[Proposal] Update mediaCapabilitiesProber API #1472

peaBerberian · 2024-06-26T17:15:28Z

After some brainstorming related to #1470, we noticed that as is, our old mediaCapabilitiesProber experimental tool could gain by being more flexible.

Especially, the getCompatibleDRMConfigurations method checked multiple EME MediaKeySystemConfiguration at once and returned information on actual configurations compatible to them once all of them have been checked.

We wondered whether this method would not be more useful by making it less powerful: we'll now only check for a single MediaKeySystemConfiguration per method call (and rename that method to checkDrmConfiguration).
Doing this signals to an application that it can just combine multiple calls if it wants the old behavior (e.g. through a Promise.all), use Promise.race to just obtain the first one etc. (this would still be possible with the previous API, but seemed unnatural to application developers).

I also added a timeout argument to it, but realized that it might be very complex to maintain, as theoretically checkDrmConfiguration could call multiple browser API sequentially in the future (as other mediaCapabilitiesProber's methods are doing).
In that case, where are we supposed to apply the timeout: to each browser API call or globally on the method?

I also profited from this work to rewrite the implementation of other mediaCapabilitiesProber's method without changing their API, as I found that a lot of its data-processing code was unneeded and difficult to follow.

doc/api/Tools/MediaCapabilitiesProber.md

src/experimental/tools/mediaCapabilitiesProber/probers/isTypeSupportedWithFeatures.ts

src/experimental/tools/mediaCapabilitiesProber/probers/HDCPPolicy.ts

src/experimental/tools/mediaCapabilitiesProber/probers/__tests__/decodingInfo.test.ts

src/experimental/tools/mediaCapabilitiesProber/probers/__tests__/mediaContentType.test.ts

src/experimental/tools/mediaCapabilitiesProber/probers/__tests__/mediaDisplayInfos.test.ts

src/experimental/tools/mediaCapabilitiesProber/probers/isTypeSupportedWithFeatures.ts

src/experimental/tools/mediaCapabilitiesProber/api.ts

This reverts commit fdd172a.

github-actions · 2025-04-22T18:57:26Z

✅ Automated performance checks have passed on commit 086705d2356d97175bb1b90a86b91cda7dd6d43f with the base branch dev.

Details

Performance tests 1st run output

No significative change in performance for tests:

Name	Mean	Median
loading	19.88ms -> 20.18ms (-0.295ms, z: 1.52915)	27.30ms -> 27.45ms
seeking	15.44ms -> 18.22ms (-2.784ms, z: 0.89372)	11.40ms -> 11.40ms
audio-track-reload	26.90ms -> 27.19ms (-0.291ms, z: 2.58275)	39.45ms -> 39.60ms
cold loading multithread	48.31ms -> 47.38ms (0.936ms, z: 10.74112)	71.10ms -> 69.90ms
seeking multithread	21.37ms -> 29.35ms (-7.981ms, z: 0.64715)	10.65ms -> 10.50ms
audio-track-reload multithread	26.78ms -> 26.79ms (-0.012ms, z: 1.02091)	39.30ms -> 39.30ms
hot loading multithread	15.87ms -> 15.80ms (0.072ms, z: 3.22256)	23.10ms -> 22.80ms

github-actions · 2025-05-07T14:26:15Z

✅ Automated performance checks have passed on commit 2440b2cabee15ee9da1a3a89ad633377aa1a230d with the base branch dev.

Details

Performance tests 1st run output

No significative change in performance for tests:

Name	Mean	Median
loading	19.30ms -> 19.65ms (-0.357ms, z: 1.92629)	26.40ms -> 26.55ms
seeking	15.26ms -> 15.86ms (-0.600ms, z: 2.58705)	11.10ms -> 11.10ms
audio-track-reload	26.26ms -> 26.15ms (0.104ms, z: 1.71632)	38.25ms -> 37.95ms
cold loading multithread	46.57ms -> 45.71ms (0.860ms, z: 13.05967)	67.95ms -> 66.75ms
seeking multithread	21.91ms -> 20.56ms (1.343ms, z: 0.22689)	10.35ms -> 10.35ms
audio-track-reload multithread	26.12ms -> 25.93ms (0.191ms, z: 2.06951)	38.10ms -> 37.95ms
hot loading multithread	15.32ms -> 15.02ms (0.301ms, z: 7.45733)	22.05ms -> 21.60ms

src/experimental/tools/mediaCapabilitiesProber/api.ts

Florent-Bouisset · 2025-05-15T15:08:41Z

src/experimental/tools/mediaCapabilitiesProber/probers/HDCPPolicy.ts

-export default function probeHDCPPolicy(
-  config: IMediaConfiguration,
-): Promise<[ProberStatus]> {
+export default async function probeHDCPPolicy(


I feel that this function can either throw or return "NotSupported"
I'm wondering if we should catch any errors and return "NotSupported"

Complex to say.

