Summary:
When a unit test throws an exception, provide more data (type, trace) in the
test failure message.
Previously, we would show only the message itself, which may not be very useful
in debugging test failures.
Test Plan: Ran "arc unit" on a test which throws, got a stack trace.
Reviewers: edward, btrahan, jungejason, nh, tuomaspelkonen, aran
Reviewed By: nh
CC: aran, nh
Differential Revision: 1123
Test Plan:
Wrote an event listener modifying the commit message and the message was
successfully changed.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, jungejason, epriestley
Differential Revision: 1103
Test Plan:
Added a test event listener, added an event dispatch in diff workflow, run the
workflow, the listener was called.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, jungejason, epriestley
Differential Revision: 1102
Summary:
We used square distance that optimized for the wrong thing. Making the same
markers spread out instead of being together
Also added a very little cost for switching type. That will make diff types
stick together a bit more
Task ID: #623
Blame Rev:
Test Plan:
Ran my new unit test. And tested a few diff in phabricator
Revert Plan:
Tags:
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley
Differential Revision: 1112
Summary:
just a little try / catch action in ArcanistDiffWorkflow.
"Ideally, it would be good to flag these changes somehow so that "arc patch" can
issue the inverse warning (e.g., "This patch could not be completely applied
because some binary data was not uploaded.") but that hasn't come up in a real
use case yet."
I think a change w/ filetype of FILE_IMAGE || FILE_BINARY *and* no phid means
that the upload failed so there's no additional flag needed. (True?) however,
it would be easy enough to store metadata that explicilty stated whether or not
the file upload succeeded.
(also / related - i looked through the arc patch workflow a bit and i don't
understand how the svn codepath loads up the actual binary files... for git,
toGitPatch => buildBinaryChange => getBlob is the right path )
Test Plan:
- set 'storage.mysql-engine.max-size' to 0 in my conf, uploaded diffs with
files
- noted in differential that it correctly detected images versus binary despite
file upload failing
- noted in differential that images had some empty UI
- reverted conf change, uploaded diffs with files
- noted in differential file showed up
- ran arc patch with DX, where DX had broken files
- noted "Downloading binary data...
Exception:
ERR-BAD-PHID: No such file exists."
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley
Differential Revision: 1100
Summary:
If arc amend is run on a rev that isn't accepted, it runs arc mark-committed
--finalize. The rev shouldn't be marked as committed if it hasn't been accepted,
so this diff adds in that check.
Test Plan:
Ran arc amend on a rev that hasn't been accepted, checked that it didn't get
marked as completed.
Reviewers: jungejason, epriestley
Reviewed By: epriestley
CC: aran, epriestley, nh
Differential Revision: 1104
Summary: 'patch' chokes on hunks with too much trailing stuff. 'git apply'
chokes on overlapping hunks. Make them both happy. Write some test cases so I
stop breaking this stuff.
Test Plan:
- Applied a previously-failing patch via SVN.
- Applied a previously-failing-then-succeeding patch via Git.
- Ran unit tests.
Reviewers: jungejason, btrahan, nh, tuomaspelkonen, aran
Reviewed By: jungejason
CC: aran, jungejason
Differential Revision: 1099
Summary: (At least my build of) PHP tries to do readline magic on the terminal
when invoked from the shell, even inside bash's command substitution operator.
Work around it by piping the output of ##echo## to ##php##, so the input isn't a
terminal.
Test Plan: ##bash$ arc li<TAB>## now gives me possible completions rather than
breaking and inserting a literal tab
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley
Differential Revision: 1096
Summary:
In D1079, I added "--color never", but this flag is provided by the "color"
extension, which is why I missed it originally, because it doesn't show up until
you enable that extension. Providing it causes installs which don't have it
enabled (disabled is the default) to fail.
Use "--config" to disable color instead. This sets a configuration setting and
works regardless of whether the color extension is present.
