactly strlen(ptr). Let's
simplify the interface and just take the string. This saves
callers specifying it (and in some cases handling a NULL
prefix).
In a handful of cases we had the length already without
calling strlen, so this is technically slower. But it's not
likely to matter (after all, if the prefix is non-empty
we'll allocate and copy it into a buffer anyway).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The hash-object command uses prefix_filename() without
duplicating its return value. Since that function returns a
static buffer, the value is overwritten by subsequent calls.
This can cause incorrect results when we use --path along
with hashing a file by its relative path, both of which need
to call prefix_filename(). We overwrite the filename
computed for --path, effectively ignoring it.
We can fix this by calling xstrdup on the return value. Note
that we don't bother freeing the "vpath" instance, as it
remains valid until the program exit.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git difftool --dir-diff" used to die a controlled death giving a
"fatal" message when encountering a locally modified symbolic link,
but it started segfaulting since v2.12. This has been fixed.
* js/difftool-builtin:
difftool: handle modified symlinks in dir-diff mode
t7800: cleanup cruft left behind by tests
t7800: remove whitespace before redirect
The string after_subject is added to a strbuf by pp_title_line() if
it's not NULL. Adding an empty string has the same effect as not
adding anything, but the latter is easier, so don't bother changing
the context member from NULL to "".
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Instead of counting the arguments to see if there are any and then
building the full command use a single loop and add the hook command
just before the first argument. This reduces duplication and overall
code size.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Commit 0281e487fd ("grep: optionally recurse into submodules")
added functions grep_submodule() and grep_submodule_launch() which
use "struct work_item" which is defined only when thread support
is available.
The original implementation of grep_submodule() used the "struct
work_item" in order to gain access to a strbuf to store its output which
was to be printed at a later point in time. This differs from how both
grep_file() and grep_sha1() handle their output. This patch eliminates
the reliance on the "struct work_item" and instead opts to use the
output function stored in the output field of the "struct grep_opt"
object directly, making it behave similarly to both grep_file() and
grep_sha1().
Reported-by: Rahul Bedarkar <rahul.bedarkar@imgtec.com>
Signed-off-by: Brandon Williams <bmwill@google.com>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When initializing a submodule set the submodule.<name>.active config to
true if the module hasn't already been configured to be active by some
other means (e.g. a pathspec set in submodule.active).
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach clone --recurse-submodules to optionally take a pathspec argument
which describes which submodules should be recursively initialized and
cloned. If no pathspec is provided, --recurse-submodules will
recursively initialize and clone all submodules by using a default
pathspec of ".". In order to construct more complex pathspecs,
--recurse-submodules can be given multiple times.
This also configures the 'submodule.active' configuration option to be
the given pathspec, such that any future invocation of `git submodule
update` will keep up with the pathspec.
Additionally the switch '--recurse' is removed from the Documentation as
well as marked hidden in the options array, to streamline the options
for submodules. A simple '--recurse' doesn't convey what is being
recursed, e.g. it could mean directories or trees (c.f. ls-tree) In a
lot of other commands we already have '--recurse-submodules' to mean
recursing into submodules, so advertise this spelling here as the
genuine option.
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach `submodule init` to initialize submodules which have been
configured to be active by setting 'submodule.active' with a pathspec.
Now if no path arguments are given and 'submodule.active' is configured,
`init` will initialize all submodules which have been configured to be
active. If no path arguments are given and 'submodule.active' is not
configured, then `init` will retain the old behavior of initializing all
submodules.
This allows users to record more complex patterns as it saves retyping
them whenever you invoke update.
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git revert -m 0 $merge_commit" complained that reverting a merge
needs to say relative to which parent the reversion needs to
happen, as if "-m 0" weren't given. The correct diagnosis is that
"-m 0" does not refer to the first parent ("-m 1" does). This has
been fixed.
* jk/cherry-pick-0-mainline:
cherry-pick: detect bogus arguments to --mainline
From a working tree of a repository, a new option of "rev-parse"
lets you ask if the repository is used as a submodule of another
project, and where the root level of the working tree of that
project (i.e. your superproject) is.
* sb/rev-parse-show-superproject-root:
rev-parse: add --show-superproject-working-tree
The experimental "split index" feature has gained a few
configuration variables to make it easier to use.
* cc/split-index-config: (22 commits)
Documentation/git-update-index: explain splitIndex.*
Documentation/config: add splitIndex.sharedIndexExpire
read-cache: use freshen_shared_index() in read_index_from()
read-cache: refactor read_index_from()
t1700: test shared index file expiration
read-cache: unlink old sharedindex files
config: add git_config_get_expiry() from gc.c
read-cache: touch shared index files when used
sha1_file: make check_and_freshen_file() non static
Documentation/config: add splitIndex.maxPercentChange
t1700: add tests for splitIndex.maxPercentChange
read-cache: regenerate shared index if necessary
config: add git_config_get_max_percent_split_change()
Documentation/git-update-index: talk about core.splitIndex config var
Documentation/config: add information for core.splitIndex
t1700: add tests for core.splitIndex
update-index: warn in case of split-index incoherency
read-cache: add and then use tweak_split_index()
split-index: add {add,remove}_split_index() functions
config: add git_config_get_split_index()
...
When using the --recurse-submodules flag with a relative pathspec which
includes "..", an error is produced inside the child process spawned for a
submodule. When creating the pathspec struct in the child, the ".." is
interpreted to mean "go up a directory" which causes an error stating that the
path ".." is outside of the repository.
While it is true that ".." is outside the scope of the submodule, it is
confusing to a user who originally invoked the command where ".." was indeed
still inside the scope of the superproject. Since the child process launched
for the submodule has some context that it is operating underneath a
superproject, this error could be avoided.
This patch fixes the bug by passing the 'prefix' to the child process. Now
each child process that works on a submodule has two points of reference to the
superproject: (1) the 'super_prefix' which is the path from the root of the
superproject down to root of the submodule and (2) the 'prefix' which is the
path from the root of the superproject down to the directory where the user
invoked the git command.
With these two pieces of information a child process can correctly interpret
the pathspecs provided by the user as well as being able to properly format its
output relative to the directory the user invoked the original command from.
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When using the --recurse-submodules flag with a relative pathspec which
includes "..", an error is produced inside the child process spawned for
a submodule. When creating the pathspec struct in the child, the ".."
is interpreted to mean "go up a directory" which causes an error stating
that the path ".." is outside of the repository.
While it is true that ".." is outside the scope of the submodule, it is
confusing to a user who originally invoked the command where ".." was
indeed still inside the scope of the superproject. Since the child
process launched for the submodule has some context that it is operating
underneath a superproject, this error could be avoided.
This patch fixes the bug by passing the 'prefix' to the child process.
Now each child process that works on a submodule has two points of
reference to the superproject: (1) the 'super_prefix' which is the path
from the root of the superproject down to root of the submodule and (2)
the 'prefix' which is the path from the root of the superproject down to
the directory where the user invoked the git command.
With these two pieces of information a child process can correctly
interpret the pathspecs provided by the user as well as being able to
properly format its output relative to the directory the user invoked
the original command from.
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Use the 'is_submodule_initialized()' helper to check for configured
submodules instead of manually checking for the submodule's URL in the
config.
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The definition of which submodules are of interest by the user
is tied to the configuration submodule.<name>.url; when it is
set to a non-empty string, it is of interest. We'd want to be
able to later change this definition, but there are many places
that explicitly check this condition in the scripted Porcelain.
Introduce the "is-active" subcommand to "submodule--helper", so
that the exact definition of what submodule is of interest can
be centrally defined (and changed in later steps). In a few
patches that follow, this helper is used to replace the explicit
checks of the configuration variable in scripts.
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A new known failure mode is introduced[1], which is actually not
a failure but a feature in read-tree. Unlike checkout for which
the recursive submodule tests were originally written, read-tree does
warn about ignored untracked files that would be overwritten.
For the sake of keeping the test library for submodules generic, just
mark the test as a failure.
[1] KNOWN_FAILURE_SUBMODULE_OVERWRITE_IGNORED_UNTRACKED
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This exposes a flag to recurse into submodules
in builtin/checkout making use of the code implemented
in prior patches.
A new failure mode is introduced in the submodule
update library, as the directory/submodule conflict
is not solved in prior patches.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The final() function accepts a NULL value for certain
parameters, and falls back to writing into a reusable "name"
buffer, and then either:
1. For "keep_name", requiring all uses to do "keep_name ?
keep_name : name.buf". This is awkward, and it's easy
to accidentally look at the maybe-NULL keep_name.
2. For "final_index_name" and "final_pack_name", aliasing
those pointers to the "name" buffer. This is easier to
use, but the aliased pointers become invalid after the
buffer is reused (this isn't a bug now, but it's a
potential pitfall).
One way to make this safer would be to introduce an extra
pointer to do the aliasing, and have its lifetime match the
validity of the "name" buffer. But it's still easy to
accidentally use the wrong name (i.e., to use
"final_pack_name" instead of the aliased pointer).
Instead, let's use three separate buffers that will remain
valid through the function. That makes it safe to alias the
pointers and use them consistently. The extra allocations
shouldn't matter, as this function is not performance
sensitive.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In several places we write the name of the pack filename
into a fixed-size buffer using snprintf(), but do not check
the return value. As a result, a very long object directory
could cause us to quietly truncate the pack filename
(potentially leading to a corrupted repository, as a newly
written packfile could be missing its .pack extension).
We can use odb_pack_name() to do this with a strbuf (and
shorten the code, as well).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The odb_pack_keep() function generates the name of a .keep
file and opens it. This has two problems:
1. It requires a fixed-size buffer to create the filename
and doesn't notice when the result is truncated.
2. Of the two callers, one sometimes wants to open a
filename it already has, which makes things awkward (it
has to do so manually, and skips the leading-directory
creation).
Instead, let's have odb_pack_keep() just open the file.
Generating the name isn't hard, and a future patch will
switch callers over to odb_pack_name() anyway.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We need the gentle version in a later patch. As we have just one caller,
migrate the caller.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Remove code fragment from module_clone that duplicates functionality
of connect_work_tree_and_git_dir in dir.c
Signed-off-by: Valery Tolstov <me@vtolstov.org>
Reviewed-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
All of these options do the same thing "--foo" iterates over
the "foo" refs, and "--foo=<glob>" does the same with a
glob. We can factor this into its own function to avoid
repeating ourselves.
