Issue metadata
Sign in to add a comment
|
Regression: Ubertray is seen faded and not visible when any overlay is open in OOBE Screen |
||||||||||||||||||||
Issue descriptionChrome Version: 70.0.3538.13/11021.10.0 dev channel Daisy,Kip,Reks OS: Chrome OS What steps will reproduce the problem? (1)Recover build>> Observe Ubertray in OOBE screen, grey background is missing (2)Open Join WiFi overlay and observe ubertray is seen faded and is not visible Actual: Ubertray is seen faded and not visible when any overlay is open in OOBE Screen Expected: No such issue should be seen This is a Regression issue as same is working fine in 70.0.3538.0/11035.0.0 dev Attaching screencasts for reference...
,
Sep 10
jdufault@: Could you take a look? Thank you.
,
Sep 10
,
Sep 10
,
Sep 10
I bisected and found the source of the regression: https://chromium-review.googlesource.com/c/chromium/src/+/1197730 manucornet@, would you mind taking a look at this?
,
Sep 11
Friendly ping. Checking on the progress of these RBBs, please update with an ETA on fixes if this issue is still an RBB. Otherwise please remove the RBB label. Thanks.
,
Sep 14
Checking on the progress of these RBBs, please update with an ETA on fixes if this issue is still an RBB. Otherwise please remove the RBB label. Thanks.
,
Sep 17
Moving to RBS.
,
Sep 18
Sorry I didn't see this earlier, I was mostly focused on bugs filed against the shelf component. Ccing Sébastien. The videos are pretty blurry so I'm not completely sure what's happening here, but it seems like the problem is coming from the fact that the text and the background don't have very different colors. The added clear background used to create some contrast, but the new shelf makes it so that this background is a lot more subtle and the text becomes unreadable. It seems like we can either 1) revert the change in this mode (e.g. not show the rounded highlight unless we're in an active session), or 2) darken the text. 1) Seems a little safer (going backwards rather than forward) but again I'm not super sure what's happening here from the videos, anyone please feel free to educate me :-) Sébastien, what do you think?
,
Sep 19
We have a full spec for this for which the implementation is tracked in bug 874290 Spec: https://gallery.googleplex.com/projects/MCHbtQVoQ2HCZXS6M2rSjiqy/files/MCGWDM8eaujQbvFrcLi35FSwwMSJbxAVNpw @wzang, could this be related to your current work?
,
Sep 19
It seems that this bug is related to ubertray (I haven't touched the code there) instead of shelf. I believe yamaguchi@ implemented the ubertry portion of bug 874290. So maybe yamaguchi@ or tetsui@ have more contexts?
,
Sep 19
I am not familiar with the status area background, because I just have tweaked one of the icons there. Tetsui would have more context. So far I confirmed the icons and texts in the status tray results in #80868b on the screen, regardless of if it's shaded by the system-modal dialog exists or not. By this reason I think relevant component would be the way we show a system-modal dialog box darkening the other windows and backgrounds. Which component is it?
,
Sep 19
Now I've found the icons and text are drawn in front of the dialog (and shade). See the attached screenshot. Perhaps by the same reason, the unified system tray can be opened while the network popup is open, and the network popup can be opened nestedly. So we should keep the status area behind the popup and its shade. My two cents: In #12, the color of icons may look different between the two images (dark blue and grey), but it's an optical illusion of human eyes. The both are #80868b.
,
Sep 19
Okay, this sounds like something for Tetsui san? Let me know if you think I should take care of this. Once M-70 is out the door I'll be happy to take over the status tray :-)
,
Sep 28
It does not affect usability at all so removing RBS & lowering to P3. BTW, I realized Shelf in OOBE is not properly implemented according to the spec #10.
,
Oct 22
,
Nov 13
,
Dec 13
Bulk-assigning unassigned status area bugs to me. |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by tetsui@chromium.org
, Sep 10