Summary: Ref T8099. I think these colors make sense, but if any seem wrong, lmk. Colors the type of file change.
Test Plan: Test a few diffs locally, read carefully, ask epriestley.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: epriestley, Korvin
Maniphest Tasks: T8099
Differential Revision: https://secure.phabricator.com/D13502
Summary: Ref T8650. This should stop the problem, but isn't a root cause fix. See discussion on the task.
Test Plan: Made some local diffs, but this is a bit hard to reproduce reliably.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8650
Differential Revision: https://secure.phabricator.com/D13441
Summary:
Ref T8096. Various tweaks here:
- Sort result lists by importance (even lint -- "errors first" seems better than "alphabetical by file", I think?).
- Do sane stuff with display limits.
- Add a "view all" view.
- Don't show a huge table of passing tests in Differential.
- Link to full results.
Test Plan: See screenshots.
Reviewers: chad, btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8096
Differential Revision: https://secure.phabricator.com/D13407
Summary: Ref T8096. No functional changes, just a bit less code.
Test Plan: Viewed some revisions, saw the same stuff as before.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8096
Differential Revision: https://secure.phabricator.com/D13404
Summary: Fixes T8095. Still needs UI/UX work (see T8096) but this has all the core features now.
Test Plan: Saw Harbormaster lint/unit data as though it was Differential lint-unit data.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8095
Differential Revision: https://secure.phabricator.com/D13401
Summary:
Ref T8096. Fixes a few bugs and glitches.
- Set build completion time when handling a message.
- Format duration information in a more human-readable way.
- Use a table for build variables.
- Fix up container PHIDs on diffs (a touch hacky, should be OK for now though).
Test Plan: Browsed around the UI.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8096
Differential Revision: https://secure.phabricator.com/D13382
Summary:
Ref T8095. When build results are reported for a target, allow them to include unit and lint results.
There is no real way to see this stuff in the UI yet, either in Harbormaster or Differential.
Test Plan: Manually called this method with some results, saw Harbormaster update appropriately.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8095
Differential Revision: https://secure.phabricator.com/D13380
Summary: Ref T8095. This weird grey table has no remaining callsites and can be removed.
Test Plan: Grepped for symbols.
Reviewers: chad, btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8095
Differential Revision: https://secure.phabricator.com/D13379
Summary:
Ref T8095. Same as D13377, but for unit results.
This is a bit rough and there's some duplication between this and unit results. I'll likely merge them later, but I think some of it is superficial since these iterations are still a little crude.
Test Plan: {F523499}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8095
Differential Revision: https://secure.phabricator.com/D13378
Summary:
Ref T8095. Render lint results in a future-ready way.
This makes the renderer accept `HarbormasterBuildLintMessage` objects. If we have legacy data instead, it converts it into `HarbormasterBuildLintMessage` objects.
Design is a bit rough but will be cleaned up later after T7739.
This moves away from "postponed linters", which are obsolete after Harbormaster (and were only ever used by Facebook).
Test Plan: {F523429}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8095
Differential Revision: https://secure.phabricator.com/D13377
Summary: Ref T8099, This adds a consistent background color to object and policy tags, and highlights them when they deviate from the normal. Still likely worth revamping 'closed' and 'review' state colors.
Test Plan: Review lots of diffs and tasks.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: epriestley, Korvin
Maniphest Tasks: T8099
Differential Revision: https://secure.phabricator.com/D13399
Summary:
Fixes T6787. I'm kind of cheating a little bit here by not unifying default selection with `initializeNew(...)` methods, but I figure we can let this settle for a bit and then go do that later. It's pretty minor.
Since we're not doing templates I kind of want to swap the `'template'` key to `'type'` so maybe I'll do that too at some point.
@chad, freel free to change these, I was just trying to make them pretty obvious. I //do// think it's good for them to stand out, but my approach is probably a bit inconsistent/heavy-handed in the new design.
Test Plan:
{F525024}
{F525025}
{F525026}
{F525027}
Reviewers: btrahan, chad
Reviewed By: btrahan
Subscribers: johnny-bit, joshuaspence, chad, epriestley
Maniphest Tasks: T6787
Differential Revision: https://secure.phabricator.com/D13387
Summary:
Ref T8095. Two general problems:
- I want Harbormaster to own all lint and unit test results.
