Summary:
The behavior (and name) of this function was changed in
D16405, but the documentation was not updated to reflect the new
contract.
Test Plan: Untested; pure doc changes.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin
Differential Revision: https://secure.phabricator.com/D16425
Summary:
Since 737f5c0df9 arcanist shows [y/N] two times, when arcanist asks you, if you want to proceed, if you want to update a not owned revision. Ths patch fixes
that, so that Arcanist shows [y/N] only once, like at other situations, when Arcanist asks you a question.
Ref T11489
Test Plan: Updated a not owned revision.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley
Tags: #arcanist
Maniphest Tasks: T11489
Differential Revision: https://secure.phabricator.com/D16415
Summary:
Ref T7148. When uploading files from the CLI with a particular view policy, we may not respect it if the file is unique (so the data isn't already known) and small (so it doesn't invoke the chunker).
This is rare (and may never have happened outside of testing) because:
- production dumps are always larger than the minimum chunk size;
- only cluster stuff uses `setViewPolicy()`;
- the default policy is "Administrators" anyway, which is safe.
However, I caught it in local testing, so fix it up.
Test Plan: Used `bin/host upload --file ...` to upload a small, unique file. Verified it uploaded with the correct custom view policy ("No One") rather than the default view policy ("Administrators").
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T7148
Differential Revision: https://secure.phabricator.com/D16408
Summary:
The previous parser failed when only one of the old/new filenames was
quoted, as with a rename of a Unicode filename to a non-Unicode
filename, or vice versa. It also incorrectly parsed custom prefixes,
even going to far as to encode this logic in the tests: a diff of
"src/file dst/file" which is not a rename should not be recorded as
changing "src/file", but rather "file".
Switch to using the "rename from" / "rename to" as the source of truth
for old and current filenames when complex renames are done. This
matches the logic that git itself uses to parse patches; the contents
of the `diff --git` line are merely viewed as a simplest-case
prediction, with renames handled later should it not match.
The diff parser already had logic to parse "rename from" / "rename to"
lines and extract their information. As such, this diff consists
primarily of removing logic from the `splitGitDiffPaths` method, and
allowing it to quietly fail.
This resolves two ambiguity mentioned in comments and tests: in
renaming "old file old" to "file", as well as the renaming of
"old file with spaces" to "new file with spaces" when neither are
quoted.
Test Plan:
`arc unit`. Many of the existing test cases no longer
applied to the `splitGitDiffPaths` method; they were moved into a new
test method which also supplies values for "rename from" and "rename
to" lines.
As noted in the summary, this alters one of the expected values of an
existing test. Specifically, the following diff is a file addition of
`file` with custom prefixes, and not the addition of "dst/file":
```
diff --git src/file dst/file
new file mode 100644
index 0000000..1269488
--- /dev/null
+++ dst/file
@@ -0,0 +1 @@
+data
```
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D16405
Summary:
`parseGitRawDiff` dealt with the `A`, `M`, and `D` status flags from
`git diff --raw`, for file additions, modifications, and deletions
respectively. However, it failed to cope with `C` and `R` flags, for
copies and renames. Git version 2.9 and above default to resolving
renames, even in `git diff --raw` output, making this lack of support
only salient now (though users with Git's `diff.rename` set
encountered it previously).
Those two flags differ from the other three in that they offer both the
source and destination filename, separated by a tab. As
`parseGitRawDiff` was not aware of this property, it returned a
"filename" of `"oldfile\tnewfile"`. This is surfaced in several
places, including as passed to linters as a filename to check.
Needless to say, this file is nearly guaranteed to never exist on
disk.
Detect both the `C` and `R` flag types, and generate either a file
addition, or a pair of addition/deletion entries.
Test Plan:
Renamed a file, with a linter that printed each file it was
called with.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: jboning, Korvin
Differential Revision: https://secure.phabricator.com/D16387
Summary:
Fixes T11435. This isn't a perfect solution since there's a little code duplication, but a perfect solution is probably a bit more involved.
See T11435 for some discussion. In particular, most `git diff` commands already get this flag via `ArcanistGitAPI->getDiffBaseOptions()`.
Test Plan: Will land this change.
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T11435
Differential Revision: https://secure.phabricator.com/D16375
Summary: Depends on D16364. See that revision for discussion.
Test Plan: Ran `arc get-config history.immutable` in `arcanist/`, saw setting.
