Skip unknown channel configurations when parsing session config.

Skip unknown channel configurations when parsing session config.

We may need to add new transport types in the future. Current config
parsing code fails to parse configs with transport types it doesn't 
understand which will make it hard to add new transport type without
breaking backward compatibility.

BUG=137135
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=148679

[simonmorris, 2012-07-26 16:11:06]

http://codereview.chromium.org/10831022/diff/1/remoting/protocol/content_desc...
File remoting/protocol/content_description.cc (right):

http://codereview.chromium.org/10831022/diff/1/remoting/protocol/content_desc...
remoting/protocol/content_description.cc:240: child = child->NextNamed(tag);
The loop will go infinite if ParseChannelConfig ever returns false.

A unit test would have caught this error.

[Sergey Ulanov, 2012-07-26 22:54:45]

Added unittests for this code

http://codereview.chromium.org/10831022/diff/1/remoting/protocol/content_desc...
File remoting/protocol/content_description.cc (right):

http://codereview.chromium.org/10831022/diff/1/remoting/protocol/content_desc...
remoting/protocol/content_description.cc:240: child = child->NextNamed(tag);
On 2012/07/26 16:11:06, simonmorris wrote:
> The loop will go infinite if ParseChannelConfig ever returns false.

Fixed.

> 
> A unit test would have caught this error.

And added unittest this time.

[simonmorris, 2012-07-26 23:08:50]

lgtm with optional suggestions.

http://codereview.chromium.org/10831022/diff/4001/remoting/protocol/content_d...
File remoting/protocol/content_description.cc (right):

http://codereview.chromium.org/10831022/diff/4001/remoting/protocol/content_d...
remoting/protocol/content_description.cc:262: const XmlElement* child = NULL;
You could move the declaration of child down to where it's initialized.

http://codereview.chromium.org/10831022/diff/4001/remoting/protocol/content_d...
remoting/protocol/content_description.cc:272: return
scoped_ptr<ContentDescription>(NULL);
Remove NULL here (or add NULL to the initialization of the value you return in
the earlier error case), just for consistency.

[Sergey Ulanov, 2012-07-26 23:34:13]

http://codereview.chromium.org/10831022/diff/4001/remoting/protocol/content_d...
File remoting/protocol/content_description.cc (right):

http://codereview.chromium.org/10831022/diff/4001/remoting/protocol/content_d...
remoting/protocol/content_description.cc:262: const XmlElement* child = NULL;
On 2012/07/26 23:08:51, simonmorris wrote:
> You could move the declaration of child down to where it's initialized.

Done.

http://codereview.chromium.org/10831022/diff/4001/remoting/protocol/content_d...
remoting/protocol/content_description.cc:272: return
scoped_ptr<ContentDescription>(NULL);
On 2012/07/26 23:08:51, simonmorris wrote:
> Remove NULL here (or add NULL to the initialization of the value you return in
> the earlier error case), just for consistency.

Done.

[commit-bot: I haz the power, 2012-07-26 23:34:27]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sergeyu@chromium.org/10831022/9001

[commit-bot: I haz the power, 2012-07-27 00:54:45]

Change committed as 148679