add support for deviceId in gUM constraints

RESOLVED FIXED in Firefox 42

Status

()

P1
normal
Rank:
15
RESOLVED FIXED
5 years ago
3 years ago

People

(Reporter: blassey, Assigned: jib)

Tracking

unspecified
mozilla42
x86
Mac OS X
Points:
---
Dependency tree / graph

Firefox Tracking Flags

(firefox42 fixed, fennec-)

Details

(URL)

Attachments

(12 attachments, 2 obsolete attachments)

40 bytes, text/x-review-board-request
jib
: review+
Details
40 bytes, text/x-review-board-request
jib
: review+
Details
40 bytes, text/x-review-board-request
jib
: review+
Details
40 bytes, text/x-review-board-request
jib
: review+
Details
40 bytes, text/x-review-board-request
jib
: review+
Details
40 bytes, text/x-review-board-request
jib
: review+
Details
40 bytes, text/x-review-board-request
jib
: review+
Details
40 bytes, text/x-review-board-request
jib
: review+
Details
40 bytes, text/x-review-board-request
jib
: review+
Details
40 bytes, text/x-review-board-request
jib
: review+
Details
40 bytes, text/x-review-board-request
jib
: review+
Details
108.25 KB, patch
jib
: review+
Details | Diff | Splinter Review
Created attachment 8454389 [details] [diff] [review]
sourceId.patch

This is my best reading of the spec, wouldn't be surprised if I got something wrong though.
Attachment #8454389 - Flags: review?(rjesup)
Comment on attachment 8454389 [details] [diff] [review]
sourceId.patch

Review of attachment 8454389 [details] [diff] [review]:
-----------------------------------------------------------------

Hmmm. IIRC, SourceID's are supposed to be unique strings such as guids, which are used for selecting a source.   What's the source of the SourceID?  In this case, we want it to be unique-per-run (not persistent across runs).

reassigning review to jib, who has tracked this more closely
Attachment #8454389 - Flags: review?(rjesup) → review?(jib)
Comment on attachment 8454389 [details] [diff] [review]
sourceId.patch

(In reply to Randell Jesup [:jesup] from comment #1)
> Hmmm. IIRC, SourceID's are supposed to be unique strings such as guids,
> which are used for selecting a source.   What's the source of the SourceID? 
> In this case, we want it to be unique-per-run (not persistent across runs).

Actually a bit trickier.

From http://dev.w3.org/2011/webrtc/editor/getusermedia.html#iana-registrations :

 "The application-unique identifier for this source. The same identifier MUST be
  valid between sessions of this application, but MUST also be different for
  other applications. Some sort of GUID is recommended for the identifier."
Attachment #8454389 - Flags: review?(jib) → review-
(In reply to Jan-Ivar Bruaroey [:jib] from comment #2)
> Comment on attachment 8454389 [details] [diff] [review]
> sourceId.patch
> 
> (In reply to Randell Jesup [:jesup] from comment #1)
> > Hmmm. IIRC, SourceID's are supposed to be unique strings such as guids,
> > which are used for selecting a source.   What's the source of the SourceID? 
> > In this case, we want it to be unique-per-run (not persistent across runs).
> 
> Actually a bit trickier.
> 
> From
> http://dev.w3.org/2011/webrtc/editor/getusermedia.html#iana-registrations :
> 
>  "The application-unique identifier for this source. The same identifier
> MUST be
>   valid between sessions of this application, but MUST also be different for
>   other applications. Some sort of GUID is recommended for the identifier."

As this patch stands, passing "window" as the sourceId will result in the tab picker popping up. I plan to follow up with a patch to allow you to pass window:<window id> which will skip that dialog for privledged callers. If that's going down the wrong path, let me know.
(In reply to Brad Lassey [:blassey] (use needinfo?) from comment #3)
> If that's going down the wrong path, let me know.

I think so. If all we need is a boolean we should perhaps not overload sourceId with this as it might confuse people who are waiting for the actual sourceId behavior. We could make our own proprietary key and/or map out a proposal for standardization, unless there's a rationale behind doing this overload I'm missing. Also, why do we need the boolean? What happens differently for true and false?
Flags: needinfo?(martin.thomson)
Yeah, this is going to be difficult to get right.  For tab sharing, it is probably best if we have a separate constraint for tab/window selection.  For those, source identifier probably just needs to be something fixed but non-colliding, like "tab".

