IMO, just make sure your BNF produces a parser that parses all the inputs correctly and then use parens to weed out the uninteresting parts as needed.
On Thu, Dec 22, 2016 at 5:38 PM, hovercraft-google <[email protected]> wrote: > Thank you very much for your attention and references. Actually, I already > saw and continue to study some of it. In essence, what I am looking for is > a most efficient development tool for DSLs in several application areas. > Specifics is in that these kinds of DSL doesn't require strict parsing of > entire input. Instead, they must discover a known distinctive pattern, like > a global structure of document, and extract only pieces of information > based on it. The simple pattern search is not helpful there. I need the BNF > grammar, but it should be able to bypass areas of 'not interesting' input. > Naturally, writing the strict grammar for those uninteresting parts would > be an overkill. > My current approach to do it with Marpa: I just make Marpa believe it > parses entire the input by substituting simplified dummy parts of grammar > for uninteresting places, and I find it more elegant than making perl code > exits. > But I think, Marpa can have special means to work with such a grammars. > Regards, > Al > > -- > 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/d/optout. > -- 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/d/optout.
