Add unittests for secure resource over secure and insecure QUIC.

Add unittests for secure resource over secure and insecure QUIC.

Add two unittests to test the condition that triggers
ERR_REQUEST_FOR_SECURE_RESOURCE_OVER_INSECURE_QUIC.

BUG=

Committed: https://crrev.com/50883590a541dd7c786d1c9f4ea21c69ea2dce34
Cr-Commit-Position: refs/heads/master@{#329471}

[Bence, 2015-05-12 16:36:35]

Patchset #2 (id:20001) has been deleted

[Bence, 2015-05-12 17:50:53]

bnc@chromium.org changed reviewers:
+ rch@chromium.org

[Bence, 2015-05-12 17:50:53]

Ryan, PTAL.  Thanks.

[Ryan Hamilton, 2015-05-12 18:15:56]

lgtm, but please add a comment to describe why the first test is actually
impossible to hit in production, if you think it's worthwhile having.

https://codereview.chromium.org/1135103003/diff/1/net/quic/quic_network_trans...
File net/quic/quic_network_transaction_unittest.cc (right):

https://codereview.chromium.org/1135103003/diff/1/net/quic/quic_network_trans...
net/quic/quic_network_transaction_unittest.cc:1227: }
Hm. This is actually unreachable in reality, right? Said different, we never
should encounter the ERR_REQUEST_FOR_SECURE_RESOURCE_OVER_INSECURE_QUIC error
unless we have a *programming* bug. It's almost a DCHECK.

I don't mind having a test like this to make sure that code can be emitted, but
we should probably explain why the test is unrealistic.

In particular the crypto client stream will not confirm a handshake for secure
QUIC unless the proof has been verified.

[Bence, 2015-05-12 19:51:56]

Thanks.

https://codereview.chromium.org/1135103003/diff/1/net/quic/quic_network_trans...
File net/quic/quic_network_transaction_unittest.cc (right):

https://codereview.chromium.org/1135103003/diff/1/net/quic/quic_network_trans...
net/quic/quic_network_transaction_unittest.cc:1227: }
On 2015/05/12 18:15:56, Ryan Hamilton wrote:
> Hm. This is actually unreachable in reality, right? Said different, we never
> should encounter the ERR_REQUEST_FOR_SECURE_RESOURCE_OVER_INSECURE_QUIC error
> unless we have a *programming* bug. It's almost a DCHECK.
> 
> I don't mind having a test like this to make sure that code can be emitted,
but
> we should probably explain why the test is unrealistic.
> 
> In particular the crypto client stream will not confirm a handshake for secure
> QUIC unless the proof has been verified.

Done.

[Bence, 2015-05-12 19:51:59]

The CQ bit was checked by bnc@chromium.org

[Bence, 2015-05-12 19:52:00]

The patchset sent to the CQ was uploaded after l-g-t-m from rch@chromium.org
Link to the patchset: https://codereview.chromium.org/1135103003/#ps40001
(title: "Add comment.")

[commit-bot: I haz the power, 2015-05-12 19:52:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1135103003/40001

[commit-bot: I haz the power, 2015-05-12 20:10:40]

Committed patchset #2 (id:40001)

[commit-bot: I haz the power, 2015-05-12 20:11:35]

Patchset 2 (id:??) landed as
https://crrev.com/50883590a541dd7c786d1c9f4ea21c69ea2dce34
Cr-Commit-Position: refs/heads/master@{#329471}