Summary:
Ref T7447. Ref T7870. When you "reply" to a ghost inline, make it work properly.
This exact behavior is arguable. In particular, when you reply to a ghost inline, we //could// put the reply on the same diff as the original.
I suspect it aligns better with user exepectation to put the new inline on the current (visible) diff instead, and generally for inlines to flow forward through time and all of the ghosts to pretty much be older than all of the non-ghosts in most cases. We can see how it feels and adjust things if this turns out to not make sense.
Test Plan:
- Replied to ghost inlines, got new inlines on the proper display line.
- Replied to normal inlines, got normal behavior.
- Made some new inlines.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7870, T7447
Differential Revision: https://secure.phabricator.com/D12492
Summary:
Ref T7447. This might be overkill, but I want to over-explain things until we have more confidence that this is rarely confusing.
NOTE: I'm playing it a bit fast and loose with `setIsGhost()` (passing a dictionary) because making API changes requires changing the interface and Diffusion, which is a pain. I'll clean this up at the end once the interface is more final. This is well-contained for now.
Test Plan:
- Viewed "base vs 2" in a diff with 3 diffs, saw some "older comments" and some "newer comments".
- Hovered the tags for an explanation of comment spookiness.
{F377703}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7447
Differential Revision: https://secure.phabricator.com/D12490
Summary:
Ref T7447. This ports comments forward and backward in the best case:
- The old comment is on a changeset with the same filename.
- The old and new files are pretty much the same, line-for-line.
This will fail to port a lot of comments around and probably port a lot of comments into goofy places. We can see how bad it is in practice.
Errata:
- Design is me cobbling something together, probably worth tweaking.
- "Old Comment" should, at a minimum, say "Newer Comment" sometimes, or we should come up with some better name for this stuff.
Test Plan: {F377214}
Reviewers: chad, btrahan
Reviewed By: btrahan
Subscribers: johnny-bit, yelirekim, epriestley
Maniphest Tasks: T7447
Differential Revision: https://secure.phabricator.com/D12484
Summary:
Ref T4100. Ref T5595. This implements these fields in one mega-field:
- Projects
- Not in projects
- In any project
- Include results in no projects
- In users' projects
Hopefully, this is a step in the right direction.
Test Plan: {F375555}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: joshuaspence, chad, epriestley
Maniphest Tasks: T4100, T5595
Differential Revision: https://secure.phabricator.com/D12463
Summary:
Ref T4100. Ref T5595. This allows PolicyAwareQuery to write all the logic for AND, OR, NOT, and NULL (i.e., "not in any projects") queries against any edge type.
It accepts an edge type and a list of constraints (which are basically just operator-value pairs, like `<NOT, PHID-X-Y>`, meaning the results must not have an edge connecting them to `PHID-X-Y`).
This doesn't actually do anything yet; see future diffs.
Test Plan: `arc unit --everything`
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T4100, T5595
Differential Revision: https://secure.phabricator.com/D12455
Summary:
Ref T4100. Ref T5595. These functions are trivial for now, but move us toward being able to define more default query behavior by default.
Future changes will give these methods meaningful, nontrivial behaviors.
Test Plan: `arc unit --everything`
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T5595, T4100
Differential Revision: https://secure.phabricator.com/D12454
Summary:
Ref T4100. Ref T5595.
To support a unified "Projects:" query across all applications, a future diff is going to add a set of "Edge Logic" capabilities to `PolicyAwareQuery` which write the required SELECT, JOIN, WHERE, HAVING and GROUP clauses for you.
With the addition of "Edge Logic", we'll have three systems which may need to build components of query claues: ordering/paging, customfields/applicationsearch, and edge logic.
For most clauses, queries don't currently call into the parent explicitly to get default components. I want to move more query construction logic up the class tree so it can be shared.
For most methods, this isn't a problem, but many subclasses define a `buildWhereClause()`. Make all such definitions protected and consistent.
This causes no behavioral changes.
Test Plan: Ran `arc unit --everything`, which does a pretty through job of verifying this statically.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: yelirekim, hach-que, epriestley
Maniphest Tasks: T4100, T5595
Differential Revision: https://secure.phabricator.com/D12453
Summary: Ref T5750. Update the Almanac service query to be browsable.
Test Plan:
- Browsed and reordered Diffusion.
- Browsed and reordered services in Almanac.
{F373735}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T5750
Differential Revision: https://secure.phabricator.com/D12433
Summary:
Ref T7803. Prior to this change sequence, Query classes conflated paging values (the actual thing that goes in a "x > 3" clause) with cursor values (arbitrary identifiers which track where the user is in a result list).
Although the two can sometimes be the same, the vast majority of implementations are simpler and better when object IDs are used as cursors and paging values are derived from them.
The new stuff handles this in a consistent way, so we're free to separate getPagingValue() from paging. The new method is essentially getResultCursor().
This also implements getPageCursors(), which allows queries to return directional cursors. The inability to do this was a practical limitation blocking the implementation of T7803.
Test Plan:
- Browsed a bunch of results and paged through queries.
- Grepped for removed methods.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7803
Differential Revision: https://secure.phabricator.com/D12383
Summary:
Ref T7803. Currently, available high-level orders are spread across Query and SearchEngine classes and implemented separately for each application.
Lift the concept of "builtin" (high-level, user-facing, named) orders (similar to "builtin" queries in ApplicationSearch) into the root Query class, and let it drive the SearchEngine implementation. This allows you to define a new order in one place and have it automatically work across the entire stack.
This will also let Conduit expose this information in a straightforward way.
Test Plan:
- Used ApplicationSearch in Diffusion.
- Used all result orderings.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7803
Differential Revision: https://secure.phabricator.com/D12379
Summary: Ref T7803. Remove these in favor of more generalized paging and ordering.
