[net/dns] Serve requests from HOSTS file if possible.

[net/dns] Serve requests from HOSTS file if possible.

If the DnsConfig is available, we can serve the requests synchronously in Resolve. However, we might not receive the DnsConfig until the Job is placed in the queue. In that case, we attempt to serve the jobs once DnsConfig is available.

BUG=114826
TEST=./net_unittests --gtest_filter=HostResolverImplTest.ServeFromHosts

Original review: https://chromiumcodereview.appspot.com/9667025/


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=126697

[szym, 2012-03-14 15:50:54]

I looked at other uses of NetAddressToString and found that in all unittests it
is allowed to fail (return ""). Should we just do this, or is there a better
way?

http://codereview.chromium.org/9705017/diff/1/net/base/host_resolver_impl_uni...
File net/base/host_resolver_impl_unittest.cc (right):

http://codereview.chromium.org/9705017/diff/1/net/base/host_resolver_impl_uni...
net/base/host_resolver_impl_unittest.cc:741: EXPECT_EQ("[2001:db8::1]:5555",
FirstAddressToString(addrlist));
Will not work on systems not supporting IPv6, e.g., Windows XP.
NetAddressToString (and IPEndPoint.ToString) return "" on such systems.

http://codereview.chromium.org/9705017/diff/1/net/base/host_resolver_impl_uni...
net/base/host_resolver_impl_unittest.cc:1894: EXPECT_EQ("[::1]:80",
FirstAddressToString(req3.addrlist()));
As above.

http://codereview.chromium.org/9705017/diff/1/net/base/host_resolver_impl_uni...
net/base/host_resolver_impl_unittest.cc:1903: EXPECT_EQ("::1",
NetAddressToString(addr));
Ditto.

http://codereview.chromium.org/9705017/diff/1/net/base/host_resolver_impl_uni...
net/base/host_resolver_impl_unittest.cc:1916: EXPECT_EQ("[::1]:80",
FirstAddressToString(req6.addrlist()));
And again.

[mmenke, 2012-03-14 16:11:04]

Hmm...  Could just allow them to fail.  Alternatively, could just skip those
lines on XP only.

On 2012/03/14 15:50:54, szym wrote:
> I looked at other uses of NetAddressToString and found that in all unittests
it
> is allowed to fail (return ""). Should we just do this, or is there a better
> way?
> 
>
http://codereview.chromium.org/9705017/diff/1/net/base/host_resolver_impl_uni...
> File net/base/host_resolver_impl_unittest.cc (right):
> 
>
http://codereview.chromium.org/9705017/diff/1/net/base/host_resolver_impl_uni...
> net/base/host_resolver_impl_unittest.cc:741: EXPECT_EQ("[2001:db8::1]:5555",
> FirstAddressToString(addrlist));
> Will not work on systems not supporting IPv6, e.g., Windows XP.
> NetAddressToString (and IPEndPoint.ToString) return "" on such systems.
> 
>
http://codereview.chromium.org/9705017/diff/1/net/base/host_resolver_impl_uni...
> net/base/host_resolver_impl_unittest.cc:1894: EXPECT_EQ("[::1]:80",
> FirstAddressToString(req3.addrlist()));
> As above.
> 
>
http://codereview.chromium.org/9705017/diff/1/net/base/host_resolver_impl_uni...
> net/base/host_resolver_impl_unittest.cc:1903: EXPECT_EQ("::1",
> NetAddressToString(addr));
> Ditto.
> 
>
http://codereview.chromium.org/9705017/diff/1/net/base/host_resolver_impl_uni...
> net/base/host_resolver_impl_unittest.cc:1916: EXPECT_EQ("[::1]:80",
> FirstAddressToString(req6.addrlist()));
> And again.

[szym, 2012-03-14 17:19:04]

How about this solution?

[mmenke, 2012-03-14 17:21:35]

On 2012/03/14 17:19:04, szym wrote:
> How about this solution?

LGTM

[commit-bot: I haz the power, 2012-03-14 17:48:47]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/szym@chromium.org/9705017/5003

[commit-bot: I haz the power, 2012-03-14 19:09:28]

Change committed as 126697