Remove additional dependencies from InternetOptionsHandler

Remove additional dependencies from InternetOptionsHandler

Removes UIData from NetworkState.

BUG=279351
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=282821

[pneubeck (no reviews), 2014-07-11 20:35:07]

again a tremendous simplification!

Please make the commit message a bit more detailed (e.g. removing UIData from
NetworkState).

https://codereview.chromium.org/383013002/diff/1/chrome/browser/ui/webui/opti...
File chrome/browser/ui/webui/options/chromeos/internet_options_handler.cc
(right):

https://codereview.chromium.org/383013002/diff/1/chrome/browser/ui/webui/opti...
chrome/browser/ui/webui/options/chromeos/internet_options_handler.cc:511:
FindPolicyByGUID(LoginState::Get()->primary_user_hash(),
once we refer to the 'right user' more than once, i'd put it into a helper
function. So that this decision what the right user is, is kept in a single
place.

https://codereview.chromium.org/383013002/diff/1/chrome/browser/ui/webui/opti...
chrome/browser/ui/webui/options/chromeos/internet_options_handler.cc:729: //
TODO(stevenjb): Find a better way to populate initial data before
i think, other WebUIs have a kind of handshake:

1) If JS is done loading, it calls to C++.
2) C++ calls back to JS with initial data.

(IIRC the internet_details page does this already?)

https://codereview.chromium.org/383013002/diff/1/chromeos/network/network_sta...
File chromeos/network/network_state.h (left):

https://codereview.chromium.org/383013002/diff/1/chromeos/network/network_sta...
chromeos/network/network_state.h:66: const NetworkUIData& ui_data() const {
return ui_data_; }
Yeah!!! Defeating the beast finally :-)

https://codereview.chromium.org/383013002/diff/1/chromeos/network/network_uti...
File chromeos/network/network_util.cc (right):

https://codereview.chromium.org/383013002/diff/1/chromeos/network/network_uti...
chromeos/network/network_util.cc:168: std::string userhash =
LoginState::Get()->primary_user_hash();
It seems to me that this is unexpected and will eventually lead to surprises.
Could you pass in the user_hash instead of defaulting to the primary user (even
if it's conditionally required if debugging_properties=true)?

[stevenjb, 2014-07-11 21:26:10]

https://codereview.chromium.org/383013002/diff/1/chrome/browser/ui/webui/opti...
File chrome/browser/ui/webui/options/chromeos/internet_options_handler.cc
(right):

https://codereview.chromium.org/383013002/diff/1/chrome/browser/ui/webui/opti...
chrome/browser/ui/webui/options/chromeos/internet_options_handler.cc:511:
FindPolicyByGUID(LoginState::Get()->primary_user_hash(),
On 2014/07/11 20:35:06, pneubeck wrote:
> once we refer to the 'right user' more than once, i'd put it into a helper
> function. So that this decision what the right user is, is kept in a single
> place.
That's kind of what LoginState::Get()->primary_user_hash() is supposed to be.
Given that this code is extremely temporary, I'd just as soon not do too much
re-factoring here. The recent NetworkingPrivateApi re-factor I did handles
profiles correctly, and I plan to use that class for NetworkConfig.cc also,
using the profile associated with the WebUI, so everything will be correct and
primary_user_hash() can go away (at least for networking code).

https://codereview.chromium.org/383013002/diff/1/chrome/browser/ui/webui/opti...
chrome/browser/ui/webui/options/chromeos/internet_options_handler.cc:729: //
TODO(stevenjb): Find a better way to populate initial data before
On 2014/07/11 20:35:07, pneubeck wrote:
> i think, other WebUIs have a kind of handshake:
> 
> 1) If JS is done loading, it calls to C++.
> 2) C++ calls back to JS with initial data.
> 
> (IIRC the internet_details page does this already?)

This is being done so that the initial layout doesn't jump. The settings page
takes so long to load that if we wait for the JS to complete we see the layout
shift. I'm planning to dive deeper into the WebUI code to do this properly in
one of the next stages.

https://codereview.chromium.org/383013002/diff/1/chromeos/network/network_sta...
File chromeos/network/network_state.h (left):

https://codereview.chromium.org/383013002/diff/1/chromeos/network/network_sta...
chromeos/network/network_state.h:66: const NetworkUIData& ui_data() const {
return ui_data_; }
On 2014/07/11 20:35:07, pneubeck wrote:
> Yeah!!! Defeating the beast finally :-)

Acknowledged.

https://codereview.chromium.org/383013002/diff/1/chromeos/network/network_uti...
File chromeos/network/network_util.cc (right):

https://codereview.chromium.org/383013002/diff/1/chromeos/network/network_uti...
chromeos/network/network_util.cc:168: std::string userhash =
LoginState::Get()->primary_user_hash();
On 2014/07/11 20:35:07, pneubeck wrote:
> It seems to me that this is unexpected and will eventually lead to surprises.
> Could you pass in the user_hash instead of defaulting to the primary user
(even
> if it's conditionally required if debugging_properties=true)?

I'll rip this out for now, it's handy but not worth making a major change to
support at this point. We also have the "Get managed properties" checkbox now in
chrome://network.

[pneubeck (no reviews), 2014-07-12 09:09:37]

lgtm

[pneubeck (no reviews), 2014-07-12 09:10:35]

interestingly I can update the commit message myself... :-)

[pneubeck (no reviews), 2014-07-12 09:11:19]

The CQ bit was checked by pneubeck@chromium.org

[commit-bot: I haz the power, 2014-07-12 09:12:11]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/stevenjb@chromium.org/383013002/20001

[commit-bot: I haz the power, 2014-07-12 12:54:53]

Change committed as 282821