[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Font naming rears its ugly head again



Pierre MacKay writes:
>     In your raw encoding you have left out all the accented letters.
>     There are two important reasons why I think they should be encoded
>     in the raw font:
> 
> This may well be a necessary alternative coding, but it has to
> be an alternative coding.  The reason is that once you stick
> Aacute into the raw (-p flag) encoding, it pre-empts the
> creation of a VPL composite recipe.  
> 
> Adobe Garamond and Caslon may create the accented glyphs
> as simple (non-composite) characters, but the vast majority
> of fonts from the vast majority of vendors use composites.
> 
> Special cases require special files.
> 
> ...
> 
> We must be grateful for the existence of carefully designed
> fonts such as JMR alerts us to, but I suspect we must also
> allow that the vast majority of available fonts will not
> be created with such care.

I don't quite get your point, Pierre.  The way I do it works for _all_
PostScript fonts which have the ``standard'' set of 228 glyphs.  It
makes no difference if the accented letter has a ``seac'' instruction
in its type1 font program (that's what Adobe refers to as a composite
character, and that's what the CC in the AFM files stands for), if it
uses ``callsubr'' instructions to put the base letter and the accent
together or if it's all done inline in the type1 font program.  It's
true that afm2tfm will use the accented letter from the font, if you
put it into the PostScript encoding vector, rather than generating a
VPL composite recipe.  The accented letters from the fonts usually
look better (even if they aren't specially designed, especially at
low resolution) than what you can produce with a VPL composite recipe.
So, I can see no reason not to use the accented letters from the font.
But if you for some special reason don't want to do that you can edit
the VPL font by hand (you'll have to do that anyway if you e.g. want
to position the accents differently) or you can use another PostScript
encoding for that font.

Jan Michael Rynning			Email: jmr@nada.kth.se
Department of Numerical Analysis
  and Computing Science			Voice: +46-8-7906288
Royal Institute of Technology		Fax: +46-8-7900930
S-100 44 Stockholm
Sweden					Normal host: jmr.nada.kth.se