Thanks Alex and Amos,
The Audit gave me a lot!
I hope to sit on it today and if not today later next week.
For now I am using this patch in a testing machine to make sure it does
what needed for a reasonable period of time without bad side effects.
How long it is before committing a patch to trunk?(after all the fixes)
it would need approval of the others and 10 more days?
Thanks,
Eliezer
On 12/6/2012 9:04 AM, Alex Rousskov wrote:
On 12/05/2012 06:20 PM, Amos Jeffries wrote:
- for store_id it is not clear from the name what it is exactly.
"Storage of StoreID for the specific cases that the request not avalile"
leaves me just as much in the dark about what this variable holds.
Agreed, especially since this is a request member so request is
naturally available. This sounds like a comment cut-and-paste error of
some sorts because the patch always sets this data member after a
successful helper response.
+ Prefer something like: "The ID string used internally by Squid to
uniquely de-duplicate this requests URL with other URLs stored objects."
I find "uniquely de-duplicate" too puzzling. If you want to briefly
define what store ID is here, how about this:
If defined, store_id_program mapped requested URL to this ID.
Store uses this ID (and not the URL) to find and store entries,
avoiding caching duplicate entries when different URLs point to
"essentially the same" cachable resource.
If you do not want to define the concept itself here, then the first
line would suffice IMO.
BTW, when you do not think the patch is ready for commit, please use
[PREVIEW] Subject prefix or no prefix at all. The [PATCH] prefix is for
patches that you think should be committed essentially without changes.
HTH,
Alex.
--
Eliezer Croitoru
https://www1.ngtech.co.il
sip:ngt...@sip2sip.info
IT consulting for Nonprofit organizations
eliezer <at> ngtech.co.il
