- From: Jerry Smith (WPT) <jdsmith@microsoft.com>
- Date: Wed, 5 Oct 2016 00:04:55 +0000
- To: "Jerry Smith (WPT)" <jdsmith@microsoft.com>, David Dorwin <ddorwin@google.com>, Mark Watson <watsonm@netflix.com>
- CC: Paul Cotton <Paul.Cotton@microsoft.com>, Matt Wolenetz <wolenetz@google.com>, Philippe Le Hégaret <plh@w3.org>, "public-html-media@w3.org" <public-html-media@w3.org>
- Message-ID: <SN2PR03MB04740E9943937769E788EE6A4C40@SN2PR03MB047.namprd03.prod.outlook.com>
I can only run https on the public test harness link (https://w3c-test.org/tools/runner/index.html) which means without polyfills unless we merge them there.
- https://rawgit.com/w3c/test-results/gh-pages/encrypted-media/all.html
- https://rawgit.com/w3c/test-results/gh-pages/encrypted-media/less-than-2.html
- https://rawgit.com/w3c/test-results/gh-pages/encrypted-media/complete-fails.html
I don't get dramatically more passes using https, but there are also new tests added, so the numbers flexed from that as well.
- All:
o Was: Completely failed files: undefined; Completely failed subtests: 40; Failure level: 40/268 (14.93%)
o Now: Completely failed files: undefined; Completely failed subtests: 43; Failure level: 43/277 (15.52%)
- Less than 2:
o Was: Test files without 2 passes: 36; Subtests without 2 passes: 76; Failure level: 76/277 (27.44%)
o Now: Test files without 2 passes: 40; Subtests without 2 passes: 80; Failure level: 80/268 (29.85%)
- Complete fail:
o Was: Test files: 63; Total subtests: 268 (Polyfills double the test count, so this is actually 138 subtests total)
o Now: Test files: 72; Total subtests: 277
"Now" results are from the new pass using https. I do see these tests now passing on CH55 (at least) that didn't previously:
/encrypted-media/drm-mp4-playback-temporary-multikey.html (1/1, 100.00%, 0.37% of total)<http://www.w3c-test.org/encrypted-media/drm-mp4-playback-temporary-multikey.html>
drm, temporary, mp4, playback, multikey audio/video
/encrypted-media/drm-mp4-playback-temporary-multisession.html (1/1, 100.00%, 0.37% of total)<http://www.w3c-test.org/encrypted-media/drm-mp4-playback-temporary-multisession.html>
drm, temporary, mp4, playback with multiple sessions, multikey video
/encrypted-media/drm-mp4-playback-temporary-setMediaKeys-after-src.html (1/1, 100.00%, 0.37% of total)<http://www.w3c-test.org/encrypted-media/drm-mp4-playback-temporary-setMediaKeys-after-src.html>
drm, temporary, mp4, playback, setMediaKeys after setting video.src
/encrypted-media/drm-mp4-playback-temporary-setMediaKeys-after-update.html (1/1, 100.00%, 0.37% of total)<http://www.w3c-test.org/encrypted-media/drm-mp4-playback-temporary-setMediaKeys-after-update.html>
drm, temporary, mp4, playback, setMediaKeys after updating session
/encrypted-media/drm-mp4-playback-temporary-two-videos.html (1/1, 100.00%, 0.37% of total)<http://www.w3c-test.org/encrypted-media/drm-mp4-playback-temporary-two-videos.html>
drm, temporary, mp4, playback two videos
/encrypted-media/drm-mp4-playback-temporary-waitingforkey.html (1/1, 100.00%, 0.37% of total)<http://www.w3c-test.org/encrypted-media/drm-mp4-playback-temporary-waitingforkey.html>
drm, successful playback, temporary, mp4, waitingforkey event, 1 key
/encrypted-media/drm-mp4-playback-temporary.html (1/1, 100.00%, 0.37% of total)<http://www.w3c-test.org/encrypted-media/drm-mp4-playback-temporary.html>
drm, temporary, mp4, playback, single key
It seems from here out we should run tests on the public harness. Do we want to merge in the polyfilled tests (and update them each merge)? Mark: What's your take? My read is that the Polyfills are mostly affecting persistent sessions that will likely be put at risk.
