Language class had a code snippet to verify whether a text is valid
UTF-8 though that could not be used from another place. The snippet use
mb_check_encoding() and fallback to some regex whenever mbstring is not
available.
* introduce StringUtils::isUtf8() which is mostly code moved out of the
language class.
* Enhance regex readability by using an expanded regex (//x)
* Made the regex to recognize longer sequences
* Add some unit tests to the mbstring and the PHP native implementation
* An optional second parameter can be passed to isUtf8() to force the
use of our PHP implementation. This is used for unit testing.
Change-Id: I4cf4dfe2eb02f046db1726f4654ba649e01419f2
Also made file/class documentation more consistent and removed a duplicate comment from SpecialPageFactory.php in SpecialPage.php.
Change-Id: I99dd2de7fe461f2fad4e0bd315ebc2899958a90f
* Split link placeholder/replacement handling into a separate object, LinkHolderArray.
* Remove Title objects from LinkCache, they apparently weren't being used at all. Same unconstrained memory usage as the former $parser->mLinkHolders.
* Introduced ExplodeIterator -- a workalike for explode() which doesn't use a significant amount of memory
* Introduced StringUtils::explode() -- select whether to use the simulated or native explode() depending on how many items there are
* Migrated most instances of explode() in Parser.php to StringUtils::explode()
* Renamed some variables in Parser::doBlockLevels()
* In Parser.php: $fname => __METHOD__, Parser => self/__CLASS__, to support Parser_DiffTest more easily
* Doc update in includes/MessageCache.php for r39412
* MW_TITLECACHE_MAX => Title::CACHE_MAX, nicer name, easier to access from another module