Related articles |
---|
requiring balanced parens in a regexp? petermichaux@gmail.com (Peter Michaux) (2006-11-10) |
Re: requiring balanced parens in a regexp? petermichaux@gmail.com (Peter Michaux) (2006-11-10) |
Re: requiring balanced parens in a regexp? mailbox@dmitry-kazakov.de (Dmitry A. Kazakov) (2006-11-10) |
Re: requiring balanced parens in a regexp? martin@gkc.org.uk (Martin Ward) (2006-11-10) |
Re: requiring balanced parens in a regexp? haberg@math.su.se (2006-11-10) |
Re: requiring balanced parens in a regexp? alexc@TheWorld.com (Alex Colvin) (2006-11-10) |
Re: requiring balanced parens in a regexp? cfc@shell01.TheWorld.com (Chris F Clark) (2006-11-10) |
From: | Alex Colvin <alexc@TheWorld.com> |
Newsgroups: | comp.compilers |
Date: | 10 Nov 2006 18:38:36 -0500 |
Organization: | The World : www.TheWorld.com : Since 1989 |
References: | 06-11-039 06-11-040 |
Keywords: | lex |
Posted-Date: | 10 Nov 2006 18:38:36 EST |
>Thanks for the info. My problem is even bigger as there might be mixed
>quotes with various balance in relation to the parens.
>I think that I will need to write a real tokenizer and parser to get my
>job done properly. I haven't done that before and it is something I
>have always wanted to do but it seems like a *big* task.
You may not really need a tokenizer and parser, but you probably do need
a stack. How about keeping a stack of things that must be matched
(paremtheses, quotes, ...) to get you back to the part you need to test.
Just chew up input characters, pushing and popping this stack as
necessary, and doing your test where it's empty.
--
mac the naïf
Return to the
comp.compilers page.
Search the
comp.compilers archives again.