I currently use a program to create correct joining letters for handwriting script that uses an OT font with all necessary glyphs. I have embarked on a project to have a 'font-only' solution by incorporating 'features'. But it has thrown up a few fundamental questions I need answering so I can understand the logical sequence of how features work.
I established that 'calt' feature is the most suitable for Contextual Alternates, but it triggers a replacement only of the first instance of a marked letter. As in the case of double s in the word 'session'....
http://www.clubtype.co.uk/fontlab/sessions.gifQuestions are:
Is 'calt' the only feature that is necessary to get the desired result?
If there are more features required, which are they?
Should one feature come before another?
If more than one feature IS required, then how do I ensure they get turned on by Default when the font is installed?
Basic joins (which don't include s to s)...
feature calt { # contextual alternate
#----------all first-to-second letter joins (apart from s to s)
sub a' [ a b c d f g h i j k l m n o p q r t u v w y ] by uniE002;
sub b' [ a b c d f g h i j k l m n o p q r t u v w y ] by uniE003;
sub c' [ a b c d f g h i j k l m n o p q r t u v w y ] by uniE004;
#etc
#--------initial s to a-z
sub s' [ a b c d f g h i j k l m n o p q r t u v w y ] by uniE014;
#etc
# ---------half s at end of word to space (this works fine)
sub s' [space] by uniE732;
} calt;
Whatever I do to make a double s does not work. See 'session' graphic third example.
#---------first half s to next s
sub s' [ s ] by uniE48D;
Neither can I affect the join from first s to e.
#---------
sub s' [uniE268] by uniE48D;
Should I be using lookups like this article suggests...?
http://www.typophile.com/node/3788How would that look?
I'd be most grateful for expert advice.
Thanks,
Adrian
