- Nov 25, 2017
-
-
Kevin Bullock authored
-
- Nov 02, 2017
-
-
Kevin Bullock authored
-
- Dec 24, 2015
-
-
Sean Farley authored
-
- May 15, 2015
-
-
Sean Farley authored
To help improve the speed Mercurial operations, we shouldn't leave the whole repository in the draft phase. The only hueristic git has for 'public' is remote HEAD. Therefore, if remote HEAD exists then we use it to mark the corresponding local commit as public. For now, we'll put this behind a config flag for easier rollback.
-
- Nov 02, 2017
-
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
- Oct 24, 2017
-
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
- Nov 02, 2017
-
-
Kevin Bullock authored
-
- Oct 24, 2017
-
-
Kevin Bullock authored
Provides a place to register the defaults for all config options in one place, and a helper to read them. This will let us eliminate deprecation warnings on hg 4.4 (in a future change) while remaining compatible with older versions.
-
- Nov 02, 2017
-
-
Kevin Bullock authored
Since updating run-tests.py in e085b381e8e2, a default date of '0 0' is no longer set automatically. Restore this by adding it to our own test suite's default hgrc.
-
- Oct 24, 2017
-
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
Before this change, setting git.similarity to a value out of range would result in a traceback because 'util' is hg-git's own module. Adds a test demonstrating the fixed behavior.
-
- Oct 23, 2017
-
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
It's been three and a half years or so, so I'd say we're due. This brings us up to the version from hg @ 274811627808 (during the 4.4 code freeze). In particular, we'll need support for optional output lines added in 6025cac3d02f.
-
- Oct 05, 2017
-
-
Durham Goode authored
Upstream changed the representation of this structure, to make it not a named tuple. So let's update to match it. The tests caught this issue.
-
Durham Goode authored
Some upstream code now relies on ctx.repo() to access the repo object, since hggit stored ctx.repo as a field instead of a function, that failed. Let's move it to be a function. For overlaychangectx we can't just store the repo in self._repo because that causes other parts of the base class to act differently and causes tests to fail, so we store it as a new field.
-
- Oct 01, 2017
-
-
Kevin Bullock authored
This is a bit of a simplification of an earlier patch from Jun Wu <quark@fb.com>.
-
- Aug 17, 2017
-
-
Sean Farley authored
-
- Aug 10, 2017
-
-
Kevin Bullock authored
-
Kevin Bullock authored
-
Kevin Bullock authored
-
- Aug 04, 2017
-
-
Katsunori FUJIWARA authored
Before this revision, invalid pattern in hgignore file causes unintentional failure for UnboundLocalError of ignorefunc, if hggit is used with Mercurial 3.5 or later. In such case: - checking source of invalid pattern at failure uses "pats" list for hgignore files, but - "pats" list is empty, if ignoremod is None (= Mercurial 3.5 or later) - therefore, checking with matchmod.match() overlooks invalid pattern Then, "return ignorefunc" is executed without assignment to ignorefunc, and causes UnboundLocalError. To show pattern error in hgignore file as expected even with Mercurial 3.5 or later, this revision puts '(FILE, ["include: FILE"])' tuples into "pats" (to avoid code duplication, putting into allpats is shared, too). This makes checking source of invalid pattern at failure work as expected for hgignore files. Fixes #197
-
- Aug 10, 2017
-
-
Kevin Bullock authored
-
Kevin Bullock authored
The test changes demonstrate that the hostname gets URL-encoded before being passed to ssh(1).
-
Kevin Bullock authored
-
- Aug 04, 2017
-
-
Sean Farley authored
-