- I don't want users to have to configure anything for `arc` to keep working automatically.
These are in conflict because generic lint/unit test ownership in Harbormaster requires that build targets exist which we can attach build results to. However, we can't currently create build targets on demand: Harbormaster assumes it is responsible for creating targets, then running code or making third-party service calls to actually run the builds.
I considered two broad approaches to let `arc` push results into Harbormaster without requiring administrators to configure some kind of "arc results" build plan:
# Add magic target PHIDs like `PHID-MAGIC-this-is-really-arc-unit`.
# Add new code to build real targets with real PHIDs.
(1) is probably a bit less work to get off the ground, but I think it's worse overall and very likely to create more problems in the long run. I particularly worry that it will lead to a small amount of special casing in a very large number of places, which seems more fragile.
(2) is more work upfront but I think does a better job of putting all the special casing in one place that we can, e.g., more reasonably unit test, and letting the rest of the code rarely/never care about this case since it's just dealing with normal plans/steps/targets as far as it can tell.
This diff introduces "autoplans", which are source templates for plans/steps. This let us "push" these targets into Harbormaster. Hypthetically, any process "like" arc can use autoplans to upload test/lint/etc results. In practice, probably only `arc` will ever use this, but I think it's still quite a bit cleaner than the alternative despite all the generality.
Workflow is basically:
- `arc` creates a diff.
- `arc` calls `harbormaster.queryautotargets`, passing the diff PHID and saying "I have some lint and unit results I want to stick on this thing".
- Harbormaster builds the plan, steps, and targets (if any of them don't already exist), and hands back the target PHIDs so `arc` has a completely standard-looking place to put results.
- `arc` uploads the test results to the right targets, as though Harbormaster had asked it to run unit/lint in the first place.
(This doesn't actually do any of that yet, just sets things up.)
I'll maybe doc turn that ^^^^^^ into a doc for posterity since I think it's hard to guess what an "autotarget" is, but I'm going to grab some lunch first.
Test Plan:
- Added unit tests to make sure we can build these things properly.
- Used `harbormaster.queryautotargets` to build autotargets for a bunch of diffs.
- Verified targets come up in "waiting for message" state.
- Verified plans and steps are not editable.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: hach-que, epriestley
Maniphest Tasks: T8095
Differential Revision: https://secure.phabricator.com/D13345
Summary: Ref T8099, Now that we have cleaner headers, we can add more pop to important items, like the state of an Object. This makes it easier to just note the color and generally understand the state of the object (closed, returned, accepted, open). @epriestley, I think you previously thought this was a bug, but if it still feels bad, let me know.
Test Plan:
Review Differential, Maniphest, Polls for various object states.
{F520973}
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T8099
Differential Revision: https://secure.phabricator.com/D13369
Summary: Ref T8099. This adds a new class which all search engines return for layout. I thought about this a number of ways, and I think this is the cleanest path. Each Engine can return whatever UI bits they needs, and AppSearch or Dashboard picks and lays the bits out as needed. In the AppSearch case, interfaces like Notifications, Calendar, Legalpad all need more custom layouts. I think this also leaves a resonable path forward for NUX as well. Also, not sure I implemented the class correctly, but assume thats easy to fix?
Test Plan: Review and do a search in each application changed. Grep for all call sites.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: epriestley, Korvin
Maniphest Tasks: T8099
Differential Revision: https://secure.phabricator.com/D13332
Summary:
Ref T8575. We run a big "(A) UNION (B)" query on the home page and on the main Differential page.
"A" can always be improved by using `%Ls`, so it can use the second half of the `(authorPHID, status)` key.
"B" can sometimes be improved if the fraction of open revisions is smaller than the fraction of revisions you are reviewing. This is true for me on secure.phabricator.com (I'm a reviewer, either directly or via 'Blessed Reviewers', on about 80% of revisions, but <5% are open). In these cases, a `(status, phid)` key is more efficient.
