6
Quote:
By replacing accented characters by html charcaters translators (Kris & Dugris) allows to users to use same language files for ISO site as well for an UTF-8 site.
It doesn't work for all the files. eg mail templates.
There you have to do the conversion yourself by eg Notepad++.
The html encoding is making all texts nearly unreadable, because mosttimes numerical encoding is used (which may lead to translation errors with some character sets) and has resemblence with the stone age of computing, where computers used hexadecimal.
The only proper way for language files would be by supplying two sets of files, one in an utf directory and with UTF-8 (without BOM) encoding for use on UTF-8 websites with UTF-8 databases and one in an iso directory in ansi encoding for use with Latin1 databases and sites with ISO-8859-1 or ISO-8859-15.
A full module (or core) archive should have the UTF-8 files inside and provide separatly the two directories of the language file, for making an overwrite to the desired format as needed.
An example is this Dutch translation of
IndexScan.
