[boost] Re: Re: Re: Re: Regexp code bloat

From: Eugene Lazutkin (eugene.lazutkin_at_[hidden])
Date: 2002-10-26 15:49:17

"William E. Kempf" <wekempf_at_[hidden]> wrote in message
news:1338.192.168.1.1.1035648554.squirrel_at_frodo.kempf-ville.com...
>
> This is the crux of the matter.  If you'd required full regexp
> functionality, I doubt you could have hand coded it down much below the
> 50K reported.

Most probably you are right.

> But I don't see your argument as having any relevance to the discussion of
> whether or not Boost.RegExp or GRETA are overly bloated.  Unless you can
> prove that such functionality can be provided in less than the reported
> 50K, the only thing we have evidence of is that full RegExp engines are
> not a universal solution for parsing... which everyone should know
> already.

You are right. I am not trying to prove that regexp++/Greta are "overly
bloated" implementing "full regexp functionality". I say that "full regexp
functionality" is not necessary in most cases. If I want to use some simple
subset, I will have "full regexp functionality", which I don't need.

It is not problem with a library per se. It is a problem with approach,
which forces reimplementing the same stuff all over again.

Thanks,

Eugene