The return value of write_in_full() is either "-1", or the
requested number of bytes[1]. If we make a partial write
before seeing an error, we still return -1, not a partial
value. This goes back to f6aa66cb95 (write_in_full: really
write in full or return error on disk full., 2007-01-11).
So checking anything except "was the return value negative"
is pointless. And there are a couple of reasons not to do
so:
1. It can do a funny signed/unsigned comparison. If your
"len" is signed (e.g., a size_t) then the compiler will
promote the "-1" to its unsigned variant.
This works out for "!= len" (unless you really were
trying to write the maximum size_t bytes), but is a
bug if you check "< len" (an example of which was fixed
recently in config.c).
We should avoid promoting the mental model that you
need to check the length at all, so that new sites are
not tempted to copy us.
2. Checking for a negative value is shorter to type,
especially when the length is an expression.
3. Linus says so. In d34cf19b89 (Clean up write_in_full()
users, 2007-01-11), right after the write_in_full()
semantics were changed, he wrote:
I really wish every "write_in_full()" user would just
check against "<0" now, but this fixes the nasty and
stupid ones.
Appeals to authority aside, this makes it clear that
writing it this way does not have an intentional
benefit. It's a historical curiosity that we never
bothered to clean up (and which was undoubtedly
cargo-culted into new sites).
So let's convert these obviously-correct cases (this
includes write_str_in_full(), which is just a wrapper for
write_in_full()).
[1] A careful reader may notice there is one way that
write_in_full() can return a different value. If we ask
write() to write N bytes and get a return value that is
_larger_ than N, we could return a larger total. But
besides the fact that this would imply a totally broken
version of write(), it would already invoke undefined
behavior. Our internal remaining counter is an unsigned
size_t, which means that subtracting too many byte will
wrap it around to a very large number. So we'll instantly
begin reading off the end of the buffer, trying to write
gigabytes (or petabytes) of data.
Signed-off-by: Jeff King <peff@peff.net>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We ask to write 41 bytes and make sure that the return value
is at least 41. This is the same "dangerous" pattern that
was fixed in the prior commit (wherein a negative return
value is promoted to unsigned), though it is not dangerous
here because our "41" is a constant, not an unsigned
variable.
But we should convert it anyway to avoid modeling a
dangerous construct.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The return type of write_in_full() is a signed ssize_t,
because we may return "-1" on failure (even if we succeeded
in writing some bytes). But "len" itself is may be an
unsigned type (the function takes a size_t, but of course we
may have something else in the calling function). So while
it seems like:
if (write_in_full(fd, buf, len) < len)
die_errno("write error");
would trigger on error, it won't if "len" is unsigned. The
compiler sees a signed/unsigned comparison and promotes the
signed value, resulting in (size_t)-1, the highest possible
size_t (or again, whatever type the caller has). This cannot
possibly be smaller than "len", and so the conditional can
never trigger.
I scoured the code base for cases of this, but it turns out
that these two in git_config_set_multivar_in_file_gently()
are the only ones. Here our "len" is the difference between
two size_t variables, making the result an unsigned size_t.
We can fix this by just checking for a negative return value
directly, as write_in_full() will never return any value
except -1 or the full count.
There's no addition to the test suite here, since you need
to convince write() to fail in order to see the problem. The
simplest reproduction recipe I came up with is to trigger
ENOSPC:
# make a limited-size filesystem
dd if=/dev/zero of=small.disk bs=1M count=1
mke2fs small.disk
mkdir mnt
sudo mount -o loop small.disk mnt
cd mnt
sudo chown $USER:$USER .
# make a config file with some content
git config --file=config one.key value
git config --file=config two.key value
# now fill up the disk
dd if=/dev/zero of=fill
# and try to delete a key, which requires copying the rest
# of the file to config.lock, and will fail on write()
git config --file=config --unset two.key
That final command should (and does after this patch)
produce an error message due to the failed write, and leave
the file intact. Instead, it silently ignores the failure
and renames config.lock into place, leaving you with a
totally empty config file!
Reported-by: demerphq <demerphq@gmail.com>
Signed-off-by: Jeff King <peff@peff.net>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Following are several fixes for duplicated words ("of of") and one
case where an extra article ("a") slipped in.
Signed-off-by: Evan Zacks <zackse@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
While git doesn't track empty directories, git archive can be tricked
into putting some into archives. One way is to construct an empty tree
object, as t5004 does. While that is supported by the object database,
it can't be represented in the index and thus it's unlikely to occur in
the wild.
