> I guess that printafm program, now bundled with ghostscript, was
> used It prints afm for a Type 1 font.
Aah, excellent.
> When I have run
>
> printafm Symbol-Slanted > symbolsl7.afm
>
> the resulting file had only minor differences with symbolsl.afm from
> groff.
Can you provide a Makefile rule for that, together with a comment?
> But the problem of the unknown reasons behind changes done by
> symbol.sed remains.
Well, these changes are easy to explain! To assure correct stacking
of brackets, parentheses, and braces which are vertically extensible
and thus built up from elements, symbol.sed assures a unified height
and depth -- eqn uses the \n[rst] and \n[rsb] registers to get this
information (cf. src/preproc/eqn/delim.cpp). It also adds
mathematical positioning information for the integral sign.
I've just slightly extended the comment in symbol.sed.
Werner
_______________________________________________
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/listinfo/groff
