RE: [squid-users] squid 2.6-rproxy: assertion store_client.c:211: "sc->cmp_offset == copy_offset"

> -----Original Message-----
> Sure, early adopters might be bitten a bit by bugs caused by changes
> to HEAD, and at times (such as the last months) functionality of
> development version of the patch or even the HEAD version of Squid
> may even be broken, but that is life. This decision is intentional to
> ensure a good progress of the development without forking.

I was thinking of something that tracked it very very closely... not
even stable, just buildable, so as to attract early adopters who don't
know C... but I see your point that if HEAD itself is broken for
extended periods of time then tracking very closely with a stable
version becomes all but impossible.
 
> Correct me if I am wrong, but I strongly suspect that if there was a
> stable version of the patch easily provided for free you would not
> have spent any time on trying to help making the development version
> work?

True.

> And also been very disappointed if later finding that you
> cannot upgrade to a later version of Squid just because nobody has
> bothered upgrading the patches you need.

Well I don't know if I would go that far. That's not too different from
the current state, so it probably would have deferred my efforts into
the future.

On another note, if I submitted a patch to HEAD that changed
READ_AHEAD_GAP to be a config file parameter, would that patch be
accepted?

Thanks,
Jeff

--
NearlyFreeSpeech.NET 
$1/1GB Web Hosting, no minimums, no monthly fees, no kidding.
http://www.nearlyfreespeech.net/