third_party/WebKit/LayoutTests/imported/wpt/fullscreen/api/document-exit-fullscreen-timing-manual.html - Issue 2564543002: Add tests for fullscreenchange and fullscreenerror event timing

Side by Side Diff: third_party/WebKit/LayoutTests/imported/wpt/fullscreen/api/document-exit-fullscreen-timing-manual.html

Issue 2564543002: Add tests for fullscreenchange and fullscreenerror event timing (Closed)

Patch Set: Created 4 years ago

Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.

Jump to:

« no previous file with comments | « third_party/WebKit/LayoutTests/imported/wpt/MANIFEST.json ('k') | third_party/WebKit/LayoutTests/imported/wpt/fullscreen/api/element-request-fullscreen-timing-manual.html » ('j') | third_party/WebKit/LayoutTests/imported/wpt/fullscreen/api/element-request-fullscreen-timing-manual.html » ('J')
Toggle Intra-line Diffs ('i') | Expand Comments ('e') | Collapse Comments ('c') | Hide Comments ('s')

OLD	NEW
(Empty)
	1 <!DOCTYPE html>

	2 <title>Document#exitFullscreen() timing</title>

	3 <script src="/resources/testharness.js"></script>

	4 <script src="/resources/testharnessreport.js"></script>

	5 <script src="../trusted-click.js"></script>

	6 <div id="log"></div>

	7 <script>

	8 async_test(t => {

	9 const div = document.querySelector('div');

	10 trusted_request(div);

	11

	12 document.onfullscreenchange = t.step_func(() => {

	13 // We are now in fullscreen. Exit again.

	14 assert_equals(document.fullscreenElement, div);

	15 document.exitFullscreen();

	16

	17 // If fullscreenchange is an animation frame event, then animation frame

	18 // callbacks should be run after it is fired, before the timer callback.

	19 // The resize event, if any, should fire before the fullscreenchange event.

	20 let resized = false;

	21 document.onfullscreenchange = t.step_func(() => {

	22 if (!resized)

	23 console.info('resize event timing not tested');
	mlamouri (slow - plz ping) 2016/12/08 14:23:24 Should that fail? Should that fail? foolip 2016/12/08 14:33:03 I did it this way because of https://crbug.com/672 Show quoted text On 2016/12/08 14:23:24, mlamouri wrote: > Should that fail? I did it this way because of https://crbug.com/672436 But sure, this is for upstream WPT and this'll look strange, so I'll make the test fail instead, and make a copy of the test in the coming CL that skips the resize check with reference to that bug. mlamouri (slow - plz ping) 2016/12/08 15:12:10 You can also have an -expected.txt file if you hav Show quoted text On 2016/12/08 at 14:33:03, foolip wrote: > On 2016/12/08 14:23:24, mlamouri wrote: > > Should that fail? > > I did it this way because of https://crbug.com/672436 > > But sure, this is for upstream WPT and this'll look strange, so I'll make the test fail instead, and make a copy of the test in the coming CL that skips the resize check with reference to that bug. You can also have an -expected.txt file if you have expected failures. foolip 2016/12/08 15:16:57 Yes, but this failure is early, and so would mask Show quoted text On 2016/12/08 15:12:10, mlamouri wrote: > On 2016/12/08 at 14:33:03, foolip wrote: > > On 2016/12/08 14:23:24, mlamouri wrote: > > > Should that fail? > > > > I did it this way because of https://crbug.com/672436 > > > > But sure, this is for upstream WPT and this'll look strange, so I'll make the > test fail instead, and make a copy of the test in the coming CL that skips the > resize check with reference to that bug. > > You can also have an -expected.txt file if you have expected failures. Yes, but this failure is early, and so would mask later failures. I'll try to instead record the order of events and assert_array_equals after the setTimeout instead, that should change the expected output if the first element of the array changes. mlamouri (slow - plz ping) 2016/12/08 15:21:55 In case of it can help, testharness.js has an `Eve Show quoted text On 2016/12/08 at 15:16:57, foolip wrote: > On 2016/12/08 15:12:10, mlamouri wrote: > > On 2016/12/08 at 14:33:03, foolip wrote: > > > On 2016/12/08 14:23:24, mlamouri wrote: > > > > Should that fail? > > > > > > I did it this way because of https://crbug.com/672436 > > > > > > But sure, this is for upstream WPT and this'll look strange, so I'll make the > > test fail instead, and make a copy of the test in the coming CL that skips the > > resize check with reference to that bug. > > > > You can also have an -expected.txt file if you have expected failures. > > Yes, but this failure is early, and so would mask later failures. I'll try to instead record the order of events and assert_array_equals after the setTimeout instead, that should change the expected output if the first element of the array changes. In case of it can help, testharness.js has an `EventWatcher`. foolip 2016/12/08 15:57:45 Yeah, I've been wanting to make use of that, but h Show quoted text On 2016/12/08 15:21:55, mlamouri wrote: > On 2016/12/08 at 15:16:57, foolip wrote: > > On 2016/12/08 15:12:10, mlamouri wrote: > > > On 2016/12/08 at 14:33:03, foolip wrote: > > > > On 2016/12/08 14:23:24, mlamouri wrote: > > > > > Should that fail? > > > > > > > > I did it this way because of https://crbug.com/672436 > > > > > > > > But sure, this is for upstream WPT and this'll look strange, so I'll make > the > > > test fail instead, and make a copy of the test in the coming CL that skips > the > > > resize check with reference to that bug. > > > > > > You can also have an -expected.txt file if you have expected failures. > > > > Yes, but this failure is early, and so would mask later failures. I'll try to > instead record the order of events and assert_array_equals after the setTimeout > instead, that should change the expected output if the first element of the > array changes. > > In case of it can help, testharness.js has an `EventWatcher`. Yeah, I've been wanting to make use of that, but have had to stay away for many fullscreen tests because it uses promises and thus would get in the way of verifying the exact timing of when events are fired, which is the point of this test. I'm sure there are some fullscreen tests that I could simplify with EventWatcher though, and leave the exact timing to tests like this.
	24 window.onresize = t.unreached_func('resize event');

	25 setTimeout(t.unreached_func('timer callback'));

	26 requestAnimationFrame(t.step_func_done());

	27 });

	28 window.onresize = t.step_func(() => resized = true);

	29 });

	30 }, document.title + ' of fullscreenchange event');

	31 </script>

OLD	NEW