Summary:
Fixes T4163. Re-word the question so its the same as in "arc diff". Draw the line though and don't automagically do anything if the user says "Y", 'cuz amending / adding files last minute in 'arc land' and other workflows is batshit insane. Assuming infinite growth in the future, I think its best to get this language consistent now.
I changed the shouldAmend member variable to a shouldAmend() function with a static inside. Previously shouldAmend was getting set as a side effect and it was kind of weird. I thought maybe it was written this way because the calls to the vcs are a little slow or something. As such, I figured caching it in the static was a good idea? Didn't seem awful but maybe a premature optimization with whatever the performance reality turns out to be.
Also a modest amount of bonus pht.
Test Plan: this very diff. i'm going to arc land it laters too.
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: chad, epriestley, Korvin
Maniphest Tasks: T4163
Differential Revision: https://secure.phabricator.com/D8753
Summary: Fixes T4291. Also pht user-facing strings.
Test Plan:
made a branch `foo` off master. made a commit and a diff in `foo`. switched backed to master and cowboy committed some thing. went back to branch`foo`, did an arc land, and saw the error message. went back to master, did a git resert --hard HEAD^1, went back to branch `foo`, and then successfully arc landed.
also ran arc help land and things looked good
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: epriestley, Korvin
Maniphest Tasks: T4291
Differential Revision: https://secure.phabricator.com/D8738
Summary:
I changed the rendering of the bar color for the `priority` column when running `arc tasks` to match the `priorityColor` property.
If the `priorityColor` property is one of the basic colors already supported by ansi, then the bar is set to that color, otherwise it is set to white.
This will allow the user to customize maniphest priorities and then set their own colors and have those colors display correctly when running `arc tasks`
Fixed some linting errors
Test Plan:
Run `arc tasks` and ensure that:
* the priority column is displayed with a colored bar
* the priority column bar is the correct color (or white if it is an unsupported color)
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley, Korvin
Differential Revision: https://secure.phabricator.com/D8735
Summary: I changed the way `arc tasks` was checking to see if a maniphest task was closed from looking for a non-empty status to looking for the isClosed property submitted in D8731.
Test Plan: Run `arc tasks` and check that the tasks show open/closed correctly.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley, Korvin
Differential Revision: https://secure.phabricator.com/D8732
Summary:
We recently moved our HEAD and it caused some issues on `arc patch` with git-svn repos. The base revision is incorrect and patch will fail. Add the check in such case to make it work.
The check was there before but removed in change b202158. The reason wasn't mentioned there though.
Test Plan: Tried it on svn.
Reviewers: lifeihuang, JoelB, #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley, Korvin
Differential Revision: https://secure.phabricator.com/D8717
Summary: Ref T4670.
Test Plan: arc patch D8685; arc land --hold; verified i got a nice message asking me to be sure i wanted to land epriestley's code
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: epriestley, Korvin
Maniphest Tasks: T4670
Differential Revision: https://secure.phabricator.com/D8709
Summary: Remove any-author flag from 'arc which' make any-author be the default behavior. Annotate revision with the owners username.
Test Plan:
Apply a patch that you don't own (arc patch Dxxx) run 'arc which' verify that:
1. You see the revision
2. You see the original authors username next to the revision (owned by sjobs) etc.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: epriestley, Korvin
Differential Revision: https://secure.phabricator.com/D8690
Summary: Of note this now forces the pytest-cov plugin to be installed unless they turn it off.
Test Plan:
```
../arcanist/bin/arc unit
COVERAGE REPORT
97% changes/models/test.py
100% tests/changes/api/serializer/models/test_testcase.py
100% changes/api/serializer/models/testcase.py
UNIT OKAY No unit test failures.
Updated an existing Differential revision:
Revision URI: https://tails.corp.dropbox.com/D43387
Included changes:
M changes/api/serializer/models/testcase.py
M changes/models/test.py
M tests/changes/api/serializer/models/test_testcase.py
```
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley, Korvin
Differential Revision: https://secure.phabricator.com/D8667
Summary:
Fixes T4596. I misunderstood this issue and D8512 was not correct. Specifically:
- The `hg log` needs to be escaped, since otherwise "arcpatch-x" is interpreted as a revset.
- The `hg update` does not need to be escaped, since updating to a revset doesn't make sense and the command never treats its argument as a revset.