Test Plan: Ran "arc diff" in a mercurial working copy with the color extension
enabled and disabled.
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: nh
CC: aran, nh
Differential Revision: 1092
Summary: Make this text a little more informative, this stuff is fairly stable
now and has actual documentation available.
Test Plan: Read the document.
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 1086
Summary: We ostensibly now support Mercurial at least mostly, so don't
underpromise quite so much.
Test Plan: Ran "arc diff" in a Mercurial working copy without being warned.
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 1085
Summary: We currently may generate patches which have hunks that include overlapping context. Git was okay with this until at least 1.7.3.4, but started rejecting these patches some time later.
We already attempt to detect and prevent this condition, we just don't do a very good job of it. Fix the check so that we avoid generating overlapping hunks.
Test Plan:
- Cleanly applied troublesome patches under Git 1.7.7.2.
- Used "arc export --git" to verify patches generate without overlapping sections.
- Locally changed $context to 2 and 1, verified patch behavior was reasonable.
Reviewers: kdeggelman, bizrad6, jonathanhester, jungejason, nh, tuomaspelkonen, aran
CC:
Differential Revision: 1084
Summary: Mercurial has a --color flag which disables the use of ANSI colors. Use
it to prevent "hg diff" from giving us colorized diff output.
Test Plan: Ran "hg diff --color never", verified it disabled ANSI color in diff
output.
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: aran
CC: aran
Differential Revision: 1079
Summary:
Adds a secret, undoucmented "encoding" key to ".arcconfig" which makes a very
half-hearted effort to convert encodings. This is probably good enough that
Differential can be used for code review, but there will be issues with 'arc
patch', 'arc export', paste, maybe conduit stuff, Diffusion, and whatever else I
haven't thought of.
This also doesn't store the original encoding so anything converted like this
won't reasonably be able to be made to work with all that stuff in the future.
See T452 for a broader discussion of the issues involved.
Test Plan:
Short circuited the UTF-8 detection to always fail, had my files "converted"
from ISO-8859-1 to UTF-8.
@davidreuss: you can test this by applying this patch to arcanist/, adding
'"encoding" : "ISO-8859-1"' to your .arcconfig, touching some non-ASCII file,
and then running "arc diff".
Reviewers: davidreuss, jungejason, tuomaspelkonen, aran
Reviewed By: davidreuss
CC: aran, davidreuss, epriestley, nshamg123
Differential Revision: 812
Summary:
Save commited revision id in workflow, so it can be used in Arcanist
didRunWorkflow hook.
Test Plan:
Running getRevisionID on the workflow should return id of the committed
revision.
Reviewers: epriestley, jungejason
Reviewed By: epriestley
CC: aran, epriestley, mareksapota
Differential Revision: 1078
Summary:
D1061 introduced a 'text file' check, but it fails under SVN for new
directories.
- Revert D1061 (This reverts commit b2cd18252701be2093b52652fb3d1d94c5df571e.)
- Make getChangedLines() return null to indicate that the operation doesn't
make sense. I think this was the intent of the code in the lint engine.
- Fix a bug where running "arc lint" on a change in an SVN working copy from a
subdirectory would fail.
- Fix a bug where warnings with no line information were incorrectly
discarded.
Test Plan:
- Ran "arc lint" in an SVN working copy with a new directory (no failure).
- Forced FilenameLinter to always raise a warning. Added a binary file and ran
"arc lint". The warning was reported for the new binary file, a new text file,
and a new directory.
Reviewers: jungejason, andrewjcg, nh, tuomaspelkonen, aran
Reviewed By: andrewjcg
CC: aran, andrewjcg, epriestley
Differential Revision: 1076
Summary: Apparently I just never tested this or something. Make it work
correctly.
Test Plan: Ran "arc diff" in a Mercurial working copy with a binary file in
outgoing.
Reviewers: Makinde
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 1074
Summary: We incorrectly add the "Uncommitted" flag to untracked files, which
causes them to raise various prompts and not respect "--allow-untracked".