There are two subtleties to note:
- the original called for_each_branch_ref(), etc, in the
non-glob case. Now we will call for_each_ref_in("refs/heads/")
which is exactly what for_each_branch_ref() did under
the hood.
- for --glob, we'll call for_each_glob_ref_in() with a
NULL "prefix" argument. Which is exactly what
for_each_glob_ref() was doing already.
So both cases should behave identically, and it seems
reasonable to assume that this will remain the same. The
functions we are calling now are the more-generic ones, and
the ones we are dropping are just convenience wrappers.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We can't just use a bare skip_prefix() for these cases,
because we need to match both the "--foo" form and the
"--foo=<value>" form (and tell the difference between the
two in the caller).
We can wrap this in a simple helper which has two obvious
callsites, and will gain some more in the next patch.
Note that the error output for abbrev-ref changes slightly,
as we don't keep our original "arg" pointer. However, the
new output should hopefully be more clear:
[before]
fatal: unknown mode for --abbrev-ref=foo
[after]
fatal: unknown mode for --abbrev-ref: foo
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Using skip_prefix lets us avoid manually-counted offsets
into the argument string. This patch converts the simple and
obvious cases.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The cherry-pick and revert commands use OPT_INTEGER() to
parse --mainline. The stock parser is smart enough to reject
non-numeric nonsense, but it doesn't know that parent
counting starts at 1.
Worse, the value "0" is indistinguishable from the unset
case, so a user who assumes the counting is 0-based will get
a confusing message:
$ git cherry-pick -m 0 $merge
error: commit ... is a merge but no -m option was given.
Let's use a custom callback that enforces our range.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Detect the null object ID for symlinks in dir-diff so that difftool can
detect when symlinks are modified in the worktree.
Previously, a null symlink object ID would crash difftool.
Handle null object IDs as unknown content that must be read from
the worktree.
Helped-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: David Aguilar <davvid@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git branch --list" takes the "--abbrev" and "--no-abbrev" options
to control the output of the object name in its "-v"(erbose)
output, but a recent update started ignoring them; this fixes it
before the breakage reaches to any released version.
* kn/ref-filter-branch-list:
branch: honor --abbrev/--no-abbrev in --list mode
"git push" had a handful of codepaths that could lead to a deadlock
when unexpected error happened, which has been fixed.
* jk/push-deadlock-regression-fix:
send-pack: report signal death of pack-objects
send-pack: read "unpack" status even on pack-objects failure
send-pack: improve unpack-status error messages
send-pack: use skip_prefix for parsing unpack status
send-pack: extract parsing of "unpack" response
receive-pack: fix deadlock when we cannot create tmpdir
"git fetch" that requests a commit by object name, when the other
side does not allow such an request, failed without much
explanation.
* mm/fetch-show-error-message-on-unadvertised-object:
fetch-pack: add specific error for fetching an unadvertised object
fetch_refs_via_pack: call report_unmatched_refs
fetch-pack: move code to report unmatched refs to a function
"git branch @" created refs/heads/@ as a branch, and in general the
code that handled @{-1} and @{upstream} was a bit too loose in
disambiguating.
* jk/interpret-branch-name:
checkout: restrict @-expansions when finding branch
strbuf_check_ref_format(): expand only local branches
branch: restrict @-expansions when deleting
t3204: test git-branch @-expansion corner cases
interpret_branch_name: allow callers to restrict expansions
strbuf_branchname: add docstring
strbuf_branchname: drop return value
interpret_branch_name: move docstring to header file
interpret_branch_name(): handle auto-namelen for @{-1}
Git v2.12 was shipped with an embarrassing breakage where various
operations that verify paths given from the user stopped dying when
seeing an issue, and instead later triggering segfault.
* js/realpath-pathdup-fix:
real_pathdup(): fix callsites that wanted it to die on error
t1501: demonstrate NULL pointer access with invalid GIT_WORK_TREE
All callers of add_blame_entry() allocate and copy the second argument.
Let the function do it for them, reducing code duplication.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When "git submodule init" decides that the submodule in the working
tree is its upstream, it now gives a warning as it is not a very
common setup.
* sb/submodule-init-url-selection:
submodule init: warn about falling back to a local path
When the "branch --list" command was converted to use the --format
facility from the ref-filter API, we forgot to honor the --abbrev
setting in the default output format and instead used a hardcoded
"7".
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In some situations it is useful to know if the given repository
is a submodule of another repository.
Add the flag --show-superproject-working-tree to git-rev-parse
to make it easy to find out if there is a superproject. When no
superproject exists, the output will be empty.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In 4ac9006f83 (real_path: have callers use real_pathdup and
strbuf_realpath, 2016-12-12), we changed the xstrdup(real_path())
pattern to use real_pathdup() directly.
The problem with this change is that real_path() calls
strbuf_realpath() with die_on_error = 1 while real_pathdup() calls
it with die_on_error = 0. Meaning that in cases where real_path()
causes Git to die() with an error message, real_pathdup() is silent
and returns NULL instead.
The callers, however, are ill-prepared for that change, as they expect
the return value to be non-NULL (and otherwise the function died
with an appropriate error message).
Fix this by extending real_pathdup()'s signature to accept the
die_on_error flag and simply pass it through to strbuf_realpath(),
and then adjust all callers after a careful audit whether they would
handle NULLs well.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The err_fd descriptor passed to the unpack() function is
intended to be handed off to the child index-pack, and our
async muxer will read until it gets EOF. However, if we
encounter an error before handing off the descriptor, we
must manually close(err_fd). Otherwise we will be waiting
for our muxer to finish, while the muxer is waiting for EOF
on err_fd.
We fixed an identical deadlock already in 49ecfa13f
(receive-pack: close sideband fd on early pack errors,
2013-04-19). But since then, the function grew a new
early-return in 722ff7f87 (receive-pack: quarantine objects
until pre-receive accepts, 2016-10-03), when we fail to
create a temporary directory. This return needs the same
treatment.
Reported-by: Horst Schirmeier <horst@schirmeier.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Prepare to reuse this code in transport.c for "git fetch".
While we're here, internationalize the existing error message.
Signed-off-by: Matt McCutchen <matt@mattmccutchen.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When we parse "git checkout $NAME", we try to interpret
$NAME as a local branch-name. If it is, then we point HEAD
to that branch. Otherwise, we detach the HEAD at whatever
commit $NAME points to.
We do the interpretation by calling strbuf_branchname(), and
then blindly sticking "refs/heads/" on the front. This leads
to nonsense results when expansions like "@{upstream}" or
"@" point to something besides a local branch. We end up
with a local branch name like "refs/heads/origin/master" or
"refs/heads/HEAD".
Normally this has no user-visible effect because those
branches don't exist, and so we fallback to feeding the
result to get_sha1(), which resolves them correctly.
But as the new test in t3204 shows, there are corner cases
where the effect is observable, and we check out the wrong
local branch rather than detaching to the correct one.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We use strbuf_branchname() to expand the branch name from
the command line, so you can delete the branch given by
@{-1}, for example. However, we allow other nonsense like
"@", and we do not respect our "-r" flag (so we may end up
deleting an oddly-named local ref instead of a remote one).
We can fix this by passing the appropriate "allowed" flag to
strbuf_branchname().
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The interpret_branch_name() function converts names like
@{-1} and @{upstream} into branch names. The expanded ref
names are not fully qualified, and may be outside of the
refs/heads/ namespace (e.g., "@" expands to "HEAD", and
"@{upstream}" is likely to be in "refs/remotes/").
This is OK for callers like dwim_ref() which are primarily
interested in resolving the resulting name, no matter where
it is. But callers like "git branch" treat the result as a
branch name in refs/heads/. When we expand to a ref outside
that namespace, the results are very confusing (e.g., "git
branch @" tries to create refs/heads/HEAD, which is
nonsense).
Callers can't know from the returned string how the
expansion happened (e.g., did the user really ask for a
branch named "HEAD", or did we do a bogus expansion?). One
fix would be to return some out-parameters describing the
types of expansion that occurred. This has the benefit that
the caller can generate precise error messages ("I
understood @{upstream} to mean origin/master, but that is a
remote tracking branch, so you cannot create it as a local
name").
However, out-parameters make the function interface somewhat
cumbersome. Instead, let's do the opposite: let the caller
tell us which elements to expand. That's easier to pass in,
and none of the callers give more precise error messages
than "@{upstream} isn't a valid branch name" anyway (which
should be sufficient).
The strbuf_branchname() function needs a similar parameter,
as most of the callers access interpret_branch_name()
through it.
We can break the callers down into two groups:
1. Callers that are happy with any kind of ref in the
result. We pass "0" here, so they continue to work
without restrictions. This includes merge_name(),
the reflog handling in add_pending_object_with_path(),
and substitute_branch_name(). This last is what powers
dwim_ref().
2. Callers that have funny corner cases (mostly in
git-branch and git-checkout). These need to make use of
the new parameter, but I've left them as "0" in this
patch, and will address them individually in follow-on
patches.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add the email-style subject prefix (e.g. "Subject: [PATCH] ") directly
when it's needed instead of letting log_write_email_headers() prepare
it in a static buffer in advance. This simplifies storage ownership and
code flow.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This function will be used in a following commit to get the expiration
time of the shared index files from the config, and it is generic
enough to be put in "config.c".
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When users are using `git update-index --(no-)split-index`, they
may expect the split-index feature to be used or not according to
the option they just used, but this might not be the case if the
new "core.splitIndex" config variable has been set. In this case
let's warn about what will happen and why.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Also use the functions in cmd_update_index() in
builtin/update-index.c.
These functions will be used in a following commit to tweak
our use of the split-index feature depending on the setting
of a configuration variable.
Signed-off-by: Christian Couder <chriscool@tuxfamily.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When a submodule is initialized, the config variable 'submodule.<name>.url'
is set depending on the value of the same variable in the .gitmodules
file. When the URL indicates to be relative, then the url is computed
relative to its default remote. The default remote cannot be determined
accurately in all cases, such that it falls back to 'origin'.
The 'origin' remote may not exist, though. In that case we give up looking
for a suitable remote and we'll just assume it to be a local relative path.
This can be confusing to users as there is a lot of guessing involved,
which is not obvious to the user.
So in the corner case of assuming a local autoritative truth, warn the
user to lessen the confusion.
This behavior was introduced in 4d6893200 (submodule add: allow relative
repository path even when no url is set, 2011-06-06), which shared the
code with submodule-init and then ported to C in 3604242f08 (submodule:
port init from shell to C, 2016-04-15).