- The `hg bookmark` does not need to be escaped, for similar reasons.
Test Plan:
- Ran these commands in isolation and got sensible, consistent results.
- Ran `arc patch` several times in a row and got proper bookmark names.
Reviewers: btrahan, durham, rvanvelzen
Reviewed By: rvanvelzen
Subscribers: epriestley
Maniphest Tasks: T4596
Differential Revision: https://secure.phabricator.com/D8661
Summary: Fixes T4649. The issue in that task is caused because we're submitting a block of text including comments. We've probably been doing this for a long time, but maybe were more liberal in parsing before. Instead, strip them.
Test Plan: Ran `arc diff --verbatim` and got "Dxxx" prefilled correctly.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T4649
Differential Revision: https://secure.phabricator.com/D8658
Summary: Ref T4697. This is similar to the existing stuff, but `svnlook diff ...` can also produce a "Copied" header. Currently, we choke on it in Herald when running pre-commit rules.
Test Plan: Added and ran tests. In the next diff, used this in a real system.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T4697
Differential Revision: https://secure.phabricator.com/D8653
Summary:
see https://github.com/facebook/phabricator/issues/546 - arc complete blows up when not
in a workdir.
There's no "is ArcanistWorkingCopyIdentity object valid" method, but getVCSType() looks like the
closest match.
git grep for `getProjectRoot` didn't reveal any more problmatic call sites.
Test Plan: `arc [tab] [tab]`
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley, Korvin
Differential Revision: https://secure.phabricator.com/D8578
Summary:
Fixes T4063. The `git format-patch` command produces a special header
and footer which we need to detect, strip, and parse.
Test Plan:
- Added and ran unit tests.
- Submitted a diff with `git format-patch HEAD^ --stdout | arc diff --raw`.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T4063
Differential Revision: https://secure.phabricator.com/D8547
Summary:
Fixes T4603. We fire `arc close-revision --finalize` implicitly from `arc land`, which may close a corresponding Differential revision.
We want to close if the repository is not present in Phabricator (i.e., we'll never be able to close in response to the commit message, since we'll never see it). Historically, we used Arcanist Project -> "Tracked" to make this determination. Instead, just check if the working copy is associated with a repository. This is simpler, easier, and works better.
Test Plan: Ran `arc close-revision --finalize`.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: aran, epriestley
Maniphest Tasks: T4603
Differential Revision: https://secure.phabricator.com/D8523
Summary:
Ref T4603. This workflow predates `arc land` and doesn't make much sense in modern Phabricator/Arcanist. It is surprising that `arc amend` will sometimes close accepted revisions, and we're better at detecting that repositories are tracked, and tracking repositories is easier.
Also fix some inaccuracies and old claims in the documenation and help.
Test Plan: Ran `arc amend`.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: aran, epriestley
Maniphest Tasks: T4603
Differential Revision: https://secure.phabricator.com/D8522
Summary: Fixes T4596. I couldn't immediately reproduce this, but the `hgsprintf()` version is clearly more correct.
Test Plan: Used `arc patch --trace` to examine hg commands.
Reviewers: btrahan, durham
Reviewed By: durham
Subscribers: aran, epriestley
Maniphest Tasks: T4596
Differential Revision: https://secure.phabricator.com/D8512
Summary:
We recently tried to advance the arcanist HEAD in our release branch but failed, due to an exception in SVN pre-commit hook like this:
abort: Commit blocked by pre-commit hook (exit code 1) with output:
LINT 1.3s FacebookWebJSLintLinter (1 file)
Exception
Some linters failed:
- FacebookWebCopyrightLinter: BadMethodCallException: Call to a member function getConfigFromAnySource() on a non-object
(Run with --trace for a full exception trace.)
However `arc lint` works just fine. By searching the change history, it looks related to a few commits D7271, D7377, D7382, especially D7377, where configuration manager is added to the lint engine. Add it to the SVN precommit hook workflow too.
Test Plan: I am not quite sure how to test it out easily. Any suggestions?
Reviewers: lifeihuang, JoelB, epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: aran, epriestley, mikemag, Korvin
Differential Revision: https://secure.phabricator.com/D8492
Summary:
pretty straight-forward stuff here. Note that in other events we use "fields" or "specification" rather than "revision"; I think "revision" is best particularly in this context where it is in fact the revision being landed.
