Re: [MERGE] Support large response headers

From: Adrian Chadd <adrian@dont-contact.us>
Date: Mon, 7 Apr 2008 16:01:32 +0800

On Mon, Apr 07, 2008, Henrik Nordstrom wrote:
> m??n 2008-04-07 klockan 13:58 +0800 skrev Adrian Chadd:
> > The first snippet of refactoring, which has been committed:
> >
> > http://www.squid-cache.org/Versions/v2/HEAD/changesets/12032.patch
>
> Yes...
>
> > Shuffling the storeClientReadHeaders() stuff out of client_side.c and into
> > store_client.c in preparation to teach all the other store clients about it:
> >
> > http://www.creative.net.au/diffs/20080407-squid-2.HEAD-storeclient-headers-1.diff
> >
> > Comments please!
>
> Not much to say other than there where it's supposed to be.
>
> I wonder why I added the function to client_side.c and not
> store_client.c... but I guess there was some client struct dependency in
> very early versions..

Well, I'm happy to commit this fix to Squid-2.HEAD, and then begin reworking each
of the store client users to use this for fetching reply headers, which each of
them wants to do. This is the "Correct" way of doing things anyway and should be
done before we sit back and think about how to "fix" things, as it makes a lot
of stuff quite a bit easier to do (eg, if we choose now to remap position 0 as
relevant to the reply body rather than incl. headers.)

I'm trying to avoid doing the much simpler solution and ignoring the rest -
ie, just having a local buffer, reading the reply from the server-side into it
until we have all the headers, and then just copying the first 4k out.
Thats a possible stopgap, especially for Squid-2.7, but I don't want that
lingering in the codebase.

Adrian

-- 
- Xenion - http://www.xenion.com.au/ - VPS Hosting - Commercial Squid Support -
- $25/pm entry-level VPSes w/ capped bandwidth charges available in WA -
Received on Mon Apr 07 2008 - 01:44:03 MDT

This archive was generated by hypermail 2.2.0 : Wed Apr 30 2008 - 12:00:07 MDT