In case of submodule-add, this behavior makes sense in some use cases[1],
however for submodule-init there does not seem to be an immediate obvious
use case to fall back to a local submodule. However there might be, so
warn instead of die here.
While adding the warning, also clarify the behavior of relative URLs in
the documentation.
[1] e.g. http://stackoverflow.com/questions/8721984/git-ignore-files-for-public-repository-but-not-for-private
"store a secret locally in a submodule, with no intention to publish it"
Reported-by: Shawn Pearce <spearce@spearce.org>
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git remote rm X", when a branch has remote X configured as the
value of its branch.*.remote, tried to remove branch.*.remote and
branch.*.merge and failed if either is unset.
* rl/remote-allow-missing-branch-name-merge:
remote: ignore failure to remove missing branch.<name>.merge
"git update-ref -d" and other operations to delete references did
not leave any entry in HEAD's reflog when the reference being
deleted was the current branch. This is not a problem in practice
because you do not want to delete the branch you are currently on,
but caused renaming of the current branch to something else not to
be logged in a useful way.
* km/delete-ref-reflog-message:
branch: record creation of renamed branch in HEAD's log
rename_ref: replace empty message in HEAD's log
update-ref: pass reflog message to delete_ref()
delete_ref: accept a reflog message argument
The "--git-path", "--git-common-dir", and "--shared-index-path"
options of "git rev-parse" did not produce usable output. They are
now updated to show the path to the correct file, relative to where
the caller is.
* js/git-path-in-subdir:
rev-parse: fix several options when running in a subdirectory
rev-parse tests: add tests executed from a subdirectory
The code to parse the command line "git grep <patterns>... <rev>
[[--] <pathspec>...]" has been cleaned up, and a handful of bugs
have been fixed (e.g. we used to check "--" if it is a rev).
* jk/grep-no-index-fix:
grep: treat revs the same for --untracked as for --no-index
grep: do not diagnose misspelt revs with --no-index
grep: avoid resolving revision names in --no-index case
grep: fix "--" rev/pathspec disambiguation
grep: re-order rev-parsing loop
grep: do not unnecessarily query repo for "--"
grep: move thread initialization a little lower
Some warning() messages from "git clean" were updated to show the
errno from failed system calls.
* nd/clean-preserve-errno-in-warning:
clean: use warning_errno() when appropriate
"git show-branch" expected there were only very short branch names
in the repository and used a fixed-length buffer to hold them
without checking for overflow.
* jk/show-branch-lift-name-len-limit:
show-branch: use skip_prefix to drop magic numbers
show-branch: store resolved head in heap buffer
show-branch: drop head_len variable
A "gc.log" file left by a backgrounded "gc --auto" disables further
automatic gc; it has been taught to run at least once a day (by
default) by ignoring a stale "gc.log" file that is too old.
* dt/gc-ignore-old-gc-logs:
gc: ignore old gc.log files
"git rebase -i" starts using the recently updated "sequencer" code.
* js/rebase-helper:
rebase -i: use the rebase--helper builtin
rebase--helper: add a builtin helper for interactive rebases
The gitattributes machinery is being taught to work better in a
multi-threaded environment.
* bw/attr: (27 commits)
attr: reformat git_attr_set_direction() function
attr: push the bare repo check into read_attr()
attr: store attribute stack in attr_check structure
attr: tighten const correctness with git_attr and match_attr
attr: remove maybe-real, maybe-macro from git_attr
attr: eliminate global check_all_attr array
attr: use hashmap for attribute dictionary
attr: change validity check for attribute names to use positive logic
attr: pass struct attr_check to collect_some_attrs
attr: retire git_check_attrs() API
attr: convert git_check_attrs() callers to use the new API
attr: convert git_all_attrs() to use "struct attr_check"
attr: (re)introduce git_check_attr() and struct attr_check
attr: rename function and struct related to checking attributes
attr.c: outline the future plans by heavily commenting
Documentation: fix a typo
attr.c: add push_stack() helper
attr: support quoting pathname patterns in C style
attr.c: plug small leak in parse_attr_line()
attr.c: tighten constness around "git_attr" structure
...
Clean-up and updates to command line completion (in contrib/).
* sg/completion: (22 commits)
completion: restore removed line continuating backslash
completion: cache the path to the repository
completion: extract repository discovery from __gitdir()
completion: don't guard git executions with __gitdir()
completion: consolidate silencing errors from git commands
completion: don't use __gitdir() for git commands
completion: respect 'git -C <path>'
rev-parse: add '--absolute-git-dir' option
completion: fix completion after 'git -C <path>'
completion: don't offer commands when 'git --opt' needs an argument
completion: list short refs from a remote given as a URL
completion: don't list 'HEAD' when trying refs completion outside of a repo
completion: list refs from remote when remote's name matches a directory
completion: respect 'git --git-dir=<path>' when listing remote refs
completion: fix most spots not respecting 'git --git-dir=<path>'
completion: ensure that the repository path given on the command line exists
completion tests: add tests for the __git_refs() helper function
completion tests: check __gitdir()'s output in the error cases
completion tests: consolidate getting path of current working directory
completion tests: make the $cur variable local to the test helper functions
...
"git tag" did not leave useful message when adding a new entry to
reflog; this was left unnoticed for a long time because refs/tags/*
doesn't keep reflog by default.
* cw/tag-reflog-message:
tag: generate useful reflog message
Optimizes resource usage while enumerating refs from alternate
object store, to help receiving end of "push" that hosts a
repository with many "forks".
* jk/alternate-ref-optim:
receive-pack: avoid duplicates between our refs and alternates
receive-pack: treat namespace .have lines like alternates
receive-pack: fix misleading namespace/.have comment
receive-pack: use oidset to de-duplicate .have lines
add oidset API
fetch-pack: cache results of for_each_alternate_ref
for_each_alternate_ref: replace transport code with for-each-ref
for_each_alternate_ref: pass name/oid instead of ref struct
for_each_alternate_ref: use strbuf for path allocation
for_each_alternate_ref: stop trimming trailing slashes
for_each_alternate_ref: handle failure from real_pathdup()
The code to list branches in "git branch" has been consolidated
with the more generic ref-filter API.
* kn/ref-filter-branch-list: (21 commits)
ref-filter: resurrect "strip" as a synonym to "lstrip"
branch: implement '--format' option
branch: use ref-filter printing APIs
branch, tag: use porcelain output
ref-filter: allow porcelain to translate messages in the output
ref-filter: add an 'rstrip=<N>' option to atoms which deal with refnames
ref-filter: modify the 'lstrip=<N>' option to work with negative '<N>'
ref-filter: Do not abruptly die when using the 'lstrip=<N>' option
ref-filter: rename the 'strip' option to 'lstrip'
ref-filter: make remote_ref_atom_parser() use refname_atom_parser_internal()
ref-filter: introduce refname_atom_parser()
ref-filter: introduce refname_atom_parser_internal()
ref-filter: make "%(symref)" atom work with the ':short' modifier
ref-filter: add support for %(upstream:track,nobracket)
ref-filter: make %(upstream:track) prints "[gone]" for invalid upstreams
ref-filter: introduce format_ref_array_item()
ref-filter: move get_head_description() from branch.c
ref-filter: modify "%(objectname:short)" to take length
ref-filter: implement %(if:equals=<string>) and %(if:notequals=<string>)
ref-filter: include reference to 'used_atom' within 'atom_value'
...
"git repack --depth=<n>" for a long time busted the specified depth
when reusing delta from existing packs. This has been corrected.
* jk/delta-chain-limit:
pack-objects: convert recursion to iteration in break_delta_chain()
pack-objects: enforce --depth limit in reused deltas
"git describe" and "git name-rev" have been taught to take more
than one refname patterns to restrict the set of refs to base their
naming output on, and also learned to take negative patterns to
name refs not to be used for naming via their "--exclude" option.
* jk/describe-omit-some-refs:
describe: teach describe negative pattern matches
describe: teach --match to accept multiple patterns
name-rev: add support to exclude refs by pattern match
name-rev: extend --refs to accept multiple patterns
doc: add documentation for OPT_STRING_LIST
Convert the remaining uses of unsigned char [20] to struct object_id.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert each_loose_object_fn and each_packed_object_fn to take a pointer
to struct object_id. Update the various callbacks. Convert several
40-based constants to use GIT_SHA1_HEXSZ.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Make each_reflog_ent_fn take two struct object_id pointers instead of
two pointers to unsigned char. Convert the various callbacks to use
struct object_id as well. Also, rename fsck_handle_reflog_sha1 to
fsck_handle_reflog_oid.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert various uses of unsigned char [20] to struct object_id. Rename
replace_object_sha1 to replace_object_oid. Finally, specify a constant
in terms of GIT_SHA1_HEXSZ.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
There are a few leaf functions in various files that call
resolve_refdup. Convert these functions to use struct object_id
internally to prepare for transitioning resolve_refdup itself.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Additionally convert several uses of the constant 40 into
GIT_SHA1_HEXSZ.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert several functions to use struct object_id, and rename them so
that they no longer refer to SHA-1.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert most of the code to use struct object_id, including struct
origin_data and struct merge_parents. Convert several instances of
hardcoded numbers into references to GIT_SHA1_HEXSZ.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In addition to converting to struct object_id, write some hardcoded
buffer sizes in terms of GIT_SHA1_RAWSZ.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert the functions in this file and struct commit_name to struct
object_id.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert most leaf functions to struct object_id. Change several
hardcoded numbers to uses of parse_oid_hex. In doing so, verify that we
when we want two trees, we have exactly two trees.
Finally, in stdin_diff_commit, avoid accessing the byte after the NUL.
This will be a NUL as well, since the first NUL was a newline we
overwrote. However, with parse_oid_hex, we no longer need to increment
the pointer directly, and can simply increment it as part of our check
for the space character.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
It is not all too unusual for a branch to use "branch.<name>.remote"
without "branch.<name>.merge". You may be using the 'push.default'
configuration set to 'current', for example, and do
$ git checkout -b side colleague/side
$ git config branch.side.remote colleague
However, "git remote rm" to remove the remote used in such a manner
fails with
"fatal: could not unset 'branch.<name>.merge'"
because it assumes that a branch that has .remote defined must also
have .merge defined. Detect the "cannot unset because it is not set
to begin with" case and ignore it.
Signed-off-by: Ross Lagerwall <rosslagerwall@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Renaming the current branch adds an event to the current branch's log
and to HEAD's log. However, the logged entries differ. The entry in
the branch's log represents the entire renaming operation (the old and
new hash are identical), whereas the entry in HEAD's log represents
the deletion only (the new sha1 is null).