Test Plan: Sorted and paged results in various applications.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7803
Differential Revision: https://secure.phabricator.com/D12378
Summary:
Ref T7803. The ApplicationSearch integration is still a little rough here, but it seems to have the correct behavior.
The rest of this is now at least relatively sane, cohesive, and properly behaved.
Test Plan:
- Used all grouping and ordering queries in Maniphest. Pagingated results.
- Used custom field ordering in Maniphest. Paginated results.
- Paginated through the `null` section of "Assigned" and "Projects" group-by queries. Pagingation now works correctly (it does not work at HEAD).
- Ran unit tests covering priority changes.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7803
Differential Revision: https://secure.phabricator.com/D12372
Summary:
Ref T7803. Removes some getReversePaging().
This also fixes `null` column handling, by adding an explicit `'null'` key with possible values "head" (put NULL before other values) or "tail" (put NULL after other values).
Maniphest has some glitchiness in paging through NULLs right now, but I believe it's all pre-existing and will be resolved when it fully converts. Diffusion is fully converted and pages through NULL correctly.
Test Plan:
- Failed to identify any reason for ChangesetQuery to reverse paging.
- Paged thorugh Diffusion.
- Paged through Maniphest.
- Maniphest has some issues when paging inside a NULL section, but these issues are preexisting and will be resolved later in this change sequence.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7803
Differential Revision: https://secure.phabricator.com/D12371
Summary:
Ref T7803. Some Query subclasses implement getPagingColumn() in a trivial way, usually to provide a table alias.
Formalize the concept of a primary table alias, and remove obsoleted getPagingColumn() implementations.
Test Plan: Issued affected queries.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7803
Differential Revision: https://secure.phabricator.com/D12356
Summary:
Ref T7803. Ordering and paging are inherently intertwined, but they often aren't driven by the same data right now.
Start driving them through the same data:
- `getOrderableColumns()` defines orderable and pageable columns.
- `getPagingValueMap()` reads values from a cursor.
This is generally sufficient to implement both paging and ordering.
Also, add some more sanity checks to try to curtail the number of ambiguous/invalid orderings applications produce, since these cause subtle/messy bugs.
Test Plan:
- Paged through pastes and a few other object types.
- Intentionally changed defaults to be invalid and hit some of the errors.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7803
Differential Revision: https://secure.phabricator.com/D12355
Summary: Ref T7803. Reduce the amount of code we're trusting to build SQL queries.
Test Plan:
- Paged through results in Maniphest, Differential and Diffusion.
- Some of the NULLable groups in Maniphest are a bit funky but this was preexisting.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7803
Differential Revision: https://secure.phabricator.com/D12353
Summary:
Ref T7803. Queries currently have a single `getPagingColumn()`, which is oversimplified and insufficient to describe many ordering operations. Frequently, orders must span multiple columns.
Move toward an "order vector", which is a list of orderable values like "name, id". These map directly to columns, and are sufficient to actually describe orders. The more modern Query classes (Maniphest, Repository) essentially do this manually anyway.
Test Plan:
- Added and executed unit tests.
- Browsed around, verified the correct ORDER BY clauses were generated.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7803
Differential Revision: https://secure.phabricator.com/D12352
Summary:
Ref T7803. Instead of trusting subqueries to provide safe values, escape them explicitly.
(We'll probably have a few cases somewhere where this doesn't work, but can make them the exception rather than the rule.)
Test Plan: Issued all "order" queries in Diffusion.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7803
Differential Revision: https://secure.phabricator.com/D12351
Summary: didn't quite get there in D12309
Test Plan: made a revert commit and inspected my feed and it was correct (screenshot shortly)
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D12333
Summary: Add some more translations to make the interface very wow (Thanks to Robert Calaceto)
Test Plan: Squinted my eyes and stared at the UI until stuff made sense.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin, chad, epriestley
Differential Revision: https://secure.phabricator.com/D12312
Summary: I think D11212 missed the feed variants. add 'em.
Test Plan: hope and pray and push to prod and see
Reviewers: joshuaspence, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D12309
Summary: In its current form, this file is not being linted. This doesn't seem to be intentional.
Test Plan: Introduced a linter error and ran `arc lint`.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D12282
Summary: These arrays looks a little odd, most likely due to the autofix applied by `ArcanistXHPASTLinter::LINT_ARRAY_SEPARATOR`. See D12296 in which I attempt to improve the autocorrection from this linter rule.
Test Plan: N/A
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D12281
Summary:
Fixes T7484. There's a bunch of spooky mystery here but the current behavior can probably cause problems in at least some situations.
Also moves a couple callsigns to monograms (see T4245).
Test Plan:
- Faked a short lock length to hit the exception.
- Updated normally.
- Grepped for other use sites, none seemed suspicious or likely to overflow the lock length.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7484
Differential Revision: https://secure.phabricator.com/D12263
Summary: Fixes T7720. We currently leak the "draft" state of checkboxes; never treat checkboxes as drafts if you can't mark them.
Test Plan: Checked a box, reloaded page in other browser. Previously, the draft state partially propagated. Now, it no longer does.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T7720
Differential Revision: https://secure.phabricator.com/D12262
Summary: Added Submit translation for Very Wow Locale
Test Plan: Look at the code and hope it works
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D12254
Summary: Ref T7689. Ref T4100. This advances the goals of removing `loadViewerHandles()` (only 67 callsites remain!) and letting tokenizers some day take token functions like `viewer()` and `members(differential)`.
Test Plan:
- Sent a new message; used "To".
- I simplified the cancel URI construction slightly because it's moot in all normal cases.
- Edited a thread; used "Add Participants".
- Searched rooms; used "Participants".
