Closed Bug 1450163 Opened 2 years ago Closed 2 years ago

Remove BrowserTestUtils.waitForSessionStoreUpdate from browser/base/content/test/webrtc/browser_devices_get_user_media_multi_process.js

Categories

(Firefox :: Site Permissions, enhancement, P3, trivial)

61 Branch
enhancement

Tracking

()

RESOLVED FIXED
Firefox 61
Tracking Status
firefox61 --- fixed

People

(Reporter: arai, Assigned: arai)

References

Details

Attachments

(1 file)

Remainder of bug 1444007 (and maybe bug 1447180)
BrowserTestUtils.waitForSessionStoreUpdate is used in browser/base/content/test/webrtc/browser_devices_get_user_media_multi_process.js to avoid mixing event between tabs after removing tab.
it should wait for something else more appropriate.
Rank: 29
Flags: needinfo?(jib)
Priority: -- → P3
Tooru, we'd be happy to take patches or advice here. Are there specific questions for the core WebRTC team? Otherwise these tests are from the front-end team I think. Re-categorizing. Also, do we use tracking flags for tests?
Component: WebRTC → Device Permissions
Flags: needinfo?(jib) → needinfo?(arai.unmht)
Product: Core → Firefox
Version: Trunk → 61 Branch
Here's some more context about the bug.

previously BrowserTestUtils.removeTab was returning a promise that is resolved when SessionStore information is updated,
but it's removed in bug 1442465, because in most case tests don't have to wait for it, and also it was misleading what they're actually waiting for.

Then, browser_devices_get_user_media_multi_process.js was also using BrowserTestUtils.removeTab.
Just to stop waiting on the Promise returned by BrowserTestUtils.removeTab breaks the test,
because the test had to wait for something after removing tab, to make sure the tab removal is reflected to internal data or UI, and SessionStore update event was late enough for it.

So, in bug 1442465 I switched it to explicitly use BrowserTestUtils.waitForSessionStoreUpdate, to wait after removing tab, as a temporary workaround.
the test should wait for more appropriate thing instead of non-related SessionStore update.


The question here is what those tests have to wait for.
I'm not sure whether it's core WebRTC thing or front-end thing tho.

if tracking flags are not appropriate, feel free to remove.
Flags: needinfo?(arai.unmht)
Comment on attachment 8965559 [details] [diff] [review]
Wait for the end of the second tab streams before checking the number of streams and indicator in browser/base/content/test/webrtc/browser_devices_get_user_media_multi_process.js.

Thanks!
Attachment #8965559 - Flags: review?(dao+bmo) → review+
https://hg.mozilla.org/integration/mozilla-inbound/rev/bfc452f47794b2d9b38a0dfe2b755858235048ff
Bug 1450163 - Wait for the end of the second tab streams before checking the number of streams and indicator in browser/base/content/test/webrtc/browser_devices_get_user_media_multi_process.js. r=dao
https://hg.mozilla.org/mozilla-central/rev/bfc452f47794
Status: ASSIGNED → RESOLVED
Closed: 2 years ago
Resolution: --- → FIXED
Target Milestone: --- → Firefox 61
You need to log in before you can comment on or make changes to this bug.