Constructing a Language object in order to initialize the
BlobStoreFactory service causes a circular dependency
(see T231866).
SqlBlobStore was using the Language object to all iconv.
But nothing language specific is done in Language::iconv,
so we can just inline the call.
Bug: T231866
Change-Id: I90c25decbcff10ea762a2c7474a12fd2041b3abc
If the rows were obtained using RevisionStore::getQueryInfo with
'page' flags, the revision row already contains the fields needed
to construct the Title without an additional database query.
Change-Id: Ie36c85871a8996a5706c80d286854a9c8363905f
getContentBlobsForBatch provides a more low level interface than
newRevisionsFromBatch, allowing bulk access to serialized content
without the need to instantiate Title, RevisionRecord, RevisionSlots,
SlotRecord, and Content objects.
Bug: T228675
Bug: T234034
Change-Id: I8481ad211e2d9f11bc10ea10c16e78b74538d95b
This removes most of the pre-actor user and user_text columns, and the
$wgActorTableSchemaMigrationStage setting that used to determine
whether the columns were used.
rev_user and rev_user_text remain in the code, as on Wikimedia wikis the
revision table is too large to alter at this time. A future change will
combine that with the removal of rev_comment, rev_content_model, and
rev_content_format (and the addition of rev_comment_id and rev_actor).
ActorMigration's constructor continues to take a $stage parameter, and
continues to have the logic for handling it, for the benefit of
extensions that might need their own migration process. Code using
ActorMigration for accessing the core fields should be updated to use
the new actor fields directly. That will be done for in a followup.
Bug: T188327
Change-Id: Id35544b879af1cd708f3efd303fce8d9a1b9eb02
Make LoadBalancer::reallyOpenConnection() handle initializing DBO_TRX
instead of Database::__construct().
Also:
* Avoid having the "catch" block appear like it returns a
half-constructed Database.
* Use the variable name $conn instead of $db to be consistent
throughout the class. Only send Database::__construct() parameters
that it recognizes instead of mixing in setLBInfo() data.
Change-Id: Iffc3d1d0713051a164adb51a4c4ee12e4ac887c3
* Inject settings and global instances as dependencies to the
ExternalStoreMedium instances. This includes the local wiki
domain, so that wfWikiId() calls are not scattered around.
* Create ExternalStoreAccess service for read/write logic.
* Deprecate the ExternalStore wrapper methods.
* Add some exception cases for bogus store URLs are used instead
of just giving PHP warnings and failing later.
* Make moveToExternal.php require the type/protocol to decide
which ExternalStoreMedium to use instead of assuming "DB".
* Convert logging calls to use LoggerInterface.
Change-Id: I40c3b5534fc8a31116c4c5eb64ee6e4903a6197a
This introduces a way to construct a RevisionRecord based on a
known set of SlotRecords. To allow this to be used consistently
with the legacy revision schema, some tweaks had to be made
to getSlotsQueryInfo().
Bug: T220493
Change-Id: I5ea972bb07ca1cfb3a2ad8ef120aef77e460745c
3e36ba655e added an option for passing a page ID to this method of
Revision, and e03787afd9 switched it to a Title and made it mandatory.
This behavior propagated to the method in RevisionStore. As far as I
can tell, the parameter does not help anything, but it can add a
database query to get the page ID if it's not cached, and impedes
conversion to LinkTarget. I can't figure out any reason to not
completely drop it. I've noted it as deprecated but still supported it
for now for compatibility -- I found one extension that does pass it.
(It's ignored, though, which theoretically would be a behavior change if
someone was passing a Title that didn't match the revision ID.)
While I was at it, I added the method to RevisionLookup, although it's
only used in later patches. Properly I should move that piece to a later
patch, but it didn't seem worth the effort.
I didn't change the Revision method, because the whole Revision class is
deprecated anyway.
Change-Id: I26ef5f2bf828f0f450633b7237d26d888e2c8773
Probably good to start testing this in CI sometime soon.
This also updates a bunch of tests that were forcing an older stage to
force SCHEMA_COMPAT_NEW instead, or to test both ways (until a future
patch removes the _OLD version).
Bug: T188327
Change-Id: Icb9b55cb9d754f2d30d6883005658b9670834756
Created class RevisionStoreCacheRecord to avoid returning stale
cached revision visibility and actor data when changes were
made after the cache was populated for that revision.
Bug: T216159
Change-Id: Iabed80e06a08ff0793dfe64db881cbcd535cb13f
This removes most of the pre-CommentStore text columns, and the
$wgCommentTableSchemaMigrationStage setting that used to determine
whether the columns were used.
rev_comment remains in the code, as on Wikimedia wikis the revision
table is too large to alter at this time. A future change will combine
that with the removal of rev_user_text, rev_content_model, and
rev_content_format (and the addition of rev_comment_id and rev_actor).