- Searched countdowns; used "Authors".
- Created a diff; used "Repository".
- Edited a revision; edited "Projects"; edited "Reveiwers"; edited "Subscribers".
- Searched for revisions; edited "responsible users"; "authors"; "reviwers"; "subscribers"; "repositories".
- Added revision comments; edited "Add Reveiwers"; "Add Subscribers".
- Commented on a commit; edited "Add Auditors"; "Add subscribers".
- Edited a commit; edited "Projects".
- Edited a repository; edited "Projects".
- Searched feed, used "include Users"; "include Proejcts".
- Searched files, used "authors".
- Edited initiative; edited "Projects".
- Searched backers; used "Backers".
- Searched initiatives; used "Owners".
- Edited build plans; edited "Run Command".
- Searched Herald; used "Authors".
- Added signature exemption in Legalpad.
- Searhced legalpad; used "creators"; used "contributors".
- Searched signatures; used "documents"; used "signers".
- Created meme.
- Searched macros; used "Authors".
- Used "Projects" in Maniphest reports.
- Used Maniphest comment actions.
- Edited Maniphest tasks; edited "Assigned To"; edited "CC"; edited "projects".
- Used "parent" in Maniphest task creation workflow.
- Searched for projects; used "assigned to"; "in any projec"; "in all projects"; "not in projects"; "in users' projects"; "authors"; "subscribers".
- Edited Maniphest bug filing domains, used "Default Author".
- Searched for OAuth applications, used "Creators".
- Edited Owners pacakge; edited "Primary Owner"; edited "Owners".
- Searched for Owners packages; used "Owner".
- OMG this UI is OLD
- Edited a paste; edited "Projects".
- Searched for paste; used "Authors".
- Searched user activity log; used "Actors"; used "Users".
- Edited a mock; edited "Projects"; edited "CC".
- Searched for mocks; used "Authors".
- Edited Phortune account; edited "Members".
- Edited Phortune merchant account; edited "Members".
- Searched Phrequent; used "Users".
- Edited Ponder question; sued "projects".
- Searched Ponder; used "Authors"; used "Answered By".
- Added project members.
- Searched for projects; used "Members".
- Edited a Releeph product; edited "Pushers".
- Searched pull requests; searched "Requestors".
- Edited an arcanist project; used "Uses Symbols From".
- Searhced push logs; used "Repositories"; used "Pushers".
- Searched repositories; used "In nay project".
- Used global search; used Authors/owners/Subscribers/In Any Project.
- Edited a slowvote; used "Projects".
- Searched slovotes; used "Authors".
- Created a custom "Users" field; edited and searched for it.
- Made a whole lot of typos in this list. ^^^^^^
Did not test:
- Lint is nontrivial to test locally, I'll test it in production.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T4100, T7689
Differential Revision: https://secure.phabricator.com/D12224
Summary:
Ref T1460. Overall:
- Pass `objectOwnerPHID` consistently.
- Pass viewer consistently.
- Set the correct draft state for checkboxes on the client.
Test Plan:
- Made inline comments in Differential.
- Made inline comments in Diffusion.
Reviewers: chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T1460
Differential Revision: https://secure.phabricator.com/D12186
Summary:
Rebuilds the UI in Differential commenting. Specifically we look at the following design patterns:
**To the author:**
- The author of the diff should be able to easily identify what comments are done and not done.
- We keep undone comments yellow
- Clicking done turns comment block into 'unsubmitted state'
**To the reviewer:**
- Easier understanding of unsubmitted states
- All conversations to be yellow/important
**Todo**
- Not all color CSS states correct
- Unpulished checkbox support
Test Plan:
Test leaving comments, published and unpublished. Checking Done, unpublished and published. Check delete states.
From the Diff Author's perspective:
{F352094}
For a Diff commenter's perspective:
{F352095}
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T1460, T7660, T7503
Differential Revision: https://secure.phabricator.com/D12171
Summary:
This returns the PHID of the current revision owner, or the commit author, if one exists.
NOTE: For drafts, we currently return `null`; I'll fix that in a future change. Should be correct for submitted comments.
Test Plan: Added an inline, nothing seemed broken.
Reviewers: chad
Reviewed By: chad
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D12185
Summary: Ref T6755. I'll add some notes there about specifics.
Test Plan:
- Made connections to HTTP and HTTPS URIs.
- Added some debugging code to verify that HTTP URIs were pre-resolved.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6755
Differential Revision: https://secure.phabricator.com/D12169
Summary:
Fixes T7664. When there are a large number of tasks (400+) with the same subpriority (which can happen if the subpriority features are rarely used), it may take more than 30 seconds to rebalance them.
Make the algorithm more aggressive about rebalancing homogenous blocks of tasks.
This may need to get even fancier, but I'd guess it can process blocks 1-2 orders of magnitude larger, which should be ~all installs.
(If someone still hits issues with this, I'll make it fancier.)
Once a block is rebalanced, it doesn't need to be rebalanced again (at least, not as a whole block) so we basically just need to get over the initial hurdle here and then we're good.
In the worst case, we can provide `bin/maniphest rebalance` or similar and do the rebalance step offline.
And, in any case, we have more test coverage here now.
Test Plan:
- Existing tests.
- New tests.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7664
Differential Revision: https://secure.phabricator.com/D12166
Summary: Ref T7660. I'm not toggling "inline-state-is-draft" correctly in JS yet since it's a little tricky (you can reload to see it) but the main state should work.
Test Plan:
- Clicked "done", saw comment opacity fade with placeholder style.
Reviewers: chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T7660
Differential Revision: https://secure.phabricator.com/D12160
Summary:
Ref T1460. When a revision author updates/comments/etc on a revision, publish all their checkmarks.
This doesn't handle Diffusion/audits yet.