Reviewers: chad, fooishbar
Reviewed By: fooishbar
Differential Revision: https://secure.phabricator.com/D16365
Summary: Ref T11409. Add lint to detect using `[...]` to define arrays. This doesn't work in PHP 5.2/5.3, which some of our users run.
Test Plan:
- Ran `arc unit`.
- Ran `arc lint src/applications/harbormaster/conduit/HarbormasterQueryBuildsConduitAPIMethod.php` to detect the issue in T11409.
Reviewers: yelirekim, chad
Reviewed By: chad
Maniphest Tasks: T11409
Differential Revision: https://secure.phabricator.com/D16357
Summary: Fixes T7489. Depends on D16332, which moved this code to libphutil.
Test Plan:
```
$ arc banch --bystatus
(Assuming 'banch' is the British spelling of 'branch'.)
(Assuming '--bystatus' is the British spelling of '--by-status'.)
...
```
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T7489
Differential Revision: https://secure.phabricator.com/D16333
Summary:
The `cp` and `mv` commands, when run from a Windows
environment, error out. Use library functions from `Filesystem`
for cross-platform compatibility.
Test Plan:
Ran `arc lint` with auto-fix lint errors on both Linux and
Windows.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Spies: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D16273
Summary: Fixes T11222. This was lazy-future-proofed for Conduit SSH support, but users are boundlessly creative. Check protocols explicitly.
Test Plan:
```
$ arc install-certificate a.b:1/
Usage Exception: Server URI "a.b:1/" must include the "http" or "https" protocol. It should be in the form "https://phabricator.example.com/".
```
- Also went through a successful workflow with a URI in the form provided in the example.
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T11222
Differential Revision: https://secure.phabricator.com/D16188
Summary:
Ref T4631. Ref T10939. I don't have any good solutions here; this is perhaps the least-bad one.
- This prompt is misleading/confusing in the presence of Herald/Owners.
- This prompt is likely of very little value for experienced reviewers.
- When it works, this prompt may be of some value for new reviewers, but getting it wrong is probably more confusing than getting it right is helpful, and there is a more accurate version of the warning in the web UI that new users are likely to see.
- In the long run, this code should not live in the client.
Test Plan: Created this revision without specifying reviewers, probably didn't get prompted.
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T4631, T10939
Differential Revision: https://secure.phabricator.com/D16139
Summary: Ref T10227. This converts weird hard-codey magic to the new HTTPEngineExtension.
Test Plan: See D16090.
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T10227
Differential Revision: https://secure.phabricator.com/D16091
Summary: Ref T7643. This moves the prefix/suffix smoothing behavior back to the old one, which seems better for code diffs.
Test Plan: `arc unit --everything`
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T7643
Differential Revision: https://secure.phabricator.com/D16074
Summary:
Ref T7643. We've done the smoothing from D16068 for a long time, it just wasn't part of EditDistanceMatrix.
Since it now is, call it instead of keeping a copy of the logic around.
Previously, the unit tests were testing un-smoothed diffs. Just have them test smoothed diffs instead, since nothing actually uses unsmoothed diffs.
Test Plan: Pasted a raw diff into the web UI, got a sensible inline diff for it.
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T7643
Differential Revision: https://secure.phabricator.com/D16069
Summary: See D15828 - arc is reporting file size as `0` for unexisting files - make it stop.
Test Plan: `arc diff` with empty, deleted, added files - see size reported as `null` when appropriate.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: chad, Korvin
Differential Revision: https://secure.phabricator.com/D16059
Summary: Fixes T10431. Updates the getAliases() method to read every `arc` configuration file.
Test Plan:
Added an `arc duck` alias to `/etc/arcconfig`, ran `arc duck`, saw "quack".
Added an `arc pig` alias to `~/.arcrc` via `arc alias`, ran `arc pig`, saw "oink oink".
Reviewers: nevogd, chad, #blessed_reviewers
Reviewed By: chad, #blessed_reviewers
Subscribers: eadler, epriestley
Tags: #twitter
Maniphest Tasks: T10431
Differential Revision: https://secure.phabricator.com/D15342
Summary:
When doing svn copy, or svn mv, a SynthenticAdditionDiff is generated.
If the path is a directory, an error will occur when checking the
mime-type of the directory. Immediately after the properties check,
the function returns null if the path is a directory. Move this
check to before the properties check to avoid exiting with an error.
```
Command failed with error #1!