Jerry
From: Jerry Smith (WPT) [mailto:jdsmith@microsoft.com]
Sent: Tuesday, October 4, 2016 2:29 PM
To: David Dorwin <ddorwin@google.com>; Mark Watson <watsonm@netflix.com>
Cc: Paul Cotton <Paul.Cotton@microsoft.com>; Matt Wolenetz <wolenetz@google.com>; Philippe Le Hégaret <plh@w3.org>; public-html-media@w3.org
Subject: RE: Updated EME Test Results
Thanks. I will rerun and post uodated reports a bit later today.
Sent from my Windows 10 phone
From: David Dorwin<mailto:ddorwin@google.com>
Sent: Tuesday, October 4, 2016 1:50 PM
To: Mark Watson<mailto:watsonm@netflix.com>
Cc: Jerry Smith (WPT)<mailto:jdsmith@microsoft.com>; Paul Cotton<mailto:Paul.Cotton@microsoft.com>; Matt Wolenetz<mailto:wolenetz@google.com>; Philippe Le Hégaret<mailto:plh@w3.org>; public-html-media@w3.org<mailto:public-html-media@w3.org>
Subject: Re: Updated EME Test Results
The tests need to be run using HTTPS. I can reproduce failures on HTTP.
On Tue, Oct 4, 2016 at 1:00 PM, Mark Watson <watsonm@netflix.com<mailto:watsonm@netflix.com>> wrote:
There's still quite a lot here which "should" pass but which is failing. We may have more regressions.
...Mark
On Tue, Oct 4, 2016 at 11:40 AM, Jerry Smith (WPT) <jdsmith@microsoft.com<mailto:jdsmith@microsoft.com>> wrote:
General:
Most of the Google tests have been migrated, and all of those are now disabled in the repository. I ran the tests late yesterday, and am produced modified reports that show polyfill results IF the test fails for a given browser without the polyfill. If you see "Polyfill" in the report, it means just that. Polyfill results aren't counted as a pass, and so show up in the failure reports as well.
Test reports:
- https://rawgit.com/w3c/test-results/gh-pages/encrypted-media/all.html
- https://rawgit.com/w3c/test-results/gh-pages/encrypted-media/less-than-2.html
- https://rawgit.com/w3c/test-results/gh-pages/encrypted-media/complete-fails.html
Summary:
Total TypeError failures: 14 tests Instead of InvalidAccessError
Total Timeout issues: 17 tests Many on persistent session types or specific temporary session use cases
Total Polyfill passes: 5 tests On persistent session types
Total temporary playback with one pass: 11 tests Need to be fixed
Total error case tests: 8 tests Are these tests necessary for CR?
Total test scoring issues, not real problems: 14 tests One passing implementation because test is keySystem specific
Misc other failures: 11 tests
Total subtest failures: 80 tests
At Risk Based on Test:
1. Temporary sessions
2. Persistent-license sessions
3. Persistent-usage-record sessions
Questions :