Test Plan: {F346870}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: yelirekim, epriestley
Maniphest Tasks: T1460
Differential Revision: https://secure.phabricator.com/D12126
Summary:
Ref T1460. This just barely works, but throwing it up in case any of it sounds mechanically crazy before we build integrations/UI/etc.
Specifically, these are the behaviors:
- You can mark your own draft comments as "done" before you submit them. The intent is to let reviewers mark their stuff advisory/minor/not-important before they submit it, to hint to authors that they don't expect the feedback to necessarily be addressed (maybe it's a joke, maybe it's just discussion, maybe it's "consider..").
- You can mark others' published comments as "done" if you're the revision/commit author. The intent is to keep this lightweight by not requiring an audit trail of who marked what done when. If anyone could mark anything done, we'd have to have some way to show who marked stuff.
- When you mark stuff done (or unmark it), it goes into a "draft" state, where you see the change but others don't see it yet. The intent is twofold:
- Be consistent with how inlines work.
- Allow us to publish a "epriestley updated this revision + epriestley marked 15 inlines as done" story later if we want. This seems more useful than publishing 15 "epriestley marked one thing as done" stories.
- The actual bit where done-ness publishes isn't implemented.
- UI is bare bones.
- No integration with the rest of the UI yet.
Test Plan: Clicked some checkboxes.
Reviewers: chad, btrahan
Reviewed By: btrahan
Subscribers: paulshen, chasemp, epriestley
Maniphest Tasks: T1460
Differential Revision: https://secure.phabricator.com/D12033
Summary:
Ref T6755. This is a partial fix, but:
- Allow netblocks to be blacklisted instead of making the feature all-or-nothing.
- Default to disallow requests to all reserved private/local/special IP blocks. This should generally be a "safe" setting.
- Explain the risks better.
- Improve the errors rasied by Macro when failing.
- Removed `security.allow-outbound-http`, as it is superseded by this setting and is somewhat misleading.
- We still make outbound HTTP requests to OAuth.
- We still make outbound HTTP requests for repositories.
From a technical perspective:
- Separate URIs that are safe to link to or redirect to (basically, not "javascript://") from URIs that are safe to fetch (nothing in a private block).
- Add the default blacklist.
- Be more careful with response data in Macro fetching, and don't let the user see it if it isn't ultimately valid.
Additionally:
- I want to do this check before pulling repositories, but that's enough of a mess that it should go in a separate diff.
- The future implementation of T4190 needs to perform the fetch check.
Test Plan:
- Fetched a valid macro.
- Fetched a non-image, verified it didn't result in a viewable file.
- Fetched a private-ip-space image, got an error.
- Fetched a 404, got a useful-enough error without additional revealing response content (which is usually HTML anyway and not useful).
- Fetched a bad protocol, got an error.
- Linked to a local resource, a phriction page, a valid remote site, all worked.
- Linked to private IP space, which worked fine (we want to let you link and redierect to other private services, just not fetch them).
- Added and executed unit tests.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6755
Differential Revision: https://secure.phabricator.com/D12136
Summary:
Ref T7522. This is mostly useful in the cluster, but could be useful for external installs too.
If you want to import an instance into a test/dry-run state (in the cluster, to test an import; in the general case, to do something like test new hardware or configuration), you currently risk spamming users with a lot of duplicate notifications. In particular, if Phabricator tracks remotes, both instances will continue importing commits and sending email about them. Both instances will try to publish to mirrors, too, which could be bad news, and both instances will try to update linked services.
Instead, provide a flag to let an instance run in "silent mode", which disables all outbound messaging and data.
We need to remember to support this flag on any new outbound channels, but we add about one of those per year so I think that's reasonable.
Test Plan:
- Flipped config.
- Saw it void email, feed and mirroring.
- Didn't test SMS since it's not really in use yet and not convenient to test.
- (Can you think of any publishing I missed?)
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7522
Differential Revision: https://secure.phabricator.com/D12109
Summary:
Ref T7149. When users give us dumpfiles for import, they will almost inevitably use the `phabricator` namespace. They need to be renamed to use an instance namespace.
We can do this either by:
- importing the data first, then renaming; or
- renaming first, then importing.
This implements the second one, basically `storage renamespace --in dump.sql --from phabricator --to instancename > instance.sql`.
Renaming first is a little hackier since we have to `preg_match()` a SQL dump file, but I think it's better overall:
- With only one database, it lets you dump/import without downtime.
- If you have development stuff in a development environment in the `phabricator` namespace, you don't have to move it aside to do an import.
- No possibility that two people doing an import at the same time on the same box will collide with each other.
- You can do the rename once and then repeat the import process with the renamed dump more easily.
- No tricky stuff with modern Phabricator running against an old dump and the database names not matching up.
None of this is super important, but it just makes large dumps a bit easier to work with, and the dumpfile format is regular enough that this seems unlikely to ever really not work.
Test Plan: Renamespaced a dump, did a `diff -u`, saw all the relevant parts changed (and only those parts changed).
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7149
Differential Revision: https://secure.phabricator.com/D12105
Summary:
Ref T7522. This seems like the least-bad approach to a messy issue:
- When backfilling accounts from an imported instance, I need to write ExternalAccount rows to the instance to link instance accounts with upstream accounts.
- We do this in the daemons in some other cases, which lets us run all the code in the context of the instance. However, I really want to do this in-process here because it's way way simpler and we need to do writes to //both// the instance and the upstream, and they're interleaved, and they depend on one another.
- I can hard-code the query with `qsprintf()` but that feels like 100x worse than this.
This allows me to do this:
```
id(new PhabricatorExternalAccount())
->setForcedConnnection($instance_conn)
->...
