AudioWorklet Origin Trial

284 views
Skip to first unread message

hong...@chromium.org

unread,
Dec 11, 2017, 2:30:35 PM12/11/17
to experimentation-dev
Hello,

WebAudio team is finalizing the preparation of Origin Trial for AudioWorklet:

The code has been checked-in and the feature has been tracked by the OT feature pipeline sheet:

Could you add our feature to the sign up form?

Thanks!

-Hongchan

aro...@chromium.org

unread,
Dec 11, 2017, 4:51:57 PM12/11/17
to experimentation-dev
You listed three UseCounter values on the pipeline sheet. We would prefer to only have one counter. Which one would you say best captures overall trial usage?

Hongchan Choi

unread,
Dec 11, 2017, 5:46:01 PM12/11/17
to aro...@chromium.org, experimentation-dev
Hi Allen,

We decided to choose "AudioWorkletNodeConstructor". I edited the tracker accordingly.

Thanks!

-Hongchan

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation...@chromium.org.
To post to this group, send email to experimen...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/experimentation-dev/94f54439-64cc-4693-bc69-b6b1b5dbc728%40chromium.org.

hong...@chromium.org

unread,
Dec 12, 2017, 11:27:29 AM12/12/17
to experimentation-dev, aro...@chromium.org
Hello Allen,

What's my next step here? I would like to get the feature in the sing up form, so I can put the link to it on the blog post.

Cheers,
Hongchan


On Monday, December 11, 2017 at 2:46:01 PM UTC-8, Hongchan Choi wrote:
Hi Allen,

We decided to choose "AudioWorkletNodeConstructor". I edited the tracker accordingly.

Thanks!

-Hongchan

On Mon, Dec 11, 2017 at 1:52 PM <aro...@chromium.org> wrote:
You listed three UseCounter values on the pipeline sheet. We would prefer to only have one counter. Which one would you say best captures overall trial usage?

On Monday, December 11, 2017 at 2:30:35 PM UTC-5, hong...@chromium.org wrote:
Hello,

WebAudio team is finalizing the preparation of Origin Trial for AudioWorklet:

The code has been checked-in and the feature has been tracked by the OT feature pipeline sheet:

Could you add our feature to the sign up form?

Thanks!

-Hongchan

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

aro...@chromium.org

unread,
Dec 12, 2017, 1:41:33 PM12/12/17
to experimentation-dev, aro...@chromium.org
We usually wait until the release containing the trial has reached beta channel to reduce developer confusion. If I'm reading correctly, your trial is in M64, which has not yet been released to beta.

hong...@chromium.org

unread,
Dec 12, 2017, 1:48:26 PM12/12/17
to experimentation-dev, aro...@chromium.org
I see. Thanks for the clarification.

Hongchan Choi

unread,
Dec 15, 2017, 12:53:28 PM12/15/17
to experimentation-dev, aro...@chromium.org
Hello Allen,

On the sign-up form, I still can't see AudioWorklet in the drop down menu. Am I missing something? I believe M64 Beta rolled out and I can see "Disable Hardware Noise..." in the menu, which was registered later than AudioWorklet.

Thanks!

-Hongchan


To unsubscribe from this group and stop receiving emails from it, send an email to experimentation...@chromium.org.
To post to this group, send email to experimen...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation...@chromium.org.
To post to this group, send email to experimen...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/experimentation-dev/c41ab15f-3d8b-4b69-bc1e-65d6ee1fd329%40chromium.org.

aro...@chromium.org

unread,
Dec 15, 2017, 1:15:29 PM12/15/17
to experimentation-dev, aro...@chromium.org
Sorry for the delay. I have added it to the form.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

Hongchan Choi

unread,
Dec 15, 2017, 2:58:58 PM12/15/17
to aro...@chromium.org, experimentation-dev
Thanks so much, Allen! I can see the entry in the menu now.

-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation...@chromium.org.
To post to this group, send email to experimen...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation...@chromium.org.
To post to this group, send email to experimen...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation...@chromium.org.
To post to this group, send email to experimen...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/experimentation-dev/fbafd246-c2db-4680-b301-aa8868d7c5da%40chromium.org.

hong...@chromium.org

unread,
Jan 2, 2018, 2:54:05 PM1/2/18
to experimentation-dev, aro...@chromium.org
Hello Allen,

I do have some questions on Origin Trial frameworks. I followed the instruction here to test how OT works on AudioWorklet. I believe all the codes are checked in the latest beta (64.0.3282.39) but the testing routine does not seem to work. Here's what I did:

1. Generate a token by: tools/origin_trials/generate_token.py http://localhost:8000 AudioWorklet
2. The test web page has <meta http-equiv="origin-trial" content="GENERATED_TOKEN"> in it.
3. Launch the browser with "--origin-trial-public-key=dRCs+TocuKkocNKa0AtZ4awrt9XKH2SQCI6o4FY6BNA="

With all these steps, I still can't access AudioWorklet feature on the page.

