test: Use real mouse events with Chromium in the shell #21419
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
martinpitt
added
the
no-test
martinpitt
force-pushed
the
chromium-shell-real-mouse
branch
from
18e9f45 to
1ce54dd
Compare
|
First attempt had many random pixel failures due to the mouse being in a random place and highlighting UI elements. Plus the HostSwitching regression. With moving the mouse away it looks better -- networking and storage pass, and some other and expensive failures. |
martinpitt
force-pushed
the
chromium-shell-real-mouse
branch
from
1ce54dd to
abd3b3f
Compare
|
Now it's down to host switcher failures, the rest passes. This is due to the animations -- see that utterly compressed menu in the screenshot. We need to wait for that to settle. |
martinpitt
removed
the
no-test
martinpitt
force-pushed
the
chromium-shell-real-mouse
branch
from
abd3b3f to
7ba1bd2
Compare
|
The precise combination of affected retries here brings back this tracer crash (second instance). While that's unrelated, it makes this PR unbecomingly red, looking. Handled by cockpit-project/bots#7210 |
You can't open the session menu to log out while the dialog is still open. That cheat only works with our Chromium test driver which directly inserts `MouseEvent`s. We are about to move these to real clicks.
Replace the static sleep with a polling loop.
Opening the host switcher and the host edit view animates (expands) the hosts menu for about half a second. When doing real clicks instead of MouseEvent injection, this causes the target elements to move around underneath the test, resulting in mis-clicks. Avoid that by waiting for the animation to be over. Unfortunately `ph_count_animations()` doesn't quite catch it completely in all cases, so add an extra sleep for robustness. This is similar to what we do in the pixel tests.
BiDi mouse events work fine on the top-level document (i.e. not in frames). So let's use them to test the shell. Generalize the "move mouse position to a harmless place" hack from commit b5b8c6d, so that we make the pixel snapshots without focusing a random element on the page. We can also run `TestPages.testPageStatus` again (reverts commit 529e334). This just needs one little tweak in TestHostSwitching.testBasic's pixel test. Our current reference does not expect any dialog item to be highlighted (when the mouse is over it), so move the mouse out of the dialog.
martinpitt
force-pushed
the
chromium-shell-real-mouse
branch
from
7ba1bd2 to
c7bdffc
Compare
|
Looking good! This is a known flake, I sent cockpit-project/bots#7211 to fix it. This failure is an unrelated known flaky pixel test, #21422 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
BiDi mouse events work fine on the top-level document (i.e. not in frames). So let's use them to test the shell.