BTW, using GUID is probably not appropriate; we'll probably want to have a per-origin secret and use HMAC(per-origin-secret, device-identifier), with the secret sharing fate with cookies so that we regenerate as needed.  That likely means using indexedDB or something horrific like that.
Flags: needinfo?(martin.thomson)
tracking-fennec: ? → -
No longer blocks: 1037015
Assignee: nobody → jib
Summary: add support for sourceId in gUM constraints → add support for deviceId in gUM constraints
Created attachment 8602784 [details] [diff] [review]
Part 1: deviceId prep + move facingMode constraint

- Move device-level constraints (facingMode) down w/others (so advanced works right)
- Add deviceId WebIDL
- Cleanup id code a little.
Attachment #8454389 - Attachment is obsolete: true
Depends on: 1162412
Priority: -- → P2
Depends on: 1173255
Comment on attachment 8602784 [details] [diff] [review]
Part 1: deviceId prep + move facingMode constraint

Moved patch to Bug 1173255.
Attachment #8602784 - Attachment is obsolete: true
backlog: --- → webRTC+
Rank: 15
Priority: P2 → P1
Duplicate of this bug: 901616
Created attachment 8626251 [details]
MozReview Request: Bug 1037389 - move input validation to sync step of getUserMedia. r=jesup

Bug 1037389 - move input validation to sync step of getUserMedia
Attachment #8626251 - Flags: review?(rjesup)
Created attachment 8626252 [details]
MozReview Request: Bug 1037389 - getUserMedia calls enumerateDevices for device ids. r=jesup

Bug 1037389 - getUserMedia calls enumerateDevices for device ids
Attachment #8626252 - Flags: review?(rjesup)
Created attachment 8626253 [details]
MozReview Request: Bug 1037389 - factor applyConstraints out of enumeration. r=jesup

Bug 1037389 - factor applyConstraints out of enumeration
Attachment #8626253 - Flags: review?(rjesup)
Created attachment 8626254 [details]
MozReview Request: Bug 1037389 - make applyConstraints work again. r=jesup

Bug 1037389 - make applyConstraints work again
Attachment #8626254 - Flags: review?(rjesup)
Created attachment 8626255 [details]
MozReview Request: Bug 1037389 - make GetUserMediaDevices work again. r=jesup

Bug 1037389 - make GetUserMediaDevices work again
Attachment #8626255 - Flags: review?(rjesup)
Created attachment 8626256 [details]
MozReview Request: Bug 1037389 - add support for deviceId in gUM constraints. r=smaug, r=jesup

Bug 1037389 - add support for deviceId in gUM constraints
Attachment #8626256 - Flags: review?(rjesup)
Attachment #8626256 - Flags: review?(bugs)
Notes for jesup (smaug, for webidl change in last patch, see URL for relevant spec)

/r/12031 - Bug 1037389 - move input validation to sync step of getUserMedia

  - Spec change cleanup: https://github.com/w3c/mediacapture-main/issues/174
    Goal is to return already-rejected promises for everything that can be
    validated synchronously.

  - Removed mostly redundant GetUserMediaTask constructor

/r/12033 - Bug 1037389 - getUserMedia calls enumerateDevices for device ids

  - Streamline order in GetUserMedia() function further, for more "sync" errors.

  - note: EnumerateDevicesImpl still takes constraints in this version (gone later).

/r/12035 - Bug 1037389 - factor applyConstraints out of enumeration

  - Take constraints out of EnumerateDevicesImpl. i.e. split GetSources()
    into GetSources() and ApplyConstraints() (internal functions)
  - Remove SelectDevice() codepath, a separate mostly redundant codepath
    for applying constraints used only by privileged code.
  - GetUserMediaDevices (used by doorhangers) is gutted in this version.

/r/12037 - Bug 1037389 - make applyConstraints work again

  - Make ApplyConstraints internal function take a passed-in device list
    which is a combined audio/video list, so split it up into audio and video.

/r/12039 - Bug 1037389 - make GetUserMediaDevices work again

  - To avoid changing the doorhanger interface, GetUserMediaDevices
    still returns the device list, but no longer enumerates itself.
    Instead it returns the one already obtained by a recent getUserMedia
    request with the same inner window id. For a device list, it shouldn't
    matter which one (if there are multiple gUM requests), as long as they
    have the same origin, which should be guaranteed by the fact that
    onNavigation tears all requests down. Frames have their own window-id's I hope?