COMMAND
svn propget 'svn:mime-type' '/home/trasz/svn/ports/cad/py-pycam'@
STDOUT
(empty)
STDERR
svn: warning: W200017: Property 'svn:mime-type' not found on '/home/trasz/svn/ports/cad/py-pycam@'
svn: E200000: A problem occurred; see other errors for details
(Run with `--trace` for a full exception trace.)
```
Test Plan: Created differentials of changes with `svn copy` and `svn mv`
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin
Tags: #subversion
Differential Revision: https://secure.phabricator.com/D15985
Summary:
`arc which` is currently broken for svn repos.
Fixes T6635
(I think I independently wrote an identical change to yours)
Test Plan: `∴../../p/arcanist/bin/arc which` on a svn repo.
Reviewers: aik099, epriestley, #blessed_reviewers
Reviewed By: aik099, epriestley, #blessed_reviewers
Subscribers: aik099, Korvin
Maniphest Tasks: T6635
Differential Revision: https://secure.phabricator.com/D15922
Summary:
See rARC3ffed59bd7. Currently, when a unit test includes a syntax error, it is raised in an unclear way ("error at line 10, char 1: XHP1 Unknown lint message!").
This is because each test case only activates rules it wants to test, so we lose the ID/name for the syntax message. However, we always want to test this and the lint engine can always raise it.
To get a better error message, include it unconditionally. So a test for rule `X` really tests two rules: syntax, and `X`.
Test Plan:
Ran `arc unit` at HEAD, got a better test failure:
```
FAIL ArcanistCallTimePassByReferenceXHPASTLinterRuleTestCase::testLinter
In 'call-time-pass-by-reference.lint-test', expected lint to raise error on line 10 at char 8, but no error was raised. Actually raised:
error at line 10, char 1: XHP1 PHP Syntax Error!
```
NOTE: This doesn't pass tests yet, it just makes the test failure easier to understand. I'll see about fixing the test in the next change.
Reviewers: chad, richardvanvelzen
Reviewed By: richardvanvelzen
Differential Revision: https://secure.phabricator.com/D15819
Summary: See D15678. A node containing a return type hint is added right before the statement body node. This updates all relevant linter rules to now retrieve the body from the correct index.
Test Plan: Ran `arc unit --everything`, inspected every single message to make sure all xhpast test cases succeeded. Also grepped for `getChildByIndex(5` and `getChildOfType(5`.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin
Differential Revision: https://secure.phabricator.com/D15814
Summary: It is currently not possible to apply multiple linter standards because `$value` was used instead of `$standard`.
Test Plan: Was able to apply multiple linter standards.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: eadler, thaiphv, Korvin
Differential Revision: https://secure.phabricator.com/D15212
Summary:
Flake8 extensions are allowed to use their own letter-prefixed codes. For
example, the flake8-debugger extension emits 'T002'-tagged messages.
This change relaxes getLintCodeFromLinterConfigurationKey() to also recognize
extension codes. Otherwise, attempting to configure message severities for
e.g. 'T002' would result in an exception.
Messages from extensions continue to default to ERROR severity, as they did
before this change.
Test Plan:
Successfully reduced the severity of 'T002' to a warning via .arclint:
"severity": {
"T002": "warning"
}
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin
Differential Revision: https://secure.phabricator.com/D15810
Summary:
This is mostly just a personal quality-of-life fix. I run this command fairly often and having it return a little faster is nice.
This replaces a `git show` for each individual branch with a big `git for-each-ref` which we were already running anyway. This is quite a bit faster.
This command also occasionally hangs or segfaults for me while executing the huge pile of subprocesses. This is unreliable to reproduce, probably some bug in some PHP extension I have, and likely hard to narrow down, and this approach is better in every way anyway.
Test Plan:
- Ran `arc branch` in Git, observed faster output (in my `phabricator/`, about 2000ms -> 1200ms).
- Ran `arc feature` in Mercurial.
Reviewers: chad
Reviewed By: chad
Differential Revision: https://secure.phabricator.com/D15735
Summary:
It is common practice in Wikimedia's projects to amend a contributor's
change without taking over authorship of the change. We found that
the only enforcement of commandeering before amending is in arcanist,
not validated server-side. While it would be fairly straightforward to
maintain this as a patch to arcanist, I thought I would see if upstream
is willing to support making this optional.
With this change, amending without commandeering is enabled by a flag in
`.arcconfig` and it defaults to the old behavior.
