Remove PermissionContext factories

Remove PermissionContext factories

Currently there is no real benefit to having each PermissionContext as a keyed
service and it adds a lot of boilerplate. This CL moves PermissionContext
construction into the PermissionManager keyed service instead.

Committed: https://crrev.com/e3afee6b62380b5b09caf7513701f66d8964c6b6
Cr-Commit-Position: refs/heads/master@{#387843}

[raymes, 2016-04-11 02:45:08]

The CQ bit was checked by raymes@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-04-11 02:45:17]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1871343002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1871343002/20001

[commit-bot: I haz the power, 2016-04-11 02:56:06]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-04-11 02:56:06]

Dry run: Try jobs failed on following builders:
  linux_chromium_compile_dbg_32_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[raymes, 2016-04-11 03:40:58]

The CQ bit was checked by raymes@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-04-11 03:41:06]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1871343002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1871343002/40001

[commit-bot: I haz the power, 2016-04-11 04:22:43]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-04-11 04:22:44]

Dry run: This issue passed the CQ dry run.

[raymes, 2016-04-11 04:25:54]

Description was changed from

==========
Remove PermissionContext factories

Currently there is no real benefit to having each PermissionContext as a keyed
service and it adds a lot of boilerplate. This CL moves PermissionContext
construction into the PermissionManager keyed service instead.

BUG=
==========

to

==========
Remove PermissionContext factories

Currently there is no real benefit to having each PermissionContext as a keyed
service and it adds a lot of boilerplate. This CL moves PermissionContext
construction into the PermissionManager keyed service instead.
==========

[raymes, 2016-04-11 04:27:11]

raymes@chromium.org changed reviewers:
+ mlamouri@chromium.org

[raymes, 2016-04-11 04:27:12]

Hey Mounir, PTAL and let me know your thoughts. Thanks!

[mlamouri (slow - plz ping), 2016-04-11 14:06:06]

Thanks! This was among the low hanging fruits we needed to fix. Really great to
see that happening.

There is really a small change to do, otherwise lgtm but I would really
appreciate if you could have a look at the follow-up :)

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/chromeos...
File chrome/browser/chromeos/attestation/platform_verification_flow.cc (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/chromeos...
chrome/browser/chromeos/attestation/platform_verification_flow.cc:113:
content::PermissionType::PROTECTED_MEDIA_IDENTIFIER);
I think you could greatly improve that by doing:

PermissionStatus status =
PermissionManager::Get(Profile::FromBrowserContext(web_contents->GetBrowserContext()))->GetPermissionStatus(content::PermissionType::PROTECTED_MEDIA_IDENTIFIER);

Fully remove the usage of content settings and PermissionContextBase.

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/guest_vi...
File
chrome/browser/guest_view/web_view/chrome_web_view_permission_helper_delegate.cc
(right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/guest_vi...
chrome/browser/guest_view/web_view/chrome_web_view_permission_helper_delegate.cc:231:
->GetPermissionContext(content::PermissionType::GEOLOCATION)
As above, this could be:
PermissionManager::Get(profile)->RequestPermission(content::PermissionType::Geolocation,
[...])

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/media/me...
File chrome/browser/media/media_permission.cc (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/media/me...
chrome/browser/media/media_permission.cc:60: permission_type);
Could that be moved to something like:
`PermissionManager::Get(profile_)->GetPermissionStatus(permission_type);` ?

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
File chrome/browser/permissions/permission_manager.cc (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
chrome/browser/permissions/permission_manager.cc:219: make_scoped_ptr(new
MidiPermissionContext(profile));
Use base::WrapUnique() from "base/memory/ptr_util.h"

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
File chrome/browser/permissions/permission_manager.h (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
chrome/browser/permissions/permission_manager.h:34: PermissionContextBase*
GetPermissionContext(content::PermissionType type);
I would love this to go away actually. No need to do it in this CL but given
that PermissionManager is the entry point for PermissionContext, I think we
should be able to force PermissionManager to be used instead of getting the
PermissionContext. WDYT? Maybe you could add a TODO here?

(Obviously, when I say go away, I mean becomes private for internal and test
usage.)

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
chrome/browser/permissions/permission_manager.h:102: permission_contexts_;
What about:
std::unordered_map<content::PermissionType,
std::unique_ptr<PermissionContextBase>> ?

