- Apr 12, 2019
-
-
timeless authored
irccloud and others may perform content negotation and reject (406) if the accept header does not match the mime type of the object
-
Augie Fackler authored
As far as I can tell it should be fine to unconditionally skip _prepareabortorcontinue if we're in the process of raising an Abort here. Differential Revision: https://phab.mercurial-scm.org/D6226
-
Augie Fackler authored
Differential Revision: https://phab.mercurial-scm.org/D6225
-
- Apr 06, 2019
-
-
Pierre-Yves David authored
While looking into adding error output in this test, I did some cleanup.
-
Pierre-Yves David authored
I am sitting next to Joerg Sonnenberger and we are discussion his experience with repoview. This first effect of this discussion is this documentation clarification.
-
- Apr 05, 2019
-
-
Jordi Gutiérrez Hermoso authored
This is where all the action happens for the status-related revsets, and a little documentation doesn't hurt.
-
- Apr 11, 2019
-
-
Georges Racinet authored
Instead of accessing `undecided` directly for ui display purposes, we introduce a `stats()` method that could be extended in the future with more interesting information. This is in preparation for a forthcoming Rust version of this object. Indeed, attributes and furthermore properties are a bit complicated for classes in native code. We could go further and rename `undecided` to mark it private, but `_undecided` is already taken as support for `_undecided` lazyness.
-
Martin von Zweigbergk authored
The paths passed into the matcher are normalized (this applies to include patterns and regular patterns, and to both glob kind and path kind), so the regex for input "foo/" ended up being "foo(?:/|$)". Once we have a (recursive) pattern kind only for directories, we could switch to that here and remove the "mfiles[0] == path" check. Until then, let's at least make it not misleading. Differential Revision: https://phab.mercurial-scm.org/D6224
-
Martin von Zweigbergk authored
Before this patch, test-rebase-inmemory.t would stop erroring out about the conflict if you added a "cd a" before line 252. That was because a glob matcher (which are relative) was unintentionally used. That happened because the matcher was given "include" patterns (not regular patterns), and "include" patterns are always glob by default (i.e. unless you write them including the kind prefix). IOW, the "default='path'" argument passed to ctx.match() was ignored. Differential Revision: https://phab.mercurial-scm.org/D6223
-
- Apr 10, 2019
-
-
Martin von Zweigbergk authored
I'm pretty sure this is our preferred term. Differential Revision: https://phab.mercurial-scm.org/D6222
-
- Apr 11, 2019
-
-
Pierre-Yves David authored
When using re2, we call test_match() instead of match() on the compiled regex object. While match() returns a matcher object or None, test_match() returns True or False. So since 2e2699af5649 running test on a machine with a re2 install fails in many places. Instead we make the code a bit more general and everything goes back to normal.
-
- Apr 09, 2019
-
-
Pulkit Goyal authored
# skip-blame as just b'' prefixes Differential Revision: https://phab.mercurial-scm.org/D6221
-
Pulkit Goyal authored
Differential Revision: https://phab.mercurial-scm.org/D6220
-
Yuya Nishihara authored
The doc doesn't state that "s#" of Py_BuildValue() is controlled by PY_SSIZE_T_CLEAN (unlike the one for PyArg_ParseTuple()), but actually it's switched to Py_ssize_t. https://docs.python.org/2/c-api/arg.html#c.Py_BuildValue https://github.com/python/cpython/blob/2.7/Python/modsupport.c#L432 Follow up for b01bbb8ff1f2 and 896b19d12c08.
-
- Apr 08, 2019
-
-
Augie Fackler authored
This fixes `hg grep -r 'wdir()'` when remotefilelog is enabled and the working directory contains uncommitted modifications. Differential Revision: https://phab.mercurial-scm.org/D6217
-
Augie Fackler authored
This demonstrates how remotefilelog breaks grepping dirtied working directories. A future change will introduce a fix. Differential Revision: https://phab.mercurial-scm.org/D6216
-
- Apr 03, 2019
-
-
Martin von Zweigbergk authored
Mercurial currently reads the .rc files specified in HGRCPATH (and the system-default paths) in directory order, which is unspecified. My team at work maintains a set of .rc files. So far there has been no overlap between them, so we had not noticed this behavior. However, we would now like to release some common .rc files and then have another one per plaform with platform-specific overrides. It would be nice if we can determine the load order by choosing names carefully. This patch enables that by loading the .rc files in lexicographical order. Before this patch, the added test case would consistently say "30" on my file system (whatever I have -- some Linux FS). Differential Revision: https://phab.mercurial-scm.org/D6193
-
- Apr 04, 2019
-
-
Martin von Zweigbergk authored
If you `hg add` a file and then delete it from disk, and then run `hg addremove`, the file ends up in the "removed" set that gets passed to the findrenames() override. We then crash because the file is not in the working copy parent. This patch fixes that. Differential Revision: https://phab.mercurial-scm.org/D6194
-
- Apr 06, 2019
-
-
Matt Harbison authored
For whatever reason, even though only python2 is on PATH, passing `python.exe` causes the later check that it's not py3 to bail out.
-
Matt Harbison authored
`capture_output` was added in 3.7. I was tempted to just check and abort in build.py, since Windows doesn't have the Linux problem where some distros only ship an older python. But this is in a library that could be used elsewhere in the future.
-
- Apr 04, 2019
-
-
Jordi Gutiérrez Hermoso authored
This isn't complete, and it would be nice to show the exact same colours that `hg diff` would show. That goal is too lofty, so this just shows some basic colours, on the premise that a little is better than nothing.
