Specify the use of RiskAdvisoryData for Wallet unmask requests on iOS.

Specify the use of RiskAdvisoryData for Wallet unmask requests on iOS.

See https://codereview.chromium.org/1058473002/ for background.

BUG=484806
Committed: https://crrev.com/bcd6c7c0db2dea4595b048df3b3fa429481dd80f
Cr-Commit-Position: refs/heads/master@{#330885}

[Justin Donnelly, 2015-05-20 22:35:04]

jdonnelly@chromium.org changed reviewers:
+ estade@chromium.org

[Evan Stade, 2015-05-20 22:44:47]

https://codereview.chromium.org/1148123003/diff/1/components/autofill/core/br...
File components/autofill/core/browser/ui/card_unmask_prompt_controller_impl.cc
(right):

https://codereview.chromium.org/1148123003/diff/1/components/autofill/core/br...
components/autofill/core/browser/ui/card_unmask_prompt_controller_impl.cc:52:
#if defined(OS_IOS)
seems like this ifdef could go in the card_unmask_delegate.cc instead (affecting
the default constructor)

[Justin Donnelly, 2015-05-20 22:59:17]

https://codereview.chromium.org/1148123003/diff/1/components/autofill/core/br...
File components/autofill/core/browser/ui/card_unmask_prompt_controller_impl.cc
(right):

https://codereview.chromium.org/1148123003/diff/1/components/autofill/core/br...
components/autofill/core/browser/ui/card_unmask_prompt_controller_impl.cc:52:
#if defined(OS_IOS)
On 2015/05/20 22:44:46, Evan Stade wrote:
> seems like this ifdef could go in the card_unmask_delegate.cc instead
(affecting
> the default constructor)

Done.

[Evan Stade, 2015-05-20 23:30:34]

https://codereview.chromium.org/1148123003/diff/40001/components/autofill/cor...
File components/autofill/core/browser/card_unmask_delegate.h (left):

https://codereview.chromium.org/1148123003/diff/40001/components/autofill/cor...
components/autofill/core/browser/card_unmask_delegate.h:37: bool
providing_risk_advisory_data = false;
const bool?

[Justin Donnelly, 2015-05-21 00:29:51]

https://codereview.chromium.org/1148123003/diff/40001/components/autofill/cor...
File components/autofill/core/browser/card_unmask_delegate.h (left):

https://codereview.chromium.org/1148123003/diff/40001/components/autofill/cor...
components/autofill/core/browser/card_unmask_delegate.h:37: bool
providing_risk_advisory_data = false;
On 2015/05/20 23:30:34, Evan Stade wrote:
> const bool?

Adding const produces the following compile error:

../../components/autofill/core/browser/ui/card_unmask_prompt_controller_impl.cc:51:21:
error: object of type 'autofill::CardUnmaskDelegate::UnmaskResponse' cannot be
assigned because its copy assignment operator is implicitly deleted
  pending_response_ = CardUnmaskDelegate::UnmaskResponse();
                    ^
../../components/autofill/core/browser/card_unmask_delegate.h:37:16: note: copy
assignment operator of 'UnmaskResponse' is implicitly deleted because field
'providing_risk_advisory_data' is of const-qualified type 'const bool'
    const bool providing_risk_advisory_data;
               ^

[Evan Stade, 2015-05-21 01:08:12]

On 2015/05/21 00:29:51, Justin Donnelly wrote:
>
https://codereview.chromium.org/1148123003/diff/40001/components/autofill/cor...
> File components/autofill/core/browser/card_unmask_delegate.h (left):
> 
>
https://codereview.chromium.org/1148123003/diff/40001/components/autofill/cor...
> components/autofill/core/browser/card_unmask_delegate.h:37: bool
> providing_risk_advisory_data = false;
> On 2015/05/20 23:30:34, Evan Stade wrote:
> > const bool?
> 
> Adding const produces the following compile error:
> 
>
../../components/autofill/core/browser/ui/card_unmask_prompt_controller_impl.cc:51:21:
> error: object of type 'autofill::CardUnmaskDelegate::UnmaskResponse' cannot be
> assigned because its copy assignment operator is implicitly deleted
>   pending_response_ = CardUnmaskDelegate::UnmaskResponse();
>                     ^
> ../../components/autofill/core/browser/card_unmask_delegate.h:37:16: note:
copy
> assignment operator of 'UnmaskResponse' is implicitly deleted because field
> 'providing_risk_advisory_data' is of const-qualified type 'const bool'
>     const bool providing_risk_advisory_data;
>                ^

well yes, you have to change how it's initialized (use the initializer list on
all platforms)

[Justin Donnelly, 2015-05-21 01:20:21]

