A previous commit changed ->tree to ->maybe_tree throughout, which may
have worked at the time, but wasn't safe, because maybe_tree is loaded
lazily. This manifested itself in crashes when using the "follow" log
feature. The proper fix is to use the correct contextual accessors
everytime we want access to maybe_tree. Thankfully, the commit.cocci
script takes care of creating mostly-correct patches that we could then
fix up, resulting in this commit here.
Fixes: 255b78f ("git: update to v2.18.0")
Reviewed-by: Christian Hesse <mail@eworm.de>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
This patch adds support for zstd [0] compressed snapshots (*.tar.zst).
We enable multiple working threads (-T0), but keep default compression
level. The latter can be influenced by environment variable.
[0] https://www.zstd.net/
Signed-off-by: Christian Hesse <mail@eworm.de>
This patch adds support for lzip [1] compressed snapshots (*.tar.lz)
[1] https://www.nongnu.org/lzip/
Signed-off-by: Hanspeter Portner <dev@open-music-kontrollers.ch>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Update to git version v2.25.0.
Upstream renamed 'init_display_notes()' to 'load_display_notes()' in
commit 1e6ed5441a61b5085978e0429691e2e2425f6846 ("notes: rename to
load_display_notes()").
Signed-off-by: Christian Hesse <mail@eworm.de>
Chances are that strace is available but not functional due to
restricted permissions:
strace: test_ptrace_get_syscall_info: PTRACE_TRACEME: Operation not permitted
strace: ptrace(PTRACE_TRACEME, ...): Operation not permitted
+++ exited with 1 +++
Just skip the tests then.
Signed-off-by: Christian Hesse <mail@eworm.de>
The function read_agefile() returns time_t, which is a signed datatime.
We should not return unsigned (negative) value here.
Reported-by: Johannes Stezenbach <js@linuxtv.org>
Signed-off-by: Christian Hesse <mail@eworm.de>
Update to git version v2.24.0.
Never use get_cached_commit_buffer() directly, use repo_get_commit_buffer()
instead. The latter calls the former anyway. This fixes segmentation fault
when commit-graph is enabled and get_cached_commit_buffer() does not return
the expected result.
Signed-off-by: Christian Hesse <mail@eworm.de>
Update to git version v2.22.0.
Upstream commit bce9db6d ("trace2: use system/global config for default
trace2 settings") caused a regression. We have to unset HOME and
XDG_CONFIG_HOME before early loading of config from trace2 code kicks in.
Signed-off-by: Christian Hesse <mail@eworm.de>
The blame operation can cause high cost in terms of CPU load for huge
repositories. Let's add a per repository override for enable-blame.
Signed-off-by: Christian Hesse <mail@eworm.de>
For testing versions the version string differs for git tag (v2.22.0-rc3)
and tarball file name (2.22.0.rc3). Let's fix validation for testing
versions.
Signed-off-by: Christian Hesse <mail@eworm.de>
Update to git version v2.21.0. Required changes follow upstream commits:
* 6a7895fd8a3bd409f2b71ffc355d5142172cc2a0
(commit: prepare free_commit_buffer and release_commit_memory for
any repo)
* e092073d643b17c82d72cf692fbfaea9c9796f11
(tree.c: make read_tree*() take 'struct repository *')
Signed-off-by: Christian Hesse <mail@eworm.de>
Reviewed-by: John Keeping <john@keeping.me.uk>
Git version v2.21.0 marks strncat() as banned (commit
ace5707a803eda0f1dde3d776dc3729d3bc7759a), so replace it.
Signed-off-by: Christian Hesse <mail@eworm.de>
Perhaps a more ideal version of this would be to not print breadcrumbs
at all for paths that don't exist in the given repo at the given oid.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Reported-by: Fydor Wire Snark <wsnark@tuta.io>
- Remove ids from anchor elements. They were unusable because they were
duplicated between files and versions of files.
- Always close span, with html().
- Fix missing / on closing tr element in cgit_ssdiff_header_end().
Signed-off-by: Chris Mayo <aklhfex@gmail.com>
The old algorithm was totally incorrect. While we're at it, use «
instead of \, since it makes more sense.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Update to git version v2.20.0. Required changes follow upstream commits:
* 00436bf1b1c2a8fe6cf5d2c2457d419d683042f4
(archive: initialize archivers earlier)
* 611e42a5980a3a9f8bb3b1b49c1abde63c7a191e
(xdiff: provide a separate emit callback for hunks)
Signed-off-by: Christian Hesse <mail@eworm.de>
Git upstream bans strcat() with commit:
banned.h: mark strcat() as banned
1b11b64b815db62f93a04242e4aed5687a448748
Signed-off-by: Christian Hesse <mail@eworm.de>
Git upstream bans strncpy() with commit:
banned.h: mark strncpy() as banned
e488b7aba743d23b830d239dcc33d9ca0745a9ad
Signed-off-by: Christian Hesse <mail@eworm.de>
Git upstream bans strcat() with commit:
banned.h: mark strcat() as banned
1b11b64b815db62f93a04242e4aed5687a448748
To avoid compiler warnings from gcc 8.1.x we get the hard way.
Signed-off-by: Christian Hesse <mail@eworm.de>
Git upstream bans sprintf() with commit:
banned.h: mark sprintf() as banned
cc8fdaee1eeaf05d8dd55ff11f111b815f673c58
Signed-off-by: Christian Hesse <mail@eworm.de>
Git upstream bans strncpy() with commit:
banned.h: mark strncpy() as banned
e488b7aba743d23b830d239dcc33d9ca0745a9ad
Signed-off-by: Christian Hesse <mail@eworm.de>
Git upstream bans strcpy() with commit:
automatically ban strcpy()
c8af66ab8ad7cd78557f0f9f5ef6a52fd46ee6dd
Signed-off-by: Christian Hesse <mail@eworm.de>
Git upstream bans sprintf() with commit:
banned.h: mark sprintf() as banned
cc8fdaee1eeaf05d8dd55ff11f111b815f673c58
Signed-off-by: Christian Hesse <mail@eworm.de>
Git upstream bans strncpy() with commit:
banned.h: mark strncpy() as banned
e488b7aba743d23b830d239dcc33d9ca0745a9ad
Signed-off-by: Christian Hesse <mail@eworm.de>
This makes the markdown filter generate anchor links for headings.
Signed-off-by: Christian Hesse <mail@eworm.de>
Tested-by: jean-christophe manciot <actionmystique@gmail.com>
Even if we find snapshot-prefix in the repo configuration, we are not
writing it out into the rc- file, so setting the value does not have any
effect.
Signed-off-by: Konstantin Ryabitsev <konstantin@linuxfoundation.org>
There's no use in giving a silly example to folks who will just copy it,
so instead try to do something slightly better.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Though SHA1 is broken, HMAC-SHA1 is still fine. But let's not push our
luck; SHA256 is more sensible anyway.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
