Proposed Resolutions: Behavior on invalid mask, clipPath or filter

I agree that we can use async resolution for this. I'll post to the issue thread as well. 

Unless I hear here or on GitHub that someone wants to go over this issue on a call, or objects to the resolution, the rules below will be resolved by the CSS WG on October 14th.

Thanks,

Alan

On 9/21/19, 11:54 AM, "Dirk Schulze" <dschulze@adobe.com> wrote:

    Hi CSS WG members,
    
    The SVG WG discussed https://github.com/w3c/fxtf-drafts/issues/130: How should UAs behave on invalid mask, clipPath or filter? As invalid, the WG looked at mask, clipPath or filter
    * having no content
    * having only child elements that do not follow the content model
    * having child elements that partly follow and not follow the content model
    * mask, clip-path, filter with missing/invalid URLs or not-yet available resources
    * circular dependencies
    
    Upfront I tested Edge, Firefox, Safari and Chrome and compared the different results: https://github.com/w3c/fxtf-drafts/issues/130#issuecomment-390981529

    
    
    The SVG WG suggests that the CSS WG resolves on the following rules (https://github.com/w3c/fxtf-drafts/issues/130#issuecomment-532981392):
    • For clipping and masking, we follow the behavior of Edge, WebKit, and Blink in the https://github.com/w3c/fxtf-drafts/issues/130#issuecomment-390981529 table
    • If a referenced filter is missing or invalid, the side effects like stacking context are still preserved.
    
    Unless there are concerns to those suggestions we don’t need to bring this up in a call on the CSS WG. In this case it would be great if we could use the async resolution process with an appropriate period to review the issue. @Alan, @Rossen?
    
    Thanks,
    Dirk

Received on Saturday, 21 September 2019 03:03:12 UTC