Another way is using the literal name of a directory in an exclude
pathspec -- its contents are are excluded, but the directory stub is
included. That's inconsistent: exclude pathspecs containing wildcards
don't leave empty directories in the archive.
Yet another way is have a few levels of nested subdirectories (e.g.
d1/d2/d3/file1) and ignoring the entries at the leaves (e.g. file1).
The directories with the ignored content are ignored as well (e.g. d3),
but their empty parents are included (e.g. d2).
As empty directories are not supported by git, they should also not be
written into archives. If an empty directory is really needed then it
can be tracked and archived by placing an empty .gitignore file in it.
There already is a mechanism in place for suppressing empty directories.
When read_tree_recursive() encounters a directory excluded by a pathspec
then it enters it anyway because it might contain included entries. It
calls the callback function before it is able to decide if the directory
is actually needed. For that reason git archive adds directories to a
queue and writes entries for them only when it encounters the first
child item -- but currently only if pathspecs with wildcards are used.
Queue *all* directories, no matter if there even are pathspecs present.
This prevents git archive from writing entries for empty directories in
all cases.
Suggested-by: Jeff King <peff@peff.net>
Signed-off-by: Rene Scharfe <l.s.r@web.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Pass DO_FOR_EACH_INCLUDE_BROKEN when iterating over replacement refs
so that the iteration does not require opening the named objects from
the object store. This avoids a dependency cycle between object access
and replace ref iteration.
Moreover the ref subsystem has not been migrated yet to access the
object store via passed in repository objects. As a result, without
this patch, iterating over replace refs in a repository other than
the_repository it produces errors:
error: refs/replace/3afabef75c627b894cccc3bcae86837abc7c32fe does not point to a valid object!
Noticed while adapting the object store (and in particular its
evaluation of replace refs) to handle arbitrary repositories.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The check_has_commit helper uses resolves a submodule entry to a
commit, when validating its existence. As a side effect this means
tolerates a submodule entry pointing to a tag, which is not a valid
submodule entry that git commands would know how to cope with.
Tighten the check to require an actual commit, not a tag pointing to a
commit.
Also improve the error handling when a submodule entry points to
non-commit (e.g., a blob) to error out instead of warning and
pretending the pointed to object doesn't exist.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The MRU cache that keeps track of recently used packs is represented
using two global variables:
struct mru packed_git_mru_storage;
struct mru *packed_git_mru = &packed_git_mru_storage;
Callers never assign to the packed_git_mru pointer, though, so we can
simplify by eliminating it and using &packed_git_mru_storage (renamed
to &packed_git_mru) directly. This variable is only used by the
packfile subsystem, making this a relatively uninvasive change (and
any new unadapted callers would trigger a compile error).
Noticed while moving these globals to the object_store struct.
Signed-off-by: Stefan Beller <sbeller@google.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Acked-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When the user tries to use '--help' option on an aliased command
information about the alias is printed as sshown below,
$ git co --help
`git co' is aliased to `checkout'
This doesn't seem correct as the user has aliased only 'co' and not
'git co'. This might even be incorrect in cases in which the user has
used an alias like 'tgit'.
$ tgit co --help
`git co' is aliased to `checkout'
So, make the message more precise.
Signed-off-by: Kaartic Sivaraam <kaarticsivaraam91196@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Callers are only allowed to pass certain flags into
ref_transaction_update, other flags are internal to it. To prevent
mistakes from the callers, strip the internal only flags out before
continuing.
This was noticed because of a compiler warning gcc 7.1.1 issued about
passing a NULL parameter as second parameter to memcpy (through
hashcpy):
In file included from refs.c:5:0:
refs.c: In function ‘ref_transaction_verify’:
cache.h:948:2: error: argument 2 null where non-null expected [-Werror=nonnull]
memcpy(sha_dst, sha_src, GIT_SHA1_RAWSZ);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from git-compat-util.h:165:0,
from cache.h:4,
from refs.c:5:
/usr/include/string.h:43:14: note: in a call to function ‘memcpy’ declared here
extern void *memcpy (void *__restrict __dest, const void *__restrict __src,
^~~~~~
The call to hascpy in ref_transaction_add_update is protected by the
passed in flags, but as we only add flags there, gcc notices
REF_HAVE_NEW or REF_HAVE_OLD flags could be passed in from the outside,
which would potentially result in passing in NULL as second parameter to
memcpy.
