On Sat, 2003-09-06 at 04:48, Owen Taylor wrote: > On Fri, 2003-09-05 at 18:16, James H. Cloos Jr. wrote: > > >>>>> "Owen" == Owen Taylor <otaylor redhat com> writes: > > > > >> Now, if I want to render the composite character 4001+4010, how > > >> should I proceed? Is there a way to map unicode sequences to > > >> actual (physical) fonts. Prefarably in the form: > > > > Owen> Not really sure what you are asking - maybe you can be less > > Owen> hypothetical? > > > > He wants to be able to use a font like the muleiap-1 fonts to display > > such combining sequences. > > > > (muleipa-1 also has tilda'ed versions of latin small letter alpha, > > latin small letter open o and latin small ligature oe that are also > > not precomposed in unicode.) > > > > (NB that muleipa-1 was just the first example I thought of; it is > > probably not the one the OP is using, merely similar.) > > There is no problem with these type of thing. Generally, what you > do with TrueType fonts, is that you simply don't put such glyphs > in the encoding vector. It's also possible to put them in the > private use area. Ok. Once I _have_ those glyphs embedded into a font and numbered them in a non-conflicting way in a private use area (may be this is not really necessary), how can I go about mapping unicode sequences to those glyphs. Do I have to write a shaper myself, or is there a generic shaper at X/Gtk/pango layers which can be configured using some kind of map files? Thanks for the quick replies. Anuradha
Attachment:
signature.asc
Description: This is a digitally signed message part