There may be a coding error; the code used for ä seems to be the UTF-8 code for ‰, as it appears from this web page (third column, second column being Nokia S80):
ä 0xE4 0xC3A4 ‰ 0x2030 0xE280B0
According to this web page, http://bytes.com/topic/postgresql/answers/173340-codepage-win1252, 0xE280B0 is also a WIN1252 code, but not for ä
Or it may be a secret Illy code.
In any case, is it not possible to reverse the code, and get something reasonable, such as this?
"ä" -> 0xE280B0 -> ae
"ö" -> 0xCB86 -> oe
"ü" -> 0xC2b8 -> ue
Thank you for the reply! Converting to something reasonable was my intention, yes. I just wanted to do that systematically. This is best possible if I know the encoding of these layer names. I received files from different customers, and they were not encoded the same way. Some were encoded in a Latin1 way and some were encoded in the way I explained above . So your assumtion of some secret illy encoding may be right. Sadly enough the documentation of the AI file format from Adobe is really poor.
You are welcome, stoeffu. Please let us hear your findings and the outcome.
I think it's the old 16bit int codes used pre-Unicode. Scott seems to have figured it out...
You can of course change layer names at any time. Just double-click on a layer name and alter it in the dialogue box. For example oe instead of ö etc.