Fixes T4565.
Test Plan: php -l
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: aran, epriestley, Korvin
Maniphest Tasks: T4565
Differential Revision: https://secure.phabricator.com/D8484
Summary:
Fixes T4559.
It looks like the code currently (at least partially) handles this by checking for `(no branch)`. I suspect that the behaviour of `git` has changed (I am running version 1.9.0) because I haven't figured out what state to be in to cause `git` to output `(no branch)`.
Test Plan: Ran `arc branch` when on a detached branch.
Reviewers: #blessed_reviewers, epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T4559
Differential Revision: https://secure.phabricator.com/D8466
Summary: See discussion in D8460. Primarily, this prints out the failing value when a true/false assertion fails, so if it was something useful (like a function result) it's visible.
Test Plan:
Added `assertTrue("quack")`:
FAIL ArcanistDiffParserTestCase::testParser
Assertion failed, expected 'true' (at ArcanistDiffParserTestCase.php:16).
ACTUAL VALUE
quack
Added `assertFalse("quack")`:
FAIL ArcanistDiffParserTestCase::testParser
Assertion failed, expected 'false' (at ArcanistDiffParserTestCase.php:16).
ACTUAL VALUE
quack
Added `assertEqual("quack", "moo")`:
FAIL ArcanistDiffParserTestCase::testParser
Assertion failed, expected values to be equal (at ArcanistDiffParserTestCase.php:16).
Expected: quack
Actual: moo
Reviewers: joshuaspence
Reviewed By: joshuaspence
CC: aran
Differential Revision: https://secure.phabricator.com/D8465
Summary:
There are quite a few tests in Arcanist, libphutil and Phabricator that do something similar to `$this->assertEqual(false, ...)` or `$this->assertEqual(true, ...)`.
This is unnecessarily verbose and it would be cleaner if we had `assertFalse` and `assertTrue` methods.
Test Plan: I contemplated adding a unit test for the `getCallerInfo` method but wasn't sure if it was required / where it should live.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Differential Revision: https://secure.phabricator.com/D8460
Summary:
I noticed that code coverage wasn't showing in Differential for
some repositories that we are using with Phabricator.
`arc unit` would should unit test coverage, but the paths were messy
(for example, `.//foo/bar.py` instead of `foo/bar.py`). As a result,
the code coverage info wasn't recognised as being for the correct
module.
I'm not sure why this logic is the way that it is... perhaps this is to
do with an older version of `nose` (I am using v1.3.0).
Test Plan:
I created a diff for an internal repository that we have, and
observed that code coverage was displayed in Differential.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran, seporaitis, avive, dctrwatson, zeeg
Differential Revision: https://secure.phabricator.com/D8433
Summary:
Fixes T4570. When a test case doesn't make any assertions, fail it:
- A tiny fraction of tests pass by not throwing. These tests can easily make a trivial assertion. It took about 5 minutes to fix them all (D8435, D8436).
- In other cases, no assertions means a test construction problem, as with T4570. In these cases, failing loudly catches a severe error.
- Fixes the no-assertion test cases in `arcanist/`
- Makes the PHP 5.4 test pass for the moment, see discussion in T4334.
Test Plan: Ran `arc unit --everything`.
Reviewers: leebyron, btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T4570
Differential Revision: https://secure.phabricator.com/D8437
Summary: Ref T4570. Testing a directory with no recognized tests currently passes, but should fail.
Test Plan:
- Ran `arc unit`.
- Removed tests from a test directory, ran `arc unit`, got test failure.
Reviewers: leebyron, btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T4570
Differential Revision: https://secure.phabricator.com/D8434
Summary: D7271 added this line but the signature of function ArcanistWorkingCopyIdentity::newFromRootAndConfigFile(...) was never changed. I have no idea why it was added here and it's causing failures in facebook. I am just removing it now.
Test Plan: No idea how to test
Reviewers: wez, lifeihuang, JoelB, #blessed_reviewers, epriestley
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Differential Revision: https://secure.phabricator.com/D8432
Summary: Ref T988. Not pretty or polished, but works fine.
Test Plan: Generated the docs, made sure they were all there.
Reviewers: chad, btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T988
Differential Revision: https://secure.phabricator.com/D8413
Summary:
Ref T2222. Discussion in T4481. Ref T2543.
