Use suffixcmp() from Git instead of reimplementing it. This is a
preparation for moving to ends_with() in Git 1.8.6.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Since the introduction of selective snapshot format configuration in
dc3c9b5 (allow selective enabling of snapshots, 2007-07-21), we allowed
seven different delimiters for snapshot formats, while the documentation
has always been clear about spaces being the only valid delimiter:
The value is a space-separated list of zero or more of the values
"tar", "tar.gz", "tar.bz2", "tar.xz" and "zip".
Supporting the undocumented delimiters makes the code unnecessarily
complex. Remove them.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
This is a preparation for replacing all prefix checks with either
strip_prefix() or starts_with() when Git 1.8.6 is released.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
* Remove the dependency on Git (which can be obtained automatically when
building, using either the Git submodule or `make get-git`).
* Use proper upstream names of dependencies.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
* Several small spelling and capitalization fixes.
* Use consistent and better-looking formatting that is compatible with
AsciiDoc (and partly compatible with RST).
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
So that people wishing to use "enable-http-clone" don't have to find
out the correct settings on their own.
Signed-off-by: Přemysl Janouch <p.janouch@gmail.com>
* Name "cgit Development Team" as copyright holder to avoid listing
every single developer.
* Update copyright ranges.
Signed-off-by: Lukas Fleischer <cgit@crytocrack.de>
Everything works just bumping the version in Makefile and commit hash
in submodule. No code changes required.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Currently line numbers look like (for blob view and sdiff respectively):
<a class='no' id='n68' name='n68' href='#n68'>68</a>
<td class='lineno'><a class='no' href='...#n1' id='n1' name='n1'>1</a></td>
name=".." is unnecessary if the id attribute is set (this even applies
to IE6), so drop it. (aside, in HTML5, the name attribute is gone.)
The line number links can be selected through their parent classes, no
need for another class "no", so drop it too.
For a file with 2000 lines, this yields a saving of 40% (29% gzipped).
While at it, fix the hover effect of line numbers: now the line number
get a black background as was intended.
Signed-off-by: Peter Wu <lekensteyn@gmail.com>
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
When outputting the Content-Type HTTP header we print the MIME type and
then append "; charset=<charset>" if the charset variable is non-null.
We don't want a charset when we have selected "application/octet-stream"
or when the user has specified a custom MIME type, since they may have
specified their own charset. To avoid this, make sure we set the page's
charset to NULL in ui-plain before we generate the HTTP headers.
Signed-off-by: John Keeping <john@keeping.me.uk>
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Instead of using our own vector implementation, use argv_array from Git
which has been specifically designed for dynamic size argv arrays.
Drop vector.h and vector.c which are no longer needed.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
This is no longer needed since commit fb3655df (use struct strbuf
instead of static buffers, 2013-04-06).
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Remove filepair_cb_raw() and all related functions. These are no longer
needed. We now use Git's internal functions for raw diff formatting
everywhere.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Use Git's internal diff_tree_sha1() function for the /rawdiff/ command
instead of trying to recreate this functionality.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Use `git rev-list --max-parents=0 HEAD` instead of `git rev-list HEAD |
tail -1` to get the root commit. This works since Git 1.7.4.2.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Add max_parents = 1 to the revision walk in order to make sure we do not
include the footer signature twice for merge commits.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Rename parameters and local variables to match those from ui-diff. Also,
convert a "char *" to "const char *".
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
This allows for specifying a revision range using the id2 parameter of
/patch/. The output that is produced is similar to
$ git format-patch --stdout id2..id
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Instead of using our own formatting, use log_tree_commit() from Git to
create patches. This removes unnecessary duplicate code and also fixes a
bug with e-mail address formatting that existed in our own
implementation.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Sync with what we do everywhere else and check the return value of
get_sha1() instead of calling sha1_object_info() to validate the object.
Note that we later call lookup_commit_reference(), which checks that
both SHA1 values refer to commits, anyway.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
This can be used to generate raw diffs between arbitrary revisions using
something like
/rawdiff/?id=v0.9&id2=v0.9.1
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
This adds a parameter to cgit_print_diff() to create raw diffs, using
the same format as `git diff <commit>`.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Move filepair_cb() from ui-patch.c to ui-shared.c and rename it to
filepair_cb_raw(). This callback will be used in ui-diff.c in a
follow-up patch.
Note that it is not straightforward to extract filepair_cb() from
ui-diff.c which is why it is not done here as well.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
When having found the first path separator occurrence at position i, we
invoked strchr() on the same position i in subsequent iterations
resulting in the same path separator being returned by strchr() over and
over again. Increase the position by one to skip the occurrence that has
just been found and advance to the next separator.
Reported-by: Konstantin Ryabitsev <mricon@kernel.org>
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Locally installed packages are usually installed to /usr/local.
Packagers can use `make prefix=/usr` to get back the old behavior.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Use struct strbuf from Git instead of fixed-size buffers to remove the
limit on the length of configuration file lines and refactor
read_config_line() to improve readability.
Note that this also fixes a buffer overflow that existed with the
original fixed-size buffer implementation.
Signed-off-by: Lukas Fleischer <cgit@cryptocrack.de>
Pages like /commit?h=wip&id=8a335ce618ba77fbf05148d6f8be17bd48ba4340
were being marked as dynamic, because of h=wip, when it should be
static, because of id=.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
We've long supported negative ttls, for infinite cache, except the
documentation incorrectly showed one of our defaults as being 5 and not
-1. As well, with a negative ttl, we were actually making the HTTP
expired header go backwards. This changes it to go ahead ten years
instead.
Further, we add an cache-about-ttl option to set a different ttl for
about pages, which are now increasingly being filtered through markdown
or just sent statically anyway.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
My dmesg is filled with the oom killer bringing down processes while the
Bingbot downloads every snapshot for every commit of the Linux kernel in
tar.xz format. Sure, I should be running with memory limits, and now I'm
using cgroups, but a more general solution is to prevent crawlers from
wasting resources like that in the first place.
Suggested-by: Natanael Copa <ncopa@alpinelinux.org>
Suggested-by: Julius Plenz <plenz@cis.fu-berlin.de>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
