Summary: Ref T13546. When "arc land" performs cascading rebases, do them in "feature1", "feature2", etc., order so they're easier to follow. The outcome is not dependent on execution order.
Test Plan: Landed a change which cascaded many other branches, saw more comprehensible update order.
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21320
Summary: Ref T13546. This has a lot of dangerously rough edges, but has managed to land at least one commit in each Git and Mercurial.
Test Plan:
- Landed one commit under ideal conditions in Git and Mercurial.
- See followups.
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21315
Summary:
Ref T13546. The new "arc land" workflow can rebase several branches per second. With branches like "feature1", "feature2", etc., this leads to out-of-order listing in "arc branch".
When two branches would otherwise sort to the same position, sort them by name.
Test Plan: Ran "arc branch" after a cascading rebase by "arc land", saw "land5", "land7", "land8", etc., instead of an arbitrary order.
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21316
Summary: Ref T13544. This flag only disables a warning and should be a prompt default, not a flag.
Test Plan: Grepped for "ignore-unsound-tests", created this revision.
Maniphest Tasks: T13544
Differential Revision: https://secure.phabricator.com/D21303
Summary:
Ref T13546. Resizing the terminal window to send SIGWINCH, or other signals, may interrupt "stream_select()" with an error which upgrades to a RuntimeException.
When "stream_select()" fails, continue and test the stream itself.
Test Plan: Waited for a prompt, resized the window. Before patch: SIGWINCH interruption. After patch: undisturbed prompt.
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21317
Summary:
Ref T13546. Introduces a more structured construct for saving and restoring local repository state.
This is similar to old behavior, except that:
- "arc.autostash" is no longer respected;
- untracked changes are stashed; and
- we do not offer to amend.
Test Plan: In future changes, saved and restored various permutations of local state.
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21314
Summary: Ref T13546. Adds modern support for "arc.land.onto", "arc.land.onto-remote", and "history.immutable".
Test Plan: Read configuration in a future change.
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21313
Summary: Ref T13546. Prepares "arc land" to use hardpoint queries to load build information.
Test Plan: Ran `arc inspect --explore revision(1234)`, got a full related object tree including build information.
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21312
Summary: Ref T13546. These are used by a future "arc land" workflow to support the "Land changes you don't own?" and "Land changes with open dependencies?" prompts.
Test Plan: Ran a future "arc land" flow, hit both prompts.
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21311
Summary: See PHI1772. This is a quick patch to make some debugging/testing tasks easier until remaining workflows modernize. See T13490.
Test Plan: Ran `scripts/arcanist.php --library ...`, got the same behavior as `scripts/arcanist.php --load-phutil-library ...`.
Differential Revision: https://secure.phabricator.com/D21323
Summary: Ref T13544. This is an obscure flag and almost never useful. You can accomplish the same goal, roughly, with "git diff | arc diff --raw -". See also PHI675. See also T13338.
Test Plan: Grepped for "less-context", created this revision.
Maniphest Tasks: T13544
Differential Revision: https://secure.phabricator.com/D21301
Summary: Ref T13544. This flag is generally questionable, likely has no actual uses, and is slated for obsoletion and replacement elsewhere (see T13338).
Test Plan: Grepped for "encoding", found no relevant hits.
Maniphest Tasks: T13544
Differential Revision: https://secure.phabricator.com/D21300
Summary:
Ref T13544. These flags change the behavior of the "arc lint" subprocess.
I believe there is no reason to ever use "arc diff --lintall". If you want to find extra warnings to fix, you can use "arc lint --lintall" to express this intent.
Use of "arc diff --only-new" almost certainly means your linters are raising messages at "error" severity which should instead be raised at "warning" severity. If you only care about fixing a particular type of error in changed code, it should be raised as a "warning". The correct remedy is to adjust the severity, not use "--only-new", which is a very broad, slow, complicated hammer.
Test Plan: Searched for "lintall" and "only-new" in this workflow. These flags still exist in "arc lint", but may be changed in the future. Generated this change.