Test Plan: Ran "arc diff --allow-untracked" in a clean working copy with
untracked files, was not fatally error'd.
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: Makinde
CC: aran, Makinde, epriestley
Differential Revision: 1073
Summary:
Allow `arc patch` without authentication if Phabricator instance has
'differential.anonymous-access' set to true.
Test Plan:
Set 'differential.anonymous-access' in Phabricator to true and run `arc patch`
without installing a certificate. `arc patch` should work as expected.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley
Differential Revision: 1069
Summary:
In D962, I switched us from using "hg summary" to "hg id" to find the working
copy revision. However, "hg id" reports the revision with a trailing "+" if the
working copy is dirty, so we fail before hitting the explicit check for this
later.
Trim off the trailing '+' if it is present.
Test Plan:
Ran "arc diff" in a dirty Mercurial working copy and got a good error message
about uncommitted changes.
~/repos/hg-working-copy $ arc diff
WARNING: Mercurial support is largely imaginary right now.
Usage Exception: You have uncommitted changes in this branch. Commit (or
revert) them before proceeding.
Working copy: /INSECURE/repos/hg-working-copy/
Uncommitted changes in working copy
hello.c
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 1070
Summary: newer PyLint includes commas and a column number for every message
after the line number. ignore that if it's present.
Test Plan: ##arc lint --trace --lintall## a file with messages with the old and
new pylint (2.5), works fine with both
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley
Differential Revision: 1062
Summary:
Currently, lint messages are only included if they are errors or
if they affect lines which the diff changed. The implementation
of this caused issues for non-text files (e.g. binaries), as line
change information is not available, and the corresponding lint
messages were dropped (for non-errors).
In this diff, only lint messages concerning text files are dropped
based on this line filtering.
Test Plan: arc lint with binary file
Reviewers: jungejason, epriestley
Reviewed By: epriestley
CC: aran, aravindn, andrewjcg, liat, epriestley
Differential Revision: 1061
Test Plan:
Using SVN make some changes to the repo, run `arc diff`. As other user accept
the revision, add the changes to your repo using `arc patch` and then run `arc
commit --revision revisionID` to commit them. Arcanist should ask if you are
sure that you want to commit this revision and if you answer `Y` it should
commit to SNV repo.
Reviewers: epriestley, jungejason
Reviewed By: epriestley
CC: aran, mareksapota, epriestley
Differential Revision: 1055
Test Plan:
Tried viewing a diff detected as binary in diffusion with and without
setting my desired encoding. Worked as expected.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley
Differential Revision: 1008
Summary: allow adding PYTHONPATHs directly in addition to the hardcoded few
already allowed.
Test Plan: ##arc lint## successfully picks up the paths listed in my .arcconfig,
and no longer errors on modules not found from those directories
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley
Differential Revision: 1050
Test Plan:
Run `arc mark-committed` on a revision that you don't own, you should see a
prompt asking you if you really want to do that and if you answer `Y` it should
mark the revision as committed.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley, mareksapota
Differential Revision: 1052
mercurial changesets
Summary: Makes the parsing slightly more liberal so a test case from @Makinde
passes.
Test Plan: Ran unit tests.
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 1054
Summary: add the ability to specify an rcfile path either absolute or relative
to the project root.
Test Plan: added one for a project, ran ##arc lint --trace## to see the expected
result
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley
Differential Revision: 1046
Summary:
- The top documentation is more readable, and the details about the severity
mapping regexps is moved to the top from inline -- they're important.
- array_merge() is used when appending command line args from
##lint.pylint.options## instead of '+'. The latter merges by key instead of
appending the two lists as intended.
- rely on the exit code instead of magic text that may or may not be there
depending on how ##pylint## is invoked.
These bugs were introduced in d762311a9d.
