On Wed, May 04, 2005 at 02:50:22PM +0100, Martin Michlmayr wrote: > Package: magicfilter > Version: 1.2-58 > Severity: normal > > It seems that magicfilter doesn't deal with files which are encoded in > UTF-8.
Yes. Magicfilter at the moment doesn't handle text encodings universally. It is in the printer files. > When I print a simple file consisting of the text "test ä", > the "ä" is printed as two characters; i.e. it seems that magicfilter > assumes the file is Latin-1 and interprets the 2-byte string as two > characters instead of one. Supporting UTF-8 files which are basically > Latin-1 should be easy because you could simply convert them with > iconv, but it would be nice if magicfilter had full UTF-8 support for > text files, so signs such as ??? (bullet) and ??? (heart) would be printed > correctly. Yes, this would be nice. In my opinion this should be done by the text output printing program, in this case enscript, and not by magicfilter itself. You could of course add an iconv step in between, but this would limit the conversion from say UTF8 to Latin1. > Do you think that's possible? FWIW, I'm using ljet4m-filter but this > shouldn't make any difference. It does. I assume it is a PS printer, so enscript is used for text handling. a2ps is unfortunately not Unicode-ready either (Bug #180236) David