Test Plan: Tweaked queries and added keys on this server, saw less derpy query plans and performance.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8575
Differential Revision: https://secure.phabricator.com/D13325
Summary:
Fixes T5138. Some of the "revision" properties are really "diff" properties, but we only show the properties for the most recent / current diff.
- Immediately, this makes it hard or impossible to review, e.g., lint/unit results for older diffs.
- Longer-term, these limits will become more problematic with more data on diffs after Harbormaster.
Instead, separate "revision" from "diff" properties.
(In the long term, it might make sense to show more diffs in this panel -- e.g., tabs for the 8 most recent updates or something -- but I went with the simplest approach for now since I don't have a clean way to deal with 100-update revisions offhand.)
Test Plan: {F500480}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: cburroughs, epriestley
Maniphest Tasks: T5138
Differential Revision: https://secure.phabricator.com/D13282
Summary: All classes should extend from some other class. See D13275 for some explanation.
Test Plan: `arc unit`
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: epriestley, Korvin
Differential Revision: https://secure.phabricator.com/D13283
Summary: This `break` statement causes `./bin/hunks migrate` to only migrate one-hunk-at-a-time, which is unnecessary and slightly misleading. Instead, allow the script to migrate //all// legacy hunks to modern storage. In particular, this means that we can recommend that installs run this command sometime before D13222 is landed.
Test Plan: It's a pain to setup the data necessary to test this, but this is identical to the change that I made on our production install when I migrated our hunk storage.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: epriestley, Korvin
Differential Revision: https://secure.phabricator.com/D13288
Summary:
Ref T5681. Ref T6860. This doesn't do anything interesting on its own, just makes the next diff smaller.
In the next diff, policies become aware of the types of objects they're acting on. We need to specify which object type all the "Default View/Edit" settings are for so they get the right rules.
For example, a rule like "Allow task author" is OK for "View Policy" on a task, and also OK for "Default View Policy" on ManiphestApplication. But it's not OK for "Can Create Tasks" on ManiphestApplication.
So annotate all the "template"/"default" policies with their types. The next diff will use these to let you select appropriate rules for the given object type.
Test Plan:
- Used `grep` to find these.
- This change has no effect.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T5681, T6860
Differential Revision: https://secure.phabricator.com/D13251
Summary:
Ref T8455. Use standard effects for revisions, instead of a custom effect.
This fixes the major issue (conduit error) in T8455 because the standard effect now performs PHID type filtering.
This retains other behaviors (in particular: not re-CC'ing explicitly removed CCs).
Test Plan:
- With a Herald rule that adds a mailing list as a CC, created a revision before the change and hit the error in T8455. After the change, saw correct behavior.
- Wrote a normal Herald rule to add CCs and created a revision, saw it fire properly.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8455
Differential Revision: https://secure.phabricator.com/D13183
Summary:
Ref T8455. The Herald code in general isn't nearly as modular as it should be, and the subscriber code particularly has some legacy cruft. This is making it fragile and causing the issue described in T8455.
Currently, each Herald adapter has essentially identical code which it uses to determine which users are subscribed to an object. Instead, share code between object types.
I removed "explicitCCs":
- The value was always identical to doing the query in the common/standard way.
- They were only used to print a diagnostic message on transcripts, which I think is no longer relevant.
- I believe it predates transactions, so when it was added you couldn't figure out the old object state by looking at the transaction history. Now, CC changes are recorded there, so there's no need to restate the CC state on the transcript.
- Even if we do want to restore this (or something similar), we can do it directly from Herald now.
Test Plan:
- Created rules that use the "CCs" field in Herald, Pholio, Maniphest and Differential.
- Updated objects in each application.
- Observed valid field reads in the tranascript.
- Grepped for `FIELD_CC`.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8455
Differential Revision: https://secure.phabricator.com/D13177
Summary: Ref T8463.
Test Plan:
- Created a new revision via web UI with a username `@mention` in the summary and no repository.
- Prior to patch, hit a "not attached" error.
- After patch, no error.
- Created a new web UI revision, as above, but with a repository; saw repository work fine.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8463
Differential Revision: https://secure.phabricator.com/D13205
Summary: Remove the `*TransactionType` classes and define the constants in the corresponding `*Transaction` class instead.
