textarea implicit min-width in flex-grow context (overflows parents for no apparent reason) - see https://tiles.alt.tata.casa
Categories
(Web Compatibility :: Site Reports, defect, P3)
Tracking
(Webcompat Priority:P3, Webcompat Score:1)
People
(Reporter: kynn, Unassigned)
References
(Depends on 1 open bug, )
Details
(Keywords: webcompat:platform-bug, webcompat:site-report)
User Story
platform:windows,mac,linux,android impact:minor-visual configuration:general affects:all branch:release diagnosis-team:layout user-impact-score:2
Attachments
(3 files)
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:129.0) Gecko/20100101 Firefox/129.0
Steps to reproduce:
Subject has been discussed in https://github.com/webcompat/web-bugs/issues/138325 ; where they redirected me here.
Steps :
- Go to https://tiles.alt.tata.casa
- Click on the home circle button
- Click "use a model"
- Click "Weekly tasks"
- Play with the responsive design by reducing browser view-width.
Actual results:
text-area width doesn't shrink below a certain value (~200px).
But no min-width has been set.
Thus, text-area overflows its parent.
Expected results:
No min-width has been set in CSS.
Thus, text-area should reduce in order to never overflow it's parent.
It should be added that Edge gives me the expected behavior on the same website.
Comment 2•2 years ago
|
||
The Bugbug bot thinks this bug should belong to the 'Core::Layout: Flexbox' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.
Comment 3•2 years ago
|
||
Testcase as reporter described - intrinsically sized textarea is larger than its parent.
Updated•2 years ago
|
Hi,
First I want to say I'm completely ignorant to the process of bug correction by the mozilla team & community.
I understand that this bug was previously confirmed by :dshin, yet nothing changed since one year and a half.
Was it ignored, judged not necessary to correct, or simply forgotten ?
Thank you in advance.
Comment 5•8 months ago
|
||
Yup, sorry (& sorry for delayed response) -- we're a small team juggling many competing projects/priorities, so bugs often get ignored for a while after initial triage, depending on severity & impact.
I think the attached testcase might've been over-reduced -- it gives me the same behavior in Chrome as well as Firefox, at first glance. But I do see the behavior-difference on the original site. I think this is a version of bug 2021158 which I just filed. I'm going to mark this as a dependency of that bug.
Side note, personally I have to connect to a VPN in order to access the https://tiles.alt.tata.casa site, because it looks like this particular site is only available over IPv6, and my regular ISP connection doesn't provide me with an IPv6 address or routing, but I can get a working IPv6 connection via e.g. Mozilla VPN.
Comment 6•8 months ago
•
|
||
Regarding "overflows parents for no apparent reason" - there is sort-of a reason, which is that children-of-a-flex-container get an "automatic minimum size", which is based on their content size. For many form controls, this is just the regular automatic width of the form control that it would have if there weren't any constraints on it (which might be bigger than the flex container, and hence the form control will overflow beyond the bounds of its flex container).
There's an exception in the spec for elements that are scroll containers, though; and Chrome/Safari seem to treat <textarea> elements as being scroll containers for the purpose of applying this exception. Firefox only applies the exception on textareas if you set overflow:auto on the textarea (or 'scroll' or 'hidden'), as in the bottom half of my testcase on bug 2021158.
If you happen to have a route to contact the web developers of this site (or you happen to be one of the developers): they can work around this Firefox issue [if you like] by either setting min-width:0 or overflow:auto on these textarea elements. (min-width: 0 is the more direct fix; the setting overflow simply makes the default min-width behave as 0, so it's a bit more direct to just set min-width to 0 directly.)
Updated•8 months ago
|
Comment 7•8 months ago
•
|
||
Here's a screencast comparing Firefox (top) to Chrome (bottom) to demonstrate the bug.
The key area is the darker-shaded aside element, to the right side of each colorful textarea. That darker-shaded element has an "anchor" icon that appears when you hover it.
As you shrink your Firefox window, that element gets crowded out (and ultimately disappears) because the textarea doesn't shrink -- so the darker element gets pushed off the right side of the container and gets clipped by an overflow:hidden ancestor.
Whereas in Chrome, the textarea shrinks and the aside element remains fully visible.
Updated•8 months ago
|
Description
•