Add AlternateProtocol to broken_altproto_list_ and map_.

Add AlternateProtocol to broken_altproto_list_ and map_.

Add AlternateProtocol (port and protocol) to
HttpServerPropertiesImpl::broken_alternate_protocol_list_ and
HttpServerPropertiesImpl::broken_alternate_protocol_list_ in preparation for
supporting multiple AlternateProtocols per host.  See draft at
https://crrev.com/665083009 for how this fits into a larger picture.

BUG=392575
Committed: https://crrev.com/9f2a7c417364a965f99c3e1948fd948d07b316cd
Cr-Commit-Position: refs/heads/master@{#314843}

[Bence, 2015-02-03 19:59:29]

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

[Bence, 2015-02-03 19:59:29]

Ryan, PTAL.  Thanks.

[Ryan Hamilton, 2015-02-03 22:39:35]

https://codereview.chromium.org/888943003/diff/1/net/http/http_server_propert...
File net/http/http_server_properties_impl.cc (right):

https://codereview.chromium.org/888943003/diff/1/net/http/http_server_propert...
net/http/http_server_properties_impl.cc:359: if (!HasAlternateProtocol(server))
Looks like this needs to be rebased on top of your other CL.

https://codereview.chromium.org/888943003/diff/1/net/http/http_server_propert...
net/http/http_server_properties_impl.cc:373: alternate_protocol.protocol);
might consider adding a 2 arg constructor(server, alternate protocol info) since
that's basically what's used each time.

https://codereview.chromium.org/888943003/diff/1/net/http/http_server_propert...
net/http/http_server_properties_impl.cc:536:
ClearAlternateProtocol(entry.server);
I think this method should now that a BrokenAPEntry instead of just a server,
right?

[Bence, 2015-02-04 14:25:33]

PTAL.

https://codereview.chromium.org/888943003/diff/1/net/http/http_server_propert...
File net/http/http_server_properties_impl.cc (right):

https://codereview.chromium.org/888943003/diff/1/net/http/http_server_propert...
net/http/http_server_properties_impl.cc:359: if (!HasAlternateProtocol(server))
On 2015/02/03 22:39:35, Ryan Hamilton wrote:
> Looks like this needs to be rebased on top of your other CL.

Done.

https://codereview.chromium.org/888943003/diff/1/net/http/http_server_propert...
net/http/http_server_properties_impl.cc:373: alternate_protocol.protocol);
On 2015/02/03 22:39:35, Ryan Hamilton wrote:
> might consider adding a 2 arg constructor(server, alternate protocol info)
since
> that's basically what's used each time.

I was thinking a lot about it, but it turns out AlternateProtocolInfo includes
probability, which we do not want to have in BrokenAlternateProtocolEntry. 
Therefore I decided against it.

I might define an AlternativeService struct down the road with (protocol, host,
port) triple, and might make AlternateServiceInfo have one of those as a member.
 We'll need to see if GetAlternativeServices, SetAlternativeServiceBroken etc.
should take the (protocol, host, port) triple as a single struct, or as three
separate arguments.  But that's gonna be another CL...

https://codereview.chromium.org/888943003/diff/1/net/http/http_server_propert...
net/http/http_server_properties_impl.cc:536:
ClearAlternateProtocol(entry.server);
On 2015/02/03 22:39:35, Ryan Hamilton wrote:
> I think this method should now that a BrokenAPEntry instead of just a server,
> right?

A server can still only have one AlternateProtocolInfo, so I think the change
you propose does not belong to this CL quite yet.  But you are right, it will
come as soon as I introduce multiple alternate protocols per server.

[Ryan Hamilton, 2015-02-05 17:32:24]

lgtm

https://codereview.chromium.org/888943003/diff/1/net/http/http_server_propert...
File net/http/http_server_properties_impl.cc (right):

https://codereview.chromium.org/888943003/diff/1/net/http/http_server_propert...
net/http/http_server_properties_impl.cc:536:
ClearAlternateProtocol(entry.server);
On 2015/02/04 14:25:33, Bence wrote:
> On 2015/02/03 22:39:35, Ryan Hamilton wrote:
> > I think this method should now that a BrokenAPEntry instead of just a
server,
> > right?
> 
> A server can still only have one AlternateProtocolInfo, so I think the change
> you propose does not belong to this CL quite yet.  But you are right, it will
> come as soon as I introduce multiple alternate protocols per server.

Ah,yes, good point.

[Bence, 2015-02-05 18:53:51]

The CQ bit was checked by bnc@chromium.org

[commit-bot: I haz the power, 2015-02-05 18:54:52]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/888943003/20001

[commit-bot: I haz the power, 2015-02-05 18:58:28]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-02-05 18:59:28]

Patchset 2 (id:??) landed as
https://crrev.com/9f2a7c417364a965f99c3e1948fd948d07b316cd
Cr-Commit-Position: refs/heads/master@{#314843}