Inverting the font and background colors with Always override option will break the newtab page content
Categories
(Firefox :: New Tab Page, defect)
Tracking
()
People
(Reporter: vlucaci, Assigned: morgan)
References
Details
(Keywords: regression, Whiteboard: access-P2)
Attachments
(1 file)
(deleted),
image/png
|
Details |
Affected versions
- 72.0b8
- 71.0
Affected platforms
- Ubuntu 18.04x64
Steps to reproduce
- Launch FF with a fresh profile.
- Got to about:preferences#general
- Go to Language&Appearence section, Colors
- Invert the Text&Background colors (Black text to White, and White background to black)
- Select "Always" in the override the colors drop down.
- Press ok.
- Open a new tab.
Expected result
- The page is properly rendered and the icons of the visited and bookmarked websites are properly displayed.
Actual result
- The contents of the visited and bookmarked pages have no icons and the page is not rendered correctly.
Regression range
- will return with regression ASAP.
Additional notes
- As long as the Always option in the "override the colors" drop down,remains active, the page will not be rendered correctly and the website icons will not be displayed even after reverting the colors to default.
*Does not occur in latest FF Nightly
Screencast to issues
Comment 1•5 years ago
|
||
Hey Morgan, do you have a sense of how important this issue is, accessibility-wise?
Assignee | ||
Comment 2•5 years ago
|
||
(In reply to Mike Conley (:mconley) (:⚙️) (Wayyyy behind on needinfos) from comment #1)
Hey Morgan, do you have a sense of how important this issue is, accessibility-wise?
Hello! I'd access-p1 this by: https://docs.google.com/document/d/1UOI65Ghjj615NOvdOTWtCW790lgikDiUsE8Ss7pkdo0/edit
Assignee | ||
Comment 3•5 years ago
|
||
Old high contrast mode prevented images from rendering; looks like that's what's happening in the screenshots. Might be worth checking out if this was changed by 1594099 landing (this allows images)
Comment 4•5 years ago
|
||
Thanks, Morgan! Going to mark P3 for now. ni?ing vlucaci for that regression range too.
Comment 5•5 years ago
|
||
I think this is a dupe of bug 1445266, but using the prefs instead of high contrast mode (which we don't support auto-detecting on Linux, cf bug 239914). Pinging Mardak to check. Either way, I don't think this is fixable in preferences.
It'd be useful to confirm if this was indeed fixed on nightly by bug 1594099 - per that bug, I think that's being held on Nightly so a fix on release isn't forthcoming just yet?
Assignee | ||
Comment 6•5 years ago
|
||
This is what I'm seeing when I follow the STR after landing backplate work in Nightly.
Backplate stuff is going to wait one more cycle in beta before being uplifted to release (so released in 73, I think?)
Comment 7•5 years ago
|
||
I have managed to find a regression for this issue here is the pushlog https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=80ff3f300e05f38f96c385b03d1973a966a2bd35&tochange=7d15bc419c6cd7e9f3b4d41370c3b0e5990c8d1b
. I want to mention that the pushlog provided is from when the new tab page was changed to the new design, also this behaviour reproduces as low as Fx 33.0a1.
Updated•5 years ago
|
Comment 9•5 years ago
|
||
I have performed a regression of the fix using mozregression and these are my results:
2020-01-10T16:41:01: INFO : Narrowed inbound regression window from [86961574, f44e6327] (3 builds) to [86961574, 6baef669] (2 builds) (~1 steps left)
2020-01-10T16:41:01: DEBUG : Starting merge handling...
2020-01-10T16:41:01: DEBUG : Using url: https://hg.mozilla.org/integration/autoland/json-pushes?changeset=6baef669e147828062e93a6e7edec259853e7304&full=1
2020-01-10T16:41:02: DEBUG : Found commit message: Bug 1539212: Add readability backplate for all elements containing text when HCM or a11y theme is active. r=dholbert,mattwoodrow,Jamie
Differential Revision: https://phabricator.services.mozilla.com/D42979
2020-01-10T16:41:02: DEBUG : Did not find a branch, checking all integration branches
2020-01-10T16:41:02: INFO : The bisection is done.