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/push_mes...
File chrome/browser/push_messaging/push_messaging_permission_context.cc (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/push_mes...
chrome/browser/push_messaging/push_messaging_permission_context.cc:50:
embedding_origin);
As above, I think we can abstract out the PermissionContext and directly use the
PermissionManager.

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/push_mes...
chrome/browser/push_messaging/push_messaging_permission_context.cc:95:
notification_permission_context->RequestPermission(
ditto

[raymes, 2016-04-12 07:44:54]

Hey Mounir, PTAL. I managed to remove all the GetPermissionContext references in
production code, except for one.

Thanks!

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/chromeos...
File chrome/browser/chromeos/attestation/platform_verification_flow.cc (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/chromeos...
chrome/browser/chromeos/attestation/platform_verification_flow.cc:113:
content::PermissionType::PROTECTED_MEDIA_IDENTIFIER);
Done. I considered this but wasn't sure whether to do it in this CL. I've done
it now :)

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/guest_vi...
File
chrome/browser/guest_view/web_view/chrome_web_view_permission_helper_delegate.cc
(right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/guest_vi...
chrome/browser/guest_view/web_view/chrome_web_view_permission_helper_delegate.cc:231:
->GetPermissionContext(content::PermissionType::GEOLOCATION)
On 2016/04/11 14:06:05, Mounir Lamouri wrote:
> As above, this could be:
>
PermissionManager::Get(profile)->RequestPermission(content::PermissionType::Geolocation,
> [...])

Done.

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/media/me...
File chrome/browser/media/media_permission.cc (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/media/me...
chrome/browser/media/media_permission.cc:60: permission_type);
It's a bit too complicated to change this one right now. It will be easier when
Tim does the media refactor and all this stuff will live inside the permission
context.

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
File chrome/browser/permissions/permission_manager.cc (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
chrome/browser/permissions/permission_manager.cc:219: make_scoped_ptr(new
MidiPermissionContext(profile));
On 2016/04/11 14:06:05, Mounir Lamouri wrote:
> Use base::WrapUnique() from "base/memory/ptr_util.h"

Done.

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
File chrome/browser/permissions/permission_manager.h (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
chrome/browser/permissions/permission_manager.h:34: PermissionContextBase*
GetPermissionContext(content::PermissionType type);
I agree. I managed to change all the occurrences in prod code except in
media_permission. I made MediaPermission a friend class (with a TODO) in order
to make this private. Alternatively I can just leave it public. Let me know.

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
chrome/browser/permissions/permission_manager.h:102: permission_contexts_;
On 2016/04/11 14:06:05, Mounir Lamouri wrote:
> What about:
> std::unordered_map<content::PermissionType,
> std::unique_ptr<PermissionContextBase>> ?

I tried this but got a bunch of link errors because there is no hash function
for PermissionType. I can look into this more tomorrow.

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/push_mes...
File chrome/browser/push_messaging/push_messaging_permission_context.cc (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/push_mes...
chrome/browser/push_messaging/push_messaging_permission_context.cc:50:
embedding_origin);
On 2016/04/11 14:06:06, Mounir Lamouri wrote:
> As above, I think we can abstract out the PermissionContext and directly use
the
> PermissionManager.

Done.

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/push_mes...
chrome/browser/push_messaging/push_messaging_permission_context.cc:95:
notification_permission_context->RequestPermission(
On 2016/04/11 14:06:05, Mounir Lamouri wrote:
> ditto

Done.

[mlamouri (slow - plz ping), 2016-04-12 11:36:40]

still lgtm with comments applied

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
File chrome/browser/permissions/permission_manager.h (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
chrome/browser/permissions/permission_manager.h:102: permission_contexts_;
On 2016/04/12 at 07:44:54, raymes wrote:
> On 2016/04/11 14:06:05, Mounir Lamouri wrote:
> > What about:
> > std::unordered_map<content::PermissionType,
> > std::unique_ptr<PermissionContextBase>> ?
> 
> I tried this but got a bunch of link errors because there is no hash function
for PermissionType. I can look into this more tomorrow.

Would that help to do `enum class PermissionType :char`? Otherwise, I guess you
could implement the hash function (should be trivial). Worse case, why not using
std::vector. It's an enum and nearly all values are used so we could simply do
random access.

