* Made autoloader work for unserialize()
* Made XmlFunctions.php and HttpFunctions.php autoloadable modules, via Http:: and Xml::
* Made Image class autoloadable, global functions moved to ImageFunctions.php where they can be loaded on every invocation.
* Removed some unnecessary require_once() calls
* Deferred $wgValidSkinNames initialisation
* Fixed a couple of silly construct/initialise splits. My idea from C++ experience, bad practice in PHP.
* Deferred skin initialisation in ParserOptions
* Removed $wgMwRedir during an attempt to make MagicWord autoloadable. Didn't complete that, but removing the global is still the right direction.
* Use mysql_real_escape_string instead of addslashes for string escaping in
the MySQL Database class. This may fix some rare breakage with binary fields.
Note that MediaWiki does not support the multibyte character sets where a
"dumb" byte replacement can be actively dangerous; UTF-8 is always safe
in this regard due to the bit patterns which make head and tail bytes
distinct.
* remove require_once() throughout whole code, yet left in few places
* move global functions in HttpUtils, ProxyTools, Credits to class methods
* php5 only: __autoload() now used, combined with class->file map and require()
* move initialization of $wgValidSkinNames to Skin::getSkinNames()
* few more changes that will surely break stuff.
* Proper handling of LIST_OR in makeList()
LinkBatch.php:
* Added isEmpty() and getSize() functions.
Query.php:
* Implemented double-redirects query (disabled until caching is done)
* Implemented backlinks query.
* Rewrote page info generator to allow both titles and pageIds to be specified.
* Security check for read-access.
* Output filtering prevents exposure of the internal data.
* Added transactional consistency (not sure its needed here)
* Ended the role of $wgLinkCache in link updates. Instead, links (and related entities) are registered in the ParserOutput object during a parse. The LinksUpdate constructor now takes a ParserOutput object as a parameter. $wgLinkCache is still used, but only as a cache of article IDs.
* Because the link list is now saved and restored in the parser cache, meta tag keywords now work on parser cache hits. Some refactoring took place in this area.
* Rendering of the HTML for category links has moved from Parser to OutputPage.
* Did some general pottering around in Article.php, such as allowing an Article object to be created with a specified revision ID, thereby optionally removing the dependence on $wgRequest. Not used at the current time.
* A few documentation tweaks.
- treat NULL properly in watchlist lookup for notification sending
- consolidate a lot of ugly code that fiddles in those tables
- use user_newtalk consistently in enotif mode; watchlist for email notifications, user_newtalk for on-screen message and status check
- clean up handling of other peoples' user_talk pages when in the talk-only limited enotif: only your own will get sent, not other people watching your page
- and others watching your page _will_ work in watchable enotif mode
- add a watch on page + talk page consistently for the auto-add of the user talk page, not just half the page
- don't unwatch the user talk page on view! that's just wacky
- removed UserTalkUpdate, now redundant
- have User::setNewTalk() apply immediately
- clear newtalk from User::clearNotification() and User::clearAllNotifications()
NOTE: Enabling this may break existing wikis, and still doesn't
work for all Unicode characters due to MySQL limitations.
The $wgDBmysql5 option will run 'SET NAMES utf8' on connection to the
server so it knows we're speaking UTF-8 to it; switching it on/off
can cause interpretation of non-blob columns to change depending
on the charset options set on the server.