Primarily, this will let me fix some of the rough edges that came out of ApplicationTransactions and state-based transitions.
I'm also adding a constant for T2543 while I'm in here.
Test Plan: Used `grep` to look for any weird special behavior, didn't see any.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2543, T2222
Differential Revision: https://secure.phabricator.com/D8400
Summary: Allow `@todo` comments to be linted as TODOs as well as `TODO` comments.
Test Plan: I added a new test case (`todo.lint-test`)
Reviewers: #blessed_reviewers, epriestley
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Differential Revision: https://secure.phabricator.com/D8389
Summary:
- Tidied up `ArcanistCSSLintLinter::getDefaultBinary`.
- Tidied up `CSSLintLinter::getDefaultFlags` function.
- Tidied up `ArcanistPhpcsLinter::getDefaultBinary` function.
- Tidied up `ArcanistPEP8Linter::getDefaultFlags` function
- Tidied up `ArcanistFlake8Linter::getDefaultFlags`.
- Tidied up `ArcanistCppcheckLinter::getLintOptions`.
- Tidied up `ArcanistCppcheckLinter::getLintPath`.
- Tidied up `ArcanistCpplintLinter::getLintOptions`.
- Tidied up `ArcanistCpplintLinter::getLintPath`.
- Removed child functions which are identical to the corresponding parent functions.
Test Plan: `arc lint` and `arc unit`.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: epriestley
CC: chad, Korvin, epriestley, aran
Differential Revision: https://secure.phabricator.com/D8385
Summary: To me, it seems that these comments add no value. Personal opinion I suppose.
Test Plan: N/A
Reviewers: #blessed_reviewers, epriestley
Reviewed By: epriestley
CC: chad, Korvin, epriestley, aran
Differential Revision: https://secure.phabricator.com/D8382
Summary: These are technically added in PHP 5.5.0 mainline, not an extension, but the two are effectively equiavlent.
Test Plan: o.O
Reviewers: btrahan, dctrwatson
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D8280
Summary: See <https://github.com/facebook/arcanist/pull/131>
It looks like the behaviour of csslint has changed since the ArcanistCSSLintLinter was written. Consequently, ArcanistCSSLintLinter does not work with the latest version of csslint (v0.10.0).
- `csslint` has a non-zero exit status
- Fixed `csslint` parsing for v0.10.0
Reviewed by: epriestley
Summary:
Do a little cleanup:
- Remove copyright header (we removed all of these a long time ago, this one just snuck through somehow).
- Remove `@group` comment (obsolete with new Diviner).
- Note support for all VCSes.
- Add pht() for translation.
- Hint `arc browse .`.
- Fail on no paths sooner.
- Raise a useful error if we can't figure out which repository we're heading to.
- Clarify "open" comment.
- Use `Filesystem::binaryExists()`.
- Some minor wordsmithing.
Test Plan: `arc browse`, `arc browse .`, `arc browse README`, `arc browse README src`, ran `arc browse` in valid working copy with no associated repo.
Reviewers: btrahan, spicyj
Reviewed By: spicyj
CC: aran
Differential Revision: https://secure.phabricator.com/D8176
Test Plan: Ran 'arc browse' in a repo that contains a .arcconfig and one without.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Differential Revision: https://secure.phabricator.com/D8173
Summary:
Especially on Windows it is hard to use "\1" type escapes in shell commands. The direct usage resulted in some undefined variables because the \1 and \2 weren't actually passed as control characters.
By passing them through the regular arguments list they get sent in the "correct way" regardless of OS
Test Plan: Executed `arc diff` in a HG repo and did not get undefined indexes back
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Differential Revision: https://secure.phabricator.com/D8129
Summary: Fixes T3929. If arc is merging it will say merging * into * vs always saying rebasing * onto *
Test Plan: Make arc do a rebase, then a merge
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T3929
Differential Revision: https://secure.phabricator.com/D8085
Summary:
Unlike git, Mercurial considers an `hg commit --amend` which doesn't change the working copy to be an error.
The current behavior for this is fairly bad, since the user gets an exception wrapping the entire command ("Command Failed! ...") and it's pretty verbose and not obvious what has happened.