Extend replace_each_worktree_head_symref(), whose only caller is
branch_rename(), to take a reflog message argument. This allows the
creation of the new ref to be recorded in HEAD's log. As a result,
the renaming event is represented by two entries (a deletion and a
creation entry) in HEAD's log.
It's a bit unfortunate that the branch's log and HEAD's log now
represent the renaming event in different ways. Given that the
renaming operation is not atomic, the two-entry form is a more
accurate representation of the operation and is more useful for
debugging purposes if a failure occurs between the deletion and
creation events. It would make sense to move the branch's log to the
two-entry form, but this would involve changes to how the rename is
carried out and to how the update flags and reflogs are processed for
deletions, so it may not be worth the effort.
Based-on-patch-by: Jeff King <peff@peff.net>
Signed-off-by: Kyle Meyer <kyle@kyleam.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Now that delete_ref() accepts a reflog message, pass the user-provided
message to delete_ref() rather than silently dropping it.
Signed-off-by: Kyle Meyer <kyle@kyleam.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When the current branch is renamed with 'git branch -m/-M' or deleted
with 'git update-ref -m<msg> -d', the event is recorded in HEAD's log
with an empty message. In preparation for adding a more meaningful
message to HEAD's log in these cases, update delete_ref() to take a
message argument and pass it along to ref_transaction_delete().
Modify all callers to pass NULL for the new message argument; no
change in behavior is intended.
Note that this is relevant for HEAD's log but not for the deleted
ref's log, which is currently deleted along with the ref. Even if it
were not, an entry for the deletion wouldn't be present in the deleted
ref's log. files_transaction_commit() writes to the log if
REF_NEEDS_COMMIT or REF_LOG_ONLY are set, but lock_ref_for_update()
doesn't set REF_NEEDS_COMMIT for the deleted ref because REF_DELETING
is set. In contrast, the update for HEAD has REF_LOG_ONLY set by
split_head_update(), resulting in the deletion being logged.
Signed-off-by: Kyle Meyer <kyle@kyleam.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Convert most leaf functions to use struct object_id.
Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In addition to making git_path() aware of certain file names that need
to be handled differently e.g. when running in worktrees, the commit
557bd833bb (git_path(): be aware of file relocation in $GIT_DIR,
2014-11-30) also snuck in a new option for `git rev-parse`:
`--git-path`.
On the face of it, there is no obvious bug in that commit's diff: it
faithfully calls git_path() on the argument and prints it out, i.e. `git
rev-parse --git-path <filename>` has the same precise behavior as
calling `git_path("<filename>")` in C.
The problem lies deeper, much deeper. In hindsight (which is always
unfair), implementing the .git/ directory discovery in
`setup_git_directory()` by changing the working directory may have
allowed us to avoid passing around a struct that contains information
about the current repository, but it bought us many, many problems.
In this case, when being called in a subdirectory, `git rev-parse`
changes the working directory to the top-level directory before calling
`git_path()`. In the new working directory, the result is correct. But
in the working directory of the calling script, it is incorrect.
Example: when calling `git rev-parse --git-path HEAD` in, say, the
Documentation/ subdirectory of Git's own source code, the string
`.git/HEAD` is printed.
Side note: that bug is hidden when running in a subdirectory of a
worktree that was added by the `git worktree` command: in that case, the
(correct) absolute path of the `HEAD` file is printed.
In the interest of time, this patch does not go the "correct" route to
introduce a struct with repository information (and removing global
state in the process), instead this patch chooses to detect when the
command was called in a subdirectory and forces the result to be an
absolute path.
While at it, we are also fixing the output of --git-common-dir and
--shared-index-path.
Lastly, please note that we reuse the same strbuf for all of the
relative_path() calls; this avoids frequent allocation (and duplicated
code), and it does not risk memory leaks, for two reasons: 1) the
cmd_rev_parse() function does not return anywhere between the use of
the new strbuf instance and its final release, and 2) git-rev-parse is
one of these "one-shot" programs in Git, i.e. it exits after running
for a very short time, meaning that all allocated memory is released
with the exit() call anyway.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"ls-files" run with pathspec has been micro-optimized to avoid
having to memmove(3) unnecessary bytes.
* rs/ls-files-partial-optim:
ls-files: move only kept cache entries in prune_cache()
ls-files: pass prefix length explicitly to prune_cache()
All these warning() calls are preceded by a system call. Report the
actual error to help the user understand why we fail to remove
something.
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We make several starts_with() calls, only to advance
pointers. This is exactly what skip_prefix() is for, which
lets us avoid manually-counted magic numbers.
Helped-by: Pranit Bauva <pranit.bauva@gmail.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
git-grep has always disallowed grepping in a tree (as
opposed to the working directory) with both --untracked
and --no-index. But we traditionally did so by first
collecting the revs, and then complaining when any were
provided.
The --no-index option recently learned to detect revs
much earlier. This has two user-visible effects:
- we don't bother to resolve revision names at all. So
when there's a rev/path ambiguity, we always choose to
treat it as a path.
- likewise, when you do specify a revision without "--",
the error you get is "no such path" and not "--untracked
cannot be used with revs".
The rationale for doing this with --no-index is that it is
meant to be used outside a repository, and so parsing revs
at all does not make sense.
This patch gives --untracked the same treatment. While it
_is_ meant to be used in a repository, it is explicitly
about grepping the non-repository contents. Telling the user
"we found a rev, but you are not allowed to use revs" is
not really helpful compared to "we treated your argument as
a path, and could not find it".
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We resolve HEAD and copy the result to a fixed-size buffer
with memcpy, never checking that it actually fits. This bug
dates back to 8098a178b (Add git-symbolic-ref, 2005-09-30).
Before that we used readlink(), which took a maximum buffer
size.
We can fix this by using resolve_refdup(), which duplicates
the buffer on the heap. That also lets us just check
for a NULL pointer to see if we have resolved HEAD, and
drop the extra head_p variable.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We copy the result of resolving HEAD into a buffer and keep
track of its length. But we never actually use the length
for anything besides the copy. Let's stop passing it around.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If we are using --no-index, then our arguments cannot be
revs in the first place. Not only is it pointless to
diagnose them, but if we are not in a repository, we should
not be trying to resolve any names.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We disallow the use of revisions with --no-index, but we
don't actually check and complain until well after we've
parsed the revisions.
This is the cause of a few problems:
1. We shouldn't be calling get_sha1() at all when we aren't
in a repository, as it might access the ref or object
databases. For now, this should generally just return
failure, but eventually it will become a BUG().
2. When there's a "--" disambiguator and you're outside a
repository, we'll complain early with "unable to resolve
revision". But we can give a much more specific error.
3. When there isn't a "--" disambiguator, we still do the
normal rev/path checks. This is silly, as we know we
cannot have any revs with --no-index. Everything we see
must be a path.
Outside of a repository this doesn't matter (since we
know it won't resolve), but inside one, we may complain
unnecessarily if a filename happens to also match a
refname.
This patch skips the get_sha1() call entirely in the
no-index case, and behaves as if it failed (with the
exception of giving a better error message).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If we see "git grep pattern rev -- file" then we apply the
usual rev/pathspec disambiguation rules: any "rev" before
the "--" must be a revision, and we do not need to apply the
verify_non_filename() check.
But there are two bugs here:
1. We keep a seen_dashdash flag to handle this case, but
we set it in the same left-to-right pass over the
arguments in which we parse "rev".
So when we see "rev", we do not yet know that there is
a "--", and we mistakenly complain if there is a
matching file.
We can fix this by making a preliminary pass over the
arguments to find the "--", and only then checking the rev
arguments.
2. If we can't resolve "rev" but there isn't a dashdash,
that's OK. We treat it like a path, and complain later
if it doesn't exist.
But if there _is_ a dashdash, then we know it must be a
rev, and should treat it as such, complaining if it
does not resolve. The current code instead ignores it
and tries to treat it like a path.
This patch fixes both bugs, and tries to comment the parsing
flow a bit better.
It adds tests that cover the two bugs, but also some related
situations (which already worked, but this confirms that our
fixes did not break anything).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We loop over the arguments, but every branch of the loop
hits either a "continue" or a "break". Surely we can make
this simpler.
The final conditional is:
if (arg is a rev) {
... handle rev ...
continue;
}
break;
We can rewrite this as:
if (arg is not a rev)
break;
... handle rev ...
That makes the flow a little bit simpler, and will make
things much easier to follow when we add more logic in
future patches.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When running a command of the form
git grep --no-index pattern -- path
in the absence of a Git repository, an error message will be printed:
fatal: BUG: setup_git_env called without repository
This is because "git grep" tries to interpret "--" as a rev. "git grep"
has always tried to first interpret "--" as a rev for at least a few
years, but this issue was upgraded from a pessimization to a bug in
commit 59332d1 ("Resurrect "git grep --no-index"", 2010-02-06), which
calls get_sha1 regardless of whether --no-index was specified. This bug
appeared to be benign until commit b1ef400 ("setup_git_env: avoid blind
fall-back to ".git"", 2016-10-20) when Git was taught to die in this
situation. (This "git grep" bug appears to be one of the bugs that
commit b1ef400 is meant to flush out.)
Therefore, always interpret "--" as signaling the end of options,
instead of trying to interpret it as a rev first.
Signed-off-by: Jonathan Tan <jonathantanmy@google.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Originally, we set up the threads for grep before parsing
the non-option arguments. In 53b8d931b (grep: disable
threading in non-worktree case, 2011-12-12), the thread code
got bumped lower in the function because it now needed to
know whether we got any revision arguments.
That put a big block of code in between the parsing of revs
and the parsing of pathspecs, both of which share some loop
variables. That makes it harder to read the code than the
original, where the shared loops were right next to each
other.
Let's bump the thread initialization until after all of the
parsing is done.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
A server can end up in a state where there are lots of unreferenced
loose objects (say, because many users are doing a bunch of rebasing
and pushing their rebased branches). Running "git gc --auto" in
this state would cause a gc.log file to be created, preventing
future auto gcs, causing pack files to pile up. Since many git
operations are O(n) in the number of pack files, this would lead to
poor performance.
Git should never get itself into a state where it refuses to do any
maintenance, just because at some point some piece of the maintenance
didn't make progress.
Teach Git to ignore gc.log files which are older than (by default)
one day old, which can be tweaked via the gc.logExpiry configuration
variable. That way, these pack files will get cleaned up, if
necessary, at least once per day. And operators who find a need for
more-frequent gcs can adjust gc.logExpiry to meet their needs.