Fix both the compiler warning, and make the interface safer for its
users by stripping the internal flags out.
Suggested-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
For count, sort and format, only the argument names were listed under
OPTIONS, not the option names.
Add the option names to make it clear the options exist
Signed-off-by: Kevin Daudt <me@ikke.info>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The synopsis and description inconsistently add a '=' between the
argument name and it's value. Make this consistent.
Signed-off-by: Kevin Daudt <me@ikke.info>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We run `git rev-parse` though the shell, and quote its
argument only with single-quotes. This prevents most
metacharacters from being a problem, but misses the obvious
case when $name itself has single-quotes in it. We can fix
this by applying the usual shell-quoting formula.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Refnames can contain shell metacharacters which need to be
passed verbatim to sub-processes. Using safe_pipe_capture
skips the shell entirely.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The git-cvsserver script is old and largely unmaintained
these days. But git-shell allows untrusted users to run it
out of the box, significantly increasing its attack surface.
Let's drop it from git-shell's list of internal handlers so
that it cannot be run by default. This is not backwards
compatible. But given the age and development activity on
CVS-related parts of Git, this is likely to impact very few
users, while helping many more (i.e., anybody who runs
git-shell and had no intention of supporting CVS).
There's no configuration mechanism in git-shell for us to
add a boolean and flip it to "off". But there is a mechanism
for adding custom commands, and adding CVS support here is
fairly trivial. Let's document it to give guidance to
anybody who really is still running cvsserver.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This makes the script pass arguments that are derived from end-user
input in safer way when invoking subcommands.
Reported-by: joernchen <joernchen@phenoelit.de>
Signed-off-by: joernchen <joernchen@phenoelit.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
As a preparation for replacing `command` with a call to this
function from outside GITCVS::updater package, move it to the main
package.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The handshake_capabilities() function first advertises the set of
capabilities it supports, so that the other side can pick and choose
which ones to use and ask us to enable in its response. Then we
read the response that tells us what choice the other side made. If
we saw something that we never advertised, that indicates one of two
things. The other side, i.e. the "upgraded" filter, is not paying
attention of the capabilities advertisement, and asking something
its correct operation relies on, but we are not capable of giving
that unknown feature and operate without it, so after that point the
exchange of data is a garbage-in-garbage-out. Or the other side
wanted to ask for one of the capabilities we advertised, but the
code has typo and their wish to enable a capability that its correct
operation relies on is not understood on this end. The result is
the same garbage-in-garbage-out.
Instead of sweeping such a potential bug under the rug, die loudly
when we see a request for an unsupported capability in order to
force sloppily-written filter scripts to get corrected.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If we push a branch and a tag pointing to the HEAD of this branch,
then Travis CI would run the build twice. This wastes resources and
slows the testing.
Add a function to detect this situation and skip the build the branch
if appropriate. Invoke this function on every build.
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Lars Schneider <larsxschneider@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Most of the Travis CI commands are in the '.travis.yml'. The yml format
does not support functions and therefore code duplication is necessary
to run commands across all builds.
To fix this, add a library for common CI functions. Move all Travis CI
code into dedicated scripts and make them call the library first.
Signed-off-by: Lars Schneider <larsxschneider@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git branch -M a b" while on a branch that is completely unrelated
to either branch a or branch b misbehaved when multiple worktree
was in use. This has been fixed.
* nd/worktree-kill-parse-ref:
branch: fix branch renaming not updating HEADs correctly
In addition to "cc: <a@dd.re.ss> # cruft", "cc: a@dd.re.ss # cruft"
was taught to "git send-email" as a valid way to tell it that it
needs to also send a carbon copy to <a@dd.re.ss> in the trailer
section.
* mm/send-email-cc-cruft:
send-email: don't use Mail::Address, even if available
send-email: fix garbage removal after address
The codepath to call external process filter for smudge/clean
operation learned to show the progress meter.
* ls/convert-filter-progress:
convert: display progress for filtered objects that have been delayed
Message and doc updates.
* ma/up-to-date:
treewide: correct several "up-to-date" to "up to date"
Documentation/user-manual: update outdated example output
Doc fix.
* mg/format-ref-doc-fix:
Documentation/git-for-each-ref: clarify peeling of tags for --format
Documentation: use proper wording for ref format strings
Jeff King