* Mostly whitespace in callers: $('foo').bar(baz,quux) => $( 'foo' ).bar( baz, quux )
* Also several occurrences of mixes spaces and tabs in the indention in front of a line, converted to tabs.
* And double spaces -> single spaces at random.
- jQuery changed to $ in some files because there is a closure that creates a locally scoped $, but the jQuery var is globally scoped, meaning using jQuery instead of $ inside that closure could result in interacting with a different instance of jQuery than the uses of $ in that same closure.
- In mwExtension wrap the code inside a closure which it is missing. Also take this chance to fix the whitespace style `fn( arg )` instead of `fn(arg)` on the isArray I added.
This is partially a followup to r94331.
Note: The jquery plugins inside the jquery/ folder look fine for use of jQuery within closures, except for mockjax.
Patch from mybugs.mail
If a link is added inside of a collapsible toggle and a user click on it, the
target page should be opened.
…
The attached patch was tested locally and fixes this bug (clicking outside the
link toogles the collapsed content, and clicking on the link opens the target
page).
PS: It is the same code which was used to fix the problem on Commons.
http://commons.wikimedia.org/w/index.php?diff=45383826&oldid=45265712
* Pre 1.6, jQuery returned an empty string for attributes that were not set. Although in a way that was wrong, it was the way it was. From jQuery documentation: "As of jQuery 1.6, the .attr() method returns <code>undefined</code> for attributes that have not been set." [1]
** Fixing makeCollapsible by removing empty-string checks and casting to boolean instead
* Wrapped a long line
[1] http://api.jquery.com/attr/
(jQuery 1.6.1 upgrade was in r89866)
* Check for validness of $defaultToggle only needed once (Follow-up r83309 CR)
* The new jQuery sortable plugin caused a minor breakage in other actions bound within table cells. Since, in contrary to the old wikibits script, the entire <th> is clickable and the triangle is a background image. This means anything that is clickable inside the <th> (such as [<a>Collapse</a>] propagates/bubbles up to the <th> and causes the column to be re-sorted. In other words, collapsing a table caused the last column to be sorted, and when expanded again in the opposite order. Fixed by adding event.stopPropagation(); (note, not "return false" as there may be custom events bound to the anchor tag which should be fired as well. We just need to stop propagation not all other event, such as for customToggles).
* The instantHide implementation (added in r82471) didn't cover the case where a table is collapsed by default, it needs to pass the toggle to the function so that the row it is in can be excluded.
* The bug was found by user Helder.wiki, also reproducable on the demonstration page (I guess it was cached for everyone, since nobody noticed?)
* Demonstration page has been refreshed and all tests are passed now.
The problem is that when one of the toggles is clicked the classname is only changed on the toggle that was clicked. The others will stil have the wrong class. On of the areas in which this may lead to weird behaviour is the :hover styles for the up- and downarrow cursors.
Previously would bind to all of them :
<pre>
= .find( '> .mw-collapsible-toggle' );
// ^ may match multiple ones
</pre>
... which is good. But in the click handler:
<pre>
[..].bind( 'click.mw-collapse', function( e ){
toggleLinkPremade( this, e );
} );
</pre>
... only called the function with 'this'. This has been changed to pass all of instead.
This problem was discovered in the following scenario. Where an .mw-collapsible has three children. The first two are clickable and are the togglers for the third child.
http://translatewiki.net/w/i.php?title=Sandbox&oldid=2692006
* Bug fixed in which a single customtoggle should trigger multiple customcollapsible elements
If no .mw-collapsible-content element is found when the toggleElement() function is called, assume that the user wants the mw-collapsible itself to be toggled. This enables custom use with table rows (in which we dont want something inside to toggle but the entire table row) - with a remote toggle.
This change makes it possible to have a togglable table-row.
The intial "collapsed" state for a mw-customtoggle is no longer executed by performing a click on the mw-customtoggle. Instead by calling the toggle function directly.
Reason being that an mw-customtoggle could have any number of special functions bound to it (be it the function to toggle another mw-customcollapsible, or perhaps a totally different function).
I considered using this method for 'normal' default togglers as well. However default togglers have more actions involved aside from just 'toggling' the element. Namely changing the innerText to the localized message and swapping classes. Those stay as they are by simply simulating a click() event on them (if the initial state should be collapsed).
* Fixed JSLint errors
* Added support for custom collapsibles that can have togglelink(s) anywhere in the document
* To prevent people from making secret toggle links (vandalism?) for things like '#content', prefix 'mw-customcollapsible-' is required
* Merged a few comments (bit less verbose)
* Adding extra check when fixing li-value. Value-attribute of an ordered list items defaults to '-1' instead of '' in Mozilla Firefox 3.
* Setting toggle function in a variable to easier reuse it through the plugin
* Replace "-hide" and "-show" classes with "expanded" and "collapsed" respectively
* For more flexibility, if there's a toggle-element already, bind the function to it and use it instead. Otherwise, just create one as usual. See [[betawiki:User:Krinkle/CollapsingTestpageKr#footer|CollapsingTestpageKr#Combination example]]
** It's more likely that the source has a custom position of the togglelink waiting for it to be bound, then it to be already bound.
* Some comments improved
* Message key "hide" to "collapsible-collapse" (Follow-up r78915 CR)
* Message key "show" to "collapsible-expand"
* Changed toggle-link content filling from html() to text() (message errors like <keyname> caused an HTML-element to be created)
