* Re: Expert fonts
2003-10-16 7:47 Expert fonts Thomas A.Schmitz
@ 2003-10-16 17:35 ` Adam Lindsay
2003-10-17 8:18 ` Thomas A.Schmitz
2003-10-18 3:35 ` Bruce D'Arcus
1 sibling, 1 reply; 5+ messages in thread
From: Adam Lindsay @ 2003-10-16 17:35 UTC (permalink / raw)
[-- Attachment #1: Type: text/plain, Size: 3058 bytes --]
Hello, Thomas.
I've been experimenting a little bit with this, especially with the
lovely Hoefler fonts on my own Mac. My approach would be to "synthesise"
a new encoding, semi-automatically, based on the fact that most variant
glyphs have glyph names that vary by a suffix.
How comfortable are you with the command line?
How about perl?
The attached experimental script (very primitive--hand-edit to change
parameters!) will take in an afm from a large font, a base encoding, and
a series of suffixes to search through, and output a 256-character
encoding. It's easier to have an example:
The base encoding has:
\eight
\nine
\at
\A
\H
The input suffixes to search (in decreasing precedence order) are:
"alttwo", "altone", "linefinalswash", "oldstyle"
HoeflerTextItalic.afm has (among others):
C -1 ; WX 465 ; N eight ; B 11 -21 451 681 ;
C -1 ; WX 488 ; N nine ; B -39 -225 412 455 ;
C -1 ; WX 764 ; N at ; B 44 -22 737 704 ;
C -1 ; WX 767 ; N A ; B -61 -6 680 704 ;
C -1 ; WX 808 ; N H ; B -33 -6 872 689 ;
C -1 ; WX 570 ; N eightstandard ; B 56 -22 534 700 ;
C -1 ; WX 570 ; N ninestandard ; B 34 -78 547 701 ;
C -1 ; WX 500 ; N eightoldstyle ; B 38 -21 478 682 ;
C -1 ; WX 500 ; N nineoldstyle ; B -23 -226 428 456 ;
C -1 ; WX 351 ; N eightfractiondiasuperior ; B -8 261 317 694 ;
C -1 ; WX 375 ; N ninefractiondiasuperior ; B -25 227 336 694 ;
C -1 ; WX 473 ; N Asmall ; B -25 -6 477 468 ;
C -1 ; WX 625 ; N Hsmall ; B 1 -6 626 456 ;
C -1 ; WX 1077 ; N Aaltone ; B -13 -44 1055 711 ;
C -1 ; WX 1014 ; N Haltone ; B -24 -95 1135 739 ;
C -1 ; WX 1186 ; N Halttwo ; B -14 -95 1307 739 ;
...
The output encoding would be:
\eightoldstyle
\nineoldstyle
\at
\Aaltone
\Halttwo
It kind of works. Of course, you can do this sort of thing by hand (some
examples in my OpenType for context magazine article, website currently
down), but it becomes a bit tedious.
Any thoughts?
I hope to work on this a little more, soon, but it's a busy season for
academics. :/
Disclaimers:
Just a Saturday morning's experiment.
Small caps are treated oddly.
Search is pretty much brute-force.
Absolutely no user-interface whatsoever.
Cheers,
adam
Thomas A.Schmitz said this at Thu, 16 Oct 2003 09:47:57 +0200:
>Expert fonts have been mentioned on this list several times, but I'd
>like to know if anybody could point me to e tutorial/example how to
>make them work in Context. The background of my question: on my Mac, I
>have a nice-looking truetype font "HoeflerText" which has most glyphs
>of an expert font (oldstyle numbers, small caps, additional ligatures
>etc.), and I wonder if it would be possible to split it up into several
>type 1 fonts (via pfaedit) and make them work like an expert family.
>Any suggestions about this?
>Thanks!
>Thomas
--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Adam T. Lindsay atl@comp.lancs.ac.uk
Computing Dept, Lancaster University +44(0)1524/594.537
Lancaster, LA1 4YR, UK Fax:+44(0)1524/593.608
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
[-- Attachment #2: enco.pl --]
[-- Type: application/octet-stream, Size: 933 bytes --]
#!/usr/bin/perl
open(ENC, "texnansi.enc");
#open(ENC, "ec.enc");
open(AFM, "HoeflerTextItalic.afm");
#open(AFM, "AppleChancery.afm");
my @matcher;
my @candidates = ("alttwo","altone","linefinalswash","oldstyle","");
#my @candidates = ("smallalt","small","oldstyle","");
#my @candidates = ("oldstyle",".6",".5",".4",".3",".2",".1","");
#my @candidates = ("swashwililf","swashwili","swashli","swashwi","swash","oldstyle","");
my $count = -1;
while(<ENC>) {
s/%.*$//g;
if (@matcher=/\/(\S+)/go) {
WORD: foreach my $matched (@matcher) {
foreach my $candide (@candidates) {
seek AFM, 0, 0;
while (<AFM>) {
# my $encoentry = $matched;
# if ($candide =~ /small/) {
# $encoentry =~ tr/a-z/A-Z/; }
if (/^C\s.+N\s($matched$candide)\s/g)
{ print "\\$1 % $count\n";
$count++;
next WORD;}
}
}
print "\\.notdef % $count\n";
$count++;
}}
}
close(AFM);
close(ENC);
^ permalink raw reply [flat|nested] 5+ messages in thread