Repository navigation
[css-sizing-3] max-content contribution of replaced only-aspect-ratio elements might be improvable #4218
Description
We appear to have no interop on this case:
<!DOCTYPE html>
<div style="width: max-content; border: thick double;">
<svg style="float:left;" viewBox="0 0 200 100"></svg>Chrome gives it 0x0, Firefox gives it ICB width and then ratio height. The spec says 300px width and ratio height.
We suggest to go with FF's behavior, as this seems more useful than 0, and less arbitrary than 300px.
Activity
jensimmons commentedon Aug 20, 2019
I agree.
css-meeting-bot commentedon Jan 15, 2020
The CSS Working Group just discussed
[css-sizing-3] max-content contribution of replaced only-ar elements might be improvable, and agreed to the following:RESOLVED: Specify Firefox's behaviorThe full IRC log of that discussion
<dael> Topic: [css-sizing-3] max-content contribution of replaced only-ar elements might be improvable<dael> github: https://github.com//issues/4218
<dael> fantasai: We triaged this last week. Case is we have some box which says I want width max-content
<dael> fantasai: Something inside it intrinsically sized with an aspect ratio and no size info
<dael> fantasai: No interop. Chrome is 0x0 as max content. Spec says use 300px width and ratio for height. Resolvable but arbitratry. FF uses initial containing block. That seems useful. We suggest spec FF behavior
<dael> astearns: Concerns?
<dael> astearns: Objections to Specify Firefox's behavior?
<dael> RESOLVED: Specify Firefox's behavior
[-][css-sizing-3] max-content contribution of replaced only-ar elements might be improvable[/-][+][css-sizing-3] max-content contribution of replaced only-aspect-ratio elements might be improvable[/+]fantasai commentedon Jan 16, 2020
Fixed in 57e2680