Summary:
Ref T8672. Ref T9187. Root issue in at least one case is:
- User makes a commit including a file with some non-UTF8 text (say, a Japanese file full of Shift-JIS).
- We pass the file to the TransactionEditor so it can inline or attach the patch if the server is configured for these things.
- When inlining patches, we convert them to UTF8 before inlining. We must do this since the rest of the mail is UTF8.
- When attaching patches, we send them in the original encoding (as file attachments). This is correct, and means we need to give the worker the raw patch in whatever encoding it was originally in: we can't just convert it to utf8 earlier, or we'd attach the wrong patch in some cases.
- TransactionEditor does its thing (e.g., creates the commit), then gets ready to send mail about whatever it did.
- The publishing work now happens in the daemon queue, so we prepare to queue a PublishWorker and pass it the patch (with some other data).
- When we queue workers, we serialize the state data with JSON.
So far, so good. But this is where things go wrong:
- JSON can't encode binary data, and can't encode Shift-JIS. The encoding silently fails and we ignore it.
Then we get to the worker, and things go wrong-er:
- Since the data is bad, we fatal. This isn't a permanent failure, so we continue retrying the task indefinitely.
This applies several fixes:
# When queueing tasks, fail loudly when JSON encoding fails.
# In the worker, fail permanently when data can't be decoded.
# Allow Editors to specify that some of their data is binary and needs special handling.
This is fairly messy, but some simpler alternatives don't seem like good ways forward:
- We can't convert to UTF8 earlier, because we need the original raw patch when adding it as an attachment.
- We could encode //only// this field, but I suspect some other fields will also need attention, so that adding a mechanism will be worthwhile. In particular, I suspect filenames //may// be causing a similar problem in some cases.
- We could convert task data to always use a serialize()-based binary safe encoding, but this is a larger change and I think it's correct that things are UTF8 by default, even if it makes a bit of a mess. I'd rather have an explicit mess like this than a lot of binary data floating around.
The change to make `LiskDAO` will almost certainly catch some other problems too, so I'm going to hold this until after `stable` is cut. These problems were existing problems (i.e., the code was previously breaking or destroying data) so it's definitely correct to catch them, but this will make the problems much more obvious/urgent than they previously were.
Test Plan:
- Created a commit with a bunch of Shift-JIS stuff in a file.
- Tried to import it.
Prior to patch:
- Broken PublishWorker with distant, irrelevant error message.
With patch partially applied (only new error checking):
- Explicit, local error message about bad key in serialized data.
With patch fully applied:
- Import went fine and mail generated.
Reviewers: chad
Reviewed By: chad
Subscribers: devurandom, nevogd
Maniphest Tasks: T8672, T9187
Differential Revision: https://secure.phabricator.com/D13939
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:
Ref T8424. I'm using Paste as a testbed application because Spaces make some degree of sense for it but it's also flat/simple.
This doesn't do anything interesting or useful and mostly just making the next (more interesting) diff smaller.
Test Plan:
- Ran `bin/storage upgrade -f`.
- Browsed pastes.
- Created a paste.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T8424
Differential Revision: https://secure.phabricator.com/D13154
Summary:
This was broken in D12680.
