Brian,
On Fri, Oct 22, 2004 at 09:00:27AM -0400, Brian Hechinger wrote:
> > > ok, i will volunteer to do the caching portion.
> >
> > Good! We need developer power.
>
> and i need more experience. this'll be good for me. ;)
>
> > Read the docs, and when in the code, after the normal path,
> > follow the End2End reload flag. You can also look how it is
> > handled with plugins.
>
> i will do that today, i'll let you know what i can figure out on my own.
>
> > Beware: don't dig inside the CCC or you'll ge lost. I'm
> > preparing a document on it, and without this material, it'd be
> > almost impossible to figure it out.
>
> ok, then i'll be very careful to avoid it for now. ;)
>
> > It looks like a good idea to start with. Flexible and simple.
>
> that's what i was striving for. no need to add extra complexity for no good
> reason. ;)
>
> > Caching is complex. Pragma inside meta whithin HTML is just one
> > way. HTTP has several others (for instance you'll have to start
> > reading at section 13 of rfc2068.txt (aka HTTP1.1).
>
> ok, in that case i propse to add Pragma first and once i get that working i'll
> add the other methods as well.
>
> > To make things more interesting, if an RFC-compliant user agent
> > (browser) is built, then it's very easy to abuse it. By sending
> > no-cache directives for advertising for instance.
>
> hence the need for cacherc so that we can specify which servers we will honor
> the no cache directive from. this is something we need to certain things to
> work right, but like you said, we don't want to get abused by it either.
>
> > I hope this gets you started. When in doubt, please ask.
>
> i still have to go through End2End, but real quick, how would i go about
> parsing the header looking for Pragma? what's the best place to do that?
Look inside Cache_parse_header(). :-)
--
Cheers
Jorge.-
