This reverts commit bbf96d898b96ea2d1ff19a09e545499e4f32d8d0.
Removes dependency on ruby 2.7 and isn't much slower.
See https://github.com/NixOS/nixpkgs/pull/123502 for more info
(cherry picked from commit 88c2c543bf01063adf57ca2f4ff200cf8fbbd18b)
Resolved the following conflicts:
- kernel flags between 09176d28a0ebbd8bae35f92406a48cded78ac9f7 and 2b28822d8dec38fd62ef17643ec71da3bb488619
- clojure-lsp between 3fa00685ce21fb1a4218d444d0862ebcb2b99706 and e03c068af5c9ca950214cd6c3b5cfe67f595da88
in order to be able to disable osxkeychain credential helper if no
wanted / necessary. Keep it `true` by default such that current logic is
not affected.
Signed-off-by: Andreas Schmid <service@aaschmid.de>
continuation of #109595
pkgconfig was aliased in 2018, however, it remained in
all-packages.nix due to its wide usage. This cleans
up the remaining references to pkgs.pkgsconfig and
moves the entry to aliases.nix.
python3Packages.pkgconfig remained unchanged because
it's the canonical name of the upstream package
on pypi.
This makes the git HTML manual reproducible.
It also removes the `git.info` manual, since this one is
basically entirely broken anyway. This likely happened
when we migrated from asciidoc to asciidoctor in #102020.
I don't suggest going back to asciidoc: that would
re-introduce a python2 dependency and another
reproducibility issue with the HTML docs.
Co-Authored-By: Dmitry Kalinkin <dmitry.kalinkin@gmail.com>
The install step for the darwin specific $out/etc/gitconfig file
assumed that $out/etc already existed. This assumption is no longer
true after f5e2ff1ed80dac74bfb0afa92f34641cce098a49 in #103421. This
change makes make sure the directory exists before adding to it.
This reverts commit c778945806b44d46ec16bc4302e7e7163e6bab97.
I believe this is exactly what brings the staging branch into
the right shape after the last merge from master (through staging-next);
otherwise part of staging changes would be lost
(due to being already reachable from master but reverted).
I'm sorry; I didn't notice it contained staging commits.
This reverts commit 17f5305b6c20df795c365368d2d868266519599e, reversing
changes made to a8a018ddc0a8b5c3d4fa94c94b672c37356bc075.
In that case it specified $doc output but didn't even create it.
I expect it's better to do it this way instead of creating it
as an empty directory.
(Only the failed builds get rebuilt by this commit.)
disable_test: Use \s instead of ' ' as tabs are used for indentation in
t5324-split-commit-graph.sh. It's also necessary to insert : (no op) in
front of # to prevent Bash syntax errors due to empty loops.
Two newly added tests fail due the usage of shared permissions (outside
of the build sandbox they succeed, actually even with breakpointHook
(cntr attach + cntr exec bash)):
```
t5324-split-commit-graph.sh (Wstat: 256 Tests: 29
Failed: 2)
Failed tests: 28-29
Non-zero exit status: 1
```
I'm also adding myself as maintainer since there currently doesn't seem
to be a fixed one for regular updates.
Announcement:
https://lkml.kernel.org/lkml/xmqqzh9mu4my.fsf@gitster.c.googlers.com/
This will install the HTML and text documentation into a separate output
so that users can install it without having to rebuild Git.
Previously only `doc/git/git-subtree.html` was installed (which is now
in $doc as well).
The current output sizes are as follows:
```
$ du -sh $(nix-build -A git)
47M /nix/store/wyqgalp61kmavx06rams7z8jz177nd8y-git-2.26.2
$ du -sh $(nix-build -A git.doc)
14M /nix/store/6zi22fl5xc3sg23d9shsviinvwk89wvq-git-2.26.2-doc
```
Fixes#86022 (at least partly since the output has to be installed).
@the-kenny did a good job in the past and is set as maintainer in many package,
however since 2017-2018 he stopped contributing. To create less confusion
in pull requests when people try to request his feedback, I removed him as
maintainer from all packages.
The syntax is ${parameter:-word} (i.e. previously this used
"latestTag" instead of the actual value).
(Fixes a regression from #85278.)
Also: Even though getting the latest tag isn't really security critical
(as long as Git itself is secure against untrusted input), I'd prefer to
switch from the Git to the HTTPS protocol (for authentication of the
server and encryption + uses a standard port).
The existing post-install was not successfully patching the git-gui script,
and thus was invoking the packaged osx app which uses the system tk,
which is too old to work (and is no longer supported by Apple anyway).