->save();
```
...and get a write to the instance database, which is at least not completely a minefield.
Test Plan: Backfilled instance accounts and got interleaved instance and upstream writes as expected.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7522
Differential Revision: https://secure.phabricator.com/D12098
Summary:
Ref T2009. Ref T1460.
Fixes T2618. When users hit "Delete" on inline comments, delete immediately and offer them "Undo". If they delete indirectly (e.g., by clicking "Delete" from the preview at the bottom of the page), we still prompt them, because the "Undo" action either won't be available or may not be easy to find. This is a "refdelete".
Fixes T6464. This was just a mess. Make it not as much of a mess. It should work now. Pretty sure.
Fixes T4999. We did not refresh these links often enough to find targets for them, so they could race with content. Reevaluate them after loading new changes.
Test Plan:
- Deleted and undid deletion of inlines from main view and preview.
- Clicked "View" on inlines.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6464, T4999, T2618, T1460, T2009
Differential Revision: https://secure.phabricator.com/D12032
Summary:
Ref T2009. Ref T1460. The way Diffusion and Differential load inlines is horrible garbage right now:
- Differential does an ad-hoc query to get the PHIDs, then does a real load to policy check.
- Diffusion completely fakes things. In practice this is not a policy violation, but it's dangerous.
Make TransactionCommentQuery extensible so we can subclass it and get the query building correctly in the right Query layer.
Specifically, the Diffusion and Differential subclasses of this Query will add appropriate `withX()` methods to let us express the query in SQL.
Test Plan: Loaded, previewed, edited, and submitted inlines in Differential and Diffusion
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T2009, T1460
Differential Revision: https://secure.phabricator.com/D12026
Summary:
Ref T1460. Ref T2618.
When publishing a draft inline, mark the inline it replies to (if any) as replied to.
Also, don't load deleted comments as drafts (sets the stage for T2618).
I'll make an effort to clean up the loading mess here in the next revision, and find some more appropriate home for the shared code.
Test Plan: Made and replied to comments in Differential and Diffusion. Saw comments get marked as "Has Replies" and "Is Reply".
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T2618, T1460
Differential Revision: https://secure.phabricator.com/D12025
Summary: Ref T2009. These subclasses have a mixture of similar methods, move them all to the base class.
Test Plan: Created/edited/undo/submitted comments on the left and right sides of a diff.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T2009
Differential Revision: https://secure.phabricator.com/D12024
Summary:
Ref T2009. This is another almost-identical copy of the row scaffolding, which has the same 1up/2up bugs as the 8 other copies of this code.
Turn the "undo" element into an InlineCommentView so we can scaffold it.
Then, scaffold it with the same code as everything else.
Test Plan: Hit "Undo", swapped from 1up to 2up, hit "undo" again, swapped back, tried left/right, everything rendered with proper scaffolding.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T2009
Differential Revision: https://secure.phabricator.com/D12019
Summary: Ref T2009. This has two more copies of the scaffolding.
Test Plan: Created, edited, deleted, replied to inline comments.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T2009
Differential Revision: https://secure.phabricator.com/D12018
Summary:
Ref T1460. Track and store which comments are threaded replies to other comments, vs merely appearing on the same lines.
This doesn't actually write `hasReplies` yet, since that needs to happen when we un-draft comments on submission.
Test Plan: Made inline comments in Differential and Diffusion, including replies. Replies were marked as "Is Reply".
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1460
Differential Revision: https://secure.phabricator.com/D12017
Summary:
Ref T2009. Currently, the code figures out if a comment is on the left or right by looking at the `<th />` preceeding the enclosing `<td />`.
This gets the right result in 2-up, but in 1-up rows are always `<th />`, `<th />`, `<td />`, so it always detects every inline as being in the new file.
Because "old" and "new" cells aren't inherently distingushable in the 1up view, we can't use a DOM test for this at all. Instead, just track this state explicitly.
Test Plan:
- Made left/right comments in 1up view and 2up view.
- Viewed them in 1up and 2up views.
- Hovered in 1up and 2up views.
- Diff-of-diff'd and reviewed old/new comments, then made some more.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T2009
Differential Revision: https://secure.phabricator.com/D12011
Summary: Ref T2009. Remove the 4 (!!) copies of this code.
Test Plan:
- Added, edited, and removed inline comments in 2up view.
- Stacked a bunch of comments on the same line and saw the JS place them correctly.
- Created an image diff and added, edited and removed inlines on it.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T2009
Differential Revision: https://secure.phabricator.com/D12000
Summary:
Ref T2009. Inline comments have "scaffolding", which is basically some empty table cells/rows around them to get the layout correct.
The scaffolding depends on the renderer, since the cells are different for side-by-side vs unified diffs.
This is currently duplicated all over the place:
- Edit view has 1up/2up.
- Detail view has 1up/2up.
- 1up renderer has 1up.
- 2up renderer has four separate copies of the 2up logic.
These all have subtle differences, which are mostly bugs. Start making the scaffolding more composable so we can get rid of that mess.
Test Plan: Added, edited, and removed inline comments on unified and side-by-side diffs.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T2009
Differential Revision: https://secure.phabricator.com/D11997
Summary:
Ref T2009. These classes are "Differential" now, but are used elsewhere in diff infrastructure (e.g., Diffusion).
- Rename them to "PHUIDiff".
- Move them to "src/infrastructure/".
- Give them a base class.
Test Plan: Interacted with inlines in unified and side-by-side views.
Reviewers: btrahan
Subscribers: epriestley
Maniphest Tasks: T2009
Differential Revision: https://secure.phabricator.com/D11996
Summary: Uses standard sidenav width, more spacing in labels, added background around textarea, make background work in Firefox.
Test Plan:
Test Desktop, Mobile, and Tablet break points. Test Firefox and Chrome.