There is also some cleanup: a successful manual gc, or a
warning-free auto gc with an old log file, will remove any old
gc.log files.
It might still happen that manual intervention is required
(e.g. because the repo is corrupt), but at the very least it won't
be because Git is too dumb to try again.
Signed-off-by: David Turner <dturner@twosigma.com>
Helped-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Don't throw the memory allocated for remove_dir_recursively() away after
a single call, use it for the other entries as well instead.
This change was done before in deb8e15a (rm: reuse strbuf for all
remove_dir_recursively() calls), but was reverted as a side-effect of
55856a35 (rm: absorb a submodules git dir before deletion). Reinstate
the optimization.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
prune_cache() first identifies those entries at the start of the sorted
array that can be discarded. Then it moves the rest of the entries up.
Last it identifies the unwanted trailing entries among the moved ones
and cuts them off.
Change the order: Identify both start *and* end of the range to keep
first and then move only those entries to the top. The resulting code
is slightly shorter and a bit more efficient.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Reviewed-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The function prune_cache() relies on the fact that it is only called on
max_prefix and sneakily uses the matching global variable max_prefix_len
directly. Tighten its interface by passing both the string and its
length as parameters. While at it move the NULL check into the function
to collect all cache-pruning related logic in one place.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Reviewed-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Git's interactive rebase is still implemented as a shell script, despite
its complexity. This implies that it suffers from the portability point
of view, from lack of expressibility, and of course also from
performance. The latter issue is particularly serious on Windows, where
we pay a hefty price for relying so much on POSIX.
Unfortunately, being such a huge shell script also means that we missed
the train when it would have been relatively easy to port it to C, and
instead piled feature upon feature onto that poor script that originally
never intended to be more than a slightly pimped cherry-pick in a loop.
To open the road toward better performance (in addition to all the other
benefits of C over shell scripts), let's just start *somewhere*.
The approach taken here is to add a builtin helper that at first intends
to take care of the parts of the interactive rebase that are most
affected by the performance penalties mentioned above.
In particular, after we spent all those efforts on preparing the sequencer
to process rebase -i's git-rebase-todo scripts, we implement the `git
rebase -i --continue` functionality as a new builtin, git-rebase--helper.
Once that is in place, we can work gradually on tackling the rest of the
technical debt.
Note that the rebase--helper needs to learn about the transient
--ff/--no-ff options of git-rebase, as the corresponding flag is not
persisted to, and re-read from, the state directory.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When tags are created with `--create-reflog` or with the option
`core.logAllRefUpdates` set to 'always', a reflog is created for them.
So far, the description of reflog entries for tags was empty, making the
reflog hard to understand. For example:
6e3a7b3 refs/tags/test@{0}:
Now, a reflog message is generated when creating a tag, following the
pattern "tag: tagging <short-sha1> (<description>)". If
GIT_REFLOG_ACTION is set, the message becomes "$GIT_REFLOG_ACTION
(<description>)" instead. If the tag references a commit object, the
description is set to the subject line of the commit, followed by its
commit date. For example:
6e3a7b3 refs/tags/test@{0}: tag: tagging 6e3a7b3398 (Git 2.12-rc0, 2017-02-03)
If the tag points to a tree/blob/tag objects, the following static
strings are taken as description:
- "tree object"
- "blob object"
- "other tag object"
Signed-off-by: Cornelius Weig <cornelius.weig@tngtech.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We de-duplicate ".have" refs among themselves, but never
check if they are duplicates of our local refs. It's not
unreasonable that they would be if we are a "--shared" or
"--reference" clone of a similar repository; we'd have all
the same tags.
We can handle this by inserting our local refs into the
oidset, but obviously not suppressing duplicates (since the
refnames are important).
Note that this also switches the order in which we advertise
refs, processing ours first and then any alternates. The
order shouldn't matter (and arguably showing our refs first
makes more sense).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Namely, de-duplicate them. We use the same set as the
alternates, since we call them both ".have" (i.e., there is
no value in showing one versus the other).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The comment claims that we handle alternate ".have" lines
through this function, but that hasn't been the case since
85f251045 (write_head_info(): handle "extra refs" locally,
2012-01-06).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If you have an alternate object store with a very large
number of refs, the peak memory usage of the sha1_array can
grow high, even if most of them are duplicates that end up
not being printed at all.
The similar for_each_alternate_ref() code-paths in
fetch-pack solve this by using flags in "struct object" to
de-duplicate (and so are relying on obj_hash at the core).
But we don't have a "struct object" at all in this case. We
could call lookup_unknown_object() to get one, but if our
goal is reducing memory footprint, it's not great:
- an unknown object is as large as the largest object type
(a commit), which is bigger than an oidset entry
- we can free the memory after our ref advertisement, but
"struct object" entries persist forever (and the
receive-pack may hang around for a long time, as the
bottleneck is often client upload bandwidth).
So let's use an oidset. Note that unlike a sha1-array it
doesn't sort the output as a side effect. However, our
output is at least stable, because for_each_alternate_ref()
will give us the sha1s in ref-sorted order.
In one particularly pathological case with an alternate that
has 60,000 unique refs out of 80 million total, this reduced
the peak heap usage of "git receive-pack . </dev/null" from
13GB to 14MB.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Breaking down the fields in the interface makes it easier to
change the backend of for_each_alternate_ref to something
that doesn't use "struct ref" internally.
The only field that callers actually look at is the oid,
anyway. The refname is kept in the interface as a plausible
thing for future code to want.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The `verbose` and `expire` options of the `git worktree prune`
subcommand have wrong descriptions in that they pretend to relate to
objects. But as the git-worktree(1) correctly states, these options have
nothing to do with objects but only with worktrees. Fix the description
accordingly.
Signed-off-by: Patrick Steinhardt <patrick.steinhardt@elego.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git difftool -h" reports an error:
fatal: BUG: setup_git_env called without repository
Defer repository setup so that the help option processing happens before
the repository is initialized.
Add tests to ensure that the basic usage works inside and outside of a
repository.
Signed-off-by: David Aguilar <davvid@gmail.com>
Acked-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The output of 'git rev-parse --git-dir' can be either a relative or an
absolute path, depending on whether the current working directory is
at the top of the worktree or the .git directory or not, or how the
path to the repository is specified via the '--git-dir=<path>' option
or the $GIT_DIR environment variable. And if that output is a
relative path, then it is relative to the directory where any 'git
-C <path>' options might have led us.
This doesn't matter at all for regular scripts, because the git
wrapper automatically takes care of changing directories according to
the '-C <path>' options, and the scripts can then simply follow any
path returned by 'git rev-parse --git-dir', even if it's a relative
path.
Our Bash completion script, however, is unique in that it must run
directly in the user's interactive shell environment. This means that
it's not executed through the git wrapper and would have to take care
of any '-C <path> options on its own, and it can't just change
directories as it pleases. Consequently, adding support for taking
any '-C <path>' options on the command line into account during
completion turned out to be considerably more difficult, error prone
and required more subshells and git processes when it had to cope with
a relative path to the .git directory.
Help this rather special use case and teach 'git rev-parse' a new
'--absolute-git-dir' option which always outputs a canonicalized
absolute path to the .git directory, regardless of whether the path is
discovered automatically or is specified via $GIT_DIR or 'git
--git-dir=<path>'.
Signed-off-by: SZEDER Gábor <szeder.dev@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The "core.logAllRefUpdates" that used to be boolean has been
enhanced to take 'always' as well, to record ref updates to refs
other than the ones that are expected to be updated (i.e. branches,
remote-tracking branches and notes).
* cw/log-updates-for-all-refs-really:
doc: add note about ignoring '--no-create-reflog'
update-ref: add test cases for bare repository
refs: add option core.logAllRefUpdates = always
config: add markup to core.logAllRefUpdates doc
"uchar [40]" to "struct object_id" conversion continues.
* rs/object-id:
checkout: convert post_checkout_hook() to struct object_id
use oidcpy() for copying hashes between instances of struct object_id
use oid_to_hex_r() for converting struct object_id hashes to hex strings
"git read-tree" and its underlying unpack_trees() machinery learned
to report problematic paths prefixed with the --super-prefix option.
* sb/unpack-trees-super-prefix:
unpack-trees: support super-prefix option
t1001: modernize style
t1000: modernize style
read-tree: use OPT_BOOL instead of OPT_SET_INT
Code clean-up.
* ep/commit-static-buf-cleanup:
builtin/commit.c: switch to strbuf, instead of snprintf()
builtin/commit.c: remove the PATH_MAX limitation via dynamic allocation
The remaining callers are all simple "I have N attributes I am
interested in. I'll ask about them with various paths one by one".
After this step, no caller to git_check_attrs() remains. After
removing it, we can extend "struct attr_check" struct with data
that can be used in optimizing the query for the specific N
attributes it contains.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This updates the other two ways the attribute check is done via an
array of "struct attr_check_item" elements. These two niches
appear only in "git check-attr".
* The caller does not know offhand what attributes it wants to ask
about and cannot use attr_check_initl() to prepare the
attr_check structure.
* The caller may not know what attributes it wants to ask at all,
and instead wants to learn everything that the given path has.
Such a caller can call attr_check_alloc() to allocate an empty
attr_check, and then call attr_check_append() to add attribute names
one by one.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The traditional API to check attributes is to prepare an N-element
array of "struct git_attr_check" and pass N and the array to the
function "git_check_attr()" as arguments.
In preparation to revamp the API to pass a single structure, in
which these N elements are held, rename the type used for these
individual array elements to "struct attr_check_item" and rename
the function to "git_check_attrs()".
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Brandon Williams <bmwill@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git blame --porcelain" misidentified the "previous" <commit, path>
pair (aka "source") when contents came from two or more files.
* jk/blame-fixes:
blame: output porcelain "previous" header for each file
blame: handle --no-abbrev
blame: fix alignment with --abbrev=40
It is natural that "git gc --auto" may not attempt to pack
everything into a single pack, and there is no point in warning
when the user has configured the system to use the pack bitmap,
leading to disabling further "gc".
* dt/disable-bitmap-in-auto-gc:
repack: die on incremental + write-bitmap-index
auto gc: don't write bitmaps for incremental repacks
Compression setting for producing packfiles were spread across
three codepaths, one of which did not honor any configuration.
Unify these so that all of them honor core.compression and
pack.compression variables the same way.
* jc/compression-config:
compression: unify pack.compression configuration parsing
"git fsck --connectivity-check" was not working at all.