/r/12041 - Bug 1037389 - add support for deviceId in gUM constraints

  - As I'm compiling these notes, I'm realizing I added deviceId based on how
    facingMode used to work (they're both device-level constraints) which is
    actually wrong. E.g. { deviceId: "unknownid" } will fail now when it
    shouldn't. { deviceId: { exact: "unknownid" } } is needed to fail.

    I plan to address this in a follow-up patch before landing this that will
    push deviceId handling down to where facingMode is handled today (in case
    someone uses it in advanced). Expect some refactoring needed to cover audio.
So MediaTrackConstraintSet will have deviceId even in the spec?
Attachment #8626256 - Flags: review?(bugs) → review+
Comment on attachment 8626251 [details]
MozReview Request: Bug 1037389 - move input validation to sync step of getUserMedia. r=jesup

Bug 1037389 - move input validation to sync step of getUserMedia
Comment on attachment 8626252 [details]
MozReview Request: Bug 1037389 - getUserMedia calls enumerateDevices for device ids. r=jesup

Bug 1037389 - getUserMedia calls enumerateDevices for device ids
Comment on attachment 8626253 [details]
MozReview Request: Bug 1037389 - factor applyConstraints out of enumeration. r=jesup

Bug 1037389 - factor applyConstraints out of enumeration
Comment on attachment 8626254 [details]
MozReview Request: Bug 1037389 - make applyConstraints work again. r=jesup

Bug 1037389 - make applyConstraints work again
Comment on attachment 8626255 [details]
MozReview Request: Bug 1037389 - make GetUserMediaDevices work again. r=jesup

Bug 1037389 - make GetUserMediaDevices work again
Attachment #8626256 - Flags: review+ → review?(bugs)
Comment on attachment 8626256 [details]
MozReview Request: Bug 1037389 - add support for deviceId in gUM constraints. r=smaug, r=jesup

Bug 1037389 - add support for deviceId in gUM constraints
Created attachment 8626832 [details]
MozReview Request: Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work. r=jesup

Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work
Attachment #8626832 - Flags: review?(rjesup)
Created attachment 8626833 [details]
MozReview Request: Bug 1037389 - move constraint code to helper base class to share with audio. r=jesup

move constraint code to helper base class to share with audio
Attachment #8626833 - Flags: review?(rjesup)
That should be the last patch. Works for me. Test script: comment 16.

Try - https://treeherder.mozilla.org/#/jobs?repo=try&revision=37e6717c3063
Attachment #8626256 - Flags: review?(bugs) → review+
Created attachment 8626942 [details]
MozReview Request: Bug 1037389 - fixes: unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr. r=jesup

Bug 1037389 - unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr
Attachment #8626942 - Flags: review?(rjesup)
Comment on attachment 8626251 [details]
MozReview Request: Bug 1037389 - move input validation to sync step of getUserMedia. r=jesup

https://reviewboard.mozilla.org/r/12031/#review10761

::: dom/media/MediaManager.cpp:1337
(Diff revision 2)
> +    // Initialize MediaPermissionManager before send out any permission request.

send -> sending
Attachment #8626251 - Flags: review?(rjesup) → review+
Comment on attachment 8626251 [details]
MozReview Request: Bug 1037389 - move input validation to sync step of getUserMedia. r=jesup

Bug 1037389 - move input validation to sync step of getUserMedia
Attachment #8626251 - Flags: review+ → review?(rjesup)
Comment on attachment 8626252 [details]
MozReview Request: Bug 1037389 - getUserMedia calls enumerateDevices for device ids. r=jesup

Bug 1037389 - getUserMedia calls enumerateDevices for device ids
Comment on attachment 8626253 [details]
MozReview Request: Bug 1037389 - factor applyConstraints out of enumeration. r=jesup

Bug 1037389 - factor applyConstraints out of enumeration
Comment on attachment 8626254 [details]
MozReview Request: Bug 1037389 - make applyConstraints work again. r=jesup

Bug 1037389 - make applyConstraints work again
Comment on attachment 8626255 [details]
MozReview Request: Bug 1037389 - make GetUserMediaDevices work again. r=jesup

Bug 1037389 - make GetUserMediaDevices work again
Attachment #8626256 - Flags: review+ → review?(bugs)
Comment on attachment 8626256 [details]
MozReview Request: Bug 1037389 - add support for deviceId in gUM constraints. r=smaug, r=jesup

