Add oem mount point to cros_mount_provider.

Add oem mount point to cros_mount_provider.

The new mount point will be accessible will be accessible by file manager
extension, but it won't be shown in the UI.
The mount point will contain Timescapes-ChromeOS video which will be playable in
the media player.

BUG=152216
TEST=manual (tried playing video on oem mount point using media player)


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=159856

[tbarzic, 2012-09-27 20:08:09]

Please, take a look.

Talked with Min Li and she confirmed with cmasone that it should be fine to
expose the whole oem partition.

[zel, 2012-09-27 20:11:10]

http://codereview.chromium.org/10993066/diff/1/chrome/browser/chromeos/extens...
File chrome/browser/chromeos/extensions/file_handler_util.cc (right):

http://codereview.chromium.org/10993066/diff/1/chrome/browser/chromeos/extens...
chrome/browser/chromeos/extensions/file_handler_util.cc:724: if
(url.filesystem_id() == kOemMountPointID) {
let's create a more general concept of read-only mount points than having
one-off checks like this. ideally this info should reside with the mount point
provider

[tbarzic, 2012-09-27 21:54:09]

did you mean something like this?

[zel, 2012-09-27 22:35:03]

'restricted' here is actually enforced 'read-only', you should put that
somewhere in comments.

also, this needs tests

[tbarzic, 2012-09-27 23:36:49]

On 2012/09/27 22:35:03, zel wrote:
> 'restricted' here is actually enforced 'read-only', you should put that
> somewhere in comments.

Done.
> 
> also, this needs tests

yeah, tests to come (eta tomorrow) :)

[tbarzic, 2012-09-29 01:16:11]

On 2012/09/27 23:36:49, tbarzic wrote:
> On 2012/09/27 22:35:03, zel wrote:
> > 'restricted' here is actually enforced 'read-only', you should put that
> > somewhere in comments.
> 
> Done.
> > 
> > also, this needs tests
> 
> yeah, tests to come (eta tomorrow) :)

I've added some browser tests.
There is some code duplication in the test extension, but I plan to refactor the
whole extension (there are some thing I don't like) in next week or two.. The
duplicated code should be gone then.

[zel, 2012-09-29 01:40:47]

lgtm

[kinuko, 2012-10-03 02:38:04]

http://codereview.chromium.org/10993066/diff/27001/webkit/chromeos/fileapi/cr...
File webkit/chromeos/fileapi/cros_mount_point_provider.cc (right):

http://codereview.chromium.org/10993066/diff/27001/webkit/chromeos/fileapi/cr...
webkit/chromeos/fileapi/cros_mount_point_provider.cc:194:
file_access_permissions_->GrantAccessPermission(
Do we also need to avoid calling this for the mount points whose type is
RestrictedNativeLocal?

[kinuko, 2012-10-03 02:38:38]

On 2012/10/03 02:38:04, kinuko wrote:
>
http://codereview.chromium.org/10993066/diff/27001/webkit/chromeos/fileapi/cr...
> File webkit/chromeos/fileapi/cros_mount_point_provider.cc (right):
> 
>
http://codereview.chromium.org/10993066/diff/27001/webkit/chromeos/fileapi/cr...
> webkit/chromeos/fileapi/cros_mount_point_provider.cc:194:
> file_access_permissions_->GrantAccessPermission(
> Do we also need to avoid calling this for the mount points whose type is
> RestrictedNativeLocal?

If this is not the case lgtm, sorry for belated review

[tbarzic, 2012-10-03 02:47:32]

http://codereview.chromium.org/10993066/diff/27001/webkit/chromeos/fileapi/cr...
File webkit/chromeos/fileapi/cros_mount_point_provider.cc (right):

http://codereview.chromium.org/10993066/diff/27001/webkit/chromeos/fileapi/cr...
webkit/chromeos/fileapi/cros_mount_point_provider.cc:194:
file_access_permissions_->GrantAccessPermission(
On 2012/10/03 02:38:04, kinuko wrote:
> Do we also need to avoid calling this for the mount points whose type is
> RestrictedNativeLocal?

no, if we don't call this here we won't be able to access the mount point from
file browser extension (which we need).

And it makes sense to give access to restricted mount point here, since we are
giving full access to the extension (this is done only for file browser
extension). The non-read operations will still be disabled.

[kinuko, 2012-10-03 02:51:53]

On 2012/10/03 02:47:32, tbarzic wrote:
>
http://codereview.chromium.org/10993066/diff/27001/webkit/chromeos/fileapi/cr...
> File webkit/chromeos/fileapi/cros_mount_point_provider.cc (right):
> 
>
http://codereview.chromium.org/10993066/diff/27001/webkit/chromeos/fileapi/cr...
> webkit/chromeos/fileapi/cros_mount_point_provider.cc:194:
> file_access_permissions_->GrantAccessPermission(
> On 2012/10/03 02:38:04, kinuko wrote:
> > Do we also need to avoid calling this for the mount points whose type is
> > RestrictedNativeLocal?
> 
> no, if we don't call this here we won't be able to access the mount point from
> file browser extension (which we need).
> 
> And it makes sense to give access to restricted mount point here, since we are
> giving full access to the extension (this is done only for file browser
> extension). The non-read operations will still be disabled.

Thanks for the clarification, lgtm++!

[commit-bot: I haz the power, 2012-10-03 03:09:03]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/tbarzic@chromium.org/10993066/27001

[commit-bot: I haz the power, 2012-10-03 06:07:59]

Change committed as 159856