2020-01-10T16:41:02: INFO : Stopped
Conclusion: The fix appears to be pushed for bug 1539212.
Comment 10•5 years ago
|
||
(In reply to Bodea Daniel [:danibodea] from comment #9)
I have performed a regression of the fix using mozregression and these are my results:
2020-01-10T16:41:01: INFO : Narrowed inbound regression window from [86961574, f44e6327] (3 builds) to [86961574, 6baef669] (2 builds) (~1 steps left) 2020-01-10T16:41:01: DEBUG : Starting merge handling... 2020-01-10T16:41:01: DEBUG : Using url: https://hg.mozilla.org/integration/autoland/json-pushes?changeset=6baef669e147828062e93a6e7edec259853e7304&full=1 2020-01-10T16:41:02: DEBUG : Found commit message: Bug 1539212: Add readability backplate for all elements containing text when HCM or a11y theme is active. r=dholbert,mattwoodrow,Jamie Differential Revision: https://phabricator.services.mozilla.com/D42979 2020-01-10T16:41:02: DEBUG : Did not find a branch, checking all integration branches 2020-01-10T16:41:02: INFO : The bisection is done. 2020-01-10T16:41:02: INFO : Stopped
Conclusion: The fix appears to be pushed for bug 1539212.
Sorry, I'm confused - are you saying bug 1539212 fixed this? Because you then marked it as a regressor, ie saying that bug caused this one. Those are opposite things...
Comment 11•5 years ago
|
||
Yes, that is what fixed the issue. Do you still need me to find out when it was introduced?
Comment 12•5 years ago
|
||
No, I think it's not a recent regression and there's a fix on nightly (albeit behind a pref that's not riding the trains at the moment, I think?) so that's fine for now.
Assignee | ||
Comment 13•5 years ago
|
||
(In reply to :Gijs (he/him) from comment #12)
No, I think it's not a recent regression and there's a fix on nightly (albeit behind a pref that's not riding the trains at the moment, I think?) so that's fine for now.
It [backplate]'ll be fully released in 73; if there's a way to note that this bug will be fixed then, let me know and I'll mark it :)
Updated•5 years ago
|
Comment 14•5 years ago
|
||
(In reply to Morgan Reschenberg [:morgan] from comment #13)
(In reply to :Gijs (he/him) from comment #12)
No, I think it's not a recent regression and there's a fix on nightly (albeit behind a pref that's not riding the trains at the moment, I think?) so that's fine for now.
It [backplate]'ll be fully released in 73; if there's a way to note that this bug will be fixed then, let me know and I'll mark it :)
https://searchfox.org/mozilla-central/source/modules/libpref/init/StaticPrefList.yaml#801 is still marked EARLY_BETA_OR_EARLIER, as is the beta copy of that file. Is there a separate bug on file for enabling it on late beta / release ? Or something else tracking that? Because otherwise, I don't think this is shipping in terms of being user-visible...
Assignee | ||
Comment 15•5 years ago
|
||
(In reply to :Gijs (he/him) from comment #14)
(In reply to Morgan Reschenberg [:morgan] from comment #13)
(In reply to :Gijs (he/him) from comment #12)
No, I think it's not a recent regression and there's a fix on nightly (albeit behind a pref that's not riding the trains at the moment, I think?) so that's fine for now.
It [backplate]'ll be fully released in 73; if there's a way to note that this bug will be fixed then, let me know and I'll mark it :)
https://searchfox.org/mozilla-central/source/modules/libpref/init/StaticPrefList.yaml#801 is still marked EARLY_BETA_OR_EARLIER, as is the beta copy of that file. Is there a separate bug on file for enabling it on late beta / release ? Or something else tracking that? Because otherwise, I don't think this is shipping in terms of being user-visible...
Uplift and pref-flip are bug 1608889 ! I'm working on it; had a misunderstanding on the release date but should uplift the pref change today.
Updated•5 years ago
|
Updated•5 years ago
|
Description
•