```
EXCEPTION: (PhutilJSONParserException) Parse error on line 0 at column 0: 'null' is not a valid JSON object. at [<phutil>/src/parser/PhutilJSONParser.php:41]
#0 PhutilJSONParser::parse(string) called at [<phutil>/src/utils/utils.php:1062]
#1 phutil_json_decode(string) called at [<phabricator>/src/infrastructure/storage/lisk/LiskDAO.php:1640]
#2 LiskDAO::applyLiskDataSerialization(array, boolean) called at [<phabricator>/src/infrastructure/storage/lisk/LiskDAO.php:1386]
#3 LiskDAO::willReadData(array) called at [<phabricator>/src/infrastructure/storage/lisk/PhabricatorLiskDAO.php:214]
#4 PhabricatorLiskDAO::willReadData(array) called at [<phabricator>/src/infrastructure/storage/lisk/LiskDAO.php:608]
#5 LiskDAO::loadFromArray(array) called at [<phabricator>/src/infrastructure/storage/lisk/LiskDAO.php:652]
#6 LiskDAO::loadAllFromArray(array) called at [<phabricator>/src/applications/transactions/query/PhabricatorApplicationTransactionQuery.php:62]
#7 PhabricatorApplicationTransactionQuery::loadPage() called at [<phabricator>/src/infrastructure/query/policy/PhabricatorPolicyAwareQuery.php:227]
#8 PhabricatorPolicyAwareQuery::execute() called at [<phabricator>/src/infrastructure/query/policy/PhabricatorCursorPagedPolicyAwareQuery.php:143]
#9 PhabricatorCursorPagedPolicyAwareQuery::executeWithCursorPager(AphrontCursorPagerView) called at [<phabricator>/src/applications/base/controller/PhabricatorController.php:577]
#10 PhabricatorController::buildTransactionTimeline(PhabricatorPaste, PhabricatorPasteTransactionQuery) called at [<phabricator>/src/applications/paste/controller/PhabricatorPasteViewControll
```
Test Plan: No exception shown.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D12714
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 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:
Ref T6881. This is part 1 of my 35-step plan to support subscriptions that bill monthly.
Expanding the capabilities of counters will let me use them to create a logical clock on time-based event updates, build a daemon on top of that, and eventually get time-based triggers.
Test Plan: Added and executed unit tests.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: chad, epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11395
Summary: Ref T6822. There are a bunch of places where we call `$something->generatePHID(...)` externally (outside of the class). Therefore, these methods need to be `public`.
Test Plan: I wouldn't expect //increasing// method visibility to break anything.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T6822
Differential Revision: https://secure.phabricator.com/D11363
Summary:
Ref T1191. Currently if a developer forgot to specify a column type, `storage adjust` aborts explosively mid-stream. Instead:
- Make this a formal error with an unambiugous name/description instead of something you sort of infer by seeing "<unknown>".
- Make this error prevent generation of adjustment warnings, so we don't try to `ALTER TABLE t CHANGE COLUMN c <unknown>`, which is nonsense.
- When schemata errors exist, surface them prominiently in `storage adjust`.
Overall:
- Once `storage upgrade` runs `storage adjust` automatically (soon), this will make it relatively difficult to miss these errors.
- Letting these errors slip through no longer escalates into a more severe issue.
Test Plan:
Commented out the recent `mailKey` spec and ran `storage adjust`:
```
$ ./bin/storage adjust --force
Verifying database schemata...
Found no adjustments for schemata.
Target Error
phabricator2_phriction.phriction_document.mailKey Column Has No Specification
SCHEMATA ERRORS
The schemata have serious errors (detailed above) which the adjustment
workflow can not fix.
If you are not developing Phabricator itself, report this issue to the
upstream.
If you are developing Phabricator, these errors usually indicate that your
schema specifications do not agree with the schemata your code actually
builds.
```
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10771
Summary:
Ref T1191. When changing the column type of an AUTO_INCREMENT column, we currently may lose the autoincrement attribute.
Instead, support it. This is a bit messy because AUTO_INCREMENT columns interact with PRIMARY KEY columns (tables may only have one AUTO_INCREMENT column, and it must be a primary key). We need to migrate in more phases to avoid this issue.
Introduce new `auto` and `auto64` types to represent autoincrement IDs.
Test Plan:
- Saw autoincrement show up correctly in web UI.
- Fixed an autoincrement issue on the XHProf storage table with `bin/storage adjust` safely.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10607
Summary:
Ref T1191. Nothing too notable here:
- Allow a Lisk object to specify that there's no expectation that a table exists. We have one Harbormaster object and one Token object like this.
- Removed BuildPlanTransactionComment because it's currently unused.
Test Plan:
- Saw ~200 fewer warnings; just ~800 left.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10583
Summary:
Ref T1191. Three parts:
- The old way of getting key information only got primary / unique / foreign keys, not all keys. Use `SHOW INDEXES` to get all keys instead.
- Track key uniqueness and raise warnings about it.
- Add a new "all issues" view to show an expanded, flat view of all issues. This is just an easier way to get a list so you don't have to dig around in the hierarchical view.