{F331201}
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11993
Summary:
Ref T2009. This reduces how buggy inlines are. They're still buggy.
Specifically, the inline endpoint didn't know how to scaffold inlines before, so some of them ended up rendering in the wrong rows or breaking layouts.
This passes the current renderer through to the inline editor endpoint, so it can at least get the layout correct.
Test Plan: Interacted with inlines in unified and side-by-side views.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T2009
Differential Revision: https://secure.phabricator.com/D11988
Summary:
Ref T2009. Currently, we do not persist view parameters when making context rendering requests.
The big one is the renderer (1up vs 2up). This makes context on unified diffs come in with too many columns.
However, it impacts other parameters too. For example, at HEAD, if you change highlighting to "rainbow" and then load more context, the context uses the original highlighter instead of the rainbow highlighter.
This moves context loads into ChangesetViewManager, which maintains view parameters and can provide them correctly.
- This removes "ref"; it is no longer required, as the ChangesetViewManager tracks it.
- This removes URI management from `behavior-show-more`; it is no longer required, since the ChangesetViewManager knows how to render.
- This removes "whitespace" since this is handled properly by the view manager.
Test Plan:
- Used "Show Top" / "Show All" / "Show Bottom" in 1-up and 2-up views.
- Changed file highlighting to rainbow, loaded stuff, saw rainbow stick.
- Used "Show Entire File" in 1-up and 2-up views.
- Saw loading chrome.
- No loading chrome normally.
- Made inlines, verified `copyRows()` code runs.
- Poked around Diffusion -- it is missing some parameter handling, but works OK.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T2009
Differential Revision: https://secure.phabricator.com/D11977
Summary:
Ref T2009. This basically copy/pastes them for now. Plans is:
- Make this actually work all the way.
- Add test coverage after D11970.
- Move 2-up here after test coverage.
Clicking the links does not work yet, because they use the 2-up renderer. I'll fix this in the next diff.
Test Plan: Viewed diffs in unified, saw links to show more.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: chad, epriestley
Maniphest Tasks: T2009
Differential Revision: https://secure.phabricator.com/D11976
Summary:
Fixes T7422. After the recent fix for "sort" columns, we can end up with invalid SQL in some cases when running quickstart.
In particular, we do "COLLATE binary CHARACTER SET utf8_general_ci" (which is invalid).
Preprocess these so we get "COLLATE utf8 CHARACTER SET utf8_general_ci" (which is valid and correct).
Test Plan: Ran `bin/storage upgrade -f --namespace blahblhbaba` with and without `--disable-utf8mb4`.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7422
Differential Revision: https://secure.phabricator.com/D11929
Summary:
Fixes T7287. This trades off 4-byte character support for case insensitivity in these columns, which is a much better trade on the balance.
Also adds more warnings about old MySQL. Note that we already issue a warning when you run "storage adjust" (which I've made stronger) and already "strongly recommend" MySQL 5.5 or newer in the install documentation.
Test Plan:
- Ran `storage adjust --disable-utf8mb4` to go to old definitions, then ran `storage adjust` to get back to the new ones. Everything seemed OK in both cases.
- Verified that utf8mb4 data can be migrated out of these colums with `--unsafe` (which will truncate).
- Verified that manual explains this.
- Faked my way into the setup warning.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7287
Differential Revision: https://secure.phabricator.com/D11893
Summary:
Ref T7352. Make sure modern `phd stop` can still read the old PID file format and stop the daemons, at least for now.
Without this, `stop` still detects them and tells you to `stop --force`, which works, but this makes things a good deal cleaner.
Test Plan: Ran `phd stop` from master, then `phd stop` from this revision. Saw old daemons stop cleanly.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7352
Differential Revision: https://secure.phabricator.com/D11873
Summary:
Fixes T7352. This reduces the memory footprint for instances by combining these two similar daemons into one daemon which handles the responsibilities of both.
The fit isn't 100% perfect here but it's pretty close, and the GC daemon is fairly trivial.
Test Plan:
- Adjusted all the numbers to small numbers (5 second sleep, 120 second GC length).
- Added a ton of logging.
- Started trigger daemon.
- Saw it run a GC cycle.
- Saw it reschedule another cycle after 120 seconds (adjusted down from 4 hours).
- Reverted all the logging/small numbers.
- Ran `bin/phd start`, saw stable trigger daemon running.
- Grepped for removed daemon class name.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7352
Differential Revision: https://secure.phabricator.com/D11872
Summary: Ref T7352. This is pretty straightforward. I renamed `phd.start-taskmasters` to `phd.taskmasters` for clarity.
Test Plan:
- Ran `phd start`, `phd start --autoscale-reserve 0.25`, `phd restart --autoscale-reserve 0.25`, etc.
- Examined PID file to see options were passed.
- I'm defaulting this off (0 reserve) and making it a flag rather than an option because it's a very advanced feature which is probably not useful outside of instancing.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7352
Differential Revision: https://secure.phabricator.com/D11871
Summary:
Ref T7352. We were previously identifying things by `<daemonClass, overseerPID, startTime>` but that's not unique in a world where one overseer can run multiple daemons.
We already have an internal "daemonID", it just doesn't get written into the DB right now.
Start writing it, then use it to clean up `phd status`.
Test Plan: Ran `phd status`, got more accurate/useful output than previously.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7352
Differential Revision: https://secure.phabricator.com/D11865
Summary: Ref T7352. This makes `phd stop` and `phd status` produce more reasonable output with the new PID file format.
Test Plan: Ran `phd stop`, `phd status`, etc.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7352
Differential Revision: https://secure.phabricator.com/D11856
Summary:
Ref T7352. This changes `phd` to pass configuration to overseers over stdin. We still run one overseer per daemon.
