On Thu, 17 Aug 2000, Travis Bemann wrote:
> > > English.  Far better to support Unicode from the outset than have to
> > > kludge it in later.  It's not as if it's difficult.
> 
> We could always use UTF8 (read: ASCII). ;)

OK - Let's specify UTF8.
(The other option was \uXXXX as used in Java - this makes the files
human-readable).


So, how about this then, everyone:

MetaData is in '<name>=<value>' format (spaces allowed before and after
'=' or not?), encoded in UTF-8.  Lines are delimited by a single '\n'
character.  So far we have

  Content-Type
  Content-Encoding

specified as having meaning in MetaData.  Example:

Content-Type=text/plain


Steve


_______________________________________________
Freenet-dev mailing list
Freenet-dev at lists.sourceforge.net
http://lists.sourceforge.net/mailman/listinfo/freenet-dev

Reply via email to