Test Plan:
{F206351}
{F206352}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10525
Summary: Ref T1191. Nothing too exciting in these.
Test Plan: Saw more blue in UI.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10521
Summary:
Ref T1191. Notable:
- Allowed objects to remove default columns (some feed tables have no `id`).
- Added a "note" severity and moved all the charset stuff down to that to make progress more clear.
Test Plan:
Trying to make the whole thing blue...
{F205970}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10519
Summary: Ref T1191. Fills in some more of the databases. Nothing very notable here. I didn't encounter any issues or overlong keys.
Test Plan: Used web UI to click around and verify expected schemata match up against actual schemata well.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10516
Summary: Ref T1191. This fills in some more features and gets audit and auth nearly generating reasonable expected schemata.
Test Plan: See screenshots.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10500
Summary:
Ref T1191. This lays some groundwork for generating the expected schemata, so we can compare them to the actual schemata and produce a meaningful diff.
- In general, each application will subclass `PhabricatorConfigSchemaSpec` and provide a definition of the tables it expects.
- This class has helper methods to mostly-automatically build table definitions for Lisk and (in the future) edges.
- When building expected schema, we specify a "data type", like "epoch". This is the type of data the application stores in the column, from the application's point of view. The SchemaSpec converts this into the best avilable storage type: for example, "text" will translate to `utf8mb4` if it's availalbe, or `binary` if not. This gives us a layer of indirection to insulate us from craziness.
Test Plan: See screenshots.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D10497
Summary:
See some discussion here:
24a6eeb8d8 (commitcomment-7334892)
The `protected $properties;` storage parameter added to `ProjectColumn` is shadowed by `getProperties()` in the base class.
Although this works correctly for me, it's ambiguous and worth fixing. Make the base class methods explicit.
Test Plan: Used `grep` to find callers for both methods and renamed them.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D10210
Summary:
Ref T4420. This was a performance hack introduced long ago to make typeaheads for users a little cheaper. The idea was that you could load some of an object's columns and skip other ones.
We now always load users on demand, so the cost of loading the whole objects is very small. No other use cases ever arose for this, and it seems unlikely that they will in the future. Remove it all.
Test Plan:
- Grepped for `CONFIG_PARTIAL_OBJECTS`.
- Grepped for `dirtyFields`.
- Grepped for `missingFields`.
- Grepped for `resetDirtyFields`.
- Grepped for `loadColumns`.
- Grepped for `loadColumnsWhere`.
- Grepped for `loadRawDataWhere`.
- Loaded and saved some lisk objects.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T4420
Differential Revision: https://secure.phabricator.com/D9895
Summary: I'm pretty sure that `@group` annotations are useless now... see D9855. Also fixed various other minor issues.
Test Plan: Eye-ball it.
Reviewers: #blessed_reviewers, epriestley, chad
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley, Korvin, hach-que
Differential Revision: https://secure.phabricator.com/D9859
Summary: Ran `arc lint --apply-patches --everything` over rP, mainly to change double quotes to single quotes where appropriate. These changes also validate that the `ArcanistXHPASTLinter::LINT_DOUBLE_QUOTE` rule is working as expected.
Test Plan: Eyeballed it.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: epriestley, Korvin, hach-que
Differential Revision: https://secure.phabricator.com/D9431
Summary:
Ref T1191. I believe we only have three meaningful binary fields across all applications:
- The general cache may contain gzipped content.
- The file storage blob may contain arbitrary binary content.
- The Passphrase secret can store arbitrary binary data (although it currently never does).
This adds Lisk config for binary fields, and uses `%B` where necessary.
Test Plan:
- Added and executed unit tests.
- Forced file uploads to use MySQL, uploaded binaries.
- Disabled the CONFIG_BINARY on the file storage blob and tried again, got an appropraite failure.
- Tried to register with an account containing a G-Clef, and was stopped before the insert.
Reviewers: btrahan, arice
Reviewed By: arice
CC: arice, chad, aran
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D8316
Summary: Ref T4375. We never join this table, so this is a pretty straight find/replace.
Test Plan: Browsed around Calendar, verified that nothing seemed broken. Saw my red dot in other apps.