Test Plan: ##arc lint --trace##'d a bunch of Python source files in a project,
with and without PyLint messages. Added some ##lint.pylint.options## to see
proper appending behavior.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley, cpiro
Differential Revision: 1045
Summary:
See T584. Git renders these types of changes somewhat unusually. Ensure they are
rendered as changes.
The fact that a symlink was replaced is not explicitly rendered (e.g., "This
path was changed from a symlink to a regular file."), but can be inferred from
the 'unix:filemode' property change that will accompany the changeset. We could
also make it explicit but this type of change is rare enough that it's probably
good enough as-is.
Test Plan:
- Ran unit test, which previously failed and now passes.
- Created a diff which replaces a file with a symlink, verified it rendered a
little more sensibly.
Reviewers: aravindn, dschleimer, jungejason, nh, tuomaspelkonen
Reviewed By: nh
CC: aran, goldshlager, nh
Differential Revision: 1030
Summary: See T587. Reduce the strictness of working copy checks when using
"--show", since there's a reasonable workflow where you 'arc patch' and then
'arc amend --revision X --show | git commit -a -F -' that currently won't work.
There are other ways to accomplish the same thing but this increases flexibility
overall.
Test Plan: Ran 'arc amend --show' with a dirty working copy, didn't get yelled
at.
Reviewers: jungejason, nh, tuomaspelkonen, aran
Reviewed By: nh
CC: aran, nh
Differential Revision: 1033
changes which add or alter binary files
Summary: See test cases, derived from "hg diff --git" and "git diff --binary".
Test Plan: Test cases work now.
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 1047
Summary:
This is separated from D812. See D812, D1008, D1009. Separation allows us to
land these patches safely.
See T452 for a broader discussion of the issues involved.
Test Plan: See D812.
Reviewers: davidreuss, jungejason, nh, tuomaspelkonen, aran
Reviewed By: davidreuss
CC: aran, davidreuss
Differential Revision: 1040
(Committing on behalf of @cpiro)
Summary: there's a check beneath to see if it worked, but @require_once will
fail silently if libphutil isn't available
Test Plan: tried it with libphutil findable and not findable, got expected
behavior
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley
Differential Revision: 1018
Test Plan:
Go to /differential/diff/create and try to upload a patch that was created with
`git diff` when diff.mnemonicprefix was set to true.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, epriestley
Differential Revision: 1022
Summary:
D935 missed one place of parseGitRelativeCommit() in
ArcanistExportWorkflow.
Test Plan: ran arc export and verify that it worked.
Reviewers: epriestley, tuomaspelkonen, aran
Reviewed By: aran
CC: aran
Differential Revision: 1015
Summary: A minor bug in the parser prevented it from handling git diffs where an
empty file appears at the end of the diff. Since git appends an extra newline,
we failed to jump into the '$line === null' block.
Test Plan: Ran unit tests. Generated a revision which only deleted an empty
file.
Reviewers: jungejason, nh, tuomaspelkonen, aran
Reviewed By: nh
CC: aran, nh, epriestley
Differential Revision: 998
Summary:
See D945. We have this kludgy "remote_hooks_installed" mess right now, but we
have enough information on the server nowadays to figure this out without it.
Also reduce code duplication by sharing the "mark-committed" workflow.
This causes "arc merge" to effect commit marks.
Test Plan:
In Git, Mercurial and SVN working copies ran like a million
amend/merge/commit/mark-committed commands with and without --finalize in
various states of revision completion.
This change is really hard to exhaustively test because of the number of
combinations of VCS, revision state, command, command flags, repository state
and tracking state. All the reasonable tests I could come up with worked
correctly, though.
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: jungejason
CC: aran, jungejason
Differential Revision: 967
Summary:
Expand 12-character hashes to 40-character hashes so other things will work
properly.
Also use "hg id" instead of "hg summary" to figure out where the working copy
is, since it's substantially simpler.
Test Plan: Ran "arc diff" and got properly 40-character hashes.
