httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Christian Parpart <tra...@gentoo.org>
Subject libapreq within mod_transform (was: Re: ap_unescape_url() and the '+' within an argument)
Date Sun, 29 Jan 2006 00:15:45 GMT
On Sunday 29 January 2006 00:59, Joe Schaefer wrote:
> Christian Parpart <trapni@gentoo.org> writes:
> > On Saturday 28 January 2006 23:36, Dirk-Willem van Gulik wrote:
> >> On Sat, 28 Jan 2006, Christian Parpart wrote:
> >> > wich works so far, however, firefox keeps adding '+' chars instead of
> >> > %20 to
> >>
> >> Which is perfectly OK/right - so we should recognize these and handlte
> >> htem correctly.
> >
> > what exactly did you mean by "handle them correctly"?
> >
> > So you assume this is a mod_transform bug, which (before unesacping the
> > URL query) it needs to translate each '+' char into the right ' ' space
> > then?
>
> Right;  mod_transform really should be using apreq for providing
> access to form data.  It's a good match if you can afford the
> extra dependency.

my problem is not the extra depend on apreq, but what I can't find really 
neat, is, that it seems to be a perl module (based on mod_perl somehow) and 
IIRC, mod_transform shouldn't be overbloated by depends that itself depend on 
too much the actual project (here: mod_transform) doesn't need.

Sorry for being a n00b regarding libapreq, so here my quick question: can you 
build it w/o the mod_perl depend? At least the homepage[1] looks promising in 
its feature set as it (in fact) does what we need right now.

secondly: is libapreq able to retrieve POST arguments from a filter module, 
specifically mod_transform?

If both applies, I'd be willing to try my best in incorporating it then :)

Best regards,
Christian Parpart.

[1] http://httpd.apache.org/apreq/

Mime
View raw message