WikitextContent methods replaceSection() (used when adding a section
to an existing page) and addSectionHeader() (used when creating a new
page) behaved inconsistently - the former would omit the heading
syntax when the section title was empty, but the latter would not.
They both omit it now.
Some callers of addSectionHeader() handled this case, but others did
not, which caused T311489. (I am removing the checks now, since the
change makes them redundant.)
Bug: T311489
Change-Id: Icd59bcf2b75bf50865f19fac92bddabe0c183dcc
Now largely automated:
VARS=$(grep -o "'[A-Za-z0-9_]*'" includes/MainConfigNames.php | \
tr "\n" '|' | sed "s/|$/\n/;s/'//g")
sed -i -E "s/'($VARS)'/MainConfigNames::\1/g" \
$(grep -ERIl "'($VARS)'" includes/)
Then git add -p with lots of error-prone manual checking. Then
semi-manually add all the necessary "use" lines:
vim $(grep -L 'use MediaWiki\\MainConfigNames;' \
$(git diff --cached --name-only --diff-filter=M HEAD^))
I didn't bother fixing lines that were over 100 characters unless they
were over 120 and triggered phpcs.
Bug: T305805
Change-Id: I74e0ab511abecb276717ad4276a124760a268147
Make phan stricter about null types by setting null_casts_as_any_type to
false (the default in mediawiki-phan-config)
Remaining false positive issues are suppressed.
The suppression and the setting change can only be done together
Bug: T242536
Bug: T301991
Change-Id: I0f295382b96fb3be8037a01c10487d9d591e7e01
Update/Create override classes of ContentHandler.
Soft-deprecate and remove method from Content and classes that override them.
Bug: T287158
Change-Id: Idfcfbfe1a196cd69a04ca357281d08bb3d097ce2
This helps phan to detect unreachable code and also impossible types
after the functions.
It helps phan to avoid false positives for array keys
when the keys are checked before
Bug: T240141
Change-Id: I895f70e82b3053a46cd44135b15437e6f82a07b2
Update ContentTransformer to access ContentHandler::preLoadTransform through the service.
Prepare object to hold a data that required for ContentHandler::preLoadTranform params.
This is a fully backwards compatible change.
We are doing hard deprecation via MWDebug::detectDeprecatedOverride.
However, with the ContentHandler calling Content and
Content calling ContentHandler, it doesn't matter whether
callers use Content or ContentHandler. This will allow us
to naturally convert all callers.
Bug: T287157
Change-Id: I89537e1e7d24c6e15252b2b51890a0bd81ea3e6b
Create ContentTransformer to access ContentHandler::preSaveTransform through the service.
Prepare object to hold a data that required for ContentHandler::preSaveTranform params.
This will require making a semi-backwards-incompatible
change no matter what, we don't really have a great way
of hard-deprecating overriding methods.
However, with the ContentHandler calling Content and
Content calling ContentHandler, and with the ProxyContent
trick to stop infinite recursion, it doesn't matter whether
callers use Content or ContentHandler. This will allow us
to naturally convert all callers. But won't really allow
hard-deprecation.
Bug: T287156
Change-Id: If6a2025868ceca3a3b6f11baec39695e47292e40
This avoids failures due to WikitextContent::isCountable() erroring
out when no title is set for the main RequestContext.
Bug: T257009
Change-Id: I70274e3876c5b7c273110f9afa713a85edcc6339
For compliance with the new version of the table interface policy
(T255803).
This patch was created by an automated search & replace operation
on the includes/ directory.
Bug: T257789
Change-Id: If560596f5e1e0a3da91afc36e656e7c27f040968
This annotates classes that can safely be instantiated by
extensions, per the Stable Interface Policy.
Bug: T247862
Change-Id: Ia280f559874fc0750265ddeb7f831e65fd7d7d6a
Migrate all callers of Hooks::run() to use the new
HookContainer/HookRunner system.
General principles:
* Use DI if it is already used. We're not changing the way state is
managed in this patch.
* HookContainer is always injected, not HookRunner. HookContainer
is a service, it's a more generic interface, it is the only
thing that provides isRegistered() which is needed in some cases,
and a HookRunner can be efficiently constructed from it
(confirmed by benchmark). Because HookContainer is needed
for object construction, it is also needed by all factories.
* "Ask your friendly local base class". Big hierarchies like
SpecialPage and ApiBase have getHookContainer() and getHookRunner()
methods in the base class, and classes that extend that base class
are not expected to know or care where the base class gets its
HookContainer from.
