Hi jkl,
> You seem to be saying there's some significance to groff's internal
> representation, or lack thereof. Why? I doubt it makes any
> measurable difference. Parsing likely dominates performance in
> languages that execute relatively few loops.
I agree lexing often dominates once attention has been paid to
everything else, but if there's many invocations of macros, and each
macro is re-read byte by byte, then they are together similar to many
iterations around a loop.
If I had time, I'd use
perf stat -e instructions troff -Tps -mbar foo.tr >foo.dit
for some sufficiently complex document and macro set, with and without
stripping, to see if it could be measured.
There's also the open academic question of whether better source
compression can give significant gains. :-)
--
Cheers, Ralph.
