* Make ResourceLoaderUserOptionsModule and ResourceLoaderUserTokensModule depend on mw.user
* Load mw.user.{tokens,options,groups} load as TYPE_COMBINED instead of TYPE_SCRIPT. The latter wouldn't wrap the code in mw.loader.implement()
** ...but make sure 'user' (user scripts) is excluded, that one needs to not be wrapped in a closure
* Make TYPE_COMBINED actually work in makeResourceLoaderLink()
* Add a comment in makeModuleResponse() to explain what the weird is_array( $scripts ) stuff is all about
* Add FIXME about how mw.user.options should split off the CSS part into a separate module
Solving syntax problems, performance improvements and applying code conventions:
* Replaced sprite image with separate images and letting ResourceLoader embed them with @embed (@embed means 0 http requests, less maintenance, none of the known limitations with sprites, and more readable code (named files rather than pixel offsets)
* Many functions were floating in the global namespace (like window.makeCaseInsensitive). A statement ends after a semi-colon(;). All functions declared after "catUrl" were assigned to the window object. I've instead turned the semi-colons back into comma's, merged some other var statements and moved them to the top of the closure. Changed local function declarations into function expressions for clarity.
* fetchSuggestions is called by $.fn.suggestions like ".call( $textbox, $textbox.val() )". So the context (this) isn't the raw element but the jQuery object, no need to re-construct with "$(this)" or "$(that)" which is slow and shouldn't even work. jQuery methods can be called on it directly. I've also replaced "$(this).val()" with the value-argument passed to fetchSuggestions which has this exact value already.
* Adding more function documentation. And changing @since to 1.19 as this was merged from js2-branch into 1.19-trunk and new features aren't backported to 1.18.
* Optimizing options/default construction to just "options = $.extend( {}, options )". Caching defaultOptions is cool, but doesn't really work if it's in a context/instance local variable. Moved it up to the module closure var statements, now it's static across all instances.
* In makeSuggestionBox(): Fixing invalid html fragments passed to jQuery that fail in IE. Shortcuts (like '<foo>' and '<foo/>') are only allowed for createElement triggers, not when creating longer fragments with content and/or attributes which are created through innerHTML, in the latter case the HTML must be completely valid and is not auto-corrected by IE.
* Using more jQuery chaining where possible.
* In buildRegex(): Using $.map with join( '|' ), (rather than $.each with += '|'; and substr).
* Storing the init instance of mw.ajaxCategories in mw.page for reference (rather than local/anonymous).
* Applied some best practices and "write testable code"
** Moved some of the functions created on the fly and assigned to 'this' into prototype (reference is cheaper)
** Making sure at least all 'do', 'set' and/or 'prototype' functions have a return value. Even if it's just a simple boolean true or context/this for chain-ability.
** Rewrote confirmEdit( .., .., .., ) as a prototyped method named "doConfirmEdit" which takes a single props-object with named valuas as argument, instead of list with 8 arguments.
* Removed trailing whitespace and other minor fixes to comply with the code conventions.
** Removed space between function name and caller: "foo ()" => foo())
** Changing "someArray.indexOf() + 1" into "someArr.indexOf() !== -1". We want a Boolean here, not a Number.
** Renamed all underscore-variables to non-underscore variants.
== Bug fixes ==
* When adding a category that is not already on the page as-is but of which the clean() version is already on the page, the script would fail. Fixed it by moving the checks up in handleCategoryAdd() and making sure that createCatLink() actually returned something.
* confirmEdit() wasn't working properly and had unused code (such as submitButton), removed hidden prepending to #catlinks, no need to, it can be dialog'ed directly from the jQuery object without being somewhere in the document.
* in doConfirmEdit() in submitFunction() and multiEdit: Clearing the input field after adding a category, so that when another category is being added it doesn't start with the previous value which is not allowed to be added again...
Fix 19052 which was only reporting the issue for U+3000 IDEOGRAPHIC SPACE.
Covers both external links and images links. See parser tests for examples.
Unicode 'Zs' includes all characters from the 'separator, space' category.
Characters part of this category are:
Char Name
U+0020 SPACE
U+00A0 NO-BREAK SPACE
U+1680 OGHAM SPACE MARK
U+180E MONGOLIAN VOWEL SEPARATOR
U+2000 EN QUAD
U+2001 EM QUAD
U+2002 EN SPACE
U+2003 EM SPACE
U+2004 THREE-PER-EM SPACE
U+2005 FOUR-PER-EM SPACE
U+2006 SIX-PER-EM SPACE
U+2007 FIGURE SPACE
U+2008 PUNCTUATION SPACE
U+2009 THIN SPACE
U+200A HAIR SPACE
U+202F NARROW NO-BREAK SPACE
U+205F MEDIUM MATHEMATICAL SPACE
U+3000 IDEOGRAPHIC SPACE
TEST PLAN:
$ php parserTests.php --quiet
This is MediaWiki version 1.19alpha (r93258).
