Textarea won't resize properly in translate.google.com
Categories
(Core :: Layout: Form Controls, defect, P1)
Tracking
()
Tracking | Status | |
---|---|---|
firefox-esr60 | --- | unaffected |
firefox-esr68 | --- | unaffected |
firefox68 | --- | unaffected |
firefox69 | --- | unaffected |
firefox70 | + | verified |
People
(Reporter: euthanasia_waltz, Assigned: emilio)
References
(Regression)
Details
(Keywords: nightly-community, regression)
Attachments
(2 files, 1 obsolete file)
STR:
- Open https://translate.google.com
- Enter some long sentence in textarea
ER:
The sentence is displayed fully.
AR:
The sentence is displayed partially. First 3 lines only.
mozregression Pushlog:
https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=fc14e15221007541122d68dadb6e47d083a8463f&tochange=97e691afc504c62ba2295bb3e9c3d33204b15d60
Regressed by bug 1456052.
Assignee | ||
Comment 1•5 years ago
|
||
Thanks for finding this, I'll take a look.
Assignee | ||
Comment 2•5 years ago
|
||
Assignee | ||
Updated•5 years ago
|
Updated•5 years ago
|
Assignee | ||
Comment 3•5 years ago
|
||
[Tracking Requested - why for this release]: Don't want to let this fall onto the cracks.
Assignee | ||
Comment 4•5 years ago
|
||
Assignee | ||
Comment 5•5 years ago
|
||
This is observable via CSSOM APIs, and websites rely on them being inflatted by
the padding-box in the it to make textareas auto-sizable...
I think this is wrong though, since if I make overflow-clip-box a paint-only
thing, then the overflow areas match the ones that we get before this patch.
Which means I'd need a textarea-specific patch that does the inflation...
This doesn't make any reftest fail either, and would need a test if we go for
this, probably in the lines of the test-case attached to the bug... I'll comment
with alternatives on the bug.
Assignee | ||
Comment 6•5 years ago
|
||
So this patch changes behavior in a way that's partially undesired, in the sense that it makes scrollHeight
differ between overflow-clip-box: padding-box
and overflow-clip-box: padding-box content-box
. See the test-case attached for an example.
So here are the options as I see them:
-
Taking the attached patch (plus a test). I think it's a bit unfortunate, but it may be fine to say that we always get the
content-box
scrollable area if any of the directions specifiescontent-box
? A bit awkward. -
Add a textarea-specific code-path that takes this path. Even more awkward.
-
Taking this patch and changing behavior for all elements too (that is, probably "fixing" bug 748518 / bug 1527949) in a follow-up (maybe fine?).
-
Not taking this patch and asking Google to fix it? Though it seems plausible that other sites are affected.
-
Reverting the regressing bug. The regressing bug was not a huge priority to get fixed, and it seems we could live without fixing it, at least for now...
Mats, any thoughts?
Assignee | ||
Comment 7•5 years ago
|
||
Err, see comment above. +dholbert too :)
Updated•5 years ago
|
Updated•5 years ago
|
Assignee | ||
Comment 9•5 years ago
|
||
I'm waiting from feedback from mats@ / dholbert@ (see comment 7).
We can back out the regressing bug as needed, I suspect.
Comment 10•5 years ago
|
||
(In reply to Emilio Cobos Álvarez (:emilio) from comment #9)
I'm waiting from feedback from mats@ / dholbert@ (see comment 7).
We can back out the regressing bug as needed, I suspect.
If we can't fix that bug before the beta merge, that's a likely option yes :)
Comment 11•5 years ago
|
||
Sorry, I'd forgotten about the pending needinfo.
I tend to think we should backout the regressing bug until we've got a more complete solution that we feel good about (perhaps involving bug 1527949?).
Comment 12•5 years ago
|
||
I agree that backing out bug 1456052 seems like the best short-term fix.
Updated•5 years ago
|
Comment 13•5 years ago
|
||
Pushed by emilio@crisal.io: https://hg.mozilla.org/integration/autoland/rev/ff602e12fd45 Back out changeset 97e691afc504c62ba2295bb3e9c3d33204b15d60 (bug 1456052) for causing this bug.
Updated•5 years ago
|
Comment 14•5 years ago
|
||
bugherder |
Updated•5 years ago
|
Updated•5 years ago
|
Comment 15•5 years ago
|
||
Hello!
Reproduced the issue with Firefox 70.0a1 (20190725215157) on Windows 10x64.
The issue is verified fixed with 70.0b7 (20190916074538) on Windows 10x64 following STR from comment 0.
Updated•2 years ago
|
Description
•