The original link coloring array used 0 for broken links, and an empty() check which would return true for both missing entries and entries which had been stored explicitly as 0.
r27759 switched it to use class names, but the change of the check to !isset() no longer got the explicit case.
11 previously failing test(s) now PASSING! :)
* Piped link to namespace [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
* Link containing % as a single hex sequence interpreted to char [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
* Link containing double-single-quotes '' in text embedded in italics (bug 4598 sanity check) [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
* Nonexistant template [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
* Template with thumb image (with link in description) [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
* BUG 561: {{/Subpage}} [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
* Transclusion of nonexistent MediaWiki message [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
* Say the magic word [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
* Parents of subpages, one level up, not named [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
* RAW magic word [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
* red link from nonexistent article [Fixed between 27-Nov-2007 19:23:59, 1.12alpha (r27890) and 27-Nov-2007 20:42:42, 1.12alpha (r27900)]
Use x option on regex so we can break it up for legibility.
PHP doesn't seem to let you concat multiple strings in a constant definition, which is kind of lame, so this seems to be the cleanest way to break it over lines. :P
* Made it possible to configure the parser class being used, via $wgParserConf.
* Moved defines from the top of Parser.php to either class constants or Defines.php
* Added Parser_DiffTest, a differential parser class for regression testing
* Added Parser_OldPP, a parser class which operates like the parser before this commit. I made one breaking change: a bugfix to avoid losing whitespace when adding MWTEMPLATESECTION markers.
* Made internal tidy work with PHP 5
* Added the ability to supply a hook for template fetching via ParserOptions. This is handy for testing.
* Updated parserTests.txt to account for the various breaking changes I made. Removed a few parser tests that no longer test for anything useful.
* Uses hardcoded magic numbers extensively, which is poor practice
* Adds two hooks with no documentation
* Dropping $class unencoded into the HTML output feels like bad practice to me
* A link-by-link coloring plugin sounds like it could be very expensive to begin with; I'm a bit leery of adding in such overhead.
* Removed the hardcoding of the link parameter ($1), let the message control the destination.
* Removed the hardcoded talk page links. They can be easily re-added if desired, though, and added per-wiki.
* Removed the $3 and $4 parameters
* Fixed wfEscapeWikiText() to do "]" as well as "[", so sigs with "]" won't break links
* Updated a parser test case for the escaping change
The parser now sets a "vary-revision" flag on the ParserOutput when using {{REVISIONID}}; Article then reparses the page after insertion if and only if required.
*Default signatures can be changed in [[MediaWiki:Signature]] and [[MediaWiki:Signature-ip]]
* Signatures for anonymous users link to Special:Contributions page rather than user page
* Adds five undocumented hooks
* Unclear purpose for change to SpecialPage
* Big search-and-replace in EditPage for title usage which seems to still leave things inconsistent; a bit confused what all that's mean to be doing.
Giving this a |nowiki option to wrap the returned path in <nowiki>s for non-linked standalone use.
Added File::getFullURL() function for cleaner calls; using full path to ensure consistency here.
We could probably all benefit from implementing a general rule *not* to call random non-static functions statically. This style is very vulnerable to breakage on later refactoring.
The height had been simply dropped off the face of the earth, with only the width taken into account anymore.
The infrastructure for parameters doesn't seem to gracefully handle multiple values coming from one magic word match, so this is an ugly hack.
* Split "is parsing" and "respect bad images" concepts.
* Call media handler parser hook
* Use the $linkAttribs parameter instead of putting an <a> tag around the whole media output.
* JavaScript video player based loosely on Greg Maxwell's player
* Image page text snippet customisation
* Abstraction of transform parameters in the parser. Introduced Linker::makeImageLink2().
* Made canRender(), mustRender() depend on file, not just on handler. Moved width=0, height=0 checking to ImageHandler::canRender(), since audio streams have width=height=0 but should be rendered.
Also:
* Automatic upgrade for oldimage rows on image page view, allows media handler selection based on oi_*_mime
* oi_*_mime unconditionally referenced, REQUIRES SCHEMA UPGRADE
* Don't destroy file info for missing files on upgrade
* Simple, centralised extension message file handling
* Made MessageCache::loadAllMessages non-static, optimised for repeated-call case due to abuse in User.php
* Support for lightweight parser output hooks, with callback whitelist for security
* Moved Linker::formatSize() to Language, to join the new formatTimePeriod() and formatBitrate()
* Introduced MagicWordArray, regex capture trick requires that magic word IDs DO NOT CONTAIN HYPHENS.
* Clean up the mess in ParserOutput
* Reject (ignore) custom titles which don't normalise to the same as the current page -- THIS IS IMPORTANT OTHERWISE LINKING GOES TO POT (and not the good kind of pot)
[WARNING: Touches parser version. Old caches will be expired. You might wish to undo this and add some temporarily backwards-compatibility for a few days.]
{{CURRENT*}} functions now consistently use UTC as intended, while
{{LOCAL*}} functions return local time per server config or $wgLocaltimezone.
Signature dates for Japanese and other languages including weekday now show
the correct day to match the rest of the time in local time.
