httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jim Jagielski <>
Subject What are we trying to solve? (Was: Re: another response to Roy :-))
Date Sat, 01 Jul 2000 13:32:14 GMT
Fielding, Roy wrote:
> It doesn't introduce them -- they are there already.  One of the goals of
> the filter concept was to fix the bird's nest of interconnected side-effect
> conditions that allow buff to perform well without losing the performance.
> That's why there is so much trepidation about anyone messin with 1.3.x buff.
> Your patch does nothing to change that situation.  Bucket brigades are a
> replacement for buff in addition to a solution for filters.

I don't understand this... Although it's true that Greg's patch
maintains some level of BUFF compatibility with 1.3, it also provides
for a bucket approach as well. 
> I know you have buckets, or at least a data type that could one day
> be buckets.  But you aren't using them like I did.  My buckets are passed
> around in list-queues (really just lists with front and back pointers).

But the latest version of Greg's patch appears to be readily
open to implementing various forms of bucket brigades.

> My buckets carry data and metadata and meta-metadata.  My buckets are
> used to indicate stream-end, and the filter configuration itself is
> determined by the stream content.  It probably sounds weird, but the
> effects of this interface are completely different than mere content
> filters.  They simplify everything.

On this, I have to look deeper... Does the actual implementation of
how bucket lists are handled affect being able to accomplish the
above, or does the simple fact of providing a "bucket interface"
do that?

> What can I say?  That is totally unclear from the code.  Yes, I have read
> all of your patches, just as I have read all of Ryan's patches.  You aren't
> solving the same problem that I am trying to solve

I'm gonna plead ignorance here... could someone explain the "problem"?
Maybe I just don't understand the core and key issue that is trying
to be solved. I must be missing something, because I don't see anything
inherently wrong in Greg's design as far as solving the problem
_as I understand it_.

   Jim Jagielski   [|]   [|]
                "Are you suggesting coconuts migrate??"

View raw message