On 2015/05/21 01:08:12, Evan Stade wrote:
> On 2015/05/21 00:29:51, Justin Donnelly wrote:
> >
>
https://codereview.chromium.org/1148123003/diff/40001/components/autofill/cor...
> > File components/autofill/core/browser/card_unmask_delegate.h (left):
> > 
> >
>
https://codereview.chromium.org/1148123003/diff/40001/components/autofill/cor...
> > components/autofill/core/browser/card_unmask_delegate.h:37: bool
> > providing_risk_advisory_data = false;
> > On 2015/05/20 23:30:34, Evan Stade wrote:
> > > const bool?
> > 
> > Adding const produces the following compile error:
> > 
> >
>
../../components/autofill/core/browser/ui/card_unmask_prompt_controller_impl.cc:51:21:
> > error: object of type 'autofill::CardUnmaskDelegate::UnmaskResponse' cannot
be
> > assigned because its copy assignment operator is implicitly deleted
> >   pending_response_ = CardUnmaskDelegate::UnmaskResponse();
> >                     ^
> > ../../components/autofill/core/browser/card_unmask_delegate.h:37:16: note:
> copy
> > assignment operator of 'UnmaskResponse' is implicitly deleted because field
> > 'providing_risk_advisory_data' is of const-qualified type 'const bool'
> >     const bool providing_risk_advisory_data;
> >                ^
> 
> well yes, you have to change how it's initialized (use the initializer list on
> all platforms)

Yeah, I assumed that's what you meant, but the error is the same without the
platform-specific part. If the constructor is simply:

CardUnmaskDelegate::UnmaskResponse::UnmaskResponse()
    : should_store_pan(false),
      providing_risk_advisory_data(false) {}

the compiler still complains. I don't claim to understand all the subtleties of
C++ object initialization, though, so if you think there's a way to make it
work, let me know.

[Evan Stade, 2015-05-21 01:59:16]

On 2015/05/21 01:20:21, Justin Donnelly wrote:
> On 2015/05/21 01:08:12, Evan Stade wrote:
> > On 2015/05/21 00:29:51, Justin Donnelly wrote:
> > >
> >
>
https://codereview.chromium.org/1148123003/diff/40001/components/autofill/cor...
> > > File components/autofill/core/browser/card_unmask_delegate.h (left):
> > > 
> > >
> >
>
https://codereview.chromium.org/1148123003/diff/40001/components/autofill/cor...
> > > components/autofill/core/browser/card_unmask_delegate.h:37: bool
> > > providing_risk_advisory_data = false;
> > > On 2015/05/20 23:30:34, Evan Stade wrote:
> > > > const bool?
> > > 
> > > Adding const produces the following compile error:
> > > 
> > >
> >
>
../../components/autofill/core/browser/ui/card_unmask_prompt_controller_impl.cc:51:21:
> > > error: object of type 'autofill::CardUnmaskDelegate::UnmaskResponse'
cannot
> be
> > > assigned because its copy assignment operator is implicitly deleted
> > >   pending_response_ = CardUnmaskDelegate::UnmaskResponse();
> > >                     ^
> > > ../../components/autofill/core/browser/card_unmask_delegate.h:37:16: note:
> > copy
> > > assignment operator of 'UnmaskResponse' is implicitly deleted because
field
> > > 'providing_risk_advisory_data' is of const-qualified type 'const bool'
> > >     const bool providing_risk_advisory_data;
> > >                ^
> > 
> > well yes, you have to change how it's initialized (use the initializer list
on
> > all platforms)
> 
> Yeah, I assumed that's what you meant, but the error is the same without the
> platform-specific part. If the constructor is simply:
> 
> CardUnmaskDelegate::UnmaskResponse::UnmaskResponse()
>     : should_store_pan(false),
>       providing_risk_advisory_data(false) {}
> 
> the compiler still complains. I don't claim to understand all the subtleties
of
> C++ object initialization, though, so if you think there's a way to make it
> work, let me know.

oh ok, I see the problem. Didn't read the error message closely enough. Could
get around it by using heap allocation at the callsite but that doesn't seem
worth it. I guess we can live without const. lgtm

[Justin Donnelly, 2015-05-21 02:39:04]

The CQ bit was checked by jdonnelly@chromium.org

[commit-bot: I haz the power, 2015-05-21 02:39:24]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1148123003/40001

[commit-bot: I haz the power, 2015-05-21 04:36:54]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-05-21 04:41:07]

Patchset 3 (id:??) landed as
https://crrev.com/bcd6c7c0db2dea4595b048df3b3fa429481dd80f
Cr-Commit-Position: refs/heads/master@{#330885}