Some alternatives are:
1. Detect this condition and raise a more tailored exception, like a UsageException.
2. Detect this condition and succeed.
Although I tend to think (1) is the right approach in general (that is, `arc x` should usually behave like `git x` or `hg x`), I went with (2) here because we have a handful of amend callsites and they all assume git semantics (no-op amends are successful), and because I think Mercurial's behavior is a little silly (the working copy ends up in the correct / expected state, which seems fairly clearly like a success to me).
Test Plan:
- Had reporting user verify patch.
- Ran `arc amend --revision Dxxx` twice in a Mercurial working copy.
The old output looked like this:
$ arc amend --revision 922
Amending commit message to reflect revision D922: Improve performance of PhutilSymbolLoader::selectAndLoadSymbols().
Exception
Command failed with error #1!
COMMAND
HGPLAIN=1 hg commit --amend -l '/private/var/folders/8k/c3vkmjy5335gcxdzxkhwq82w0000gn/T/71k8q057844c4g84/3539-gfkvV4'
STDOUT
nothing changed
STDERR
(empty)
(Run with --trace for a full exception trace.)
The new output looks like this:
$ arc amend --revision 922
Amending commit message to reflect revision D922: Improve performance of PhutilSymbolLoader::selectAndLoadSymbols().
Done.
Reviewers: btrahan, richardvanvelzen
Reviewed By: richardvanvelzen
CC: aran
Differential Revision: https://secure.phabricator.com/D8083
Summary:
- The modern name for the config is "project.name".
- Missing parameter in a pht().
- When the value is set, but not valid, we gave you a misleading error message.
Test Plan: Ran `arc which`.
Reviewers: talshiri, btrahan
Reviewed By: talshiri
CC: aran
Differential Revision: https://secure.phabricator.com/D8081
Summary:
Fixes T4349. Two issues:
- As discussed in T4349, we would trim the entire output and then require spaces when matching. This choked incorrectly if the last line of a file contained only whitespace. Use `phutil_split_lines()` instead, and regexp things more reasonably.
- We were capturing the line text, not the commit, as "revision". This isn't actually used elsewhere, but was obviously wrong. Make this consistent with Git/SVN.
Test Plan: Rigged a call up and saw reasonable output after the patch, on a working copy which threw before the patch.
Reviewers: durham, btrahan
Reviewed By: durham
CC: aran
Maniphest Tasks: T4349
Differential Revision: https://secure.phabricator.com/D8078
Summary:
A couple of things actually still care about this:
- The code that detects if we're running `arc` on `arc`, and then runs the `arc` we're running on instead of the `arc` we ran, checks for the project being "arcanist" (`reenter_if_this_is_arcanist_or_libphutil()`).
- Some of the symbol stuff still relies on it.
- Removing it was really more to prove it works (which it seems to) than to accomplish any concrete goals.
Test Plan: looked at the file I guess
Reviewers: btrahan, talshiri
Reviewed By: talshiri
CC: aran
Differential Revision: https://secure.phabricator.com/D8076
Summary:
Ref T4343. Continues the process of reducing the prominence of Arcanist Projects. Primarily:
- Query Phabricator to identify the working copy based on explicit configuration, or guess based on heuristics.
- Enhance `arc which` to explain the process to the user.
- The `project_id` key is no longer required in `.arcconfig`.
Minor/cleanup changes:
- Rename `project_id` to `project.name` (consistency, clarity).
- Rename `conduit_uri` to `phabricator.uri` (consistency, clairty).
- These both need documentation updates.
- Add `repository.callsign` to explicitly bind to a repository.
- Updated `.arcconfig` for the new values.
- Fix a unit test which broke a while ago when we fixed a rare definition of "unstaged".
- Make `getRepositoryUUID()` generic so we can get rid of one `instanceof`.
Test Plan:
- Ran `arc which`.
- Ran `arc diff`.
- This doesn't really change anything, so the only real risk is version compatibility breaks. This //does// introduce such a break, but the window is very narrow: if you upgrade `arc` after this commit, and try to diff against a Phabricator which was updated after yesterday (D8068) but before D8072 lands, the lookup will work so we'll add `repositoryPHID` to the `differential.creatediff` call, but it won't exist in Phabricator yet. This window is so narrow that I'm not going to try to fix it, as I'd guess there is a significant chance that no users will be affected. I don't see a clever way to fix it that doesn't involve a lot of work, either.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T4343
Differential Revision: https://secure.phabricator.com/D8073
