Last modified: 2005-07-23 04:36:21 UTC
Currently the charinsert extension enters HTML entities for all wikis. But UTF-8
wikis prefer all input to be in UTF-8. Certainly the English Wiktionary does. Another
problem is that searching does not treat HTML entites and UTF-8 plain text the same,
even with the new search engine.
The search discrepancy should be a separate report.
Bug 1036: "Search doesn't handle special characters correctly" already covers the search
discrepancy I mention though it's not very detailed and hasn't received any attention.
Charinsert inserts UTF-8 characters if that's what you give it.
Resolving as WORKSFORME.
Yes it was my mistake. There was no such problem.