Text is not parsed when containing special characters

[Former user]

We run into the following issue: Text from our Confluence instance is parsed perfectly, unless it contains special characters (like german umlauts: ü,ä,ö). If special characters are contained, nothing is parsed.

I've tried to include a character map in the stylesheet or use other ways of escaping the umlauts, but without success.

Replacing the special characters in the source by the html entities (i.e. the confluence page) does not work (and would be a no-go for us anyway, since we cannot tell all our colleagues to use this workaround).