pyragony54 wrote on Jul 31^st, 2014 at 5:38pm:


All my forums are using latest version of YaBB SE which uses MySQL and PHP... Less problems than with flat files and Perl.

And internally Perl uses Unicode.

But we still have a problem with multiple encodings in data. And that's where going to UTF-8 will make sense. The non-Latin-1 encoding is still very much a work in progress.

Monni wrote on Jul 31^st, 2014 at 2:01pm:


In terms of the Language packs - we're in transition. We needed to support old data encodings while transitioning to UTF-8 (Which we're going to need when YaBB develops a version using 'real' databases.) That's what 3.0 was supposed to do but didn't do well.

Our big issue with transitioning is that there's currently no good way to convert encoding on old data that has multiple embedded encodings. Once we have that licked (if it can be) then converting old encodings will become part of the 'convert old boards' process. Admins won't have to worry about 'should' or 'will'.

@ThM - your observations are great - it tells us where our explanation of new features needs to be better.

Guys - It's actually a mystery as to why a mixed encoding should break 'Male' and 'Female' in the vars files. Logically, those should both be simple ASCII and should encode the same in ISO-8859-1 and UTF-8. But obviously that assumption is wrong. (JonB and I went into the encoding issue thinking it would be simple. We now both feel we've entered uncharted territory where every native tells us something different - and they all want us to fall over a cliff.)

It was not my intention to create a problem what maybe does not exists, but I thought it is important to know that there are difficulties here and there, and the user is in an important topic to point out what they like to do.
thanks again ....

@ThM There is two versions of language packs for all languages except English... One for cases where there is only one language pack installed, and one for cases where there is multiple language packs installed. UTF-8 works best when there is multiple language packs with conflicting primary character encoding.

I think we need to make the warning about not mixing character encodings even stronger. 

I do have a YaBB language converter that can convert ISO-8869-1 encoded forums to UTF-8 (and it seems to do an okay job). BUT, mixed encoding forums like YaBB Forum don't do well with automatic conversion utiities - even Notepad++ chokes on some of the files there.