* jk/fsck-connectivity-check-fix:
fsck: lazily load types under --connectivity-only
fsck: move typename() printing to its own function
t1450: use "mv -f" within loose object directory
fsck: check HAS_OBJ more consistently
fsck: do not fallback "git fsck <bogus>" to "git fsck"
fsck: tighten error-checks of "git fsck <head>"
fsck: prepare dummy objects for --connectivity-check
fsck: report trees as dangling
t1450: clean up sub-objects in duplicate-entry test
Rewrite a scripted porcelain "git difftool" in C.
* js/difftool-builtin:
difftool: hack around -Wzero-length-format warning
difftool: retire the scripted version
difftool: implement the functionality in the builtin
difftool: add a skeleton for the upcoming builtin
"git show-ref HEAD" used with "--verify" because the user is not
interested in seeing refs/remotes/origin/HEAD, and used with
"--head" because the user does not want HEAD to be filtered out,
i.e. "git show-ref --head --verify HEAD", did not work as expected.
* vp/show-ref-verify-head:
show-ref: remove a stale comment
show-ref: remove dead `if (verify)' check
show-ref: detect dangling refs under --verify as well
show-ref: move --quiet handling into show_one()
show-ref: allow -d to work with --verify
show-ref: accept HEAD with --verify
With anticipatory tweaking for remotes defined in ~/.gitconfig
(e.g. "remote.origin.prune" set to true, even though there may or
may not actually be "origin" remote defined in a particular Git
repository), "git remote rename" and other commands misinterpreted
and behaved as if such a non-existing remote actually existed.
* js/remote-rename-with-half-configured-remote:
remote rename: more carefully determine whether a remote is configured
remote rename: demonstrate a bogus "remote exists" bug
"git tag" and "git verify-tag" learned to put GPG verification
status in their "--format=<placeholders>" output format.
* st/verify-tag:
t/t7004-tag: Add --format specifier tests
t/t7030-verify-tag: Add --format specifier tests
builtin/tag: add --format argument for tag -v
builtin/verify-tag: add --format to verify-tag
ref-filter: add function to print single ref_array_item
gpg-interface, tag: add GPG_VERIFY_OMIT_STATUS flag
"git submodule push" learned "--recurse-submodules=only option to
push submodules out without pushing the top-level superproject.
* bw/push-submodule-only:
push: add option to push only submodules
submodules: add RECURSE_SUBMODULES_ONLY value
transport: reformat flag #defines to be more readable
Implement the '--format' option provided by 'ref-filter'. This lets the
user list branches as per desired format similar to the implementation
in 'git for-each-ref'.
Add tests and documentation for the same.
Mentored-by: Christian Couder <christian.couder@gmail.com>
Mentored-by: Matthieu Moy <matthieu.moy@grenoble-inp.fr>
Signed-off-by: Karthik Nayak <karthik.188@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Port branch.c to use ref-filter APIs for printing. This clears out
most of the code used in branch.c for printing and replaces them with
calls made to the ref-filter library.
Introduce build_format() which gets the format required for printing
of refs. Make amendments to print_ref_list() to reflect these changes.
The strings included in build_format() may not be safely quoted for
inclusion (i.e. it might contain '%' which needs to be escaped with an
additional '%'). Introduce quote_literal_for_format() as a helper
function which takes a string and returns a version of the string that
is safely quoted to be used in the for-each-ref format which is built
in build_format().
Change calc_maxwidth() to also account for the length of HEAD ref, by
calling ref-filter:get_head_discription().
Also change the test in t6040 to reflect the changes.
Before this patch, all cross-prefix symrefs weren't shortened. Since
we're using ref-filter APIs, we shorten all symrefs by default. We also
allow the user to change the format if needed with the introduction of
the '--format' option in the next patch.
Mentored-by: Christian Couder <christian.couder@gmail.com>
Mentored-by: Matthieu Moy <matthieu.moy@grenoble-inp.fr>
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Jeff King <peff@peff.net>
Helped-by: Ramsay Jones <ramsay@ramsayjones.plus.com>
Signed-off-by: Karthik Nayak <karthik.188@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Call ref-filter's setup_ref_filter_porcelain_msg() to enable
translated messages for the %(upstream:tack) atom. Although branch.c
doesn't currently use ref-filter's printing API's, this will ensure
that when it does in the future patches, we do not need to worry about
translation.
Written-by: Matthieu Moy <matthieu.moy@grenoble-inp.fr>
Mentored-by: Christian Couder <christian.couder@gmail.com>
Mentored-by: Matthieu Moy <matthieu.moy@grenoble-inp.fr>
Signed-off-by: Karthik Nayak <karthik.188@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Switch to dynamic allocation with strbuf, so we can avoid dealing
with magic numbers in the code and reduce the cognitive burden from
the programmers. The original code is correct, but programmers no
longer have to count bytes needed for static allocation to know that.
As a side effect of this change, we also reduce the snprintf()
calls, that may silently truncate results if the programmer is not
careful.
Helped-by: René Scharfe <l.s.r@web.de>
Helped-by: Junio C Hamano <gitster@pobox.com>
Helped-by: Jeff King <peff@peff.net>
Signed-off-by: Elia Pinto <gitter.spiros@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When core.logallrefupdates is true, we only create a new reflog for refs
that are under certain well-known hierarchies. The reason is that we
know that some hierarchies (like refs/tags) are not meant to change, and
that unknown hierarchies might not want reflogs at all (e.g., a
hypothetical refs/foo might be meant to change often and drop old
history immediately).
However, sometimes it is useful to override this decision and simply log
for all refs, because the safety and audit trail is more important than
the performance implications of keeping the log around.
This patch introduces a new "always" mode for the core.logallrefupdates
option which will log updates to everything under refs/, regardless
where in the hierarchy it is (we still will not log things like
ORIG_HEAD and FETCH_HEAD, which are known to be transient).
Based-on-patch-by: Jeff King <peff@peff.net>
Signed-off-by: Cornelius Weig <cornelius.weig@tngtech.com>
Reviewed-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
string_list_clear() handles empty lists just fine, so remove the
redundant check.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Patch generated by Coccinelle and contrib/coccinelle/object_id.cocci.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Apply the semantic patch swap.cocci to convert hand-rolled swaps to use
the macro SWAP. The resulting code is shorter and easier to read, the
object code is effectively unchanged.
The patch for object.c had to be hand-edited in order to preserve the
comment before the change; Coccinelle tried to eat it for some reason.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The break_delta_chain() function is recursive over the depth
of a given delta chain, which can lead to possibly running
out of stack space. Normally delta depth is quite small, but
if there _is_ a pathological case, this is where we would
find and fix it, so we should be more careful.
We can do it without recursion at all, but there's a little
bit of cleverness needed to do so. It's easiest to explain
by covering the less-clever strategies first.
The obvious thing to try is just keeping our own stack on
the heap. Whenever we would recurse, push the new entry onto
the stack and loop instead. But this gets tricky; when we
see an ACTIVE entry, we need to care if we just pushed it
(in which case it's a cycle) or if we just popped it (in
which case we dealt with its bases, and no we need to clear
the ACTIVE flag and compute its depth).
You can hack around that in various ways, like keeping a
"just pushed" flag, but the logic gets muddled. However, we
can observe that we do all of our pushes first, and then all
of our pops afterwards. In other words, we can do this in
two passes. First dig down to the base, stopping when we see
a cycle, and pushing each item onto our stack. Then pop the
stack elements, clearing the ACTIVE flag and computing the
depth for each.
This works, and is reasonably elegant. However, why do we
need the stack for the second pass? We can just walk the
delta pointers again. There's one complication. Popping the
stack went over our list in reverse, so we could compute the
depth of each entry by incrementing the depth of its base,
which we will have just computed. To go forward in the
second pass, we have to compute the total depth on the way
down, and then assign it as we go.
This patch implements this final strategy, because it not
only keeps the memory off the stack, but it eliminates it
entirely. Credit for the cleverness in that approach goes to
Michael Haggerty; bugs are mine.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since 898b14c (pack-objects: rework check_delta_limit usage,
2007-04-16), we check the delta depth limit only when
figuring out whether we should make a new delta. We don't
consider it at all when reusing deltas, which means that
packing once with --depth=250, and then again with
--depth=50, the second pack may still contain chains larger
than 50.
This is generally considered a feature, as the results of
earlier high-depth repacks are carried forward, used for
serving fetches, etc. However, since we started using
cross-pack deltas in c9af708b1 (pack-objects: use mru list
when iterating over packs, 2016-08-11), we are no longer
bounded by the length of an existing delta chain in a single
pack.
Here's one particular pathological case: a sequence of N
packs, each with 2 objects, the base of which is stored as a
delta in a previous pack. If we chain all the deltas
together, we have a cycle of length N. We break the cycle,
but the tip delta is still at depth N-1.
This is less unlikely than it might sound. See the included
test for a reconstruction based on real-world actions. I
ran into such a case in the wild, where a client was rapidly
sending packs, and we had accumulated 10,000 before doing a
server-side repack. The pack that "git repack" tried to
generate had a very deep chain, which caused pack-objects to
run out of stack space in the recursive write_one().
This patch bounds the length of delta chains in the output
pack based on --depth, regardless of whether they are caused
by cross-pack deltas or existed in the input packs. This
fixes the problem, but does have two possible downsides:
1. High-depth aggressive repacks followed by "normal"
repacks will throw away the high-depth chains.
In the long run this is probably OK; investigation
showed that high-depth repacks aren't actually
beneficial, and we dropped the aggressive depth default
to match the normal case in 07e7dbf0d (gc: default
aggressive depth to 50, 2016-08-11).
2. If you really do want to store high-depth deltas on
disk, they may be discarded and new delta computed when
serving a fetch, unless you set pack.depth to match
your high-depth size.
The implementation uses the existing search for delta
cycles. That lets us compute the depth of any node based on
the depth of its base, because we know the base is DFS_DONE
by the time we look at it (modulo any cycles in the graph,
but we know there cannot be any because we break them as we
see them).
There is some subtlety worth mentioning, though. We record
the depth of each object as we compute it. It might seem
like we could save the per-object storage space by just
keeping track of the depth of our traversal (i.e., have
break_delta_chains() report how deep it went). But we may
visit an object through multiple delta paths, and on
subsequent paths we want to know its depth immediately,
without having to walk back down to its final base (doing so
would make our graph walk quadratic rather than linear).