I think I might have an issue: AudioWorklet depends on Worklet, which is also experimental. I naively assumed Origin Trial takes care of the dependency but it might not work that way. Is there any way to activate the dependent feature when AudioWorklet OT is granted?

Cheers,
Hongchan

-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

hong...@chromium.org

unread,
Jan 2, 2018, 3:10:29 PM1/2/18
to experimentation-dev, aro...@chromium.org
FWIW, the following command line activates AudioWorklet correctly on the latest beta.

--enable-blink-features=AudioWorklet

Not sure why OT does not work in the same way.

Marijn Kruisselbrink

unread,
Jan 2, 2018, 3:14:19 PM1/2/18
to hong...@chromium.org, experimentation-dev, aro...@chromium.org
Looking at the code, at least AudioWorklet::Create seems to check for RuntimeEnabledFeatures::AudioWorkletEnabled() rather than OriginTrials::audioWorkletEnabled, and thus will always return null if only the origin trial is enabled.

-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/experimentation-dev/dcd5cfd4-c64d-46ea-b5a9-5b456d07a88c%40chromium.org.

hong...@chromium.org

unread,
Jan 2, 2018, 3:19:22 PM1/2/18
to experimentation-dev, hong...@chromium.org, aro...@chromium.org
Oops. You're right. It is sad that I couldn't catch this in the code review. Perhaps merging the fix to M64 beta might be too late at this point.
-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/experimentation-dev/fbafd246-c2db-4680-b301-aa8868d7c5da%40chromium.org.

Jason Chase

unread,
Jan 2, 2018, 3:39:46 PM1/2/18
to hong...@chromium.org, experimentation-dev, aro...@chromium.org
Yes, Marijn correctly pointed out the problem. I'll add more detail.

Generally, when moving to an origin trial, all native code checks for "RuntimeEnabledFeatures::<my feature>Enabled()" should be replaced with calls to "OriginTrials::<my feature>Enabled()".

The OriginTrials methods include a check for the related RuntimeEnabledFeature. That allows client code can make one call, and work if either the runtime fiag OR origin trial are enabled. For example, see the generated code for OriginTrials::audioWorkletEnabled().

I looked at the usages of the AudioWorklet runtime flag, and noticed some other potential issues:
IDL not checking trial
  • BaseAudioContext.idl should probably be changed to use [OriginTrialEnabled=AudioWorklet] instead of [RuntimeEnabled=AudioWorklet]
  • In BaseAudioContext.cpp, there's a redundant check for both the runtime flag and origin trial
  • In AudioDestination.cpp, there's a DCHECK that uses the runtime flag. I guess this would only be an issue in debug builds.
The code integration for origin trials is described in:
(although it is not explicit as it could be for this scenario).

Thanks,
Jason

-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

hong...@chromium.org

unread,
Jan 2, 2018, 4:08:30 PM1/2/18
to experimentation-dev, hong...@chromium.org, aro...@chromium.org
Thanks for the clarification, Jason.

I fixed all of them but still can't get AudioWorklet activated. Are these steps correct for local testing?

1. Generate a token by: tools/origin_trials/generate_token.py http://localhost:8000 AudioWorklet
2. The test web page has <meta http-equiv="origin-trial" content="GENERATED_TOKEN"> in it.
3. Launch the browser with "--origin-trial-public-key=dRCs+TocuKkocNKa0AtZ4awrt9XKH2SQCI6o4FY6BNA="

For the local testing, I can use the public key as shown in the documentation. Right? The generator also shows "Signature (base 64)", but I believe this is not a token that should be on the page.

-Hongchan
-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

Jason Chase

unread,
Jan 2, 2018, 5:14:51 PM1/2/18
to hong...@chromium.org, experimentation-dev, aro...@chromium.org
Hongchan,

Those steps look correct for local testing. You should be able to use the public key, and you're correct that "Signature (base 64)" is *not* a token for the page.

You could try validating the token, by using the script: tools/origin_trials/check_token.py. If you copy the token from the webpage, you can at least verify no copy/paste issues from command-line to saving in html.

I'll try some local testing of my own and get back to you. I want to verify that nothing has broken since we last updated the documentation.

Thanks,
Jason

-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/experimentation-dev/0cf0ceea-a185-4268-a23d-4c0d3899b94e%40chromium.org.

hong...@chromium.org

unread,
Jan 2, 2018, 5:25:13 PM1/2/18
to experimentation-dev, hong...@chromium.org, aro...@chromium.org
Unfortunately this needs further digging. I fixed all the issues discussed in this thread and my local testing procedure is correct, but I still can't get the feature activated.
I also tried check_token.py against the generated token and it verifies correctly.

Thank you so much for your help on this, Jason!