For background see [wmf T121751](https://phabricator.wikimedia.org/T121751)
Test Plan:
* ran `arc patch D146` to locally apply a revision that I did not author,
* made a trivial change and amended the commit.
* ran `arc diff --update D146 HEAD^` to send the update to differential
* Saw that https://phabricator.wikimedia.org/D146 updated as it should.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: greggrossmeier, aklapper, Luke081515.2, Korvin, dereckson
Maniphest Tasks: T10584
Differential Revision: https://secure.phabricator.com/D15468
Summary:
Fixes T10707. Currently, `arc backout` creates a commit message which includes questionably-helpful "tips" in the message itself.
Strip these out.
Test Plan:
Used `arc backout` to revert any commit, then `git show` to see the generated message.
- Before patch: included tips.
- After patch: no tips.
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T10707
Differential Revision: https://secure.phabricator.com/D15573
Summary:
Remove couple of references to callsigns:
- `arc which` now prints repository name
- `getShouldAmend()` can now use new format of commit name
a quick git-grep looks like the remaining references are all about `repository.callsign` config.
Ref T4245
Test Plan:
- `arc which` on a repository with no callsign
- trigger `requireCleanWorkingCopy()`, see both "Do you want to amend this change" and "Do you want to create a new commit" prompts.
- fire this diff with new code.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin
Maniphest Tasks: T4245
Differential Revision: https://secure.phabricator.com/D15472
Summary:
Ref T10093. Right now, Phabricator kind of guesses that `arc` probably pushed stuff to the staging area.
This can cause confusing/misleading errors later, if it didn't actually push.
Instead, tell Phabricator that we pushed, so we can raise more tailored messages in the web UI (e.g., make "Land Revision" say "this wasn't pushed to the staging area" instead of "whoops, error!!~").
Test Plan:
Ran `arc diff` a few times, then looked in the database for properties.
{F1161655}
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T10093
Differential Revision: https://secure.phabricator.com/D15426
Summary:
Fixes T10509. Pushing changes to staging can be inefficient. What happens, roughly, is:
- Master is at commit "W" -- "W" is the most recent published commit in the main repository.
- The local working copy has one change on top of that, "X", so its history is commits "A, B, C, D, E, F, ..., U, V, W, X".
- The remote has some other previous changes that I or other users have made, maybe like "A, B, C, ..., S, T, U, Y" and "A, B, C, ..., T, U, V, Z", from previous pushes to staging areas.
- "X", "Y" and "Z" will never actually make it to master, because they'll be squash-merged/amended by `arc land`.
So the local says "I want to push 'X'", and the remote says "I know about 'Y' and 'Z', are those in the history of 'W'? You only need to send me new stuff if they are".
But they aren't, so the local says "nope, so here's the whole history for you". This is slow and sends a ton of data that the remote already has over the network.
In theory, Git could use a slightly different algorithm to tell the local about more commits, but this is hard, rarely useful, and not the kind of thing I'd be excited about changing if I was the Git upstream.
Instead, when pushing "X", also push "W", to trick Git into telling future clients about it.
Now, the remote should say "I know about 'W', 'Y' and 'Z'", and the local will say "oh, great, 'W' is in history, here's just the changes since then".
Also, fail `arc diff` if the push to staging fails, and tell users to use `--skip-staging`. This code has been in production for a while and doesn't seem to have any issues, and a failed push to staging prevents builds, lands, etc.
Test Plan:
- Ran `arc diff`, saw two changes push.
- Ran `arc diff --base arc:empty`, saw only one change push.
- Ran `arc diff` with an intentionally broken staging area, saw an error.
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T10509
Differential Revision: https://secure.phabricator.com/D15424
Summary:
With the old shebang of `#!/usr/bin/env python` on machines with python 3 as the default python it would fail.
Prefer an explicit python2 like PEP 394 suggests.
Test Plan: ran ./arc anoid and played a game
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin
Differential Revision: https://secure.phabricator.com/D15408
Summary: Fixes T10511. If you `arc browse --branch x/y/z`, we do not encode the URI properly.
Test Plan:
Ran `arc browse --branch x/y/z/ something.c`.
Before, got an error about "x" does not exist. This is wrong; the error should be about "x/y/z".
After, got the proper error:
{F1141096}
Reviewers: chad, avivey
Reviewed By: avivey
Maniphest Tasks: T10511
Differential Revision: https://secure.phabricator.com/D15397