Bug 1037389 - add support for deviceId in gUM constraints
Comment on attachment 8626832 [details]
MozReview Request: Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work. r=jesup

Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work
Comment on attachment 8626833 [details]
MozReview Request: Bug 1037389 - move constraint code to helper base class to share with audio. r=jesup

move constraint code to helper base class to share with audio
Comment on attachment 8626942 [details]
MozReview Request: Bug 1037389 - fixes: unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr. r=jesup

Bug 1037389 - unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr
Created attachment 8628233 [details]
MozReview Request: Bug 1037389 - centralize some constraints code in a MediaEngine helper. r=jesup

Bug 1037389 - centralize some constraints code in a MediaEngine helper
Attachment #8628233 - Flags: review?(rjesup)
Created attachment 8628234 [details]
MozReview Request: Bug 1037389 - set fake streams using pref in tree. r=smaug, r=jesup

Bug 1037389 - set fake streams using pref in tree
* * *
Bug 1037389 - more encompassing fake-stream enabling in tree using pref
Attachment #8628234 - Flags: review?(rjesup)
Comment on attachment 8626256 [details]
MozReview Request: Bug 1037389 - add support for deviceId in gUM constraints. r=smaug, r=jesup

Didn't I review this already?
Attachment #8626256 - Flags: review?(bugs) → review+
Yes, sorry, not sure why Review Board thinks there are changes where there are none.

Once I stop using mq I hope review board will be a nicer experience.
oh, maybe it is because I just r+ in bugzilla, and mozreview might not know about it.
I'm sure that's it. I think Review board only emits to Bugzilla, not the other way.
Comment on attachment 8626251 [details]
MozReview Request: Bug 1037389 - move input validation to sync step of getUserMedia. r=jesup

Bug 1037389 - move input validation to sync step of getUserMedia
Comment on attachment 8626252 [details]
MozReview Request: Bug 1037389 - getUserMedia calls enumerateDevices for device ids. r=jesup

Bug 1037389 - getUserMedia calls enumerateDevices for device ids
Comment on attachment 8626253 [details]
MozReview Request: Bug 1037389 - factor applyConstraints out of enumeration. r=jesup

Bug 1037389 - factor applyConstraints out of enumeration
Comment on attachment 8626254 [details]
MozReview Request: Bug 1037389 - make applyConstraints work again. r=jesup

Bug 1037389 - make applyConstraints work again
Comment on attachment 8626255 [details]
MozReview Request: Bug 1037389 - make GetUserMediaDevices work again. r=jesup

Bug 1037389 - make GetUserMediaDevices work again
Attachment #8626256 - Flags: review+ → review?(bugs)
Comment on attachment 8626256 [details]
MozReview Request: Bug 1037389 - add support for deviceId in gUM constraints. r=smaug, r=jesup

Bug 1037389 - add support for deviceId in gUM constraints
Comment on attachment 8626832 [details]
MozReview Request: Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work. r=jesup

Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work
Comment on attachment 8626833 [details]
MozReview Request: Bug 1037389 - move constraint code to helper base class to share with audio. r=jesup

move constraint code to helper base class to share with audio
Comment on attachment 8626942 [details]
MozReview Request: Bug 1037389 - fixes: unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr. r=jesup

Bug 1037389 - unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr
Comment on attachment 8628233 [details]
MozReview Request: Bug 1037389 - centralize some constraints code in a MediaEngine helper. r=jesup

Bug 1037389 - centralize some constraints code in a MediaEngine helper
Comment on attachment 8628234 [details]
MozReview Request: Bug 1037389 - set fake streams using pref in tree. r=smaug, r=jesup

Bug 1037389 - set fake streams using pref in tree
* * *
Bug 1037389 - more encompassing fake-stream enabling in tree using pref
Attachment #8626256 - Flags: review?(bugs) → review+
Odd, in comment 28 Review Board was smart enough to know I'd only touched one patch in the set.

I suspect the lesson (with mq at least) is never hg pull during review. Sorry for the bug spam! :(
https://reviewboard.mozilla.org/r/12031/#review10761

> send -> sending

This code just moved, but sure.
(In reply to Olli Pettay [:smaug] from comment #42)
> Didn't I review this already?