Test Plan: `grep`
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D13188
Fixes this, which only triggers on some kinds of mail:
```
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] [2015-06-04 02:56:38] EXCEPTION: (PhutilProxyException) Error while executing Task ID 902251. {>} (PhabricatorDataNotAttachedException) Attempting to access attached data on DifferentialRevision (via getActiveDiff()), but the data is not actually attached. Before accessing attachable data on an object, you must load and attach it.
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] Data is normally attached by calling the corresponding needX() method on the Query class when the object is loaded. You can also call the corresponding attachX() method explicitly. at [<phabricator>/src/infrastructure/storage/lisk/PhabricatorLiskDAO.php:166]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] arcanist(head=master, ref.master=8c589f1f759f), phabricator(head=master, ref.master=6dede2e2c513), phutil(head=master, ref.master=afc05a9a7f00)
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #0 <#2> PhabricatorLiskDAO::assertAttached(string) called at [<phabricator>/src/applications/differential/storage/DifferentialRevision.php:158]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #1 <#2> DifferentialRevision::getActiveDiff() called at [<phabricator>/src/applications/differential/customfield/DifferentialBranchField.php:72]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #2 <#2> DifferentialBranchField::updateTransactionMailBody(PhabricatorMetaMTAMailBody, DifferentialTransactionEditor, array) called at [<phabricator>/src/applications/transactions/editor/PhabricatorApplicationTransactionEditor.php:2481]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #3 <#2> PhabricatorApplicationTransactionEditor::addCustomFieldsToMailBody(PhabricatorMetaMTAMailBody, DifferentialRevision, array) called at [<phabricator>/src/applications/differential/editor/DifferentialTransactionEditor.php:1208]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #4 <#2> DifferentialTransactionEditor::buildMailBody(DifferentialRevision, array) called at [<phabricator>/src/applications/transactions/editor/PhabricatorApplicationTransactionEditor.php:2178]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #5 <#2> PhabricatorApplicationTransactionEditor::sendMailToTarget(DifferentialRevision, array, PhabricatorMailTarget) called at [<phabricator>/src/applications/transactions/editor/PhabricatorApplicationTransactionEditor.php:2152]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #6 <#2> PhabricatorApplicationTransactionEditor::sendMail(DifferentialRevision, array) called at [<phabricator>/src/applications/transactions/editor/PhabricatorApplicationTransactionEditor.php:998]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #7 <#2> PhabricatorApplicationTransactionEditor::publishTransactions(DifferentialRevision, array) called at [<phabricator>/src/applications/transactions/worker/PhabricatorApplicationTransactionPublishWorker.php:21]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #8 <#2> PhabricatorApplicationTransactionPublishWorker::doWork() called at [<phabricator>/src/infrastructure/daemon/workers/PhabricatorWorker.php:91]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #9 <#2> PhabricatorWorker::executeTask() called at [<phabricator>/src/infrastructure/daemon/workers/storage/PhabricatorWorkerActiveTask.php:162]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #10 <#2> PhabricatorWorkerActiveTask::executeTask() called at [<phabricator>/src/infrastructure/daemon/workers/PhabricatorTaskmasterDaemon.php:20]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #11 PhabricatorTaskmasterDaemon::run() called at [<phutil>/src/daemon/PhutilDaemon.php:185]
Daemon 39819 STDE [Thu, 04 Jun 2015 02:56:38 +0000] #12 PhutilDaemon::execute() called at [<phutil>/scripts/daemon/exec/exec_daemon.php:125]
```
Auditors: btrahan
Summary:
Ref T6367. Do all mail, feed, notification and search stuff from the daemons, in all editors.
There are four relatively-stateful editors (Audit, Differential, Phriction, PhortuneCart) which needed special care to move state into the daemons properly.
Beyond that, I moved mailTo/mailCC/feedRelated/feedNotify to be computed before we enter the worker:
- This is simpler, since a lot of editors rely on being able to call `$object->getReviewers()` or similar to compute them.
- This is more correct, since we want to freeze the lists at this moment in time.
Finally, I renamed `loadEdges` to `willPublish` and made it a slightly more general hook.
