User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/79.0.3945.117 Safari/537.36

Steps to reproduce:

Open this page https://jinyeng.github.io/HTMLNotificationsChromeBug/
Click on Full Screen button from inside of iframe. It works.
Click on Full Screen button from parent page.

Actual results:

When cliking fullscreen from parent page. It does not enter in to fullscreen.
This works fine in chrome and used to work in previous version of Firefox as well.

Expected results:

It should work in same way when called from parent page full screen button.
This works fine in chrome and used to work in previous version of Firefox as well.

Hello, i managed to reproduce the issue on Firefox 72.0.1, beta 73.0b7, and Nightly 74.0a1 (2020-01-19).
This problem seems to have appeared somewhere in version 71.
Note that the following message is printed in the console when the button is pressed:
"Request for fullscreen was denied because Element.requestFullscreen() was not called from inside a short running user-generated event handler."

The priority flag is not set for this bug.
:Dolske, could you have a look please?

For more information, please visit auto_nag documentation.

Can you find a regression window?

I believe it is because of bug 1129227. We probably need bug 1578357 like Chrome does.

Hello, I've done a regression check using mozregression, here are the results:

2020-01-30T17:43:46: INFO : Narrowed inbound regression window from [88bfb2e6, 011193f6] (4 builds) to [b7dd0f12, 011193f6] (2 builds) (~1 steps left)
2020-01-30T17:43:46: DEBUG : Starting merge handling...
2020-01-30T17:43:46: DEBUG : Using url: https://hg.mozilla.org/integration/autoland/json-pushes?changeset=011193f69f2fad43eeb92f597954b52d3e7755e8&full=1
2020-01-30T17:43:48: DEBUG : Found commit message:
Bug 1585519 - Part 4: Remove unnecessary parentheses. r=Yoric CLOSED TREE

Reviewers: Yoric

Subscribers: jandem

Bug #: 1585519

Differential Revision: https://phabricator.services.mozilla.com/D48293

edgar: Do you think that the regression range (bug 1585519) is reasonable for your thought?

(In reply to Masayuki Nakano [:masayuki] (he/him)(JST, +0900) from comment #6)

edgar: Do you think that the regression range (bug 1585519) is reasonable for your thought?

No, this doesn't look like something related to bug 1585519. Bug 1585519 is to add MOZ_LIKELY/MOZ_UNLIKELY, I don't expect it will change fullscreen behavior.

Daniel, could you check again?
Note: please don't interact with iframe during the test, e.g. click, otherwise, the iframe will be considered as user-activated and allow the full-screen request. Which will probably lead to a wrong result. Thanks.

I've repeated the test, this time without interacting with the iframe whatsoever, and did indeed reach a different conclusion.
Just to make sure we're on the same page, my steps were:

1. Reach the page
2. Click the Full Screen Parent Page button

with these steps I've arrived to this build:
DEBUG : Found commit message:
Bug 1129227 - Part 3: Add test for requesting fullscreen in async callback; r=smaug

Differential Revision: https://phabricator.services.mozilla.com/D47152

edgar: Could you set proper priority for this bug? Thanks.

I am working on bug 1578357, I expect it could fix this issue.

This should be fixed by bug 1578357, but I would like to add a test for fullscreen here.

https://treeherder.mozilla.org/#/jobs?repo=try&revision=dc859f9fa75ac6d1fbd2249a0fb6594054420c2f&group_state=expanded

https://hg.mozilla.org/mozilla-central/rev/060b2b69c447

Verified - Fixed - The issue can no longer be reproduced. Checked on MacOS 10.15, Windows 10 and Ubuntu 18.04.