Re:Show HN: my regular expression example match generator
by abecedarius on Friday September 24, 2010 @06:58PM
(#1725715)
Doug McIlroy wrote an article on that: http://www.cs.dartmouth.edu/~doug/nfa.ps.gz
Re:Show HN: my regular expression example match generator
by kenjackson on Friday September 24, 2010 @08:07PM
(#1725873)
Thank you. That is actually an excellent paper. A decade late for me, but excellent.
Re:Show HN: my regular expression example match generator
by rndmcnlly0 on Friday September 24, 2010 @06:12PM
(#1725626)
"Generate all" and "efficient" seems doomed at some level. Considering /[01]+/, there are clearly an exponential number of matches given some given max length.
The expression itself provides a nice summary of the generative space it implies. Perhaps for a particular application you could apply a kind of lazy generation. Say, given /[01]+/ it would first expand it into the subspaces of /0/, /1/, /0[01]+/, and /1[01]+/. One of these branches could be expanded next depending on what was actually needed in the next step.
Re:Show HN: my regular expression example match generator
by elblanco on Saturday September 25, 2010 @12:41AM
(#1726324)
"Higher Order Perl" has a chapter on building one in Perl. http://en.wikipedia.org/wiki/Higher-Order_Perl I believe the full text is available.
I wrote one myself, but was never really satisfied that I did a good job on it. I'd love to see the code of an efficient implementation of such a thing.