---
This is a bit fragile and I'm not //thrilled// about it.
It would probably be cleaner to have separate Editor and Publisher classes (something like @fabe's D11329 did). However, I think that's quite a lot of work, and I'd like to see stronger motivation for it (either in this actually being more fragile than I think, or there being other things we get out of it). Overall, I'm comfortable with this change, just definitely not a big fan of the "save" + "load" pattern since I think it's really fragile, nonobvious, hard to debug/predict, etc.
Test Plan:
Directly updated editors:
- Created a new Phriction page, saw "Document Content".
- Edited a Phriction page, saw "Document Diff".
- Edited a revision, got normal looking mail.
- Faked in `changedPriorToCommitURI` and verified it survived the state boundary.
- Sent Audit mail.
- Sent invoice mail.
Indirect editors - for these, I just made a change and made sure the mail generated:
- Updated a paste.
- Updated an event.
- Updated a thread.
- Updated a task.
- Updated a mock.
- Updated a question.
- Updated a project.
- Updated a file.
- Updated an initiative.
- Updated a Legalpad document.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley, fabe
Maniphest Tasks: T6367
Differential Revision: https://secure.phabricator.com/D13115
Summary:
Ref T7703. See that task and inline for a bunch of discussion.
Briefly, when we run implicit policy rules ("to see a revision, you must also be able to see its repository") at query time, they don't apply to other viewers we might check later.
We do this very rarely, but when we do we're often doing it for a bunch of different viewers (for example, in Herald) so I don't want to just reload the object a million times.
Test Plan:
- Added and executed unit tests.
- Wrote a "flag everything" Herald rule, as in the original report in T7703, and no longer got "Unknown Object" flags on revisions.
- Rigged up a lot of cases in the web UI and couldn't find any inconsistencies, although this case is normally very hard to hit.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7703
Differential Revision: https://secure.phabricator.com/D13104
Summary: Ref T8387. This is now completely obsoleted by mailing list users.
Test Plan: Grepped for `mailinglist` and related symbols.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: eadler, epriestley
Maniphest Tasks: T8387
Differential Revision: https://secure.phabricator.com/D13129
Summary: Ref T7604. Remove these "arcanist project" fields from the `LiskDAO` classes, but leave the data intact (there are no more read/writes to this data).
Test Plan:
- Grepped for calls to these methods.
- Ran `./bin/storage adjust` to check for schema issues.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7604
Differential Revision: https://secure.phabricator.com/D13012
Summary: Ref T8099, No specific reason for these to be small buttons.
Test Plan: Test desktop and mobile layouts.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T8099
Differential Revision: https://secure.phabricator.com/D13081
Summary: These could cause F442758.
Test Plan: {F442779}
Reviewers: btrahan, joshuaspence
Reviewed By: joshuaspence
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D13042
Summary:
Ref T7447. Implements per-viewer comment hiding. Once a comment is obsolete or uninteresting, you can hide it completely.
This is sticky per-user.
My hope is that this will strike a better balance between concerns than some of the other approaches (conservative porting, summarization, hide-all).
Specifically, this adds a new action here:
{F435621}
Clicking it completely collapses the comment into a small icon on the previous line, and saves the comment state as hidden for you:
{F435626}
You can click the icon to reveal all hidden comments below the line.
Test Plan:
- Hid comments.
- Showed comments.
- Created, edited, deleted and submitted comments.
- Used Diffusion comments (hiding is not implemented there yet, but I'd plan to bring it there eventually if it works out in Differential).
Reviewers: btrahan, chad
Reviewed By: btrahan
Subscribers: jparise, yelirekim, epriestley
Maniphest Tasks: T7447
Differential Revision: https://secure.phabricator.com/D13009
Summary:
We don't want to support this right now, so disable it, similar to "Land To Hosted" feature.
Keep to the code as an example for advanced installs.
Ref T182, T8313
Test Plan: load a diff.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley
Maniphest Tasks: T182, T8313
Differential Revision: https://secure.phabricator.com/D13022
Summary: These format strings use `%d` instead of `%s`.
Test Plan: Eyeball it.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D12996
