MAC Cookies (patch 5 of N)

MAC Cookies (patch 5 of N)

To help folks deal with clock skew, we've changed the timestamp to the "age" of
the credential, which is the number of seconds since the cookie was set (as
opposed to the number of seconds since the epoch).  This patch updates our
implementation to match.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=85542

[abarth-chromium, 2011-05-13 21:40:02]

@cbentzel: Here's a small patch, if you have time before your vacation. 
Otherwise, no worries.  Thanks!

[abarth-chromium, 2011-05-13 21:40:37]

+rdsmith for trivial changes to the cookie monster.

[abarth-chromium, 2011-05-14 18:56:35]

rdsmith would you be willing to look at this patch since cbentzel is
out-of-the-office?  Thanks.

[Randy Smith (Not in Mondays), 2011-05-16 20:06:46]

LGTM, with the caveat that I haven't read the updated RFC, so I'm just checking
the implementation against the issue description.

http://codereview.chromium.org/6969075/diff/1/net/http/http_mac_signature_uni...
File net/http/http_mac_signature_unittest.cc (right):

http://codereview.chromium.org/6969075/diff/1/net/http/http_mac_signature_uni...
net/http/http_mac_signature_unittest.cc:25: base::Time::Now(),
Just for my curiosity, why the extra test for this case?

[abarth-chromium, 2011-05-16 20:13:15]

Thanks!

http://codereview.chromium.org/6969075/diff/1/net/http/http_mac_signature_uni...
File net/http/http_mac_signature_unittest.cc (right):

http://codereview.chromium.org/6969075/diff/1/net/http/http_mac_signature_uni...
net/http/http_mac_signature_unittest.cc:25: base::Time::Now(),
On 2011/05/16 20:06:46, rdsmith wrote:
> Just for my curiosity, why the extra test for this case?

Oh, I added a branch in the implementation.  This test just exercise that branch
(null time).

[commit-bot: I haz the power, 2011-05-16 22:18:44]

Change committed as 85542