Great, I'll put it on my queue.
On Sat, Jan 25, 2014 at 7:59 PM, Jeffrey Kegler < [email protected]> wrote: > I like the idea. Regular expressions, as parsed using special-purpose > regular expression engines, will always survive, because of the speed. But > a lot of things people do with regular expressions would benefit from more > power. -- jeffrey > > On 01/25/2014 05:06 AM, rns wrote: > > Can Verbal Expressions<https://metacpan.org/pod/Regexp::VerbalExpressions> > (stripped > of syntactic treacle) serve as such nice BNF-ish format? E.g. > > # Create an example of how to test for correctly formed URLs > start_of_line > > > "http" maybe "s" then "://" maybe "www." anything_but " " > > > end_of_line > > This can be extended as needed for advanced Perl regex features > and replace function. > > Such BNF format would also provide nice syntax highlighting hardly > possible with regex'es and, with a superset of Regexp::Common > and pattern reuse functionality turn into a DSL of its own. > > Is MarpaX::Regex::Verbal a thing worth trying? What do you think? > > On Tuesday, January 15, 2013 9:51:25 PM UTC+2, Jeffrey Kegler wrote: >> >> There are a number of projects that would, I think, be quite popular and >> useful but which I simply don't have the cycles to consider doing >> myself. One is a Regex compiler -- a compiler from some nice BNF-ish >> format, to Perl regular expressions. I'd think this could be very >> popular -- it would be very much in the comfort zone of some programmers >> who otherwise would not consider using Marpa. >> >> To be specific, this is another specialized Marpa-to-Perl compiler. The >> compiler would write a Perl regex, and the Perl regex would be what >> actually runs. The value added by Marpa would be that more complex >> regexes could be more quickly and easily written, and the output regex >> could be nicely pretty-printed and commented. >> >> Sometimes not understood is that one thing regular expressions *cannot* >> parse is the representation of a regular expression. Regular >> expressions are defined recursively, but they do not themselves deal >> with recursion. >> >> One way to think of this project is as a Marpa super-superset of >> Regexp::Common, whose functionality could be incorporated. A related >> effort within Perl was the DEFINE predicate for sub-patterns, but DEFINE >> had horrific syntax and AFAIK was little or never used. >> > -- > You received this message because you are subscribed to the Google Groups > "marpa parser" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/groups/opt_out. > > > -- > You received this message because you are subscribed to the Google Groups > "marpa parser" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/groups/opt_out. > -- You received this message because you are subscribed to the Google Groups "marpa parser" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