Maniphest Tasks: T13544
Differential Revision: https://secure.phabricator.com/D21298
Summary:
Ref T13544. Long ago, "arc diff" started prompting the user to provide "excuses" when they submitted changes with failing lint or unit tests.
At the time, "arc" was generally more heavy-handed and the review workflow had fewer points where it provided feedback about lint and test issues.
As the workflow has evolved, there is now significantly more feedback (promotion behavior from Draft in Differential, warnings on "arc land", etc). These days, these prompts feel archaic and like they're just getting in the way.
When lint/unit have Harbormaster-triggered components, this prompt is also too early (since Harbormaster tests may fail or raise lint messages later). A modern version of this would look more like putting revisions in some kind of locked state until authors explain issues. It's possible that's worth building, but I'd like to see more interest in it. I suspect this feature is largely just a "nag" feature these days with few benefits.
Test Plan: Grepped for "advice", "excuse", "handleServerMessage", "sendMessage", "getSkipExcuse", "getErrorExcuse", got no hits. Generated this revision.
Maniphest Tasks: T13544
Differential Revision: https://secure.phabricator.com/D21297
Summary:
Ref T13544. This flag was introduced in D1385 (2012) as part of a workflow which no longer exists. I can't recall anyone ever reporting an issue which involves its use and believe it is likely unused. It's not obvious to me why someone would use it in modern "arc".
(The same goal can be accomplished with "--message-file ...", although this requires more steps.)
Test Plan: Grepped for "use-commit-message" and "getCommitMessageFromCommit", ran "arc diff".
Maniphest Tasks: T13544
Differential Revision: https://secure.phabricator.com/D21296
Summary: Ref T13546. Add support for short "-x" flags and repeatable "--say moo --say quack" flags.
Test Plan: In future changes, used "arc diff -m" and "arc land --onto ... --onto ...".
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21310
Summary: Ref T13546. Future "arc land" workflows use this element to provide "hint" or "next step" suggestions to make error resolution easier.
Test Plan: Ran future "arc land" workflows, saw tips like "use --revision to do such-and-such" or "add these files to .gitignore".
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21309
Summary:
Ref T13546. Several substeps in the new "arc land" flow benefit from this. For example:
- When prompting "land revisions you don't own?", it's used to show authors.
- When prompting "land revisions in the wrong state?", it's used to show the current states.
Test Plan: Ran future "arc land" workflows, got relevant contextual information via this mechanism.
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21308
Summary:
Ref T13546. If we try to resolve several hundred hardpoint queries which execute subprocesses, we can currently hit system limits.
For now, limit resolution to 32 simultaneous futures. In the future, this should switch to `FuturePool` and possibly become more nuanced.
Test Plan: In a future change, ran `arc land --into-empty ...` to land thousands of commits. Before change, got a "proc_open()" error when launching too many simultaneous subprocesses. After change, this "worked".
Maniphest Tasks: T13546
Differential Revision: https://secure.phabricator.com/D21307
Summary: See PHI1765. You can find version information with "arc version". Remove this undocumented special-case.
Test Plan:
- Ran `arc --version`, no longer got a workflow API exception.
- Searched the documentation for references to `arc --version` (instead of `arc version`), found none.
Differential Revision: https://secure.phabricator.com/D21306
Summary: This simplifies merging a list-of-lists, which is a common pattern when asking a list of extensions to each provide some kind of list of items.
Test Plan: Used elsewhere in Piledriver.
Differential Revision: https://secure.phabricator.com/D21294
Summary:
With some frequency, code wants to assert that some "$o->m()" returns a list of objects of type X, possibly with unique values for some "getKey()"-style method result.
Existing checks via `PhutilTypeMap` and `assert_instances_of()` aren't quite powerful enough to do this while producing an easily understandable error state. We want to know that the error arose from a call to "$o->m()" in particular.
Test Plan: Currently used elsewhere, in Piledriver code.
Differential Revision: https://secure.phabricator.com/D21293
Summary: This is a path argument, and shell completion should suggest tabs.
