- Dec 09, 2020
-
-
Dan Villiom Podlaski Christiansen authored
-
- Oct 27, 2020
-
-
Dan Villiom Podlaski Christiansen authored
This fixes cloning using the old git protocol.
-
- Dec 11, 2020
-
-
muxator authored
-
- Dec 09, 2020
-
-
muxator authored
Now it is possible to invoke hg status in a repo where there is an unreadable directory and no .hgignore without crashing. This probably uncovers another (albeit minor) bug: the "Permission denied" message is printed twice.
-
muxator authored
If a repo has no .hgignore (this happens frequently on repositories that come from a git clone, which tend to have a .gitignore but no .hgignore), and a directory is not readable by the current user (for example, for a ownership or persmission problem), hg-git causes a crash in hg status instead of printing an error message. An almost complete stack trace is: $ hg status ... Traceback (most recent call last): File "/opt/mercurial/hg-git/hggit/gitdirstate.py", line 99, in _ignore fp = open(files[0], 'rb') FileNotFoundError: [Errno 2] No such file or directory: b'/tmp/example/.hgignore' During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/opt/mercurial/hg-git/hggit/gitdirstate.py", line 182, in _finddotgitignores entries = util.listdir(join(nd), stat=True, skip=skip) PermissionError: [Errno 13] Permission denied: '/tmp/example/not_readable' During handling of the above exception, another exception occurred: Traceback (most recent call last): [...] File "/opt/mercurial/hg-git/hggit/gitdirstate.py", line 102, in _ignore fns = self._finddotgitignores() File "/opt/mercurial/hg-git/hggit/gitdirstate.py", line 185, in _finddotgitignores fwarn(nd, inst.strerror) File "/opt/mercurial/hg-git/hggit/gitdirstate.py", line 128, in fwarn self._ui.warn(b'%s: %s\n' % (self.pathto(f), msg)) TypeError: %b requires a bytes-like object, or an object that implements __bytes__, not 'str'
-
- Nov 24, 2020
-
-
Dan Villiom Podlaski Christiansen authored
-
- Nov 22, 2020
-
-
Dan Villiom Podlaski Christiansen authored
-
- Nov 21, 2020
-
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
- Aug 07, 2020
-
-
Manuel Jacob authored
Fixes #239.
-
- Aug 04, 2018
-
-
Dan Villiom Podlaski Christiansen authored
This is useful for interacting with the Git repository directly, either from shell scripts, the terminal, or perhaps an alias like this: [alias] git=!GIT_DIR=$(hg debuggitdir) git "$@" ...and with that, you can easily access features not offered by Dulwich such as `hg git gc`.
-
- Oct 23, 2020
-
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
- Jul 12, 2020
-
-
Manuel Jacob authored
This correctly handles sub-options. Before, the test case failed with: dulwich.errors.RefFormatError: refs/remotes/default:pushurl/beta
-
Manuel Jacob authored
Before, the added test case failed with “abort: git remote error: fatal: 'file://$TESTTMP/gitrepo' does not appear to be a git repository”.
-
- Oct 23, 2020
-
-
Dan Villiom Podlaski Christiansen authored
-
Dan Villiom Podlaski Christiansen authored
-
- Aug 13, 2020
-
-
Tristan Seligmann authored
Pass --ff-only to git pull to avoid a warning.
-
- Oct 11, 2020
-
-
Dan Villiom Podlaski Christiansen authored
Copied from changeset 5a19d7c9129b in the stable branch, latest as of 2020-10-11. https://www.mercurial-scm.org/repo/hg/rev/5a19d7c9129b
-
- Aug 13, 2020
-
-
Tristan Seligmann authored
Pass explicit messages to git merge to avoid relying on the default.
-
- Aug 06, 2020
-
-
muxator authored
If the user modified his log template to include gitnode(), an invocation of hg incoming causes a crash with the following message: AttributeError: 'overlayrepo' object has no attribute 'githandler'
-
- Jun 12, 2020
-
-
Manuel Jacob authored
This makes this file identical with upstream hg@07b79569fdf8 (last revision from which we copied this file).
-
Manuel Jacob authored
-
Manuel Jacob authored
-
- May 10, 2020
-
-
Manuel Jacob authored
(folded with intermediate revision with new message by Georges Racinet) The previous conditional was wrong: in current Mercurial, 'co' is in principle not an integer. The condition on integer has been introduced for the port to py3, but does not change the fact that changelog.__contains__ is for integers. The original intent in a90fe3e8a8c3 was probably to exclude local repositories, but this had the side effect of setting `co=None` in all cases, hence affecting pulls having nothing to do with hg-git, as the new test demontstrates. What we really want is to force `co=None` only for remote Git repositories, for the reasons explained in the comment. Previous intermediate commit was cleanup: eliminate if statement with condition that is always true 'co' is never an integer. It’s always None or of type bytes. Apparently 'co' was of type int some years ago. A proper fix would require to fix the lookup() remote command or change hg to not require it e.g. when pulling with -r. For now, let’s remove the broken code and be honest about that we give up.
-
- Dec 23, 2019
-
-
Georges Racinet authored
This is ugly indeed, but a true cleaning would be to convert to a sensible testing system.
-
- Mar 06, 2020
-
-
Manuel Jacob authored
-
- May 04, 2020
-
-
Manuel Jacob authored
-
- May 03, 2020
-
-
Manuel Jacob authored
These files were byteified using hg’s contrib/byteify-strings.py: - contrib/hggitperf.py - hggit/compat.py - hggit/git2hg.py - hggit/gitdirstate.py - hggit/gitrepo.py - hggit/git_handler.py - hggit/hg2git.py - hggit/hgrepo.py - hggit/overlay.py - hggit/util.py - hggit/verify.py - hggit/_ssh.py - hggit/__init__.py - tests/commitextra.py Additionally, "--treat-as-kwargs opts kwargs" were passed as arguments. Some of the changes will be undone later. Because of the large amount of literals that should be changed in comparison to the few literals that should stay, it seemed better to do it this way around. Raw literals are not changed by the script. Some of them will be changed manually in a separate changeset.
-
Manuel Jacob authored
-
- Dec 23, 2019
-
-
Georges Racinet authored
-
- Mar 05, 2020
-
-
Manuel Jacob authored
* run `find . -name "*.py" -print0 -o -path ./tests/hghave | xargs -I {} -0 2to3 --fix=print --write {}` * run `find . -name "*.py" -print0 -o -path ./tests/hghave | xargs -I {} -0 2to3 --doctests_only --fix=print --write {}` * remove extra newline at end of files introduced by last command * add `from __future__ import print_function` to Python files
-
- Nov 22, 2019
-
-
timeless developer authored
-
- Mar 05, 2020
-
-
Manuel Jacob authored
After copying the files, a6ddd3f835af was reapplied.
-
- Dec 22, 2019
-
-
Georges Racinet authored
Git behaviour and output has changed, in a way that seems hard to reconcile with glob or regex patterns. we'll probably have to write an entire variant of this test. In the meanwhile, this skipping should restore the main CI runs.
-
Georges Racinet authored
This is to run some tests only if the Git version is smaller than the given one.
-
- Aug 01, 2019
-
-
Pierre-Yves David authored
The message case got fixed in 5.0. So we need to match both version.
-
- Feb 23, 2019
-
-
Kevin Bullock authored
-