The "status" stuff needs some cleanup, but it's mostly just UI/cosmetic.
Test Plan:
- Ran `phd debug`, `phd launch`, `phd start`, `phd status`, `phd stop`, etc.
- Verified PID files write in a reasonable format.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7352
Differential Revision: https://secure.phabricator.com/D11855
Summary:
We currently cache previews, but the vast majority of previews are never rendered again (e.g., they're a preview of someone partway through typing a comment).
Especially when editing large documents (Legalpad, Phriction), this can bloat the markup cache with data that will never be read and won't get purged for 30 days.
In particular, most of the data on `admin.phacility.com` is currently 1,000 previews of legalpad documents as I made minor edits to them over the course of several hours.
This isn't a big concern, but it's a very easy fix.
Test Plan:
- Previewed a legalpad document, verified that cache rows were not written as I mashed the keyboard.
- Saved the document, verified a new cache row was written.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11832
Summary:
Ref T6840. This feels a little dirty; open to alternate suggestions.
We currently have a race condition where multiple daemons may load a commit and then save it at the same time, when processing "reverts X" text. Prior to this feature, two daemons would never load a commit at the same time.
The "reverts X" load/save has no effect (doesn't change any object properties), but it will set the state back to the loaded state on save(). This overwrites any flag updates made to the commit in the meantime, and can produce the race in T6840.
In other cases (triggers, harbormaster, repositories) we deal with this kind of problem with "append-only-updates + single-consumer", or a bunch of locking. There isn't really a good place to add a single consumer for commits, since a lot of daemons need to access them. We could move the flags column to a separate table, but this feels pretty complicated. And locking is messy, also mostly because we have so many consumers.
Just exempting this column (which has unusual behavior) from `save()` feels OK-ish? I don't know if we'll have other use cases for this, and I like it even less if we never do, but this patch is pretty small and feels fairly understandable (that said, I also don't like that it can make some properties just silently not update if you aren't on the lookout).
So, this is //a// fix, and feels simplest/least-bad for the moment to me, I thiiink.
Test Plan: Added and executed unit tests.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6840
Differential Revision: https://secure.phabricator.com/D11822
Summary: Fixes T7319. These need a "%s" for the count where they had a "%d"
Test Plan: plan in D11812 is no longer a lie! (watcher added / removed strings render correctly)
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7319
Differential Revision: https://secure.phabricator.com/D11813
Summary: Fixes T7319. ...except I can't get this working in my sandbox? Changes to the translation file don't seem to show up. TEST PLAN IS A LIE
Test Plan: became a watcher, un became a watcher - saw sensical translated strings
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7319
Differential Revision: https://secure.phabricator.com/D11812
Summary:
Right now, taskmasters on empty queues sleep for 30 seconds. With a default setup (4 taskmasters), this averages out to 7.5 seconds between the time you do anything that queues something and the time that the taskmasters start work on it.
On instances, which currently launch a smaller number of taskmasters, this wait is even longer.
Instead, sleep for the number of seconds that there are taskmasters, with a random offset. This makes the average wait to start a task from an empty queue 1 second, and the average maximum load of an empty queue also one query per second.
On loaded instances this doesn't matter, but this should dramatically improve behavior for less-loaded instances without any real tradeoffs.
Test Plan: Started several taskmasters, saw them jitter out of sync and then use short sleeps to give an empty queue about a 1s delay.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11772
Summary: Ref T7152. Gives us an event hook so we can go make users a member of any instance they've been invited to as soon as they verify an email address.
Test Plan:
- Used `bin/auth verify` to trigger the event.
- Build out the invite flow in rSERVICES.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7152
Differential Revision: https://secure.phabricator.com/D11752
Summary:
Ref T7152. Ref T1139. This updates Phabricator so third-party libraries can translate their own stuff. Also:
- Hide "All Caps" when not in development mode, since some users have found this a little confusing.
- With other changes, adds a "Raw Strings" mode (development mode only).
- Add an example silly translation to make sure the serious business flag works.
- Add a basic British English translation.
- Simplify handling of translation overrides.
Test Plan:
- Flipped serious business / development on and off and saw silly/development translations drop off.
- Switched to "All Caps" and saw all caps.
- Switched to Very English, Wow!
- Switched to British english and saw "colour".
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7152, T1139
Differential Revision: https://secure.phabricator.com/D11747
Summary:
Ref T7152. Ref T3554.
- When an administrator clicks "send invites", queue tasks to send the invites.
- Then, actually send the invites.
- Make the links in the invites work properly.
- Also provide `bin/worker execute` to make debugging one-off workers like this easier.
- Clean up some UI, too.
Test Plan:
We now get as far as the exception which is a placeholder for a registration workflow.
{F291213}
{F291214}
{F291215}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T3554, T7152
Differential Revision: https://secure.phabricator.com/D11736
Summary: Minor spring cleaning, improve the visual feel of the comments table, more consistent structure.
Test Plan:
Test multiple comments, long comments, short comments, and multiple lines.
{F282462}
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: hach-que, Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11666
Summary:
Ref T7208. Now that we have approvals (new installs are safe by default), take those into account when generating this warning.
Try to soften the warning to cover the case discussed in T7208, hopefully without requiring additional measures.
Test Plan:
{F286014}
{F286015}
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T7208
Differential Revision: https://secure.phabricator.com/D11708
Summary: mysqldump output can end up having weird encoding issues when raw BLOBs are in the output, preventing the backup restoration from succeeding. This hex-encodes blobs in the dump from the backup workflow causing the output file to only contain ASCII and ensure imports are successful.