That's Bug 1175166 apparently.
Comment on attachment 8628234 [details]
MozReview Request: Bug 1037389 - set fake streams using pref in tree. r=smaug, r=jesup

Bug 1037389 - set fake streams using pref in tree
* * *
Bug 1037389 - more encompassing fake-stream enabling in tree using pref
Attachment #8628234 - Flags: review?(bugs)
This one's actually a new webidl file to review.
Attachment #8628234 - Flags: review?(bugs) → review+
Comment on attachment 8626252 [details]
MozReview Request: Bug 1037389 - getUserMedia calls enumerateDevices for device ids. r=jesup

https://reviewboard.mozilla.org/r/12033/#review10765

::: dom/media/MediaManager.cpp:1503
(Diff revision 2)
> -                           const MediaStreamConstraints& aConstraints,
> +                           const MediaStreamConstraints& aConstraintsFDGSHJ,

What does FDGSHJ mean?  Please, something a little more descriptive.  Or a comment

::: dom/media/MediaManager.cpp:1736
(Diff revision 2)
> +//!!      ApplyConstraints(devices, c);

what does this mean?  Is it in another patch?

::: dom/media/MediaManager.cpp:1637
(Diff revision 2)
> -         src == dom::MediaSourceEnum::Screen)) {
> +                 src == dom::MediaSourceEnum::Screen)) {

if those are tabs, remove.  If RB is just indicating indent change, np
Attachment #8626252 - Flags: review?(rjesup) → review+
Attachment #8626253 - Flags: review?(rjesup) → review+
Comment on attachment 8626253 [details]
MozReview Request: Bug 1037389 - factor applyConstraints out of enumeration. r=jesup

https://reviewboard.mozilla.org/r/12035/#review10945

::: dom/media/MediaManager.cpp:1900
(Diff revision 4)
> -  nsRefPtr<PledgeSourceSet> p = EnumerateDevicesImpl(aWindowId, aConstraints);
> +  ScopedDeletePtr<SourceSet> devices; // !!!

// !!!
?

Assuming this is nothing important (just update the comment), ship it
Comment on attachment 8626254 [details]
MozReview Request: Bug 1037389 - make applyConstraints work again. r=jesup

https://reviewboard.mozilla.org/r/12037/#review10947

Ship It!
Attachment #8626254 - Flags: review?(rjesup) → review+
Comment on attachment 8626255 [details]
MozReview Request: Bug 1037389 - make GetUserMediaDevices work again. r=jesup

https://reviewboard.mozilla.org/r/12039/#review10949

Ship It!

::: dom/media/MediaManager.cpp:1956
(Diff revision 4)
> -  ScopedDeletePtr<SourceSet> devices; // !!!
> +  // Ignore passed-in constraints, instead locate + return already-constrained list.

Aha, so you can ignore the question on the earlier patch
Attachment #8626255 - Flags: review?(rjesup) → review+
Comment on attachment 8626256 [details]
MozReview Request: Bug 1037389 - add support for deviceId in gUM constraints. r=smaug, r=jesup

https://reviewboard.mozilla.org/r/12041/#review10953

Ship It!
Attachment #8626256 - Flags: review?(rjesup) → review+
Attachment #8626832 - Flags: review?(rjesup) → review+
Comment on attachment 8626832 [details]
MozReview Request: Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work. r=jesup

https://reviewboard.mozilla.org/r/12161/#review10957

::: dom/media/webrtc/MediaEngineDefault.h:63
(Diff revision 3)
> -    return true;
> +    return 0;

That was an oops... Good.
Comment on attachment 8626833 [details]
MozReview Request: Bug 1037389 - move constraint code to helper base class to share with audio. r=jesup

https://reviewboard.mozilla.org/r/12163/#review10959

Ship It!
Attachment #8626833 - Flags: review?(rjesup) → review+
Comment on attachment 8626942 [details]
MozReview Request: Bug 1037389 - fixes: unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr. r=jesup

https://reviewboard.mozilla.org/r/12175/#review10961

Ship It!
Attachment #8626942 - Flags: review?(rjesup) → review+
Comment on attachment 8628233 [details]
MozReview Request: Bug 1037389 - centralize some constraints code in a MediaEngine helper. r=jesup

https://reviewboard.mozilla.org/r/12353/#review10963

Ship It!
Attachment #8628233 - Flags: review?(rjesup) → review+
Comment on attachment 8628234 [details]
MozReview Request: Bug 1037389 - set fake streams using pref in tree. r=smaug, r=jesup

https://reviewboard.mozilla.org/r/12355/#review10965

Ship It!
Attachment #8628234 - Flags: review?(rjesup) → review+
Comment on attachment 8626251 [details]
MozReview Request: Bug 1037389 - move input validation to sync step of getUserMedia. r=jesup

https://reviewboard.mozilla.org/r/12031/#review10967

Ship It!
Attachment #8626251 - Flags: review?(rjesup) → review+
https://reviewboard.mozilla.org/r/12033/#review10765

> what does this mean?  Is it in another patch?