* ProtectedHookAccessorTrait provides protected getHookContainer() and
getHookRunner() methods, getting them from the global service
container. The point of this is to ease migration to DI by ensuring
that call sites ask their local friendly base class rather than
getting a HookRunner from the service container directly.
* Private $this->hookRunner. In some smaller classes where accessor
methods did not seem warranted, there is a private HookRunner property
which is accessed directly. Very rarely (two cases), there is a
protected property, for consistency with code that conventionally
assumes protected=private, but in cases where the class might actually
be overridden, a protected accessor is preferred over a protected
property.
* The last resort: Hooks::runner(). Mostly for static, file-scope and
global code. In a few cases it was used for objects with broken
construction schemes, out of horror or laziness.
Constructors with new required arguments:
* AuthManager
* BadFileLookup
* BlockManager
* ClassicInterwikiLookup
* ContentHandlerFactory
* ContentSecurityPolicy
* DefaultOptionsManager
* DerivedPageDataUpdater
* FullSearchResultWidget
* HtmlCacheUpdater
* LanguageFactory
* LanguageNameUtils
* LinkRenderer
* LinkRendererFactory
* LocalisationCache
* MagicWordFactory
* MessageCache
* NamespaceInfo
* PageEditStash
* PageHandlerFactory
* PageUpdater
* ParserFactory
* PermissionManager
* RevisionStore
* RevisionStoreFactory
* SearchEngineConfig
* SearchEngineFactory
* SearchFormWidget
* SearchNearMatcher
* SessionBackend
* SpecialPageFactory
* UserNameUtils
* UserOptionsManager
* WatchedItemQueryService
* WatchedItemStore
Constructors with new optional arguments:
* DefaultPreferencesFactory
* Language
* LinkHolderArray
* MovePage
* Parser
* ParserCache
* PasswordReset
* Router
setHookContainer() now required after construction:
* AuthenticationProvider
* ResourceLoaderModule
* SearchEngine
Change-Id: Id442b0dbe43aba84bd5cf801d86dedc768b082c7
This works similarly to speculative rev IDs with {{REVISIONID}}.
Re-parses can be avoided if the page ID is correctly guessed.
Also make the {{PAGEID:X}} parser function set vary-page-id.
Bug: T226785
Change-Id: I0b19be45e6ddd6cde330bfcd09d243e4e5beda01
getNativeData() is under-specified - callers can do nothing with the
value returned by getNativeData without knowing the concrete Content
class. And if they know the concrete class, they can and should use
a specialized getter instead, anyway.
Basically, getNativeData is overly generic, an example of polymorphism
done poorly. Let's fix it now.
Bug: T155582
Change-Id: Id2c61dcd38ab30416a25746e3680edb8791ae8e8
This was supposed to already be the case, but it wasn't. The flag that
was set got cleared and never did anything.
Change-Id: Ide960f8cb9228f9a9d68c540369f122ada0a2a6f
We have three methods for page counting currently supported for wikitext
non-redirect pages; 'any' counts any page that exists, 'link' counts any
page that has any outbound links, and 'comma' which searches for any ','
in the text having loaded it. This last option is much slower than these
other two, and is only used on a very small number of installations. Now
by dropping support for this method we can simplify this code and so run
it more often. Note that non-wikitext pages already did not support this
count method.
Installations with this setting set to 'comma', or any other string will
now work as if it was configured with 'any'.
Bug: T188472
Change-Id: I965927edcd2485ec4b49b2d80fdf216dbf19520b
Changes:
- uses int instead of number as param and return value type,
- uses stdClass instead of stdObject
- fixes ResourceLoaderClientHtml constructor's $target param type:
it is string|null, not an array (previously misspelled as "aray")
- changes the type of references to XML parser in XMP lib to resource
instead of not existing XMLParser
Change-Id: I98c363ebc6658d1f4dcabad97a9a92f3fcd7ea8c
The code that normalizes line endings ("\r\n" and "\r" to "\n") and
trims trailing whitespace is buried in Parser::preSaveTransform(), and
was duplicated to TextContent in 96b6afb31d, as non-wikitext content
models should still be normalizing line endings.
This splits the duplicated code into
TextContent::normalizeLineEndings(), and utilize it in the Parser.
Additionally, expand the documentation of
TextContent::preSaveTransform() to document that subclasses should make
sure they normalize line endings during the PST stage.
And remove a useless rtrim() call from WikitextContent that did nothing.
