Re: some notes and help to think is needed + Test results of testing basics in store_url_rewrite.

From: Eliezer Croitoru <eliezer_at_ngtech.co.il>
Date: Sat, 06 Oct 2012 14:24:14 +0200

On 10/6/2012 2:08 PM, Amos Jeffries wrote:
>
> ?? sure we are looking at the same Line-506 of src/client_side_reply.cc
> in trunk ??
>
> " switch (varyEvaluateMatch(e, r)) {"
>
well it seems one of us have caching problem? or what?
my 506 is:
"if (strcmp(e->mem_obj->url, urlCanonical(r)) != 0) {"

I also verified using https to make sure I dont have any regular cache
problems:
https://bazaar.launchpad.net/~squid/squid/3-trunk/view/head:/src/client_side_reply.cc#L506

"if (strcmp(e->mem_obj->url, urlCanonical(r)) != 0) {"

I just wanted to post that as I was thinking it is possible to cache
with my method and it works quite well.

I managed to cache sites by just adding additional check to make sure
that we check a url mismatch only when there wasnt done any store_url.

I think it's the best method with no store_url written all over it.
the method assumes that if someone uses store_url_rewrite he knows what
he is doing and not just doing some copy && paste.

It seems to me like a very straight forward way to implement it:
store in the mem_obj->request->store_url

in the state of hexdigest to get the public hash and set the public hash
if there is a store_url by the store_url.

then later just to blow off the url mismatch check (before the Vary
check) that will never ever suppose to match in store_url_rewrite cases.

Now I just need to update my source to the latest maybe 3.3 ? add the
needed code, review the patch and apply.

Eliezer
Received on Sat Oct 06 2012 - 12:24:34 MDT

This archive was generated by hypermail 2.2.0 : Mon Oct 08 2012 - 12:00:03 MDT