httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
Subject Re: Patch review: Re: [PATCH] hook based filters take 4
Date Tue, 27 Jun 2000 04:39:43 GMT

> Disabling content-length just because there are filters in the stream
> is a blatant cop-out.  If you have to do that then the design is wrong.
> At the very least the HTTP filter/buff should be capable of discovering
> whether it knows the content length by examing whether it has the whole
> response in buffer (or fd) before it sends out the headers.
> Of course, it is also possible that the three of us are all thinking
> of wildly different architectures from parallel universes and all the
> argument is simply because we have no shared whiteboard to reach a
> common understanding of the general model we are trying to implement.
> But I still think we are making more progress than last week.

We are definately making more progress.  I am beginning to think though
that we aren't making much more progress.  I need to step back from this
for a while.  I am -1 for putting ANY patch in at all right now.  I am
also beginning to think that we may need to re-implement some pool
functions in order to do this at all properly.

I am fixing my current patch, and I will post it because I would like to
show people that it works.  I am -1 for using it, other than as a
reference for later implementations.  I think that there are some good
things in this patch now that it is working properly (Greg, you were
correct the old patch was wrong).

I am -1 for using Greg's patch for all the reasons I dislike my patch.  If
we can find some way to do it all properly, then cool.  If not, I say
ignore layering and leave it for 3.0.

I am not giving up, I'm just stepping back for a day or two.


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

View raw message