Yes it is replaced two patches down.
https://reviewboard.mozilla.org/r/12035/#review10945

> // !!!
> ?

Replaced two patches down from this patch.

Both seem used in the tree, and have different static analysis attached to them (according to Bug 1142816). Lets do what you suggest in a follow-up as I've been through a gauntlet of static analysis and try bustage to get to the current all-green try, and I'd like to land this asap.
Comment on attachment 8626251 [details]
MozReview Request: Bug 1037389 - move input validation to sync step of getUserMedia. r=jesup

Bug 1037389 - move input validation to sync step of getUserMedia. r=jesup
Attachment #8626251 - Attachment description: MozReview Request: Bug 1037389 - move input validation to sync step of getUserMedia → MozReview Request: Bug 1037389 - move input validation to sync step of getUserMedia. r=jesup
Attachment #8626251 - Flags: review+
Attachment #8626252 - Attachment description: MozReview Request: Bug 1037389 - getUserMedia calls enumerateDevices for device ids → MozReview Request: Bug 1037389 - getUserMedia calls enumerateDevices for device ids. r=jesup
Attachment #8626252 - Flags: review+
Comment on attachment 8626252 [details]
MozReview Request: Bug 1037389 - getUserMedia calls enumerateDevices for device ids. r=jesup

Bug 1037389 - getUserMedia calls enumerateDevices for device ids. r=jesup
Attachment #8626253 - Attachment description: MozReview Request: Bug 1037389 - factor applyConstraints out of enumeration → MozReview Request: Bug 1037389 - factor applyConstraints out of enumeration. r=jesup
Attachment #8626253 - Flags: review+
Comment on attachment 8626253 [details]
MozReview Request: Bug 1037389 - factor applyConstraints out of enumeration. r=jesup

Bug 1037389 - factor applyConstraints out of enumeration. r=jesup
Comment on attachment 8626254 [details]
MozReview Request: Bug 1037389 - make applyConstraints work again. r=jesup

Bug 1037389 - make applyConstraints work again. r=jesup
Attachment #8626254 - Attachment description: MozReview Request: Bug 1037389 - make applyConstraints work again → MozReview Request: Bug 1037389 - make applyConstraints work again. r=jesup
Attachment #8626254 - Flags: review+
Comment on attachment 8626255 [details]
MozReview Request: Bug 1037389 - make GetUserMediaDevices work again. r=jesup

Bug 1037389 - make GetUserMediaDevices work again. r=jesup
Attachment #8626255 - Attachment description: MozReview Request: Bug 1037389 - make GetUserMediaDevices work again → MozReview Request: Bug 1037389 - make GetUserMediaDevices work again. r=jesup
Attachment #8626255 - Flags: review+
Comment on attachment 8626256 [details]
MozReview Request: Bug 1037389 - add support for deviceId in gUM constraints. r=smaug, r=jesup

Bug 1037389 - add support for deviceId in gUM constraints. r=smaug, r=jesup
Attachment #8626256 - Attachment description: MozReview Request: Bug 1037389 - add support for deviceId in gUM constraints → MozReview Request: Bug 1037389 - add support for deviceId in gUM constraints. r=smaug, r=jesup
Attachment #8626256 - Flags: review+
Comment on attachment 8626832 [details]
MozReview Request: Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work. r=jesup

Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work. r=jesup
Attachment #8626832 - Attachment description: MozReview Request: Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work → MozReview Request: Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work. r=jesup
Attachment #8626832 - Flags: review+
Attachment #8626833 - Attachment description: MozReview Request: move constraint code to helper base class to share with audio → MozReview Request: Bug 1037389 - move constraint code to helper base class to share with audio. r=jesup
Attachment #8626833 - Flags: review+
Comment on attachment 8626833 [details]
MozReview Request: Bug 1037389 - move constraint code to helper base class to share with audio. r=jesup