-
- Apr 05, 2019
-
-
Jordi Gutiérrez Hermoso authored
Terminals will define default colours (for example, white text on black background), but curses doesn't obey those default colours unless told to do so. Calling `curses.use_default_colors` makes curses obey the default terminal colours. One of the most obvious effects is that this allows transparency on terminals that support it. This also brings chistedit closer in appearance to crecord, which also uses default colours. The call may error out if the terminal doesn't support colors, but as far as I can tell, everything still works. If we need a more careful handling of lack of colours, blame me for not doing it now.
-
- Apr 07, 2019
-
-
Denis Laxalde authored
Match function for regex pattern kind is built through _buildregexmatch() and _buildmatch() using _rematcher() that returns a re.match function, which either returns a match object or None. This does not conform to Mercurial's matcher interface for __call__() or exact(), which are expected to return a boolean value. We fix this by building a lambda around _rematcher() in _buildregexmatch(). Accordingly, we update doctest examples to remove bool() calls that are now useless.
-
Denis Laxalde authored
Arguments 'ctx', 'listsubrepos' and 'badfn' are optional in function body.
-
Denis Laxalde authored
Argument 'warn' is actually non-required, since there's a 'if warn:' check before usage. Argument 'auditor' is passed to pathutil.canonpath(), in which it is optional.
-
- Apr 08, 2019
-
-
Denis Laxalde authored
Make the docstring raw, as it now includes escape characters.
-
- Apr 06, 2019
-
-
Denis Laxalde authored
-
Denis Laxalde authored
-
Denis Laxalde authored
Doctest examples aim at illustrating how __call__() and exact() are different, depending on the pattern kind.
-
- Apr 07, 2019
-
-
Denis Laxalde authored
Make the docstring raw, since it now includes escape characters.
-
- Apr 05, 2019
-
-
Martin von Zweigbergk authored
It seems that repo.lookup(), which is what supports the "lookup" wire protocol command, should not allow the working copy revision input. This fixes both the pull test and the convert test I just added. Differential Revision: https://phab.mercurial-scm.org/D6215
-
Martin von Zweigbergk authored
Differential Revision: https://phab.mercurial-scm.org/D6214
-
Martin von Zweigbergk authored
Differential Revision: https://phab.mercurial-scm.org/D6213
-
Martin von Zweigbergk authored
The convert extension translates commit references in the commit message. We didn't have any explicit testing of this before, so let's add a test. Differential Revision: https://phab.mercurial-scm.org/D6212
-
Matt Harbison authored
I hit this trying to build the py2exe target using python3, just to see what would happen. After commenting out `py2exe.Distribution` in setup.py and pointing to a local copy of py2exe that supports python3[1], it complained that `out` was bytes, not str. [1] https://github.com/albertosottile/py2exe/releases/tag/v0.9.3.0
-
- Apr 04, 2019
-
-
Philippe Pepiot authored
File "setup.py", line 975, in rustbuild "command: %r, environment: %r" % (self.rustsrcdir, cmd, env)) NameError: global name 'cmd' is not defined
-
- Apr 02, 2019
-
-
Arun Chandrasekaran authored
This patch introduces two new keys 'g' and 'G' that helps to navigate to the top and bottom of the file/hunk/line respectively. This is inline with the shortcuts used in man, less, more and such tools that makes it convenient to navigate swiftly. 'g' or HOME navigates to the top most file in the ncurses window. 'G' or END navigates to the bottom most file/hunk/line depending on the whether the fold is active or not. If the bottom most file is folded, it navigates to that file and stops there. If the bottom most file is unfolded, it navigates to the bottom most hunk in that file and stops there. If the bottom most hunk is unfolded, it navigates to the bottom most line in that hunk. Differential Revision: https://phab.mercurial-scm.org/D6178
-
- Apr 04, 2019
-
-
Jordi Gutiérrez Hermoso authored
I'm not sure if that ever worked and it's an internal API breakage, but `"verbose": True` is not correctly parsed, as most of these options are parsed by diffopts, whereas verbose is a global option. Setting the UI to verbose instead does work and does show a verbose patch, with full commit message. It also shows all files, which unfortunately are a bit hard to read on a single line in the default verbose template. Thus, we also change the default template to use the status template, which shows one file per line as well as its modification state.
-
Denis Laxalde authored
For commit and revert commands with --interactive and explicit files given in the command line, we now skip the invite to "examine changes to <file> ? [Ynesfdaq?]". The reason for this is that, if <file> is specified by the user, asking for confirmation is redundant. In patch.filterpatch(), we now use an optional "match" argument to conditionally call the prompt() function when entering a new "header" item. We use .exact() method to compare with files from the "header" in order to only consider (rel)path patterns. Add tests with glob patterns for commit and revert, to make sure we still ask to examine files in these cases.
-
- Apr 05, 2019
-
-
Gregory Szorc authored
The upstream source distribution from PyPI was extracted. Unwanted files were removed. The clang-format ignore list was updated to reflect the new source of files. The project contains a vendored copy of zstandard 1.3.8. The old version was 1.3.6. This should result in some minor performance wins. test-check-py3-compat.t was updated to reflect now-passing tests on Python 3.8. Some HTTP tests were updated to reflect new zstd compression output. # no-check-commit because 3rd party code has different style guidelines Differential Revision: https://phab.mercurial-scm.org/D6199
-