yep....exactly..... But either way Boost has a pretty good regex
library and that's compiled.

As for someone saying "Regular expressions" are used when there is
weak structure from a computability standpoint, regexes break down to
D/NFAs whereas a parser equates to a PDA and a true semantic checker
requires a turning machine equivalent. Personally I'd say
if you thought it through regexes may get used in parsers when there
is bad grammatical structure but a regex is the easiest, simpelest,
most well defined structure there is.

-Tom

ps that mastering regular expressions book is pretty cool
pps PERL does stuf that aren't even really true regular expressions
like backreferences
ppps I learned PERL from Nat Torkington at Qualcomm so I may be biased.

On Wed, 23 Mar 2005 16:21:26 -0800, Andrew P. Lentvorski, Jr.
<[EMAIL PROTECTED]> wrote:
> 
> On Mar 23, 2005, at 2:43 PM, Brian Deacon wrote:
> 
> > I'm not often in regex world, so somebody probably phrased a better
> > version of this observation already.... but I joked with my friend that
> > regex is an example of the signal to noise ratio being too -high-.
> 
> Actually, I don't agree.  And, it seems, neither do most programmers.
> 
> The big problem is that regexes get used when there is weak structure
> to what you are parsing.  When there is weak structure, there is
> ambiguity.  Where there is ambiguity, there is complexity--eg. headers
> for email.
> 
> When you have structure, just about anything will work.  XML,
> s-expressions, grammar parsers, etc.
> 
> > Anybody know of anything like that?
> 
> flex and bison and other things of their ilk (grammar parsers and
> lexers) would probably be what you describe.
> 
> -a
> 
> --
> [email protected]
> http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
>
-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to