Reviewers: Makinde, aran
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 962
Summary: For Mercurial parsing in Diffusion, it looks like the only command we
can use to get move/copy information is "hg status -C --rev <rev>", so add a
parser for it.
Test Plan: Ran unit tests.
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 958
Summary: I renamed this at some point but missed a callsite. See T519.
Test Plan: Ran "arc patch" in an SVN working copy.
Reviewers: Girish, jungejason
Reviewed By: Girish
CC: aran, Girish
Differential Revision: 959
Summary: Move code to actually parse "hg" output into a separate class with some
tests, so I can reuse it in the import scripts. We should probably do this for
Git/SVN at some point, too.
Test Plan: Ran unit tests, used this class in Phabricator importers, grepped for
calls to removed private methods.
Reviewers: Makinde, jungejason, nh, tuomaspelkonen, aran
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 942
Summary: When a project uses a conservative history mutability doctrine, never
try to amend history.
Test Plan: Ran "arc amend" in an "immutable_history" working copy.
Reviewers: fratrik, Makinde, aran, jungejason, tuomaspelkonen
Reviewed By: Makinde
CC: aran, Makinde, epriestley
Differential Revision: 862
Summary:
This should support conservative rewrite policies in git fairly well, under an
assumed workflow of:
- Develop in local branches, never rewrite history.
- Commit with "-m" or by typing a brief, non-template commit message
describing the checkpoint.
- Provide rich information in the web console (reviewers, etc.)
- Finalize with "git checkout master && arc merge branch && git push" or some
flavor thereof.
This supports Mercurial somewhat. The major problem is that "hg merge" fails if
the local is a fastforward of the remote, at which point there's nowhere we can
throw the commit message. Oh well. Just push it and we'll do our best to link
them up based on local commit info.
I am increasingly forming an opinion that Mercurial is "saftey-scissors git".
But also maybe I have no clue what I'm doing. I just don't understand why anyone
would think it's a good idea to have a trunk consisting of ~50% known-broken
revisions, random checkpoint parts, whitespace changes, typo fixes, etc. If you
use git with branching you can avoid this by making a trunk out of merges or
with rebase/amend, but there seems to be no way to have "one commit = one idea"
in any real sense in Mercurial.
Test Plan: Execute "arc merge" in git and mercurial.
Reviewers: fratrik, Makinde, aran, jungejason, tuomaspelkonen
Reviewed By: Makinde
CC: aran, epriestley, Makinde
Differential Revision: 860
Summary:
We have some git-specific logic on main pathways which should be in the API
class, move it around so "arc lint" with an engine works under Mercurial. This
resovles the error @makinde reported:
> PHP Catchable fatal error: Argument 1 passed to
ArcanistBaseWorkflow::parseGitRelativeCommit() must be an instance of
ArcanistGitAPI, instance of ArcanistMercurialAPI given, called in
/home/makinde/.arc_install/arcanist/src/workflow/lint/ArcanistLintWorkflow.php
on line 131 and defined in
/home/makinde/.arc_install/arcanist/src/workflow/base/ArcanistBaseWorkflow.php
on line 830
Test Plan: Ran "arc diff" in git and hg working copies, plus "arc lint" with a
configured "lint_engine".
Reviewers: Makinde, aran, jungejason, nh, tuomaspelkonen
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 935
Summary:
- Use "hg parents" to figure out where outgoing changes originate from, not
"~1", since that's a new feature in Mercurial.
- Add "--style default" all over the place since hg config can override this
(similar to the date config issues we saw in git).
- Cache working copy status so we don't run full hg diffs like 30 times
(similar to git/svn APIs).
- Use full "--git" flag instead of rather cryptic "-g".
Test Plan: Ran "arc diff" in my hg working copy, got the diff I expected.
Reviewers: Makinde, aran, jungejason, nh, tuomaspelkonen
Reviewed By: Makinde
CC: aran, Makinde
Differential Revision: 934