Likewise, one could try to record the depth not from the
base, but from our starting point (i.e., start
recursion_depth at 0, and pass "recursion_depth + 1" to each
invocation of break_delta_chains()). And then when
recursion_depth gets too big, we know that we must cut the
delta chain. But that technique is wrong if we do not visit
the nodes in topological order. In a chain A->B->C, it
if we visit "C", then "B", then "A", we will never recurse
deeper than 1 link (because we see at each node that we have
already visited it).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Apply the semantic patch for converting callers that duplicate the
result of absolute_path() to call absolute_pathdup() instead, which
avoids an extra string copy to a static buffer.
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The recent fixes to "fsck --connectivity-only" load all of
the objects with their correct types. This keeps the
connectivity-only code path close to the regular one, but it
also introduces some unnecessary inefficiency. While getting
the type of an object is cheap compared to actually opening
and parsing the object (as the non-connectivity-only case
would do), it's still not free.
For reachable non-blob objects, we end up having to parse
them later anyway (to see what they point to), making our
type lookup here redundant.
For unreachable objects, we might never hit them at all in
the reachability traversal, making the lookup completely
wasted. And in some cases, we might have quite a few
unreachable objects (e.g., when alternates are used for
shared object storage between repositories, it's normal for
there to be objects reachable from other repositories but
not the one running fsck).
The comment in mark_object_for_connectivity() claims two
benefits to getting the type up front:
1. We need to know the types during fsck_walk(). (And not
explicitly mentioned, but we also need them when
printing the types of broken or dangling commits).
We can address this by lazy-loading the types as
necessary. Most objects never need this lazy-load at
all, because they fall into one of these categories:
a. Reachable from our tips, and are coerced into the
correct type as we traverse (e.g., a parent link
will call lookup_commit(), which converts OBJ_NONE
to OBJ_COMMIT).
b. Unreachable, but not at the tip of a chunk of
unreachable history. We only mention the tips as
"dangling", so an unreachable commit which links
to hundreds of other objects needs only report the
type of the tip commit.
2. It serves as a cross-check that the coercion in (1a) is
correct (i.e., we'll complain about a parent link that
points to a blob). But we get most of this for free
already, because right after coercing, we'll parse any
non-blob objects. So we'd notice then if we expected a
commit and got a blob.
The one exception is when we expect a blob, in which
case we never actually read the object contents.
So this is a slight weakening, but given that the whole
point of --connectivity-only is to sacrifice some data
integrity checks for speed, this seems like an
acceptable tradeoff.
Here are before and after timings for an extreme case with
~5M reachable objects and another ~12M unreachable (it's the
torvalds/linux repository on GitHub, connected to shared
storage for all of the other kernel forks):
[before]
$ time git fsck --no-dangling --connectivity-only
real 3m4.323s
user 1m25.121s
sys 1m38.710s
[after]
$ time git fsck --no-dangling --connectivity-only
real 0m51.497s
user 0m49.575s
sys 0m1.776s
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When an object has a problem, we mention its type. But we do
so by feeding the result of typename() directly to
fprintf(). This is potentially dangerous because typename()
can return NULL for some type values (like OBJ_NONE).
It's doubtful that this can be triggered in practice with
the current code, so this is probably not fixing a bug. But
it future-proofs us against modifications that make things
like OBJ_NONE more likely (and gives future patches a
central point to handle them).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Building with "gcc -Wall" will complain that the format in:
warning("")
is empty. Which is true, but the warning is over-eager. We
are calling the function for its side effect of printing
"warning:", even with an empty string.
Our DEVELOPER Makefile knob disables the warning, but not
everybody uses it. Let's silence the warning in the code so
that nobody reports it or tries to "fix" it.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When cf0adba788 ("Store peeled refs in packed-refs file.",
2006-11-19) made the command to die with a message on error even
when --quiet is passed, it left the comment to say it changed the
semantics. But that kind of information belongs to the log message,
not in-code comment. Besides, the behaviour after the change has
been the established one for the past 10 years ;-)
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach git-describe the `--exclude` option which will allow specifying
a glob pattern of tags to ignore. This can be combined with the
`--match` patterns to enable more flexibility in determining which tags
to consider.
For example, suppose you wish to find the first official release tag
that contains a certain commit. If we assume that official release tags
are of the form "v*" and pre-release candidates include "*rc*" in their
name, we can now find the first release tag that introduces the commit
abcdef:
git describe --contains --match="v*" --exclude="*rc*" abcdef
Add documentation, tests, and completion for this change.
Signed-off-by: Jacob Keller <jacob.keller@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach `--match` to be accepted multiple times, accumulating a list of
patterns to match into a string list. Each pattern is inclusive, such
that a tag need only match one of the provided patterns to be
considered for matching.
This extension is useful as it enables more flexibility in what tags
match, and may avoid the need to run the describe command multiple
times to get the same result.
Add tests and update the documentation for this change.
Signed-off-by: Jacob Keller <jacob.keller@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Extend git-name-rev to support excluding refs which match shell patterns
using --exclude. These patterns can be used to limit the scope of refs
by excluding any ref that matches one of the --exclude patterns. A ref
will only be used for naming when it matches at least one --refs pattern
but does not match any of the --exclude patterns. Thus, --exclude
patterns are given precedence over --refs patterns.
For example, suppose you wish to name a series of commits based on an
official release tag of the form "v*" but excluding any pre-release tags
which match "*rc*". You can use the following to do so:
git name-rev --refs="v*" --exclude="*rc*" --all
Add tests and update Documentation for this change.
Signed-off-by: Jacob Keller <jacob.keller@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach git name-rev to take multiple --refs stored as a string list of
patterns. The list of patterns will be matched inclusively, and each ref
only needs to match one pattern to be included. A ref will only be
excluded if it does not match any of the given patterns. Additionally,
if any of the patterns would allow abbreviation, then we will abbreviate
the ref, even if another pattern is more strict and would not have
allowed abbreviation on its own.
Add tests and documentation for this change. The tests expected output
is dynamically generated. This is in order to avoid hard-coding
a commit object name in the test results (as the expected output is to
simply leave the commit object unnamed).
Signed-off-by: Jacob Keller <jacob.keller@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
As show_ref() is only ever called on the path where --verify is not
specified, `verify' can never possibly be true here.
Signed-off-by: Vladimir Panteleev <git@thecybershadow.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Move detection of dangling refs into show_one(), so that they are
detected when --verify is present as well as when it is absent.
Signed-off-by: Vladimir Panteleev <git@thecybershadow.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Do the same with --quiet as was done with -d, to remove the need to
perform this check at show_one()'s call site from the --verify branch.
Signed-off-by: Vladimir Panteleev <git@thecybershadow.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Move handling of -d into show_one(), so that it takes effect when
--verify is present as well as when it is absent. This is useful when
the user wishes to avoid the costly iteration of refs.
Signed-off-by: Vladimir Panteleev <git@thecybershadow.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Previously, when --verify was specified, show-ref would use a separate
code path which did not handle HEAD and treated it as an invalid
ref. Thus, "git show-ref --verify HEAD" (where "--verify" is used
because the user is not interested in seeing refs/remotes/origin/HEAD)
did not work as expected.
Instead of insisting that the input begins with "refs/", allow "HEAD"
as well in the codepath that handles "--verify", so that all valid
full refnames including HEAD are passed to the same output machinery.
Signed-off-by: Vladimir Panteleev <git@thecybershadow.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
One of the really nice features of the ~/.gitconfig file is that users
can override defaults by their own preferred settings for all of their
repositories.
One such default that some users like to override is whether the
"origin" remote gets auto-pruned or not. The user would simply call
git config --global remote.origin.prune true
and from now on all "origin" remotes would be pruned automatically when
fetching into the local repository.
There is just one catch: now Git thinks that the "origin" remote is
configured, even if the repository config has no [remote "origin"]
section at all, as it does not realize that the "prune" setting was
configured globally and that there really is no "origin" remote
configured in this repository.
That is a problem e.g. when renaming a remote to a new name, when Git
may be fooled into thinking that there is already a remote of that new
name.
Let's fix this by paying more attention to *where* the remote settings
came from: if they are configured in the local repository config, we
must not overwrite them. If they were configured elsewhere, we cannot
overwrite them to begin with, as we only write the repository config.
There is only one caller of remote_is_configured() (in `git fetch`) that
may want to take remotes into account even if they were configured
outside the repository config; all other callers essentially try to
prevent the Git command from overwriting settings in the repository
config.
To accommodate that fact, the remote_is_configured() function now
requires a parameter that states whether the caller is interested in all
remotes, or only in those that were configured in the repository config.
Many thanks to Jeff King whose tireless review helped with settling for
nothing less than the current strategy.
This fixes https://github.com/git-for-windows/git/issues/888
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
It served its purpose, but now we have a builtin difftool. Time for the
Perl script to enjoy Florida.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This patch gives life to the skeleton added in the previous patch.
The motivation for converting the difftool is that Perl scripts are not at
all native on Windows, and that `git difftool` therefore is pretty slow on
that platform, when there is no good reason for it to be slow.
In addition, Perl does not really have access to Git's internals. That
means that any script will always have to jump through unnecessary
hoops, and it will often need to perform unnecessary work (e.g. when
reading the entire config every time `git config` is called to query a
single config value).
The current version of the builtin difftool does not, however, make full
use of the internals but instead chooses to spawn a couple of Git
processes, still, to make for an easier conversion. There remains a lot
of room for improvement, left later.
Note: to play it safe, the original difftool is still called unless the
config setting difftool.useBuiltin is set to true.
The reason: this new, experimental, builtin difftool was shipped as part
of Git for Windows v2.11.0, to allow for easier large-scale testing, but
of course as an opt-in feature.
The speedup is actually more noticable on Linux than on Windows: a quick
test shows that t7800-difftool.sh runs in (2.183s/0.052s/0.108s)
(real/user/sys) in a Linux VM, down from (6.529s/3.112s/0.644s), while on
Windows, it is (36.064s/2.730s/7.194s), down from (47.637s/2.407s/6.863s).
The culprit is most likely the overhead incurred from *still* having to
shell out to mergetool-lib.sh and difftool--helper.sh.
Still, it is an improvement.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Code clean-up in the pathspec API.
