screen.orientation.lock should be allowed when window is fullscreen and owner document isn't fullscreen
Categories
(Core :: DOM: Core & HTML, defect)
Tracking
()
People
(Reporter: m_kato, Assigned: m_kato)
References
(Blocks 1 open bug, )
Details
Attachments
(2 files)
In the spec, (https://w3c.github.io/screen-orientation/#dfn-pre-lock-conditions), per-lock-condition isn't well-defined.
Gecko only allows document is pending fullscreen (bug 1744288. This is YouTube case) or fullscreen only. But Chrome allows that window is fullscreen even if ower document isn't fullscreen.
WPT (https://searchfox.org/mozilla-central/source/testing/web-platform/tests/screen-orientation/unlock.html) has the following test
promise_test(async (t) => {
t.add_cleanup(makeCleanup());
await test_driver.bless("request full screen");
await document.documentElement.requestFullscreen();
const iframe = await attachIframe();
const promise = iframe.contentWindow.screen.orientation.lock(
getOppositeOrientation()
);
screen.orientation.unlock();
await promise_rejects_dom(
t,
"AbortError",
iframe.contentWindow.DOMException,
promise
);
}, "unlock() aborts a pending lock request across documents");
Although iframe.contentWindow
isn't fullscreen state, screen.orientation.lock
is expectedly successful since parent document is fullscreen.
Assignee | ||
Updated•2 years ago
|
Updated•2 years ago
|
Assignee | ||
Comment 1•3 months ago
|
||
Currently, our per-lock condition for screen orientation lock API is
whether current document is fullscreen.
But Chrome and some WPT allow if nested iframe is fullscreen.
So I would like to allow screen orientation lock if nested iframe is
fullscreen and it is same origin.
Also, Chrome seems to allow it even if it is cross origin. If we should
allow it, I will work on it later.
Assignee | ||
Comment 2•3 months ago
|
||
After fixing part 1, /screen-orientation/fullscreen-interactions.html in
wpt is failure. this test doesn't work well even if part 1 isn't fixed.
Before fullscreen has been entered, fullscreen operation is canceled, we
should abort screen lock with security error.
Comment 3•2 months ago
|
||
Is there a spec bug for clarifying the spec's language?
Comment 4•18 days ago
|
||
(In reply to Peter Van der Beken [:peterv] from comment #3)
Is there a spec bug for clarifying the spec's language?
I filed https://github.com/w3c/screen-orientation/issues/255.
Description
•