Resubmit CL. See 8429004.

Resubmit CL with memory leaks patched. See 8429004.
BUG=chromium-os:19399
TEST=manual + try linux_chromeos_valgrind
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=112274

[Charlie Lee, 2011-11-30 16:31:37]

Steven, please take a look. I did multiple rounds of try linux_chromeos_valgrind
to find the memory leaks. The latest run is still running, but I'm pretty sure
it will be clean.

[stevenjb, 2011-11-30 17:36:20]

+ mnissler@ who will need to rebase his changes on this.
LGTM

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
File chrome/browser/chromeos/cros/native_network_parser.cc (right):

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
chrome/browser/chromeos/cros/native_network_parser.cc:533: return &type_mapper;
nit: "type_mapper" and "type_table" sound like generic functions, we should
probably name these "network_type_mapper", etc.

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
File chrome/browser/chromeos/cros/network_library.cc (right):

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
chrome/browser/chromeos/cros/network_library.cc:3681:
DCHECK(CrosLibrary::Get()->libcros_loaded());
nit: DCHECK is unnecessary here.

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
File chrome/browser/chromeos/cros/network_library.h (right):

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
chrome/browser/chromeos/cros/network_library.h:677: // Updates the properties
map for the corresponding property index.
nit: s/the properties map/|property_map_|/

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
File chrome/browser/chromeos/cros/onc_network_parser.cc (right):

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
chrome/browser/chromeos/cros/onc_network_parser.cc:427:
wifi_network->set_eap_identity(eap_identity);
I believe that ken said something about saving EAP_IDENTITY in the property map,
you might want to sync with him on that (could be in a follow-up CL though).

[kmixter1, 2011-11-30 17:39:01]

FYI

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
File chrome/browser/chromeos/cros/onc_network_parser.cc (right):

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
chrome/browser/chromeos/cros/onc_network_parser.cc:427:
wifi_network->set_eap_identity(eap_identity);
On 2011/11/30 17:36:20, Steven Bennetts wrote:
> I believe that ken said something about saving EAP_IDENTITY in the property
map,
> you might want to sync with him on that (could be in a follow-up CL though).

No need.  The property map is updated in UpdateState.  I guess the other places
where we manually do this are where we need to translate the value itself into
one that the connection manager likes.

[Charlie Lee, 2011-11-30 18:01:21]

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
File chrome/browser/chromeos/cros/native_network_parser.cc (right):

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
chrome/browser/chromeos/cros/native_network_parser.cc:533: return &type_mapper;
On 2011/11/30 17:36:20, Steven Bennetts wrote:
> nit: "type_mapper" and "type_table" sound like generic functions, we should
> probably name these "network_type_mapper", etc.

Done.

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
File chrome/browser/chromeos/cros/network_library.cc (right):

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
chrome/browser/chromeos/cros/network_library.cc:3681:
DCHECK(CrosLibrary::Get()->libcros_loaded());
On 2011/11/30 17:36:20, Steven Bennetts wrote:
> nit: DCHECK is unnecessary here.

Done.

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
File chrome/browser/chromeos/cros/network_library.h (right):

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
chrome/browser/chromeos/cros/network_library.h:677: // Updates the properties
map for the corresponding property index.
On 2011/11/30 17:36:20, Steven Bennetts wrote:
> nit: s/the properties map/|property_map_|/

Done.

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
File chrome/browser/chromeos/cros/onc_network_parser.cc (right):

http://codereview.chromium.org/8734013/diff/11015/chrome/browser/chromeos/cro...
chrome/browser/chromeos/cros/onc_network_parser.cc:427:
wifi_network->set_eap_identity(eap_identity);
On 2011/11/30 17:39:01, kmixter1 wrote:
> On 2011/11/30 17:36:20, Steven Bennetts wrote:
> > I believe that ken said something about saving EAP_IDENTITY in the property
> map,
> > you might want to sync with him on that (could be in a follow-up CL though).
> 
> No need.  The property map is updated in UpdateState.  I guess the other
places
> where we manually do this are where we need to translate the value itself into
> one that the connection manager likes.

Yes, exactly.

[stevenjb, 2011-11-30 18:11:56]

Makes sense. LGTM.

[Charlie Lee, 2011-11-30 18:59:10]

Mattias, is this the fix for SSID you were talking about?

[Mattias Nissler (ping if slow), 2011-11-30 20:17:55]

On 2011/11/30 18:59:10, Charlie Lee wrote:
> Mattias, is this the fix for SSID you were talking about?

Yep, that fix looks good.