1138983 - RTSP not enabled in Lollipop

Status: RESOLVED FIXED

(Firefox OS Graveyard :: RTSP, defect)

Description

[Greg Grisco]

It seems RTSP has not been enabled on Lollipop. The MOZ_RTSP flag is set on Kitkat (19) bot not Lollipop (21).

Comment 1

[Mike Lee [:mlee]]

Hi Vincent & Shawn,

Can you help us find an owner for this fxOS 2.2 FL CAF blocker?

Ethan Tseng seems to be someone who's worked in this are in the past via https://bugzilla.mozilla.org/show_bug.cgi?id=929372#c9

Thanks,
Mike

Comment 2

[Vincent Chang[:vchang][changyihsin]]

Yeah, that's right. Ethan is working on this.

Comment 3

[Keven Kuo [:kkuo]]

Attachment: Patch

Comment 4

[Ethan Tseng [:ethan]]

Comment on attachment 8573010 [details]
Patch

Keven,

Thanks for your patch.
We will apply this patch and verify all RTSP function works well.
Thus, we'll take over this bug.

Thanks. :)

Comment 5

[Ethan Tseng [:ethan]]

Comment on attachment 8573010 [details]
Patch

Oops! I just wanted to cancel the review request, not meant to r-.
Sorry about that.

Comment 6

[Ethan Tseng [:ethan]]

Attachment: bug-1138983-fix.patch

what this patch does:
1. Add commit message.
2. Add code comment for the new argument of queueDiscontinuity().

Jonathan, since your already built nexus-5-l, please help to apply this patch and test RTSP function. Thanks.

Comment 7

[Keven Kuo [:kkuo]]

(In reply to Ethan Tseng [:ethan] from comment #4)
> Comment on attachment 8573010 [details]
> Patch
> 
> Keven,
> 
> Thanks for your patch.
> We will apply this patch and verify all RTSP function works well.
> Thus, we'll take over this bug.
> 
> Thanks. :)

Thanks

--
Keven

Comment 8

[Ethan Tseng [:ethan]]

Additional remark:
I don't know the appropriate situation to use "|discard|=false" when calling queueDiscontinuity(). However, at least it is right to set |disard|=true when RTSP handles seek discontinuity.

Comment 9

[howie [:howie]]

Triage: blocking

Comment 10

[Jonathan Hao (inactive) [:jhao]]

Comment on attachment 8573029 [details] [diff] [review]
bug-1138983-fix.patch

Review of attachment 8573029 [details] [diff] [review]:
-----------------------------------------------------------------

I've done some basic testing. RTSP is working well.

Comment 11

[Ethan Tseng [:ethan]]

(In reply to Jonathan Hao [:jhao] from comment #10)
> I've done some basic testing. RTSP is working well.
Glad to hear this. Thanks Jonathan!

Comment 12

[Ethan Tseng [:ethan]]

Comment on attachment 8573029 [details] [diff] [review]
bug-1138983-fix.patch

Vincent,
Please help to review this patch. Thanks.

Comment 13

[Vincent Chang[:vchang][changyihsin]]

Comment on attachment 8573029 [details] [diff] [review]
bug-1138983-fix.patch

Review of attachment 8573029 [details] [diff] [review]:
-----------------------------------------------------------------

::: netwerk/protocol/rtsp/rtsp/RTSPSource.cpp
@@ +507,5 @@
>  
>              TrackInfo *info = &mTracks.editItemAt(trackIndex);
>              sp<AnotherPacketSource> source = info->mSource;
>              if (source != NULL) {
> +#if ANDROID_VERSION >= 21

It seems no better way to prevent the compiler flag so far. Let's try to improve it in our refactor plan.

Comment 14

[Ethan Tseng [:ethan]]

The result of Treeherder:
https://treeherder.mozilla.org/#/jobs?repo=try&revision=215715029f5c

Comment 15

[Carsten Book [:Tomcat]]

 https://hg.mozilla.org/integration/b2g-inbound/rev/fcaa845b28e8

Comment 16

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/fcaa845b28e8

Comment 17

[Michael Vines [:m1] [:evilmachines]]

Comment on attachment 8573029 [details] [diff] [review]
bug-1138983-fix.patch

[Approval Request Comment]
User impact if declined: no RTSP for v2.2

Comment 18

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-b2g37_v2_2/rev/b3be5c1bc11c