* bw/pathspec-cleanup:
pathspec: rename prefix_pathspec to init_pathspec_item
pathspec: small readability changes
pathspec: create strip submodule slash helpers
pathspec: create parse_element_magic helper
pathspec: create parse_long_magic function
pathspec: create parse_short_magic function
pathspec: factor global magic into its own function
pathspec: simpler logic to prefix original pathspec elements
pathspec: always show mnemonic and name in unsupported_magic
pathspec: remove unused variable from unsupported_magic
pathspec: copy and free owned memory
pathspec: remove the deprecated get_pathspec function
ls-tree: convert show_recursive to use the pathspec struct interface
dir: convert fill_directory to use the pathspec struct interface
dir: remove struct path_simplify
mv: remove use of deprecated 'get_pathspec()'
"git blame --porcelain" misidentified the "previous" <commit, path>
pair (aka "source") when contents came from two or more files.
* jk/blame-fixes:
blame: output porcelain "previous" header for each file
blame: handle --no-abbrev
blame: fix alignment with --abbrev=40
It is natural that "git gc --auto" may not attempt to pack
everything into a single pack, and there is no point in warning
when the user has configured the system to use the pack bitmap,
leading to disabling further "gc".
* dt/disable-bitmap-in-auto-gc:
repack: die on incremental + write-bitmap-index
auto gc: don't write bitmaps for incremental repacks
"git rm" used to refuse to remove a submodule when it has its own
git repository embedded in its working tree. It learned to move
the repository away to $GIT_DIR/modules/ of the superproject
instead, and allow the submodule to be deleted (as long as there
will be no loss of local modifications, that is).
* sb/submodule-rm-absorb:
rm: absorb a submodules git dir before deletion
submodule: rename and add flags to ok_to_remove_submodule
submodule: modernize ok_to_remove_submodule to use argv_array
submodule.h: add extern keyword to functions
"git grep" has been taught to optionally recurse into submodules.
* bw/grep-recurse-submodules:
grep: search history of moved submodules
grep: enable recurse-submodules to work on <tree> objects
grep: optionally recurse into submodules
grep: add submodules as a grep source type
submodules: load gitmodules file from commit sha1
submodules: add helper to determine if a submodule is initialized
submodules: add helper to determine if a submodule is populated
real_path: canonicalize directory separators in root parts
real_path: have callers use real_pathdup and strbuf_realpath
real_path: create real_pathdup
real_path: convert real_path_internal to strbuf_realpath
real_path: resolve symlinks by hand
Adding --format to git tag -v mutes the default output of the GPG
verification and instead prints the formatted tag object.
This allows callers to cross-check the tagname from refs/tags with
the tagname from the tag object header upon GPG verification.
The callback function for for_each_tag_name() didn't allow callers to
pass custom data to their callback functions. Add a new opaque pointer
to each_tag_name_fn's parameter to allow this.
Signed-off-by: Lukas Puehringer <luk.puehringer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Callers of verify-tag may want to cross-check the tagname from refs/tags
with the tagname from the tag object header upon GPG verification. This
is to avoid tag refs that point to an incorrect object.
Add a --format parameter to git verify-tag to print the formatted tag
object header in addition to or instead of the --verbose or --raw GPG
verification output.
Signed-off-by: Santiago Torres <santiago@nyu.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Code cleanup to avoid using redundant refspecs while fetching with
the --tags option.
* jt/fetch-no-redundant-tag-fetch-map:
fetch: do not redundantly calculate tag refmap
A lazy "git push" without refspec did not internally use a fully
specified refspec to perform 'current', 'simple', or 'upstream'
push, causing unnecessary "ambiguous ref" errors.
* jc/push-default-explicit:
push: test pushing ambiguously named branches
push: do not use potentially ambiguous default refspec
"git index-pack --stdin" needs an access to an existing repository,
but "git index-pack file.pack" to generate an .idx file that
corresponds to a packfile does not.
* jk/index-pack-wo-repo-from-stdin:
index-pack: skip collision check when not in repository
t: use nongit() function where applicable
index-pack: complain when --stdin is used outside of a repo
t5000: extract nongit function to test-lib-functions.sh
Unlike "git am --abort", "git cherry-pick --abort" moved HEAD back
to where cherry-pick started while picking multiple changes, when
the cherry-pick stopped to ask for help from the user, and the user
did "git reset --hard" to a different commit in order to re-attempt
the operation.
* sb/sequencer-abort-safety:
Revert "sequencer: remove useless get_dir() function"
sequencer: remove useless get_dir() function
sequencer: make sequencer abort safer
t3510: test that cherry-pick --abort does not unsafely change HEAD
am: change safe_to_abort()'s not rewinding error into a warning
am: fix filename in safe_to_abort() error message
"git pull --rebase", when there is no new commits on our side since
we forked from the upstream, should be able to fast-forward without
invoking "git rebase", but it didn't.
* jc/pull-rebase-ff:
pull: fast-forward "pull --rebase=true"
"git commit --allow-empty --only" (no pathspec) with dirty index
ought to be an acceptable way to create a new commit that does not
change any paths, but it was forbidden, perhaps because nobody
needed it so far.
* ak/commit-only-allow-empty:
commit: remove 'Clever' message for --only --amend
commit: make --only --allow-empty work without paths
The output from "git worktree list" was made in readdir() order,
and was unstable.
* nd/worktree-list-fixup:
worktree list: keep the list sorted
worktree.c: get_worktrees() takes a new flag argument
get_worktrees() must return main worktree as first item even on error
worktree: reorder an if statement
worktree.c: zero new 'struct worktree' on allocation
"git rev-parse --symbolic" failed with a more recent notation like
"HEAD^-1" and "HEAD^!".
* jk/rev-parse-symbolic-parents-fix:
rev-parse: fix parent shorthands with --symbolic
There are two spots that call lookup_object() and assume
that a non-NULL result means we have the object:
1. When we're checking the objects given to us by the user
on the command line.
2. When we're checking if a reflog entry is valid.
This generally follows fsck's mental model that we will have
looked at and loaded a "struct object" for each object in
the repository. But it misses one case: if another object
_mentioned_ an object, but we didn't actually parse it or
verify that it exists, it will still have a struct.
It's not clear if this is a triggerable bug or not.
Certainly the later parts of the reachability check need to
be careful of this, and do so by checking the HAS_OBJ flag.
But both of these steps happen before we start traversing,
so probably we won't have followed any links yet. Still,
it's easy enough to be defensive here.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since fsck tries to continue as much as it can after seeing
an error, we still do the reachability check even if some
heads we were given on the command-line are bogus. But if
_none_ of the heads is is valid, we fallback to checking all
refs and the index, which is not what the user asked for at
all.
Instead of checking "heads", the number of successful heads
we got, check "argc" (which we know only has non-options in
it, because parse_options removed the others).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Instead of checking reachability from the refs, you can ask
fsck to check from a particular set of heads. However, the
error checking here is quite lax. In particular:
1. It claims lookup_object() will report an error, which
is not true. It only does a hash lookup, and the user
has no clue that their argument was skipped.
2. When either the name or sha1 cannot be resolved, we
continue to exit with a successful error code, even
though we didn't check what the user asked us to.
This patch fixes both of these cases.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Normally fsck makes a pass over all objects to check their
integrity, and then follows up with a reachability check to
make sure we have all of the referenced objects (and to know
which ones are dangling). The latter checks for the HAS_OBJ
flag in obj->flags to see if we found the object in the
first pass.
Commit 02976bf85 (fsck: introduce `git fsck --connectivity-only`,
2015-06-22) taught fsck to skip the initial pass, and to
fallback to has_sha1_file() instead of the HAS_OBJ check.
However, it converted only one HAS_OBJ check to use
has_sha1_file(). But there are many other places in
builtin/fsck.c that assume that the flag is set (or that
lookup_object() will return an object at all). This leads to
several bugs with --connectivity-only:
1. mark_object() will not queue objects for examination,
so recursively following links from commits to trees,
etc, did nothing. I.e., we were checking the
reachability of hardly anything at all.
2. When a set of heads is given on the command-line, we
use lookup_object() to see if they exist. But without
the initial pass, we assume nothing exists.
3. When loading reflog entries, we do a similar
lookup_object() check, and complain that the reflog is
broken if the object doesn't exist in our hash.
So in short, --connectivity-only is broken pretty badly, and
will claim that your repository is fine when it's not.
Presumably nobody noticed for a few reasons.
One is that the embedded test does not actually test the
recursive nature of the reachability check. All of the
missing objects are still in the index, and we directly
check items from the index. This patch modifies the test to
delete the index, which shows off breakage (1).
Another is that --connectivity-only just skips the initial
pass for loose objects. So on a real repository, the packed
objects were still checked correctly. But on the flipside,
it means that "git fsck --connectivity-only" still checks
the sha1 of all of the packed objects, nullifying its
original purpose of being a faster git-fsck.
And of course the final problem is that the bug only shows
up when there _is_ corruption, which is rare. So anybody
running "git fsck --connectivity-only" proactively would
assume it was being thorough, when it was not.
One possibility for fixing this is to find all of the spots
that rely on HAS_OBJ and tweak them for the connectivity-only
case. But besides the risk that we might miss a spot (and I
found three already, corresponding to the three bugs above),
there are other parts of fsck that _can't_ work without a
full list of objects. E.g., the list of dangling objects.
Instead, let's make the connectivity-only case look more
like the normal case. Rather than skip the initial pass
completely, we'll do an abbreviated one that sets up the
HAS_OBJ flag for each object, without actually loading the
object data.
That's simple and fast, and we don't have to care about the
connectivity_only flag in the rest of the code at all.
While we're at it, let's make sure we treat loose and packed
objects the same (i.e., setting up dummy objects for both
and skipping the actual sha1 check). That makes the
connectivity-only check actually fast on a real repo (40
seconds versus 180 seconds on my copy of linux.git).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This adds a builtin difftool that still falls back to the legacy Perl
version, which has been renamed to `legacy-difftool`.
The idea is that the new, experimental, builtin difftool immediately hands
off to the legacy difftool for now, unless the config variable
difftool.useBuiltin is set to true.
This feature flag will be used in the upcoming Git for Windows v2.11.0
release, to allow early testers to opt-in to use the builtin difftool and
flesh out any bugs.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
After checking connectivity, fsck looks through the list of
any objects we've seen mentioned, and reports unreachable
and un-"used" ones as dangling. However, it skips any object
which is not marked as "parsed", as that is an object that
we _don't_ have (but that somebody mentioned).
Since 6e454b9a3 (clear parsed flag when we free tree
buffers, 2013-06-05), that flag can't be relied on, and the
correct method is to check the HAS_OBJ flag. The cleanup in
that commit missed this callsite, though. As a result, we
would generally fail to report dangling trees.
We never noticed because there were no tests in this area
(for trees or otherwise). Let's add some.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
