Faking the ligatures

Primoz Peterlin peterlin@biofiz.mf.uni-lj.si
Wed, 6 Sep 2000 19:29:04 -0400

Thank you for your reply!

On Tue, 5 Sep 2000, Pierre MacKay wrote:

> Xdvi is calling out a font which has been pre-rasterized by
> MakeTeXPK, and which presumably respects the full encoding
> of the Type1 font.  The VPL file fragment you show deliberately
> bypasses the ligature and substitutes f and i.  If you can
> find out where the ligatured glyph is, you can replace the
> two lines
>       (SETCHAR D 102) (COMMENT f)
>       (SETCHAR D 105) (COMMENT i)
> with the single line
>       (SETCHAR D 256) (COMMENT fi ligature)
> and use 257 for the fl in the same way.
> That is, assuming that the glyphs are in ASE locations.
> If not, substitute the correct decimal value
> for the actual location.

I may have misunderstood this, but I believe this wouldn't have helped me
if my Type 1 font doesn't contain the `fi' ligature glyph in the first
place? Or perhaps I did have misunderstood? :)

It is not my problem that fontinst would refuse to use ligatures already
present in the PostScript font file. There are none, and I am quite happy
with them being substituted with a sequence of two letters.

I would like to learn, though, how running gsftopk differs from running
dvips. The latter does seem to be bypassing some steps compared to
gsftopk, doesn't it?

With kind regards,

Primož Peterlin,   In┼ítitut za biofiziko, Med. fakulteta, Univerza v Ljubljani
Lipičeva 2, SI-1000 Ljubljana, Slovenija.  primoz.peterlin@biofiz.mf.uni-lj.si
Tel: +386-1-5437632, fax: +386-1-4315127, http://sizif.mf.uni-lj.si/~peterlin/