* Turn normalize data into an autoloadable class and use the class name
as key instead of a partial file name with a base directory.
* This protected function is not aimed at re-use in extensions,
has no other usage known to Codesearch. Mark it internal,
and remove the compat introduced by 31283f34bf.
* Update the now ten-year-old documentation comments that described
this as "Temporary". The functionality and comment were introduced
in 2010 by Tim with r60599 (ad19c032b0).
Ref <https://bash.toolforge.org/search?p=0&q=temporary>.
* Replace the broken wikitech-l archive link with a working one,
and describe the thread for easier future reference.
Bug: T225756
Change-Id: I004da1f25c19e1095d0917349f2ba8461f9f0715
And include tests :)
This code is independent of MediaWiki, but not really large enough to be
worth librarizing right now.
Bug: T200626
Change-Id: I022c074e8a708fb5219bc4ff4b53e7e31f60dc4b
Deprecate the second argument to Maintenance::error() in favor of a new
Maintenance::fatalError() method. This is intended to make it easier to
review flow control in maintenance scripts.
Change-Id: I75699008638f7e99b11210c7bb9e2e131fca7c9e
* Delete includes/compat/normal/.gitignore, which excluded files
now located in various directories within the utfnormal repo.
* Remove "For NFC see includes/compat/normal" comments from
maintenance/language, replacing them with a note that "This
data file is used after normalizing to NFC."
Change-Id: I0ac990ec1a547b4c454ed738ce1dfffce0f53ba4
Instead of littering includes/ with stub back-compat aliases, house such
classes in includes/compat/.
Change-Id: I4c1b83e35c8d6c18777a4a3e17d81023915cfb7f
This drops support for the custom utf8 normal PHP extension in favor
of the intl extension.
Bug: T90825
Change-Id: Ifbaeb2ef684217cf6187ccc4fb4d303f89608300
This allows getting rid of serialized/serialize.php. I also moved
includes/normal/Utf8CaseGenerate.php to maintenance/language/
generateUtf8Case.php and updated it to subclass Maintenance, as
it seems to be largely unrelated to normalization.
Using version 6.0.0 of UnicodeData.txt, the updated script generates
exactly the same serialized output as was previously checked in.
Also updated the Makefile to reflect the current set of .ser files
and added some .gitignore entries.
Change-Id: I05afece3dc4505a9f43993ac4d7726b37d9c6956
2014-01-06 18:22:24 -05:00
Renamed from maintenance/language/generateNormalizerData.php (Browse further)