Reviewers: btrahan, chad
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T4375
Differential Revision: https://secure.phabricator.com/D8145
Summary:
If you do something like this:
// Missing $user->getPHID()!
$object->setUserPHID($user)->save();
...you get a very unhelpful exception:
Expected a scalar or null for %s conversion. Query: %s
This doesn't give you any hints about what's wrong. Instead, provide a more useful exception:
Unable to insert or update object of class DifferentialRevision, field 'title' has a nonscalar value.
Test Plan: {F87614}
Reviewers: hach-que, btrahan
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D7725
Summary:
Ref T603. This cleans up an existing callsite in the policy filter, and opens up some stuff in the future.
Some policy objects don't have real PHIDs:
PhabricatorTokenGiven
PhabricatorSavedQuery
PhabricatorNamedQuery
PhrequentUserTime
PhabricatorFlag
PhabricatorDaemonLog
PhabricatorConduitMethodCallLog
ConduitAPIMethod
PhabricatorChatLogEvent
PhabricatorChatLogChannel
Although it would be reasonable to add real PHIDs to some of these (like `ChatLogChannel`), it probably doesn't make much sense for others (`DaemonLog`, `MethodCallLog`). Just let them return `null`.
Also remove some duplicate `$id` and `$phid` properties. These are declared on `PhabricatorLiskDAO` and do not need to be redeclared.
Test Plan: Ran the `testEverythingImplemented` unit test, which verifies that all classes conform to the interface.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T603
Differential Revision: https://secure.phabricator.com/D7306
Test Plan: Searched for `IDS_PHID` and loaded homepage.
Reviewers: epriestley, edward
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D5619
Summary:
Remarkup rule callbacks now get SafeHTML matches instead of string matches. If they call:
$some_lisk_dao->load($matches[1]);
..as is the case with the `{F123}` rule, we reject the SafeHTML as an invalid ID and return null.
Allow load() to string convert any object (which will either succeed or fatal in an obviously-broken way).
(Long ago we threw instead of returning null here, but it meant we had to do a lot of redundant checks.)
Test Plan: `{F123}` shows an image again. `{C1}` embeds a countdown.
Reviewers: vrana, chad
Reviewed By: vrana
CC: aran
Differential Revision: https://secure.phabricator.com/D4961
Summary:
T2345
getConfig throws an Exception when the key does not exist.
Also removes dead code that throws an Exception.
Test Plan:
Reloaded the Phabricator home page. In the process, found
2 Exceptions thrown due to nonexistent keys. After addressing these problems,
the home page loads without Exceptions.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D4541
Summary: See D3912 for discussion. InnoDB may reuse autoincrement IDs after restart; provide a way to avoid it.
Test Plan: Unit tests. Scheduled and executed tasks through `drydock lease --type host` and `phd debug taskmaster`.
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Differential Revision: https://secure.phabricator.com/D3914
Summary:
This commit doesn't change license of any file. It just makes the license implicit (inherited from LICENSE file in the root directory).
We are removing the headers for these reasons:
- It wastes space in editors, less code is visible in editor upon opening a file.
- It brings noise to diff of the first change of any file every year.
- It confuses Git file copy detection when creating small files.
- We don't have an explicit license header in other files (JS, CSS, images, documentation).
- Using license header in every file is not obligatory: http://www.apache.org/dev/apply-license.html#new.
This change is approved by Alma Chao (Lead Open Source and IP Counsel at Facebook).
Test Plan: Verified that the license survived only in LICENSE file and that it didn't modify externals.
Reviewers: epriestley, davidrecordon
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2035
Differential Revision: https://secure.phabricator.com/D3886
Summary:
See comment.
This can reveal some pretty bad bugs but HPHP handles this correctly so we already know about them.
Test Plan:
Added `phlog()` to `__call()` and observed what is defined for each method (under PHP). Also:
class C {
function __call($name, $args) {
static $class;
if (!$class) {
$class = get_class($this);
}
return $class;
}
}
class D extends C {
}
class E extends C {
}
$d = new D;
$e = new E;
var_dump($d->x());
var_dump($e->x()); // Prints D under PHP!
