Regular grammars and binary matching performance

[Hynek Vychodil]

Hello,
I have just send new regexdna solution using R14's binary module:
http://shootout.alioth.debian.org/u32q/performance.php?test=regexdna
http://shootout.alioth.debian.org/u64q/performance.php?test=regexdna
It shows how big gain we would have get by limited regular expression
support in binary module. I have written limited regular expression
support in function compile_pattern/1 (very limited and poor man
solution!) as demonstration there
http://shootout.alioth.debian.org/u32q/program.php?test=regexdna&lang=hipe&id=7
for curiosity.
I think it would be nice to have support for regular grammars in
binary module or adapt http://code.google.com/p/re2/ as limited
regular expression library because those limited regular expression
(truly regular grammars, without sub-capturing may be and so) could be
feasible in real projects. For comparison, re module is 15 times
slower in this benchmark
http://shootout.alioth.debian.org/u32q/program.php?test=regexdna&lang=hipe&id=6

Do you think it would be useful? Is anybody working on it already?

Best regards
-- 
--Hynek (Pichi) Vychodil

Analyze your data in minutes. Share your insights instantly. Thrill
your boss.  Be a data hero!
Try GoodData now for free: www.gooddata.com