CommentStore's constructor continues to take a $stage parameter, and
continues to have the logic for handling it, for the benefit of
extensions that might need their own migration process.
Bug: T166733
Change-Id: I1479c73774e01ead1490adf6128f820c09bce9d4
This might hint at an edge-case in the PHP CodeSniffer sniff that should
detect if methods are separated by a single empty line. Feel free to
investigate. I, personally, can't invest more time in this than
suggesting this quick fix.
Change-Id: Ib3c60eac76f255b4fe929f7933de256222716576
These new classes provide a mechanism for defining the
behavior of slots, like the content models it supports.
This acts as an extension point for extensions that need
to define custom slots, like the MediaInfo extension
for the SDC project.
Bug: T194046
Change-Id: Ia20c98eee819293199e541be75b5521f6413bc2f
WRITE_NEW mode seems to be working well on Wikimedia sites. Let's change
the default to MIGRATION_NEW so existing installs and Wikimedia CI will
start using the new code rather than the old.
This also fixes some unit tests that were broken with MIGRATION_NEW, and
updates some that were forcing MIGRATION_OLD to force MIGRATION_NEW
instead.
Bug: T166733
Change-Id: I7bf4ad0105dd1f6cc49eba3ddcb7a51badcd5ed3
Depends-On: I30f7cdcc3875f3f7af116c1e41e88f62ab9e91d0
This makes RevisionStore a bit more robust against use with deleted
revisions.
Note that this does not make I151019e336bda redundant, since detecting
this situation early and providing meaningful error message is useful.
Bug: T208929
Change-Id: I7635ddc0176e21d873eacadebe02603b1fe51c38
When this was originally written, the plan was to read both the old and
new fields during the transition period, while stopping writes to them
midway through. It turns out that the WHERE conditions to do read-both
correctly are generally not handled well by the database and working
around that would require a lot of complicated code (see what's being
removed from ApiQueryUserContribs here, for example).
We can simplify things greatly by instead having it write both fields
during the transition period, reading from the old for the first part
and the new for the second part, as is being done for MCR.
Bug: T204669
Change-Id: I4764c1c7883dc1003cb12729455c8107319f70b1
Depends-On: I845f6ae462f2539ebd35cbb5f2ca8b5714e2c1fb
Depends-On: I88b31b977543fdbdf69f8c1158e77e448df94e11
LoadBalancer uses Database::getDomainId() for deciding which keys to use
in the foreign connection handle arrays. This method should reflect any
changes made to the DB selection.
If the query fails, then do not change domain field. This is the sort of
approach that LoadBalancer is expects in openForeignConnection(). Also,
throw an exception when selectDB() fails.
The db/schema/prefix fields of Database no longer exist in favor of just
using the newer currentDomain field.
Also:
* Add IDatabase::selectDomain() method and made selectDB() wrap it.
* Extract the DB name from sqlite files if not explicitly provided.
* Fix inconsistent open() return values from Database subclasses.
* Make a relationSchemaQualifier() method to handle the concern of
omitting schema names in queries. The means that getDomainId() can
still return the right value, rather than confusingly omitt the schema.
* Make RevisionStore::checkDatabaseWikiId() account for the domain schema.
Unlike d2a4d614fc, this does not incorrectly assume the storage is
always for the current wiki domain. Also, LBFactorySingle sets the local
domain so it is defined even in install.php.
* Make RevisionStoreDbTestBase actually set the LoadBalancer local domain.
* Make RevisionTest::testLoadFromTitle() account for the domain schema.
Bug: T193565
Change-Id: I6e51cd54c6da78830b38906b8c46789c79498ab5
During development a lot of classes were placed in MediaWiki\Storage\.
The precedent set would mean that every class relating to something
stored in a database table, plus all related value classes and such,
would go into that namespace.
Let's put them into MediaWiki\Revision\ instead. Then future classes
related to the 'page' table can go into MediaWiki\Page\, future classes
related to the 'user' table can go into MediaWiki\User\, and so on.
Note I didn't move DerivedPageDataUpdater, PageUpdateException,
PageUpdater, or RevisionSlotsUpdate in this patch. If these are kept
long-term, they probably belong in MediaWiki\Page\ or MediaWiki\Edit\
instead.
Bug: T204158
Change-Id: I16bea8927566a3c73c07e4f4afb3537e05aa04a5
2018-10-09 10:22:48 -04:00
Renamed from tests/phpunit/includes/Storage/RevisionStoreDbTestBase.php (Browse further)
