[TYPO3-50-general] OrthoScript

Elmar Hinz elmar.DOT.hinz at team.MINUS.red.DOT.net
Sun May 13 11:37:12 CEST 2007

Am Sun, 13 May 2007 10:37:01 +0200 schrieb Martin Kutschker:

> I don't get you point. The external XML strings in UTF8 are cached in 
> serialzed arrays for retrieval. Using XML makes it easy to handle the 
> actual trnalsatoin process (= creating and maintanin the text).

The point: If you want to provide translations for your extension you
either have to write PHP array and transform to XML to match the
standards or you have to write XML by hand. If you once have transformed
to XML you are forced to write XML for alterations and extendings.

A don't follow you why XML should improve creation and maintainance of the
translation process. Internally XML or TS have to be evaluated to PHP
arrays to calculate with. For translators IMHO it is much harder to provide
XML than TS. So fewer people will participate if they have to work with
XML. As I am not a translator I don't know how the translation workflow
works in detail.

> What went wrong was that TYPO3 does now a double encoding thing. That's 
> just an implementation problem which can be solved (did that already for 
> the BE), this has nothing to do with XML or UTF-8 per se.

If translations would be kept in UTF-8, there would only be one
transformation UTF-8 > ISO or none at all. Guess that would have matched
with the original one translation process without big alterations.

> What I don't like is the XML data of flexforms in the records. As Mysql 
> doesn't support XML directly I would have used serialized arrays for 
> internal storage. Guess what, I avoid TV just for that reason ;-)

That's the point. With flexforms it's just the same as with the
translations. Move them to a lighter format and it will become more 
joyfull to work with. Clearly the definition of flexforms could be
done in TS format, maybe also the data storage. But I havn't thought about
the latter.



More information about the TYPO3-project-5_0-general mailing list