Re: [csswg-drafts] [css-anchor-position-?] SVG element or coordinate as anchor (#11301)

The CSS Working Group just discussed `[css-anchor-position-?] SVG element or coordinate as anchor`, and agreed to the following:

* `RESOLVED: Defer to next level`

<details><summary>The full IRC log of that discussion</summary>
&lt;fantasai> TabAtkins: Request is to allow HTML elements to anchor to SVG sub-elements<br>
&lt;fantasai> TabAtkins: a lot of obvious use cases for this, e.g. labels in a chart<br>
&lt;fantasai> TabAtkins: use cass seem reasonable<br>
&lt;fantasai> TabAtkins: problem is, this breaks some assumptions that browsers currently make about SVG<br>
&lt;fantasai> TabAtkins: layout engine considers SVG a replaced element. Doesn't know what happens inside.<br>
&lt;fantasai> TabAtkins: so info is not surfaced to layout tree<br>
&lt;fantasai> TabAtkins: additoinal problem is, SVG is mostly laid out with transforms, not with positioning<br>
&lt;fantasai> TabAtkins: and transforms are still not taken into account with anchorpos (intentionally)<br>
&lt;fantasai> TabAtkins: we have an open issue ot think about what cases could be made to work<br>
&lt;fantasai> TabAtkins: but in general, it means SVG elements can't tell us their position<br>
&lt;bkardell> q+<br>
&lt;fantasai> TabAtkins: question about what rectangle to use, we can just use bounding box<br>
&lt;iank_> q+<br>
&lt;fantasai> TabAtkins: but other two questions make this a hard request to satisfy, even though it has good use cases<br>
&lt;fantasai> TabAtkins: I think that means we close no change for now, but leave open possibility in the future<br>
&lt;astearns> ack bkardell<br>
&lt;fantasai> bkardell: Any time someone says SVG I'm contractually obligated to say "but what about MathML?"<br>
&lt;fantasai> iank_: It should all work today<br>
&lt;fantasai> iank_: You should test it :)<br>
&lt;fantasai> TabAtkins: I think it is a full citizen of the CSS nation<br>
&lt;fantasai> ChrisL: It doesn't depend on transforms. Has an in-flow layout model.<br>
&lt;astearns> ack iank_<br>
&lt;fantasai> iank_: Also a bunch of visual effects that typically apply to SVGs<br>
&lt;fantasai> iank_: e.g. displacement filters<br>
&lt;fantasai> iank_: It's a whole can of worms<br>
&lt;TabAtkins> fantasai: i think at the point people are using filters, it's okay that the position isn't correct<br>
&lt;TabAtkins> fantasai: but transforms are integral to svg layout, we'll have to do something about those<br>
&lt;TabAtkins> TabAtkins: i want to do something about trasnforms anyway<br>
&lt;TabAtkins> fantasai: so it looks like resoliution is "defer to next level"<br>
&lt;TabAtkins> TabAtkins: yup<br>
&lt;fantasai> astearns: Proposed resolution to defer<br>
&lt;fantasai> RESOLVED: Defer to next level<br>
&lt;fantasai> astearns: though this may also be deferred from that level<br>
&lt;fantasai> TabAtkins: potentially, yeah<br>
</details>


-- 
GitHub Notification of comment by css-meeting-bot
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/11301#issuecomment-2776723183 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Thursday, 3 April 2025 19:23:34 UTC