1. When can temporary session type failures be fixed?
2. When will TypeError and promise rejection failures be fixed?
3. Are all error case tests required to prove the spec?
Details:
1. /encrypted-media/clearkey-generate-request-disallowed-input.html (7/7, 100.00%, 2.61% of total)
CH55 & FF52 Fail: TypeError expected, got InvalidAccessError (7 tests)
2. /encrypted-media/clearkey-keystatuses.html (1/1, 100.00%, 0.37% of total)
CH55 Pass, FF 52 Fail: assert_equals: expected 0 but got 2
3. /encrypted-media/clearkey-mp4-playback-destroy-persistent-license.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/clearkey-mp4-playback-persistent-license-events.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/clearkey-mp4-playback-persistent-license.ht<http://playback-persistent-license.ht>ml (1/1, 100.00%, 0.37% of total)
/encrypted-media/clearkey-mp4-playback-persistent-usage-record-events.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/clearkey-mp4-playback-persistent-usage-record.html (1/1, 100.00%, 0.37% of total)
CH55 & FF%2 Pass with Polyfill - Known implementation issue, but include persistent sessions types that are not currently planned to be supported by all browsers (5 tests)
4. /encrypted-media/clearkey-mp4-playback-retrieve-destroy-persistent-license.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/clearkey-mp4-playback-retrieve-persistent-license.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/clearkey-mp4-playback-retrieve-persistent-usage-record.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/clearkey-mp4-playback-temporary-multisession.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-events.ht<http://drm-events.ht>ml (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-expiration.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-expiration.html
CH55, FF52, ED14: Test Timeouts (7 tests)
5. /encrypted-media/drm-generate-request-disallowed-input.html (7/7, 100.00%, 2.61% of total)
CH55, ED14, FF52: Most failures expect TypeError, got InValidAccessError (7 tests)
6. /encrypted-media/drm-keystatuses-multiple-sessions.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-keystatuses.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-destroy-persistent-license.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-persistent-license-events.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-persistent-license.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-persistent-usage-record-events.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-retrieve-persistent-usage-record.html (1/1, 100.00%, 0.37% of total)
CH55, ED14, FF52: Mostly Timeout (7 tests)
7. /encrypted-media/drm-mp4-playback-temporary-encrypted-clear-sources.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary-encrypted-clear-sources.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary-multikey.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary-multisession.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary-setMediaKeys-after-src.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary-setMediaKeys-after-update.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary-setMediaKeys-immediately.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary-setMediaKeys-onencrypted.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary-two-videos.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary-waitingforkey.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary.html (1/1, 100.00%, 0.37% of total)
CH55 & ED14 Timeout, FF52 Pass: All playback tests with MP4 content (11 tests)
8. /encrypted-media/drm-temporary-license-type.html (1/1, 100.00%, 0.37% of total)
CH55 & ED14 Timeout, FF52 Fail: Error case, shouldn't play with persistent license, but does (1 test)
9. /encrypted-media/idlharness.ht<http://idlharness.ht>ml (10/58, 17.24%, 3.73% of total)
CH55 & ED14 Fail: Throws TypeError instead of rejecting promise; FF52 Pass (9 tests)
CH55, ED14, FF52 Fail: MediaKeyMessageEvent.length expected 2, but got 1 (1 test)
10. /encrypted-media/clearkey-mp4-requestmediakeysystemaccess.ht<http://requestmediakeysystemaccess.ht>ml (4/58, 6.90%, 1.49% of total)
CH55, ED14, FF52 Fail one or more RequestMediaKeySystemAccess syntax checks/error cases (4 tests)
11. /encrypted-media/drm-mp4-playback-persistent-usage-record.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-retrieve-destroy-persistent-license.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-retrieve-persistent-license.html (1/1, 100.00%, 0.37% of total)
CH55, ED14, FF52: Timeout (3 tests)
12. /encrypted-media/drm-mp4-playback-temporary-events.html (1/1, 100.00%, 0.37% of total)
/encrypted-media/drm-mp4-playback-temporary-multikey-sequential.html (1/1, 100.00%, 0.37% of total)
CH55 & ED14: Timeout (2 tests)
FF52: Fail one test for sequence length of 9, 10 expected
FF52: Fail one test for Promise still outstanding at MediaKeys shutdown
13. /encrypted-media/drm-mp4-requestmediakeysystemaccess.html (17/72, 23.61%, 6.34% of total
CH55, ED14, FF52 Fail one or more RequestMediaKeySystemAccess syntax checks/error cases (3 tests)
ED14 only Pass: Test scoring issue, not valid single implementation feature (14 tests)
Jerry
Received on Wednesday, 5 October 2016 00:05:35 UTC