Test Plan: Typed "arc liberate s<tab>", got path suggestions.
Differential Revision: https://secure.phabricator.com/D21292
Summary:
Ref T13543. Currently, the `cpplint` tests do not function because `cpplint` is passed a path which does not end in a suffix it recognizes.
Change the tempfile / path code to pass `linter path/to/example.c`-style linters a path they expect.
Then, correct some older code which was playing it fast-and-loose with "null" vs "0".
Test Plan: Ran `arc unit --everything`, got a clean bill of health on all the linters I have installed. (This is probably not all tests, since I have only a subset of linters installed locally that we have code for.)
Maniphest Tasks: T13543
Differential Revision: https://secure.phabricator.com/D21291
Summary:
Ref T13543. If a file ends in spaces and no newline, we'll emit a message suggesting removal of the spaces. This will effectively remove the line, but the code will then attempt to highlight text within the line.
Prior to D21044 this continued without raising an error and produced a reasonable result, but it now fatals. Insetad, don't try to highlight lines which no longer exist.
Test Plan: See T13543 for details.
Maniphest Tasks: T13543
Differential Revision: https://secure.phabricator.com/D21290
Summary:
Ref T13543. The "cpplint.py" script may emit messages on line 0, but Arcanist doesn't accept these messages.
This is a small piece of a whole set of broader issues, but stop the bleeding for now.
Test Plan:
- Ran `arc lint example.h` on a file with no `#ifndef` guard, and `cpplint` configured.
- Cpplint raised a message at line "0".
- Before change: arc choked when trying to render this.
- After change: arc survives rendering.
Maniphest Tasks: T13543
Differential Revision: https://secure.phabricator.com/D21289
Summary:
See PHI1735. "ConduitEngine" was once a future pool, but this has moved to "HardpointEngine". This class may no longer make much sense.
In Phacility code, "bin/host upload" depends on using the Uploader, which needs a "ConduitEngine", not a "ConduitClient". This workflow may use asymmetric key signing, which "ConduitEngine" does not support.
To unblock PHI1735, provide glue code between "Client" and "Engine". But a "more correct" change is probably removal of "Engine".
Test Plan:
- Ran `bin/host upload`, uploaded files (with additional changes to wrap the Client).
- Created this revision.
Differential Revision: https://secure.phabricator.com/D21260
Summary: Fixes T13533. This is a narrow, fragile API for a particular Kerberos use case on one install.
Test Plan:
- Set a non-scalar key, got an exception.
- Set <"duck", "quack">, got an exception from cURL that the value was invalid.
- Set a bunch of made-up options to arbitrary values, no errors. cURL accepts anything so there's nothing we can do about this.
- Set `CURLOPT_NOBODY` and saw the request behavior change, demonstrating that the call can produce effects.
Maniphest Tasks: T13533
Differential Revision: https://secure.phabricator.com/D21251
Summary: Ref T13520. In unusual cases where there are no changes in a changeset list (e.g., empty commits) we can fatal when trying to iterate over an empty list of vectors.
Test Plan:
- Created an empty commit.
- Used "git show | pbcopy" to create a diff from it.
- Viewed it in the web UI.
- Before: fatal when iterating on `null`.
- After: clean page.
Maniphest Tasks: T13520
Differential Revision: https://secure.phabricator.com/D21221
Summary:
Ref T13528. For consistency with other commands ("arc upload", "arc diff"), support a "--browse" flag to "arc paste".
Support "--input" as a more robust alternative to `x | y` (see T6996).
Test Plan: Ran `arc paste --browse --input X`, got a new paste in a browser window. Ran other variations of flags and parameters.
Maniphest Tasks: T13528
Differential Revision: https://secure.phabricator.com/D21203
Summary:
Ref T13528. Provide a "--browse" flag to open files after they are uploaded.
Update the code to use modern query strategies.
This impacts the output of "--json", which was just raw "file.info" output before and could not represent the same path being passed several times (`arc upload X X`).