Test Plan: Had issues restoring a backup from the original `mysqldump` command issued by this workflow. Ran the same command with this flag added and I was able to restore the backup.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11704
Summary:
I'm hitting this in the cluster and couldn't figure it out after staring at it for a couple minutes. Produce a better error.
This dumps a hash of each configuration key value which is set to a non-default value into the daemon log. This is much more compact than the full config, and doesn't spread secrets around, so it seems like a good balance between providing information and going crazy with it.
Test Plan: {F284139}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11699
Summary: Fixes T7111. Also nails out the TODO to show the username
Test Plan: fired up the ole phabot, chatted "P123" and saw "P123: https://secure.phabricator.com/P123 - Masterwork From Distant Lands by epriestley"...!
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7111
Differential Revision: https://secure.phabricator.com/D11664
Summary: Use `PhutilXHPASTBinary` methods instead of `xhpast_parse` functions. Depends on D11517.
Test Plan: N/A, this is a direct swap.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11612
Summary:
Ref T6881.
- Fix dead links.
- Let implementations provide more information.
- Provide more information to implementations.
Test Plan: Links work, invoices show billing periods, fewer "Subscription 6" crumbs, all is well in the world.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11601
Summary:
Ref T6881. This adds the worker, and a script to make it easier to test. It doesn't actually invoice anything.
I'm intentionally allowing the script to double-bill since it makes testing way easier (by letting you bill the same period over and over again), and provides a tool for recovery if billing screws up.
(This diff isn't very interesting, just trying to avoid a 5K-line diff at the end.)
Test Plan: Used `bin/phortune invoice ...` to get the worker to print out some date ranges which it would theoretically invoice.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11577
Summary: Fixes T1476. The body of the email should be just the output of some diff command.
Test Plan:
git diff master > text.txt; ./bin/mail receive-test --to <configured-diff-create-address> < text.txt; a diff was successfully created...! email generated had a working link to the diff.
./bin/mail receive-test --to <configured-diff-create-address> < README.md; a diff was not created as expected...! email generated had a sensical error message, telling me that the mail body should have been generated via a diff command
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: johnny-bit, Korvin, epriestley
Maniphest Tasks: T1476
Differential Revision: https://secure.phabricator.com/D11574
Summary:
Hit this locally, with an error like:
> Version <empty string> is older than 1.9, the minimum supported version.
(Where `<empty string>` was just the empty string, not literally the text `<empty string>`.)
Be more careful about parsing versions, and parse the newer string.
Test Plan: Got "unknown version" with intentionally-broken test data, then clean readout.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11558
Summary: Fixes T7078. Adds a `./bin/storage shell` command which passes through to a MySQL shell. This is slightly more convenient than running `mysql` manually.
Test Plan: Ran `./bin/storage shell` and got a MySQL shell.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Maniphest Tasks: T7078
Differential Revision: https://secure.phabricator.com/D11548
Summary: Fixes T7034. Like HTTP, proxy requests to the correct host if a repository has an Almanac service host.
Test Plan: Ran VCS requests through the proxy.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7034
Differential Revision: https://secure.phabricator.com/D11543
Summary:
Ref T7034.
In a cluster environment, when a user connects with a VCS request over SSH (like `git pull`), the receiving server may need to proxy it to a server which can actually satisfy the request.
In order to proxy the request, we need to know which repository the user is interested in accessing.
Split the SSH workflow into two steps:
# First, identify the repository.
# Then, execute the operation.
In the future, this will allow us to put a possible "proxy the whole thing somewhere else" step in the middle, mirroring the behavior of Conduit.
This is trivially easy in `git` and `hg`. Both identify the repository on the commmand line.
This is fiendishly complex in `svn`, for the same reasons that hosting SVN was hard in the first place. Specifically:
- The client doesn't tell us what it's after.
- To get it to tell us, we have to send it a server capabilities string //first//.
- We can't just start an `svnserve` process and read the repository out after a little while, because we may need to proxy the request once we figure out the repository.
- We can't consume the client protocol frame that tells us what the client wants, because when we start the real server request it won't know what the client is after if it never receives that frame.
- On the other hand, we must consume the second copy of the server protocol frame that would be sent to the client, or they'll get two "HELLO" messages and not know what to do.
The approach here is straightforward, but the implementation is not trivial. Roughly:
- Start `svnserve`, read the "hello" frame from it.
- Kill `svnserve`.
- Send the "hello" to the client.
- Wait for the client to send us "I want repository X".
- Save the message it sent us in the "peekBuffer".
- Return "this is a request for repository X", so we can proxy it.
Then, to continue the request:
- Start the real `svnserve`.
- Read the "hello" frame from it and throw it away.
- Write the data in the "peekBuffer" to it, as though we'd just received it from the client.
- State of the world is normal again, so we can continue.
Also fixed some other issues:
- SVN could choke if `repository.default-local-path` contained extra slashes.
- PHP might emit some complaints when executing the commit hook; silence those.
Test Plan: Pushed and pulled repositories in SVN, Mercurial and Git.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7034
Differential Revision: https://secure.phabricator.com/D11541
Summary:
Fixes T7019. In a cluster environment, pushes currently fail because the commit hook can't identify the instance.
For web processes, the hostname identifies the instance -- but we don't have a hostname in the hook.
For CLI processes, the environment identifies the instance -- but we don't have an environment in the hook under SVN.
Promote the instance identifier into the upstream and pack/unpack it explicitly for hooks. This is probably not useful for anyone but us, but the amount of special-purpose code we're introducing is very small.
I poked at trying to do this in a more general way, but:
- We MUST know this BEFORE we run code, so the normal subclassing stuff is useless.
- I couldn't come up with any other parameter which might ever be useful to pass in.
Test Plan: Used `git push` to push code through proxied HTTP, got a clean push.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7019
Differential Revision: https://secure.phabricator.com/D11495
