httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
Subject Re: [PATCH] filtering and canned error responses
Date Tue, 05 Sep 2000 15:16:56 GMT
> In a message dated 00-09-02 11:30:50 EDT, Ryan writes...
> > What is the problem we are trying to solve with
> > this?  To me, it seems very straightforward that the error docs are put
> > into a bucket brigade with an EOS, because know that the doc is done, and
> > we send it down the filter stack.  This doesn't try to avoid the filter
> > logic we already have and it should work.  Where is the issue?
> An infinite loop.
> If the error being returned is 'Internal Server Error' because of
> some nasty in one of the filters themselves then how does it
> know not to touch the error document sailing through the
> chain that will cause yet another error if the same filter kicks
> in again?
> In the case of generating error pages, I think you are always
> going to need some kind of 'All filters back off this one'
> flag in the request rec or something so it doesn't get
> re-filtered ( and possibly re-looped ) as it goes out the door.

I have actually had more time to think about this whole issue this
weekend.  Here is my general thought.  We have a great filtering system
available to us.  Use it.  If we want to generate an error page, then that
should be a filter.

ap_send_error_response should basically insert a filter that removes all
of the content and generates a new error page.  If more content is sent,
this filter just removes it.

If handled correctly, this could later be used as a model for doing
sub-requests and re-directions as filters.


Ryan Bloom               
406 29th St.
San Francisco, CA 94131

View raw message