Reading tests from "tests/parser/parserTests.txt"...
Reading tests from "tests/parser/extraParserTests.txt"...
Reading tests from "../mwexts/LabeledSectionTransclusion/lstParserTests.txt"...
Passed 686 of 686 tests (100%)... ALL TESTS PASSED!
Sounds good :-)
* fix Fatal error: Call to undefined method CologneBlueTemplate::makeExternalLink()
* Use static Linker functions for all other link functions
CologneBlue: check if browseLinks is an array (in a rare case it isn't, which returns a Warning, see e.g. r91432#c19211)
Patch by Aryeh Gregor, updated by Roan Kattouw, and updated again by me. I also fixed one bug with modern.css.
Tested in IE6,7,8, Chrome & FF in all skins and both LTR and RTL contexts. I tested with floating images above and below the headers and couldn't find regressions.
An existing db is not needed, since it uses a memory one.
Having this in Database group fails in DatabaseSqlite->listTables
since it is working with a MockDatabaseSqlite whose query() return true.
* Put OutputPage::$mIsArticleRelated declaration just below $mIsArticle's one since they are related
* Made DifferenceEngine consistent with action=view, i.e. setting that flag to true when displaying any version of the page, current or not
* Made Skin::getCopyright() always use history_copyright when displaying an old version, regardless to the diff parameter
* Changed some checks from namespace != NS_SPECIAL && action=view to isArticle() so that they don't get executed when e.g. throwing an Exception (or similar things)
* Fix a bug in rNNNNN where URLs like '/wiki/Foo' weren't expanded completely if $wgServer was protocol-relative. This caused bug NNNNN.
* Add an optional second parameter to wfExpandUrl(), which takes one the PROT_* constants. This allows the caller to determine which protocol should be used if the given URL is protocol-relative, or the given URL is domain-relative but $wgServer is protocol-relative. The options are PROT_HTTP (use http), PROT_HTTPS (use https), PROT_RELATIVE (keep the URL as protocol-relative), and PROT_CURRENT (use http if the current request is http, or https if the current request is https; this is the default).
* Factor the protocol/port detection part of WebRequest::detectServer() out into detectProtocolAndStdPort(), and add detectProtocol() as a wrapper. The latter is used by wfExpandUrl() in PROT_CURRENT mode.
* Rewrite the test suite to test all possible combinations of $wgServer, $defaultProto, $url and HTTP/HTTPS mode. This means the test suite now has 120 test cases rather than 4.
Front-end:
* New mw.loader method: addSource(). Call with two arguments or an object as first argument for multiple registrations
* New property in module registry: "source". Optional for local modules (falls back to 'local'). When loading/using one or more modules, the worker will group the request by source and make separate requests to the sources as needed.
* Re-arranging object properties in mw.loader.register to match the same order all other code parts use.
* Adding documentation for 'source' and where missing updating it to include 'group' as well.
* Refactor of mw.loader.work() by Roan Kattouw and Timo Tijhof:'
-- Additional splitting layer by source (in addition to splitting by group), renamed 'groups' to 'splits'
-- Clean up of the loop, and removing a no longer needed loop after the for-in-loop
-- Much more function documentation in mw.loader.work()
-- Moved caching of wgResourceLoaderMaxQueryLength out of the loop and renamed 'limit' to 'maxQueryLength
Back-end changed provided through patch by Roan Kattouw (to avoid broken code between commits):
* New method in ResourceLoader: addSource(). During construction of ResourceLoader this will be called by default for 'local' with loadScript property set to $wgLoadScript. Additional sources can be registered through $wgResourceLoaderSources (empty array by default)
* Calling mw.loader.addSource from the startup module
* Passing source to mw.loader.register from startup module
* Some new static helper methods
Use:
* By default nothing should change in core, all modules simply default to 'local'. This info originates from the getSource()-method of the ResourceLoaderModule class, which is inherited to all core ResourceLoaderModule-implementations (none override it)
* Third-party users and/or extensions can create new classes extending ResourceLoaderModule, re-implementing the getSource-method to return something else.
Basic example:
$wgResourceLoaderSources['mywiki'] = array( 'loadScript' => 'http://example.org/w/load.php' );
class MyCentralWikiModule extends ResourceLoaderModule {
function getSource(){
return 'mywiki';
}
}
$wgResourceModules['cool.stuff'] => array( 'class' => 'MyCentralWikiModule' );
More complicated example
// imagine some stuff with a ForeignGadgetRepo class, putting stuff in $wgResourceLoaderSources in the __construct() method
class ForeignGadgetRepoGadget extends ResourceLoaderModule {
function getSource(){
return $this->source;
}
}
Loading:
Loading is completely transparent, stuff like $wgOut->addModules() or mw.loader.loader/using both take it as any other module and load from the right source accordingly.
--
This commit is part of the ResourceLoader 2 project.
