Re: [csswg-drafts] [css-anchor-position] Invalidation of last successful position option in cases not in the spec (#12577)

The Houdini Task Force just discussed `[css-anchor-position] Invalidation of last successful position option in cases not in the spec`, and agreed to the following:

* `RESOLVED: last successful position option is cleared if computed values changed in: 'position', 'position-try-*', any @position-try block referenced by 'position-try', or any property that is allowed in '@position-try', e.g. width`
* `ACTION: TabAtkins: to ask anders`

<details><summary>The full IRC log of that discussion</summary>
&lt;alisonmaher> fantasai: when we do fallbacks we try to make it sticky<br>
&lt;alisonmaher> ... until it is invalidated<br>
&lt;alisonmaher> ...when we need to we invalidate and maybe choose another fallback<br>
&lt;alisonmaher> ....three postions ??<br>
&lt;ChrisL>     when the element is not absolutely positioned anymore<br>
&lt;ChrisL>     when the computed value of position-try-fallbacks for the element changes<br>
&lt;ChrisL>     when any @position-try rules referenced by the element's position-try-fallbacks changes<br>
&lt;alisonmaher> ...there are more cases that tigger invalidation<br>
&lt;alisonmaher> ...if position try order<br>
&lt;alisonmaher> ...or base style changes<br>
&lt;alisonmaher> ...when do we invalidate the fallback and can we spec it?<br>
&lt;fantasai> https://wpt.live/css/css-anchor-position/base-style-invalidation.html<br>
&lt;alisonmaher> TabAtkins: reacquainting myself<br>
&lt;fantasai> https://wpt.live/css/css-anchor-position/anchor-position-non-anchored-fallback.html<br>
&lt;alisonmaher> ...this is just about forgetting the last successful position<br>
&lt;alisonmaher> ...for position try order I have no problem adding to the the things that make us forget the last position<br>
&lt;alisonmaher> ...that makes sense<br>
&lt;alisonmaher> ...for altering the base styles that confuses me<br>
&lt;alisonmaher> ...the other changes that are possible are also part of the base style<br>
&lt;alisonmaher> ...I am confused why this test would check for other things changing<br>
&lt;alisonmaher> ...this comes from a Chromium issue<br>
&lt;alisonmaher> iank_: Rune has the context<br>
&lt;alisonmaher> fantasai: I don't know what's happening. Chrome has additional criteria we aren't aware of<br>
&lt;alisonmaher> fantasai: maybe it's reasonable but we should spec it<br>
&lt;alisonmaher> TabAtkins: if we spec tjhis any change to base styles should cause us to forget the last remebered position<br>
&lt;alisonmaher> fantasai: don't think you want that<br>
&lt;alisonmaher> ...maybe any styles in @position try or @position try values<br>
&lt;alisonmaher> TabAtkins: reasonable subset<br>
&lt;fantasai> s/@position try values/position-try properties/<br>
&lt;alisonmaher> ...I find it reasonable to amend to position try properties<br>
&lt;alisonmaher> fantasai: position try fallbacks or options<br>
&lt;fantasai> position-try-*<br>
&lt;TabAtkins> "allowed @position-try properties"<br>
&lt;fantasai> or @position-try rules that are used<br>
&lt;alisonmaher> fantasai: we should check for those<br>
&lt;alisonmaher> TabAtkins: ok with this<br>
&lt;fantasai> PROPOSED: last successful position option is cleared if computed values changed in: 'position', 'position-try-*', any @position-try block referenced by 'position-try', or any property that is allowed in '@position-try', e.g. width<br>
&lt;fantasai> RESOLVED: last successful position option is cleared if computed values changed in: 'position', 'position-try-*', any @position-try block referenced by 'position-try', or any property that is allowed in '@position-try', e.g. width<br>
&lt;fantasai> ACTION: TabAtkins: to ask anders<br>
</details>


-- 
GitHub Notification of comment by css-meeting-bot
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/12577#issuecomment-3210575175 using your GitHub account


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

Received on Thursday, 21 August 2025 13:18:20 UTC