See also D3754.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T1261
Differential Revision: https://secure.phabricator.com/D3753
Summary: For immutable objects, just use the ID as a cursor.
Test Plan:
- Analyzed commits from an empty cursor.
- Checked that cursor was good.
- Pulled some more commits.
- Analyzed commits again, verified it only hit the new ones.
- Verified the graph of "Count of CMIT" looked reasonable.
Reviewers: vrana
Reviewed By: vrana
CC: aran
Maniphest Tasks: T1866
Differential Revision: https://secure.phabricator.com/D3656
Summary: We never use this and almost certainly never will. It's been in Lisk for ~7 years but is a solution in search of a problem. It causes a conflict with any DAO that has a `version` column.
Test Plan: Browsed around, performed inserts and updates. Edited a Phriction document.
Reviewers: vrana, btrahan
Reviewed By: btrahan
CC: aran, leslie.chong
Differential Revision: https://secure.phabricator.com/D3625
Summary:
Fixes a TODO, and silences a warning introduced by D3601.
There are several cases where we load data like:
SELECT *, ... AS extraData FROM ...
...and then pass it to `loadAllFromArray()`. Currently, this causes us to set an `extraData` property on the object.
This idiom seems fairly useful and non-dangerous, so I made `loadFromArray()` just drop extra keys.
Since we hit this loop a potentially huge number of times (10,000+ for full Maniphest pages) I did some microoptimization. Lisk is hot enough that it's one of the few places where it's worthwhile (see D1291).
Test Plan: Loaded homepage, no longer got warnings about `viewerIsMember` from Project queries. Browsed ~10 apps, didn't see any issues.
Reviewers: vrana
Reviewed By: vrana
CC: aran
Differential Revision: https://secure.phabricator.com/D3606
Summary:
I make this error quite often: I forget to declare a property I am writing to or I make a typo in it.
PHP implicitly creates a public property which I don't like.
I would much rather see a linter warning me against this than this runtime check but writing it is very difficult:
- We need to explore all parents of the class we are checking.
- It is even possible that children will declare that property but it's OK to treat this as error anyway.
- We can extend also builtin or external classes.
- It's somewhat doable for `$this` but even more complex for any `$obj` because we don't know the class of it.
This should catch significant part of these errors and I'm fine with that.
I don't plan escalating to exception because this error is not fatal and should not stop the application from working.
Test Plan: Loaded homepage, checked log.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3601
Summary:
Calling `->setPHID()` or other common Lisk setters creates an implicit public property `$phid`.
I don't like implicit properties and I see them as errors.
Its public visibility also makes me nervous and is vulnerable to bypassing any setters we may create.
Test Plan: Loaded homepage, checked log.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3600
Summary:
More and more relations are going under edges and I can't work with them from Relatives framework.
This doesn't have the nice transitive property of normal relatives (loading relative objects from relatives loads all of them at once) but I can add it when I need it.
I plan to use it in D3085 (after converting relationships to edges).
Test Plan:
$task = id(new ManiphestTask())
->loadOneWhere('phid = %s', $phid);
print_r($task->loadRelativeEdges(4));
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3344
Summary:
Lisk currently behaves in two different ways if you call it like `load("cow")` (throws) versus `load(99999999)` (returns null), where neither ID exists.
This was intended to catch programming errors as distinct from missing data, but in practice the former is very rare and you have to handle the latter in most cases anyway. The case where you pass "0" is particularly confusing. See D2971 for an example.
On the balance, I think this ends up being far more confusing than helpful. Instead, just return NULL if we're sure there's no such object.
Test Plan: Reasoned about program behavior.
Reviewers: alanh, btrahan, vrana
Reviewed By: vrana
CC: aran
Differential Revision: https://secure.phabricator.com/D2977
Summary:
In unit tests which use fixtures, we open transactions on every connection we establish. However, since we don't track connections that are established with "$force_new" (currently, only GlobalLock connections) we never close these transactions normally.
Instead of not tracking these connections, track them using unique keys so we'll never get a cache hit on them.
Test Plan: Built unit tests on top of this, had them stop dying from unclosed transactions.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1162
Differential Revision: https://secure.phabricator.com/D2938