Test Plan: Ran `arc upload` with `--browse` and `--json`.
Maniphest Tasks: T13528
Differential Revision: https://secure.phabricator.com/D21202
Summary:
Ref T13528. If we start a subprocess that immediately exits and then write to it, we can get a broken pipe error.
Recent versions of PHP appear to raise this as an actual warning, and recent changes upgrade the warning to a runtime exception.
I can't find any way to tell if the RuntimeException is a broken pipe or something else, except by examining the text of the error string.
At least for now, treat this like a "blocked pipe" condition. Since the subprocess has exited and the bytes didn't write, this should generally be reasonable.
Test Plan:
- Viewed a file in Paste with an extension that Pygments does not have a lexer for.
- This causes Pygments to exit immediately with an "unrecognized lexer" error. This closes the pipe, and the next write will fail with a broken pipe error.
- Before patch: fatal on broken pipe.
- After patch: clean resolution of the future and error condition.
Maniphest Tasks: T13528
Differential Revision: https://secure.phabricator.com/D21199
Summary:
Ref T13528. When you call `$future->resolve()`, we currently guarantee it is resolved by calling `FutureIterator->resolveAll()`.
`resolveAll()` does not actually "resolve()" futures: it guarantees that they are ready to "resolve()", but does not actually call "resolve()".
In particular, this means it does not throw exceptions.
This can lead to a case where a Future has "resolve()" called directly (e.g., via a FutureProxy), uses "FutureIterator" to resolve itself, throws an exception inside "FutureIterator", the exception is captured and attached to the Futuer, then the outer future tries to access results. This fails since it's out-of-order.
This can happen in practice with syntax highlighting futures, which may proxy pygments futures.
Instead, "resolveAll()" before testing for exaceptions.
Test Plan:
- Locally, tried to highlight a Paste with an unrecognized lexer extension using Pygments.
- Before patch: fatal when trying to access results of a Future with no results (because it has an exception instead).
- After patch: resolution throws the held exception properly.
- (See also next change.)
Maniphest Tasks: T13528
Differential Revision: https://secure.phabricator.com/D21198
Summary: Fixes T13527. Some versions of PHP strictly require that we pass a string value, and reject "stringlike" objects (objects which implement "__toString()").
Test Plan: Ran unit test, although this is somewhat aspirational because my local PHP version isn't affected.
Maniphest Tasks: T13527
Differential Revision: https://secure.phabricator.com/D21193
Summary:
See D21190. The ".gitattributes" approach fails when ".gitattributes" is in a subdirectory (or global). These are probably unusual cases, but at least one is known in the wild.
Instead:
- Restore the ":(attr:filter=lfs)" test, which seems to be the fastest accurate test available in modern Git.
- If the test fails, assume the repository is not LFS. This only impacts users running very old versions of Git.
Test Plan:
- In LFS and non-LFS repositories, created diffs. Saw correct detection again.
- Broke the command on purpose, saw LFS detection conclude "no LFS", but not fail disastrously.
Subscribers: ptarjan
Differential Revision: https://secure.phabricator.com/D21192
Summary:
See PHI1718. See also <https://discourse.phabricator-community.org/t/arc-diff-fails-due-to-git-cmd-fails/3680/>.
Currently, `arc diff` detects Git LFS with `git ls-files -z -- ':(attr:filter=lfs)'` magic. This is an accurate test, but does not work on older Git.
Try a simpler, dumber test and see if that will work. If this also has issues, we can try this stuff:
- do version detection;
- pipe the whole tree to `git check-attr`;
- try a command like `git lfs ls-files` instead, which is probably a wrapper on one of these other commands.
Test Plan:
- In a non-LFS repository, ran "arc diff" and saw the repository detect as non-LFS.
- In an LFS repository, ran "arc diff" and saw the repository detect as LFS.
Differential Revision: https://secure.phabricator.com/D21190
Summary:
Ref T13520. Replace "FileTree" with a "VectorTree" that does roughly the same thing. The major goals are:
- Compress trees which contain sequences of child directories with no sibilings.
