Serialize NPN_HTTP_2 as "h2" instead of "npn-h2".

Serialize NPN_HTTP_2 as "h2" instead of "npn-h2".

This will allow properly parsing spec-compliant Alt-Svc headers.

At the same time, still parse "npn-h2" so that old persisted settings can be
read from disk.

BUG=626375
Committed: https://crrev.com/08352775a31f45a7ad5f280ee1a7b98315338844
Cr-Commit-Position: refs/heads/master@{#404264}

[Bence, 2016-07-07 19:14:37]

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

[Bence, 2016-07-07 19:14:38]

Ryan: PTAL.  Also, I would like to merge it back to M53, what do you think about
that?

https://codereview.chromium.org/2129973002/diff/1/net/http/http_server_proper...
File net/http/http_server_properties_manager_unittest.cc (right):

https://codereview.chromium.org/2129973002/diff/1/net/http/http_server_proper...
net/http/http_server_properties_manager_unittest.cc:1148:
"\"port\":1234,\"protocol_str\":\"npn-h2\","
I left this instance and some others of "npn-h2" as is to test parsing old-style
persisted settings.

[Ryan Hamilton, 2016-07-07 21:29:19]

lgtm, though should we also change some tests in spdy_ (or http_)
network_transaction_tests?

[Bence, 2016-07-07 21:53:28]

The CQ bit was checked by bnc@chromium.org

[commit-bot: I haz the power, 2016-07-07 21:54:43]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Bence, 2016-07-07 21:56:27]

Thank you.

On 2016/07/07 21:29:19, Ryan Hamilton wrote:
> lgtm, though should we also change some tests in spdy_ (or http_)
> network_transaction_tests?

I don't think so.  Codesearch for "npn-h2" for me only turns up the two files
that this CL changes, see
https://cs.chromium.org/search/?q=npn-h2&sq=package:chromium&type=cs.  Network
transaction tests generate headers using the HttpServerProperties methods that
bogusly return "npn-h2" instead of "h2", that's why we did not catch this
earlier.  Too much logic in tests!

On the other hand, I will have to rebase https://crrev.com/2115213002 on this
CL, as there I am introducing string literals with "npn-h2" hardcoded.

[Ryan Hamilton, 2016-07-07 22:04:02]

On 2016/07/07 21:56:27, Bence wrote:
>
> On the other hand, I will have to rebase https://crrev.com/2115213002 on this
> CL, as there I am introducing string literals with "npn-h2" hardcoded.

Ah! I thought that had already landed.

LGTM

[Bence, 2016-07-07 22:16:17]

On 2016/07/07 22:04:02, Ryan Hamilton wrote:
> On 2016/07/07 21:56:27, Bence wrote:
> >
> > On the other hand, I will have to rebase https://crrev.com/2115213002 on
this
> > CL, as there I am introducing string literals with "npn-h2" hardcoded.
> 
> Ah! I thought that had already landed.
> 
> LGTM

Nope, still waiting for one reviewer.

[commit-bot: I haz the power, 2016-07-07 22:34:38]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-07-07 22:37:26]

Description was changed from

==========
Serialize NPN_HTTP_2 as "h2" instead of "npn-h2".

This will allow properly parsing spec-compliant Alt-Svc headers.

At the same time, still parse "npn-h2" so that old persisted settings can be
read from disk.

BUG=626375
==========

to

==========
Serialize NPN_HTTP_2 as "h2" instead of "npn-h2".

This will allow properly parsing spec-compliant Alt-Svc headers.

At the same time, still parse "npn-h2" so that old persisted settings can be
read from disk.

BUG=626375

Committed: https://crrev.com/08352775a31f45a7ad5f280ee1a7b98315338844
Cr-Commit-Position: refs/heads/master@{#404264}
==========

[commit-bot: I haz the power, 2016-07-07 22:37:27]

Patchset 1 (id:??) landed as
https://crrev.com/08352775a31f45a7ad5f280ee1a7b98315338844
Cr-Commit-Position: refs/heads/master@{#404264}