The compatibility functions in GlobalFunctions are just wrapper for
their equivalent in the Fallback class. We should test the implementation
and we can safely ignore those wrappers.
Shell functions ignored make use of sleep() which is evil. They also
do some outputs to the console which is probably hard to test properly.
Given they are not critical, I just ignore their code coverage, we can
still test them though :)
Language code are case insensitive. The BCP 47 recommands nice
formatting nonetheless. This patch enhance our formatting:
- tags preceded by the private tag 'x' are now lower case
- 4 letters tags are now lower case with first letter uper cased
Please note the RFC seems to have a bug for az-Arab-x-AZE-derbend
which should be az-Arab-x-aze-derbend . I have changed our test
to reflect this and added a comment for later reference.
* MessageCache::singleton() calls wfGetMessageCacheStorage() directly instead of using $messageMemc, just in case this would be called before that variable is set
* Per TimStarling: also removed deprecated methods in MessageCache class: addMessages() and related, [get|set|enable|disable]Transform(), loadAllMessages(), loadMessageFile() and some others. Same for the legacyData stuff in LocalisationCache that was only used by MessageCache::addMessages() and related.
* Converted remaining extensions
Additionally, *cough* *cough*:
* Add a Title::isMainPage helper for the fairly common $title->equals( Title::newMainPage() ); test.
* Update wfMessageFallback to also accept an array of message keys instead of requiring them listed as arguments to the function.
* Move the bulk of wfMessageFallback code into Message.php instead of leaving it in GlobalFunctions.php
* Change the wfMessageFallback implementation so that the Message class handles the fallbacks themselves eliminating any side effects caused by the fact that wfEmptyMsg always used usedb=false, language=userlang when one might actually use a different language or usedb setting in the message object that actually returned the text (this may be considered a wfEmptyMsg regression in 1.18).
* Make blank "" message contents fallback like nonexistant messages do.
* Re use the new tabAction array handling used to support mainpage-nstab in the talk and view tabs instead of making wfEmptyMsg calls directly in SkinTemplate.
instantiate a variable-length constructor in php 5.1.3 and up, and falls
back to the old, ugly, manual method that was in the old wfCreateObject
function. The instances in the core have been replaced.
-in_string has a case-insensitive option
-wfClientAcceptsGzip has a force option to force resetting the static value, useful for unit tests
Unit tests for more global functions added
content_actions hooks no longer work and have been dropped from the code, the hooks that affected vector before now affect all skins.
A few logic changes were made to make for a clean merge:
- vector was using vector-???-??? messages while SkinTemplate was using '???' messages. So as a side effect of merging that together all skins now support messages like '$skinname-view-history' which will fallback to the standard message if not defined.
- For MediaWiki: pages where the page does not exist but the message does in the i18n system SkinTemplate displayed "Edit" while Vector displayed "Create"; All skins now display "Edit".
- For users without undelete permissions SkinTemplate displayed an "Undelete" tab if the user had deletedhistory and deletedtext permissions. Vector would only display the tab for users with both deletedhistory and undelete permissions; The new behavior in all skins is to always display a tab if you have deletedhistory (since Special:Undelete will always have something to display) but display a "View ... deleted" instead of "Undelete" message if you do not have undelete permissions.
- Skins no longer need to hardcode tests for the &action= to decide if they should ignore the accesskey on some tabs, tabs which should not have an accesskey in the current page now have a "tooltiponly" key set to true.
Added in wfTimestamp reading support for the three http date formats.
Increased conformance reading rfc2822 dates (read support added in r71750/r71751). We may not want full compliance with rfc2822, though.
The only provider of rfc2822 dates is probably http and that uses a subset (it's rfc822 + 4-digit years from rfc1123).
Make wfTimestamp() return false in case of a wrong input, according to CR.
Follow up r75761. Now calculation is right (passes all tests)
on php 32 bits when using PHP >= 5.2
without prejudice to PHP 32/64 bits < 5.2 on the 32 bit range
(regardless of architecture)