It fails if EME API failed to be called.

NotSupported would be if there's a high priority of the HDCP version being unsupported but here we're not so sure.

Maybe Unknown? I'm under the impression that for now Unknown is only for when the getStatusForPolicy API is not available, which is slightly different than e.g. requestMediaKeySystemAccess failing yet both would just indicate that it is unknown.....

Then, when applications handle promises, they generally also handle rejections (even linters often force that). So I'm not sure.

I removed the Unkown and it only returns "Supported" / "NotSupported" or reject

github-actions · 2025-06-16T14:58:39Z

✅ Automated performance checks have passed on commit afec1f5a5cc87c640ec77deb560a2f78318aaf89 with the base branch dev.

Details

Performance tests 1st run output

No significative change in performance for tests:

Name	Mean	Median
loading	18.97ms -> 18.92ms (0.060ms, z: 0.93514)	25.95ms -> 25.80ms
seeking	17.16ms -> 16.52ms (0.639ms, z: 1.07438)	10.95ms -> 10.95ms
audio-track-reload	25.76ms -> 26.10ms (-0.336ms, z: 0.53573)	37.65ms -> 37.65ms
cold loading multithread	46.17ms -> 45.28ms (0.886ms, z: 14.16195)	67.65ms -> 66.45ms
seeking multithread	19.15ms -> 14.46ms (4.696ms, z: 0.10779)	10.35ms -> 10.35ms
audio-track-reload multithread	25.73ms -> 25.72ms (0.003ms, z: 0.35913)	37.65ms -> 37.65ms
hot loading multithread	14.81ms -> 14.77ms (0.043ms, z: 5.09527)	21.45ms -> 21.30ms

ProbeHDCPPolicy can either, succeed with "Supported" | "NotSupported" | "Unknown" or reject. Success with "unknown" or reject both occurs when the navigator API failed to execute and the support of HDCP cannot be determined. To standardize the behavior of this function, the "Unknown" return is removed and replaced by a promise rejection.

github-actions · 2025-06-18T09:41:24Z

✅ Automated performance checks have passed on commit 8c27b3b86ebef5ef89330c718477a5680eb9cc77 with the base branch dev.

Details

Performance tests 1st run output

No significative change in performance for tests:

Name	Mean	Median
loading	19.58ms -> 19.91ms (-0.326ms, z: 0.34568)	26.85ms -> 26.85ms
seeking	19.97ms -> 16.03ms (3.946ms, z: 0.35553)	11.25ms -> 11.10ms
audio-track-reload	26.37ms -> 26.49ms (-0.116ms, z: 0.34248)	38.55ms -> 38.55ms
cold loading multithread	47.28ms -> 47.55ms (-0.271ms, z: 9.69561)	69.15ms -> 68.10ms
seeking multithread	18.59ms -> 22.58ms (-3.993ms, z: 0.60042)	10.35ms -> 10.35ms
audio-track-reload multithread	26.29ms -> 26.17ms (0.122ms, z: 1.69261)	38.55ms -> 38.25ms
hot loading multithread	15.54ms -> 15.42ms (0.113ms, z: 3.03935)	22.20ms -> 22.05ms

github-actions · 2025-06-18T10:21:00Z

✅ Automated performance checks have passed on commit b34f4c2b0a51e001ba09ccc86ae17df97d0473d2 with the base branch dev.

Details

Performance tests 1st run output

No significative change in performance for tests:

Name	Mean	Median
loading	20.26ms -> 20.29ms (-0.023ms, z: 0.28825)	27.90ms -> 28.05ms
seeking	22.85ms -> 18.93ms (3.915ms, z: 1.12495)	11.40ms -> 11.40ms
audio-track-reload	27.08ms -> 27.12ms (-0.039ms, z: 0.37983)	39.75ms -> 39.60ms
cold loading multithread	48.24ms -> 47.57ms (0.669ms, z: 8.88062)	70.95ms -> 69.75ms
seeking multithread	24.70ms -> 26.10ms (-1.397ms, z: 1.87441)	10.50ms -> 10.50ms
audio-track-reload multithread	26.87ms -> 26.82ms (0.051ms, z: 1.06271)	39.45ms -> 39.30ms
hot loading multithread	16.28ms -> 16.12ms (0.152ms, z: 2.77614)	23.55ms -> 23.40ms

peaBerberian added the proposal This Pull Request or Issue is only a proposal for a change with the expectation of a debate on it label Jun 26, 2024

peaBerberian force-pushed the misc/mcp-refacto branch from a78ca09 to bf9e326 Compare June 27, 2024 13:42

peaBerberian added this to the 4.2.0 milestone Jun 27, 2024

peaBerberian force-pushed the misc/mcp-refacto branch from bf9e326 to 52bf6ac Compare July 11, 2024 09:54

peaBerberian force-pushed the misc/mcp-refacto branch from 17e1a36 to 597dcb3 Compare July 24, 2024 16:44

peaBerberian added the Priority: 2 (Medium) This issue or PR has a medium priority. label Jul 26, 2024

peaBerberian force-pushed the misc/mcp-refacto branch 3 times, most recently from 99059eb to 538aff1 Compare September 4, 2024 16:57

peaBerberian modified the milestones: 4.2.0, 4.3.0 Oct 4, 2024

peaBerberian force-pushed the misc/mcp-refacto branch from 538aff1 to 0ea518d Compare October 8, 2024 15:02