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/media/me...
File chrome/browser/media/media_permission.cc (right):

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/media/me...
chrome/browser/media/media_permission.cc:58:
PermissionManager::Get(profile_)->GetPermissionContext(permission_type);
Add a TODO maybe?

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/permissi...
File chrome/browser/permissions/permission_manager.h (right):

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/permissi...
chrome/browser/permissions/permission_manager.h:71: // TODO(raymes): Delete this
after refactoring MediaPermission!
Link to a bug number?

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/push_mes...
File chrome/browser/push_messaging/push_messaging_permission_context_unittest.cc
(right):

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/push_mes...
chrome/browser/push_messaging/push_messaging_permission_context_unittest.cc:168:
blink::mojom::PermissionStatus::DENIED);
Should that be PROMPT?

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/push_mes...
File chrome/browser/push_messaging/push_messaging_service_unittest.cc (right):

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/push_mes...
chrome/browser/push_messaging/push_messaging_service_unittest.cc:78:
CONTENT_SETTINGS_TYPE_PUSH_MESSAGING, CONTENT_SETTING_ALLOW);
Could we have a fake PermissionManager instead? No need to be addressed in this
CL.

[raymes, 2016-04-13 00:22:12]

The CQ bit was checked by raymes@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-04-13 00:22:31]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1871343002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1871343002/100001

[raymes, 2016-04-13 00:35:07]

Thanks Mounir!

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
File chrome/browser/permissions/permission_manager.h (right):

https://codereview.chromium.org/1871343002/diff/40001/chrome/browser/permissi...
chrome/browser/permissions/permission_manager.h:102: permission_contexts_;
Unfortunately doing enum class PermissionType : char didn't work. It seems like
this is fixed in c++ 14.

For now I defined a simple hash in PermissionManager. We could move it out if it
gets used more broadly. Does that seem ok?

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/media/me...
File chrome/browser/media/media_permission.cc (right):

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/media/me...
chrome/browser/media/media_permission.cc:58:
PermissionManager::Get(profile_)->GetPermissionContext(permission_type);
On 2016/04/12 11:36:39, Mounir Lamouri wrote:
> Add a TODO maybe?

Done.

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/permissi...
File chrome/browser/permissions/permission_manager.h (right):

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/permissi...
chrome/browser/permissions/permission_manager.h:71: // TODO(raymes): Delete this
after refactoring MediaPermission!
On 2016/04/12 11:36:39, Mounir Lamouri wrote:
> Link to a bug number?

Done.

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/push_mes...
File chrome/browser/push_messaging/push_messaging_permission_context_unittest.cc
(right):

https://codereview.chromium.org/1871343002/diff/80001/chrome/browser/push_mes...
chrome/browser/push_messaging/push_messaging_permission_context_unittest.cc:168:
blink::mojom::PermissionStatus::DENIED);
On 2016/04/12 11:36:40, Mounir Lamouri wrote:
> Should that be PROMPT?

In production code, I don't think this function will ever be called with prompt.
The result of calling RequestPermission on notifications should always be
granted or denied right?

[commit-bot: I haz the power, 2016-04-13 00:35:12]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-04-13 00:35:13]

Dry run: Try jobs failed on following builders:
  chromeos_x86-generic_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_x86-ge...)
  linux_chromium_gn_chromeos_rel on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[raymes, 2016-04-13 00:39:39]

The CQ bit was checked by raymes@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-04-13 00:40:00]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1871343002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1871343002/120001

[raymes, 2016-04-13 00:47:02]

raymes@chromium.org changed reviewers:
+ bauerb@chromium.org, dgrogan@chromium.org, jkarlin@chromium.org,
lfg@chromium.org, mlerman@chromium.org, mvanouwerkerk@chromium.org,
stevenjb@chromium.org, xhwang@chromium.org

[raymes, 2016-04-13 00:47:03]

+OWNERS

stevenjb:
  chrome/browser/chromeos/attestation/platform_verification_flow.cc
  chrome/browser/notifications/notification_permission_context_factory.cc
  chrome/browser/notifications/notification_permission_context_factory.h
  chrome/browser/notifications/notifier_state_tracker_factory.cc
  chrome/browser/notifications/platform_notification_service_browsertest.cc
mlerman:
  chrome/browser/profiles/chrome_browser_main_extra_parts_profiles.cc
  chrome/browser/profiles/off_the_record_profile_impl.cc
  chrome/browser/profiles/profile_impl.cc
jkarlin:
  chrome/browser/background_sync/background_sync_permission_context_factory.cc
  chrome/browser/background_sync/background_sync_permission_context_factory.h
bauerb:
  chrome/browser/android/tab_web_contents_delegate_android.cc
dgrogan:
  chrome/browser/storage/durable_storage_permission_context_factory.cc
  chrome/browser/storage/durable_storage_permission_context_factory.h
mvanouwerkerk:
  chrome/browser/geolocation/geolocation_permission_context_factory.cc
  chrome/browser/geolocation/geolocation_permission_context_factory.h
  chrome/browser/geolocation/geolocation_permission_context_unittest.cc
  chrome/browser/push_messaging/push_messaging_permission_context.cc
  chrome/browser/push_messaging/push_messaging_permission_context.h
  chrome/browser/push_messaging/push_messaging_permission_context_factory.cc
  chrome/browser/push_messaging/push_messaging_permission_context_factory.h
  chrome/browser/push_messaging/push_messaging_permission_context_unittest.cc
  chrome/browser/push_messaging/push_messaging_service_unittest.cc
xhwang:
  chrome/browser/media/media_permission.cc
  chrome/browser/media/media_stream_camera_permission_context_factory.cc
  chrome/browser/media/media_stream_camera_permission_context_factory.h
  chrome/browser/media/media_stream_mic_permission_context_factory.cc
  chrome/browser/media/media_stream_mic_permission_context_factory.h
  chrome/browser/media/midi_permission_context_factory.cc
  chrome/browser/media/midi_permission_context_factory.h
  chrome/browser/media/protected_media_identifier_permission_context.h
  chrome/browser/media/protected_media_identifier_permission_context_factory.cc
  chrome/browser/media/protected_media_identifier_permission_context_factory.h
lfg:
 
chrome/browser/guest_view/web_view/chrome_web_view_permission_helper_delegate.cc
 
chrome/browser/guest_view/web_view/chrome_web_view_permission_helper_delegate.h

[dgrogan, 2016-04-13 00:58:02]

durable lgtm

[lfg, 2016-04-13 01:08:43]

guest_view lgtm

[commit-bot: I haz the power, 2016-04-13 01:27:37]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-04-13 01:27:38]

Dry run: This issue passed the CQ dry run.

[mlamouri (slow - plz ping), 2016-04-13 13:06:32]

+jochen@

Jochen, can you rubberstamp the chrome/ changes resulting from the
chrome/browser/permissions/ changes that I reviewed?

[mlamouri (slow - plz ping), 2016-04-13 13:06:55]

mlamouri@chromium.org changed reviewers:
+ jochen@chromium.org

[mlamouri (slow - plz ping), 2016-04-13 13:06:57]

(this is not Inbox, people don't get added magically :()

[Mike Lerman, 2016-04-13 13:09:31]

profiles LGTM

[Michael van Ouwerkerk, 2016-04-13 13:55:55]

https://codereview.chromium.org/1871343002/diff/120001/chrome/browser/push_me...
File chrome/browser/push_messaging/push_messaging_permission_context.cc (right):

https://codereview.chromium.org/1871343002/diff/120001/chrome/browser/push_me...
chrome/browser/push_messaging/push_messaging_permission_context.cc:124: if
(notification_status == blink::mojom::PermissionStatus::DENIED) {
This is a behavioral change, as ASK now leads to ALLOW being persisted. Please
revert to the old logic.

https://codereview.chromium.org/1871343002/diff/120001/chrome/browser/push_me...
File chrome/browser/push_messaging/push_messaging_permission_context.h (right):

https://codereview.chromium.org/1871343002/diff/120001/chrome/browser/push_me...
chrome/browser/push_messaging/push_messaging_permission_context.h:50:
blink::mojom::PermissionStatus notifications_content_setting);
nit: update the argument name to reflect the type change e.g.
|notifications_permission|

[Bernhard Bauer, 2016-04-13 16:05:25]

c/b/android LGTM (feel free to TBR trivial changes like that if the overall
change has been approved).