- Build hierarchies of paths where path components may include renames.
This is approximately similar to "FileTree" and similar to client logic in the new paths panel.
Test Plan: See next change.
Maniphest Tasks: T13520
Differential Revision: https://secure.phabricator.com/D21182
Summary:
Ref T13518. The result format of this call changed in PHP 7.4, which causes us to emit "-1" matches because "-1" survives `array_filter()`.
Filter results in a way that should survive both result formats.
Test Plan: Ran `arc unit --everything` under PHP 7.4.
Maniphest Tasks: T13518
Differential Revision: https://secure.phabricator.com/D21173
Summary: Ref T13507. In Phabricator, we perform a specific "Accept-Encoding: gzip" setup test and want to manually decode the result. Allow callers to disable handling of "Content-Encoding".
Test Plan: Ran all Phabricator setup checks.
Maniphest Tasks: T13507
Differential Revision: https://secure.phabricator.com/D21121
Summary:
Ref T13507. For various messy reasons we can't blindly assume the server supports "gzip" -- but if the server tells us it does, we're on firmer ground.
If the server returns an "X-Conduit-Capabilities: gzip" header and we have compression support locally, compress subsequent requests.
This restores D21073, which was reverted by D21076.
Test Plan: With a gzip-asserting server, added debugging code and ran various "arc" commands. Saw the 2nd..Nth calls hit compression code.
Maniphest Tasks: T13507
Differential Revision: https://secure.phabricator.com/D21119
Summary:
Ref T13507. Add "Accept-Encoding: gzip" to requests if we can decompress responses.
When we receive a compressed response, decompress it.
Test Plan: Added debugging code, ran some commands, saw smaller payloads over the wire and inline decompression.
Maniphest Tasks: T13507
Differential Revision: https://secure.phabricator.com/D21118
Summary:
Ref T13507. Currently, this function is a bit conservative about what it encodes, and passing it a string of binary garbage may result in an output which is not valid UTF8.
This could be refined somewhat, since it's less than ideal if the input has valid UTF8. The ideal behavior for byte sequences where all bytes are larger than 0x7F is probably a variation of "phutil_utf8ize()" that replaces bytes with "<0xXX>" instead of the Unicode error glyph.
For now, just err on the side of mangling.
Test Plan: Dumped various binary payloads in the new gzip setup check, saw sensible output in the web UI.
Maniphest Tasks: T13507
Differential Revision: https://secure.phabricator.com/D21117
Summary: Ref T13490. This code is reachable from Phabricator binaries; only inject the legacy stuff if we're in an Arcanist stack.
Test Plan: Ran `bin/conduit help` from `phabricator/`.
Maniphest Tasks: T13490
Differential Revision: https://secure.phabricator.com/D21113
Summary: Ref T13490. More-or-less straightforward upgrade to modern calls.
Test Plan: Created and viewed pastes.
Maniphest Tasks: T13490
Differential Revision: https://secure.phabricator.com/D21104
Summary:
Ref T13490. One of the biggest issues users are hitting in modern "arc" is that workflows don't appear in "arc help" until they're updated.
Since there's still some work to do and gluing them in isn't terribly difficult, at least get things connected for now.
Test Plan: Ran "arc help", "arc help diff".
Maniphest Tasks: T13490
Differential Revision: https://secure.phabricator.com/D21101
Summary: Ref T13490. These workflows don't seem worth the maintenance cost, see T13488 for discussion.
Test Plan: Grepped for methods which looked like they might only be called by these flows, only dug up the backout stuff.
Maniphest Tasks: T13490
Differential Revision: https://secure.phabricator.com/D21100
Summary: Ref T13490. The "RevisionRef" is currently based on "differential.query" data, but all calls other than the hash-based lookup can move to "differential.revision.search".
Test Plan: Ran revision workflows like `arc inspect` and `arc browse`.
Maniphest Tasks: T13490
Differential Revision: https://secure.phabricator.com/D21099