Change-Id: I9094c671d4bbd23d75436f8f1d682d6dd6e6d2fc
Avoid polluting the parser cache with skin-dependent output from
Skin::doEditSectionLink() by introducing getRawText(), which provides
access to the uncooked text. Use this in
WikitextContent::fillParserOutput() which is the primary offender
judging by the debug logs (see referenced bug).
Bug: T124356
Change-Id: Ia9e1e4a6dc3a26f88eeebc64eed023ff20c53d58
Xhprof generates this data now. Custom profiling of various
sub-function units are kept.
Calls to profiler represented about 3% of page execution
time on Special:BlankPage (1.5% in/out); after this change
it's down to about 0.98% of page execution time.
Change-Id: Id9a1dc9d8f80bbd52e42226b724a1e1213d07af7
Since Id44d566a, the text passed to the parser when parsing a
&redirect=no page no longer contains the #REDIRECT directive. For the
benefit of extensions that want to know the redirect target from various
parser hooks, record the target on the ParserOptions object associated
with the parse.
Bug: 62856
Change-Id: Icd1da9911a43eabacbd9e9a369a8326f67f270ff
* Created a ResourceLoader module for redirect page styling.
* Fixed a couple of bugs in Article code.
Keeping two of the old files for a while because they might still be
referenced by something (bug 69277 comment 11).
Bug: 69277
Change-Id: I8fa565833545e41d0232101566da8c5ffa850e36
Currently the names of associated Content classes are hardcoded,
meaning that any extension that wishes to subclass these implementations
must also re-implement that function, usually copying it exactly
with just the class name changed. Using "static" avoids that issue.
For ContentHandlers, I added a TextContentHandler::getContentClass,
which should be used when creating new Content objects.
Change-Id: I70f1a3291aec3460120ec20121a23f4cb68e04d1
- Swap "$variable type" to "type $variable"
- Added missing types
- Fixed spacing inside docs
- Makes beginning of @param/@return/@var/@throws in capital
- Changed some types to match the more common spelling
Change-Id: I7b65fe04db431342cc58b469dc48f41a50c4e891
There are so many slightly different understandings of what a
"section" is or can be. I'm aware the documentation was improved
just a few weeks ago. I still find it incomplete and confusing.
1. I renamed it to $sectionId to make it more clear what it
really is.
2. Sections are usually numbers. 0, 1 and so on. There is no
reason to disallow the use of ints or even floats (this works
because the string representation of 0.0 is "0"). The code never
disallowed numbers.
3. 'T1' never was supported, as far as I can tell. 'T-1' is
supported. See Parser::extractSections().
4. null and false and '' all mean "the whole page" in
WikiPage::replaceSectionAtRev() but for some reason this meaning got
lost in WikitextContent::replaceSection(). I made it the same again.
Change-Id: Icc3997722d2ed742bf7703cd7c06d09199225720
This hooks allows extensions to override the normal model-specific
rendering of page content. A typical use would be to provide syntax
highlighting for pages that contain scripts. In that sense,
ContentGetParserOutput is a generalization of the old ShowRawCssJs
hook.
This, together with I76412f9d, is a (hopefully) fixed version of the
reverted Ibfb2cbefea44.
Change-Id: I979e2438689648ba4c959d8083197ef14ce524e8
When pages are loaded in the edit box via preload, allow parameter
substitution. The interface-style $1 is used rather than the
template-style {{{1}}} to avoid conflicts with preloads that add template
parameters. Syntax is:
action=edit&preload=Foo&preloadparams[]=first&preloadparams[]=second
Bug: 12853
Change-Id: If02cf4b3dba9f9d22a956d8bfff224677cbce00d
Reasons for touching this are:
* "@param type $var" were mixed in a lot of places. Both works but the
MediaWiki coding conventions suggest that specific order.
* Things like String and Bool aren't objects and shouldn't be uppercase.
* Tried to fill missing types in "@param $var".
* Tried to fill missing descriptions in "@return type" when I could.
* Removed duplicate descriptions if a @see is sufficend.
* Removed useless descriptions ("isUsefull returns true if usefull").
* Removed useless @return void.
* Replaces mixed[] with array (does have the exact same meaning).
* Tried to find better replacements for "varargs", phpDocumentor
suggest $var,...
* Order should be @since, @param, @throws, @return, @see. This is the
order Doxygen renders this.
There is always more to do but I think this is already much better
than before. Please feel free to put more change sets on top of mine
or request more changes by adding comments.
Change-Id: I05262ce06caabf79f68772302524ac91bbead1c6