[xhwang, 2016-04-13 16:08:20]

chrome/browser/media/ lgtm

[stevenjb, 2016-04-13 16:22:06]

RS LGTM

[raymes, 2016-04-14 01:56:45]

https://codereview.chromium.org/1871343002/diff/120001/chrome/browser/push_me...
File chrome/browser/push_messaging/push_messaging_permission_context.cc (right):

https://codereview.chromium.org/1871343002/diff/120001/chrome/browser/push_me...
chrome/browser/push_messaging/push_messaging_permission_context.cc:124: if
(notification_status == blink::mojom::PermissionStatus::DENIED) {
On 2016/04/13 13:55:54, Michael van Ouwerkerk wrote:
> This is a behavioral change, as ASK now leads to ALLOW being persisted. Please
> revert to the old logic.

ASK should never be the result of calling RequestPermission (which is what this
is the callback for). I added a DCHECK for that.

https://codereview.chromium.org/1871343002/diff/120001/chrome/browser/push_me...
File chrome/browser/push_messaging/push_messaging_permission_context.h (right):

https://codereview.chromium.org/1871343002/diff/120001/chrome/browser/push_me...
chrome/browser/push_messaging/push_messaging_permission_context.h:50:
blink::mojom::PermissionStatus notifications_content_setting);
On 2016/04/13 13:55:54, Michael van Ouwerkerk wrote:
> nit: update the argument name to reflect the type change e.g.
> |notifications_permission|

Done.

[raymes, 2016-04-14 01:57:11]

The CQ bit was checked by raymes@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-04-14 01:57:29]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1871343002/140002
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1871343002/140002

[commit-bot: I haz the power, 2016-04-14 02:47:23]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-04-14 02:47:25]

Dry run: This issue passed the CQ dry run.

[jkarlin, 2016-04-14 11:44:13]

background_sync lgtm

[jochen (gone - plz use gerrit), 2016-04-14 13:22:33]

lgtm

[Michael van Ouwerkerk, 2016-04-14 14:59:57]

lgtm

[raymes, 2016-04-18 00:47:13]

The CQ bit was checked by raymes@chromium.org

[raymes, 2016-04-18 00:47:13]

The patchset sent to the CQ was uploaded after l-g-t-m from
mlamouri@chromium.org, bauerb@chromium.org, stevenjb@chromium.org,
mlerman@chromium.org, dgrogan@chromium.org, xhwang@chromium.org,
lfg@chromium.org
Link to the patchset: https://codereview.chromium.org/1871343002/#ps140002
(title: " ")

[commit-bot: I haz the power, 2016-04-18 00:47:20]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1871343002/140002
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1871343002/140002

[commit-bot: I haz the power, 2016-04-18 02:00:59]

Description was changed from

==========
Remove PermissionContext factories

Currently there is no real benefit to having each PermissionContext as a keyed
service and it adds a lot of boilerplate. This CL moves PermissionContext
construction into the PermissionManager keyed service instead.
==========

to

==========
Remove PermissionContext factories

Currently there is no real benefit to having each PermissionContext as a keyed
service and it adds a lot of boilerplate. This CL moves PermissionContext
construction into the PermissionManager keyed service instead.
==========

[commit-bot: I haz the power, 2016-04-18 02:01:01]

Committed patchset #9 (id:140002)

[commit-bot: I haz the power, 2016-04-18 02:02:05]

Description was changed from

==========
Remove PermissionContext factories

Currently there is no real benefit to having each PermissionContext as a keyed
service and it adds a lot of boilerplate. This CL moves PermissionContext
construction into the PermissionManager keyed service instead.
==========

to

==========
Remove PermissionContext factories

Currently there is no real benefit to having each PermissionContext as a keyed
service and it adds a lot of boilerplate. This CL moves PermissionContext
construction into the PermissionManager keyed service instead.

Committed: https://crrev.com/e3afee6b62380b5b09caf7513701f66d8964c6b6
Cr-Commit-Position: refs/heads/master@{#387843}
==========

[commit-bot: I haz the power, 2016-04-18 02:02:07]

Patchset 9 (id:??) landed as
https://crrev.com/e3afee6b62380b5b09caf7513701f66d8964c6b6
Cr-Commit-Position: refs/heads/master@{#387843}