-Hongchan
-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

hong...@chromium.org

unread,
Jan 3, 2018, 3:44:16 PM1/3/18
to experimentation-dev, hong...@chromium.org, aro...@chromium.org, Jason Chase
Hi Jason,

Here's WIP CL:

This CL makes |BaseAudioContext.audioWorklet| visible with a valid token.

But I am still experiencing other problems. When I load a test page with a token activated, I get "Uncaught ReferenceError: AudioWorkletProcessor is not defined" error from AudioWorkletGlobalScope. This means that somehow AudioWorkletProcessor class is not exposed in that worklet scope anymore. So I checked IDL files of both classes and they all have "OriginTrialEnabled=AudioWorklet" correctly:

I am not sure why AudioWorkletProcessor is not visible any more with the change ("OriginTrialEnabled=AudioWorklet") in BaseAudioContext's IDL. My suspicion is on the generated files and binding codes, but it would be great if you have any clue or a starting point.

For comparison, everything still works fine with "--enable-blink-features=AudioWorklet"

Thank you!

-Hongchan

Jason Chase

unread,
Jan 3, 2018, 4:41:01 PM1/3/18
to hong...@chromium.org, experimentation-dev, aro...@chromium.org, Jason Chase
Hi Hongchan,

I'm looking at the CL now, and the generated binding code. It would also help if you could provide the test page(s) that are generating the reference error.

Thanks,
Jason

-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

hong...@chromium.org

unread,
Jan 3, 2018, 4:43:47 PM1/3/18
to experimentation-dev, hong...@chromium.org, aro...@chromium.org, cha...@chromium.org
Jason,

I'll upload a layout test so it's easier for you to check. Thanks for looking into it!

-Hongchan

-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

Marijn Kruisselbrink

unread,
Jan 3, 2018, 4:49:33 PM1/3/18
to hong...@chromium.org, experimentation-dev, aro...@chromium.org, Jason Chase
I think part of the problem is that while WorkerGlobalScope call OriginTrialContext::AddTokens with the tokens in the creation params, WorkletGlobalScope seems to just ignore the tokens in the creation params, and thus worklets will never have origin trial tokens set.

-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/experimentation-dev/ffb772c4-2e46-4a78-9d66-1b5d91702774%40chromium.org.

hong...@chromium.org

unread,
Jan 3, 2018, 4:53:14 PM1/3/18
to experimentation-dev, hong...@chromium.org, aro...@chromium.org, cha...@chromium.org, Hiroki Nakagawa
That explains why AWGS behaves differently. Then Worklet infra change is inevitable - I am cc-ing nhiroki@ who works on Worklet infra.
-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.
To post to this group, send email to experimentation-dev@chromium.org.

Jason Chase

unread,
Jan 3, 2018, 5:08:02 PM1/3/18
to hong...@chromium.org, experimentation-dev, aro...@chromium.org, Jason Chase, Hiroki Nakagawa
Good catch Marijn, the handling of origin trial tokens should be addressed.

Hongchan, I'd still like to see the layout test, so I can rule out any issues with the bindings code. That could still be part of the problem.

Thanks,
Jason

-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsubscribe...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

hong...@chromium.org

unread,
Jan 3, 2018, 5:47:47 PM1/3/18
to experimentation-dev, hong...@chromium.org, aro...@chromium.org, cha...@chromium.org, nhi...@chromium.org
Just uploaded the new patch set with a layout test:
https://chromium-review.googlesource.com/c/chromium/src/+/849254

Currently the second test task fails with:
---
CONSOLE ERROR: line 7: Uncaught ReferenceError: AudioWorkletProcessor is not defined
CONSOLE ERROR: line 49: Uncaught (in promise) InvalidStateError: Failed to construct 'AudioWorkletNode': AudioWorkletNode cannot be created: The node name 'dummy' is not defined in AudioWorkletGlobalScope.
---

Unfortunately simply calling OT::addToken() in WorkletGlobalScope does not solve the issue.

Jason

-Hongchan

To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

--
You received this message because you are subscribed to the Google Groups "experimentation-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to experimentation-dev+unsub...@chromium.org.

To post to this group, send email to experimentation-dev@chromium.org.

hong...@chromium.org

unread,
Jan 16, 2018, 12:17:58 PM1/16/18
to experimentation-dev, hong...@chromium.org, aro...@chromium.org
Hello Allen,

I found some errors on public/internal pages:
2) Origin Trials Feature Pipeline document - "AudioWorklet" is not marked with bold font. Not sure if this is correct.

I thought the first error might confuse some developers.

WDYT?

-Hongchan

aro...@chromium.org

unread,
Jan 16, 2018, 5:08:45 PM1/16/18
to experimentation-dev, hong...@chromium.org, aro...@chromium.org
I have a PR out to update the available trials list.
Reply all
Reply to author
Forward
0 new messages