httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dean Gaudet <>
Subject Re: [PATCH] yet another slow function
Date Wed, 07 Jan 1998 12:31:21 GMT
On Wed, 7 Jan 1998, Martin Kraemer wrote:

> On Tue, Jan 06, 1998 at 03:49:28PM -0800, Dean Gaudet wrote:
> > On Tue, 6 Jan 1998, Brian Behlendorf wrote:
> > > Couldn't we have safe_path_chars created by that cute program of yours at
> > > startup time?   Wouldn't that be safer if used on other character sets, or
> > > EBCDIC?
> >
> > Yes we could... and actually, consider that a patch in progress.  I'm
> > going to nail a lot more uses of ind.
> Yeah, of course I could #ifdef CHARSET_EBCDIC the generated table; but
> I'd prefer the dynamic version too. Are there any init-hooks for various parts
> of apache? 

Yup I thought about your EBCDIC stuff when I did it... I actually would
much rather generate those tables at compile time by compiling a program
to generate the table and then #including it, but then I thought about NT. 
But this shouldn't be a hard one to do ... unlike the stuff in
regex/Makefile which was hard coded ages ago because NT lacks a real shell
with real tools.

> > strchr() is standard and faster than ind on many boxes... at least on
> > those where the vendor took the time to code it into assembly ;)
> Hmm.. Then why aren't ind() and rind() based on strchr() and strrchr()?
> We don't have to code apache in assembler ;-)

They could be... but still I'd just rather switch to strchr and strrchr. 
Some compilers inline those automatically. 


View raw message