There can be more then one <inputbox> in page. Don't reuse id
attributes.
See also r60072
Bug: 11777
Change-Id: If7c518445c966aa5c0b16929ccdf22cdf014b59d
Add an option to InputBoxes to allow them to prefill Special:MovePage with
prefixes and other similar options to pages being created
Change-Id: I1740497030b5e9872162a1a261ac38791bb1373a
Property name is $mID, certain parts were testing $mId, and then
using isset to (presumably) hide the undefined property warnings.
Change-Id: I0c2d29fbdb8314aa0c04bea387364a7307e68b9b
This prevents possible current or future key conflicts. Keys for
translations will be updated at translatewiki.net.
Change-Id: Ie94fa53340c9534f45a70cbafc4d55a7e2d58a7c
These variables are included in attributes of elements built
with Xml::openElement or Xml:element which escape them using
Sanitizer::encodeAttribute.
Change-Id: Ib34a21a6312ba88b390110115d655b46a64b9dcd
InputBoxes creates HTML which will later be cached (probably).
Right now this can lead to strange button labels eg.
https://www.wikidata.org/wiki/Wikidata:Requests_for_comment?uselang=qqx&action=purge
will cache the qqx of the createarticle button so that all users
see it (till it's purged again). To prevent this we set the parser
to cache per user language.
Change-Id: I69ade88bb43f73404965b7cb63e88b824561d988
* Replace deprecated methods.
* Remove superfluous newlines.
* Fix docs for a few methods.
* Add some type hints.
Change-Id: I09671b5e70d33ffa9746e95c965bd1bc2bbd4a3f
* instead of checking each namespace if it is given through the parameter, check each given namespace if it is a valid one
* this fixes the inconsistency where 2 namespaces are given but only 1 is valid and that one is shown (normally if there is only 1 given, it is hidden)
* fix bug 22591 (Allow i18n of Main namespace search option label)
* allow namespace aliases as well
* also trim the given namespaces, so spaces do not make them invalid
Reuse getSearchForm() for both 'search' and 'fulltext' type, IMO it should also be used for 'search2' (exact-match only), but since 'search2' is kindof strange and ignores some params (like 'break') for which I'm not sure if it's bug or feature, leaving 'search2' as separate function.
Calling it with no extra arguments will now assume that you're escaping
a whole id, not an id fragment, which is safer. Also, instead of ugly
bitfield-based options, I've changed the options to use an array of
strings. I fixed all callers in trunk. Out-of-tree callers that were
using Sanitizer::NONE will get correct behavior, while those that were
calling it with no arguments will get slightly changed behavior (an x
will be prepended). I think this is harmless enough that we can skip
back-compat cruft here.
This should cause no visible changes. No parser test regressions.
