> It seems that Unicode.sfd is very regular and cmap files will be
> easy to generate. Maybe there is no need to generate cmap files if
> just using utf8 fonts with cmap.sty?
It isn't sufficient, unfortunately. Section 5.9.1 (`Mapping Character
Codes to Unicode Values') in the PDF standard version 1.6 explains how
to map glyphs back to characters: You need a ToUnicode cmap if the
font neither uses one of the Adobe standard encodings (like
MacRomanEncoding) nor one of a small set of CMaps (like Adobe-GB1).
Therefore, in case of subfonts, you *need* ToUnicode cmaps. I agree
that the generation of *.cmap files for cmap.sty should be easy (by
reusing the various perl scripts for subfont creation). As I've said,
I'm a happy user of dvipdfmx and too lazy to produce those files for
pdftex -- volunteers welcome!
Werner
_______________________________________________
Cjk maillist - [email protected]
http://lists.ffii.org/mailman/listinfo/cjk
