Re: [svgwg] xlink:href in <animate>, <use>, etc. doesn't work in sourceless iframes (#700)

The SVG Working Group just discussed `xlink:href in <animate>, <use>, etc. doesn't work in sourceless iframes`.

<details><summary>The full IRC log of that discussion</summary>
&lt;myles> Topic: xlink:href in &lt;animate>, &lt;use>, etc. doesn't work in sourceless iframes<br>
&lt;myles> GitHub: https://github.com/w3c/svgwg/issues/700<br>
&lt;myles> krit: Do we need to discuss any non-&lt;use> elements?<br>
&lt;myles> AmeliaBR: It's not specific to &lt;animate>, it's just about xlink:href in general<br>
&lt;myles> krit: Is there a difference between &lt;use> and &lt;animate>? Or will this apply everywhere<br>
&lt;myles> AmeliaBR: It will apply everywehre<br>
&lt;myles> AmeliaBR: The issue is about URL resolution. An iframe srcdoc doesn't have its own URL. It's defined in HTML to behave as the parent frame's URL applies to the srcdoc. But if you put inline svg which needs hrefs to its own document, then that's the wrong base URL.<br>
&lt;myles> AmeliaBR: so #target inside the srcdoc doesn't work becuase it looks for one in the main page instead of the local page.<br>
&lt;myles> AmeliaBR: This is similar to SVG inside shadow dom. The rules there say, first, look in the main document. Initial implementations' SVG cross references weren't working inside the shadow dom. There is a general acceptance now that for shadow trees, hash-only URLs should look in the local trees first, but some of the details there are still undecided as well.<br>
&lt;myles> krit: For the shadow dom discussion, rniwa and annevk had different opinions, but eventually resolved on something. Would it be possible that we ask rniwa and annevk to comment on this issue as well?<br>
&lt;myles> AmeliaBR: Yes.<br>
&lt;myles> AmeliaBR: Getting feedback from HTML would be useful because this whole thing comes from an HTML situation. It's important to try to coordinate with the shadow dom behaviors. I'm not sure how much of that has actually gotten reflected into specs.<br>
&lt;myles> krit: There are similarities with shadow dom. Hopefully we can get a solution soon.<br>
&lt;myles> krit: AmeliaBR, is there something else you are looking for?<br>
&lt;myles> AmeliaBR: I can't remember.<br>
&lt;myles> AmeliaBR: one option is to go like CSS and just say if the URL starts with a hash, treat it specially. But if we can have more general rules, that would be helpful anyway.<br>
&lt;myles> AmeliaBR: At this point, if you can follow-up with anybody relevant, that would be great.<br>
</details>


-- 
GitHub Notification of comment by css-meeting-bot
Please view or discuss this issue at https://github.com/w3c/svgwg/issues/700#issuecomment-511558963 using your GitHub account

Received on Monday, 15 July 2019 20:33:43 UTC