Bug 1037389 - move constraint code to helper base class to share with audio. r=jesup
Attachment #8626942 - Attachment description: MozReview Request: Bug 1037389 - unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr → MozReview Request: Bug 1037389 - fixes: unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr. r=jesup
Attachment #8626942 - Flags: review+
Comment on attachment 8626942 [details]
MozReview Request: Bug 1037389 - fixes: unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr. r=jesup

Bug 1037389 - fixes: unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr. r=jesup
Attachment #8628233 - Attachment description: MozReview Request: Bug 1037389 - centralize some constraints code in a MediaEngine helper → MozReview Request: Bug 1037389 - centralize some constraints code in a MediaEngine helper. r=jesup
Attachment #8628233 - Flags: review+
Comment on attachment 8628233 [details]
MozReview Request: Bug 1037389 - centralize some constraints code in a MediaEngine helper. r=jesup

Bug 1037389 - centralize some constraints code in a MediaEngine helper. r=jesup
Comment on attachment 8628234 [details]
MozReview Request: Bug 1037389 - set fake streams using pref in tree. r=smaug, r=jesup

Bug 1037389 - set fake streams using pref in tree. r=smaug, r=jesup
Attachment #8628234 - Attachment description: MozReview Request: Bug 1037389 - set fake streams using pref in tree → MozReview Request: Bug 1037389 - set fake streams using pref in tree. r=smaug, r=jesup
Attachment #8628234 - Flags: review+
Keywords: checkin-needed
Attachment #8626251 - Flags: review+
Attachment #8626252 - Flags: review+
Attachment #8626253 - Flags: review+
Attachment #8626254 - Flags: review+
Comment on attachment 8626255 [details]
MozReview Request: Bug 1037389 - make GetUserMediaDevices work again. r=jesup

https://reviewboard.mozilla.org/r/12039/#review10997

Ship It!
Attachment #8626255 - Flags: review+
Comment on attachment 8626256 [details]
MozReview Request: Bug 1037389 - add support for deviceId in gUM constraints. r=smaug, r=jesup

https://reviewboard.mozilla.org/r/12041/#review10999

Ship It!
Attachment #8626256 - Flags: review+
Comment on attachment 8626832 [details]
MozReview Request: Bug 1037389 - fix plain deviceId constraint the right way, to make advanced etc work. r=jesup

https://reviewboard.mozilla.org/r/12161/#review11001

Ship It!
Attachment #8626832 - Flags: review+
Comment on attachment 8626833 [details]
MozReview Request: Bug 1037389 - move constraint code to helper base class to share with audio. r=jesup

https://reviewboard.mozilla.org/r/12163/#review11003

Ship It!
Attachment #8626833 - Flags: review+
Attachment #8626942 - Flags: review+
Comment on attachment 8626942 [details]
MozReview Request: Bug 1037389 - fixes: unremove recording-window-ended event on DENY_ACTION + avoid capturing rawptr. r=jesup

https://reviewboard.mozilla.org/r/12175/#review11005

Ship It!
Comment on attachment 8628233 [details]
MozReview Request: Bug 1037389 - centralize some constraints code in a MediaEngine helper. r=jesup

https://reviewboard.mozilla.org/r/12353/#review11007

Ship It!
Attachment #8628233 - Flags: review+
Comment on attachment 8628234 [details]
MozReview Request: Bug 1037389 - set fake streams using pref in tree. r=smaug, r=jesup

https://reviewboard.mozilla.org/r/12355/#review11009

Ship It!
Attachment #8628234 - Flags: review+
This should be good to go. Not sure why "status" is "pending". Who's loving Review Board right now?
Created attachment 8629107 [details] [diff] [review]
add support for deviceId in gUM constraints (merge of 11 patches). r=smaug, r=jesup

Merge of all patches here, for easier landing.
Flags: needinfo?(rjesup)
Attachment #8629107 - Flags: review+
Flags: needinfo?(rjesup)
https://hg.mozilla.org/mozilla-central/rev/0a8484884b6b
Status: NEW → RESOLVED
Last Resolved: 4 years ago
status-firefox42: --- → fixed
Resolution: --- → FIXED
Target Milestone: --- → mozilla42
This broke --disable-webrtc.
Whoops, fix in Bug 1180748.
Depends on: 1180748
Duplicate of this bug: 1170910
Depends on: 1046245
You need to log in before you can comment on or make changes to this bug.