Florent-Bouisset reviewed Oct 22, 2024

View reviewed changes

doc/api/Tools/MediaCapabilitiesProber.md Outdated Show resolved Hide resolved

src/experimental/tools/mediaCapabilitiesProber/probers/isTypeSupportedWithFeatures.ts Show resolved Hide resolved

Florent-Bouisset reviewed Oct 22, 2024

View reviewed changes

src/experimental/tools/mediaCapabilitiesProber/probers/HDCPPolicy.ts Outdated Show resolved Hide resolved

Florent-Bouisset reviewed Oct 22, 2024

View reviewed changes

src/experimental/tools/mediaCapabilitiesProber/probers/__tests__/decodingInfo.test.ts Show resolved Hide resolved

Florent-Bouisset reviewed Oct 22, 2024

View reviewed changes

src/experimental/tools/mediaCapabilitiesProber/probers/__tests__/mediaContentType.test.ts Show resolved Hide resolved

Florent-Bouisset reviewed Oct 22, 2024

View reviewed changes

peaBerberian force-pushed the misc/mcp-refacto branch from 0ea518d to a2ea9bf Compare November 15, 2024 17:29

peaBerberian force-pushed the misc/mcp-refacto branch from a2ea9bf to 12248ee Compare January 28, 2025 14:24

peaBerberian modified the milestones: 4.3.0, 4.4.0 Apr 7, 2025

peaBerberian force-pushed the dev branch 2 times, most recently from 00fc806 to b7216b4 Compare April 15, 2025 18:14

peaBerberian force-pushed the misc/mcp-refacto branch 2 times, most recently from 076c092 to aaec181 Compare April 22, 2025 18:19

peaBerberian added 4 commits April 22, 2025 20:25

MCP: addCheckDrmConfiguration API

14859dc

MCP: Remove confusing log methods

1fa7f61

MCP: rewrite getStatusForHDCP method

a030e1c

MCP: simplify checkDrmConfiguration implementation

f502cf5

peaBerberian and others added 11 commits April 22, 2025 20:25

MCP: re-implement getDecodingCapabilities

9c6f94c

MCP: Remove getCompatibleDRMConfigurations

3f29201

MCP: Rewrite getDisplayCapabilities

60e3bc7

Revert "MCP: Remove confusing log methods"

cf028e4

This reverts commit fdd172a.

MCP: Remove now unneeded code

35553ce

Update MediaCapabilitiesProber documentation

e99ebf8

MCP: Fix unit tests and move around files

b61104a

MCP: Remove too complex to maintain timeout option

f9ce194

MCP: fix integration tests

d554955

Last unit test fix

7c3b601

remove unnecessary eslint-disable

4156a49

peaBerberian force-pushed the misc/mcp-refacto branch from aaec181 to 4156a49 Compare April 22, 2025 18:25

canalplus deleted a comment from github-actions bot Apr 22, 2025

Improve MediaCapabilitiesProber logic after review

d3a0d1f

Florent-Bouisset reviewed May 15, 2025

View reviewed changes

mcp: minor updates after review

5420645

update doc accordingly to previous change

05fa9ab

peaBerberian merged commit 4f6e5de into dev Jun 24, 2025
11 checks passed

[Proposal] Update mediaCapabilitiesProber API #1472

[Proposal] Update mediaCapabilitiesProber API #1472

Uh oh!

Conversation

peaBerberian commented Jun 26, 2024

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-actions bot commented Apr 22, 2025

Performance tests 1st run output

Uh oh!

github-actions bot commented May 7, 2025

Performance tests 1st run output

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Florent-Bouisset May 15, 2025

Choose a reason for hiding this comment

Uh oh!

peaBerberian Jun 16, 2025

Choose a reason for hiding this comment

Uh oh!

Florent-Bouisset Jun 18, 2025

Choose a reason for hiding this comment

Uh oh!

github-actions bot commented Jun 16, 2025

Performance tests 1st run output

Uh oh!

github-actions bot commented Jun 18, 2025

Performance tests 1st run output

Uh oh!

github-actions bot commented Jun 18, 2025

Performance tests 1st run output

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants