Description
Like pages is sometimes broken with Hebrew extended names
Test cases
All tests
The title is wrong: Multiple matches for "None...None"
Roman extended name
- Pages names: "extended name" "extended wiki name" "extended page name"
http://nirs.dyndns.org/tests/extended%20wiki%20name?action=LikePages -> ok (except wrong title)
Hebrew extended name with one letter words
Some page are found, some not. It not clear what should be the results.
Hebrew extended names with real words
Seem to work, did not check if all matches appear, its not clear what is considered a match.
Non existing Hebrew page name
http://nirs.dyndns.org/tests/(d7)(a9)(d7)(9d)?action=LikePages - FAIL (backtrace)
Multiple matches for "None...None" ... /Volumes/Home/nir/Projects/moin/tests/MoinMoin/action/LikePages.py in showMatches( pagename=u'\u05e9\u05dd', request=<MoinMoin.request.RequestCGI instance at 0x62300>, start=None, end=None, matches=u'You cannot use <a class="nonexistent" href="/tests/LikePages">LikePages</a> on an extended pagename!', show_count=True )
Note the title and the wrong error message in matches...
Details
MoinMoin Version |
1.3 development |
Discussion
Plan
- Priority: Low
Assigned to: NirSoffer
- Status: fixed in patch-212. Like Pages works now for any page name in any language, by comparing the first and last words, either by wiki word rule, or by sapces. It also uses difflib to get close matches, the definition of this match is not clear.