Number of PerformanceElementTiming entries for background-image that contains multiple layers

[hawkinsw]

I hope that this is a meaningful question. If not, I apologize!

In CSS3 you can (apparently) specify multiple background images for a single element that will be used to build a layered background.

Does the second Note in Section 3.2 apply in this situation, too?

[npm1]

It does apply for it! See this test for example:
https://github.com/web-platform-tests/wpt/blob/master/element-timing/multiple-background-images.html

Do you think this needs further clarification in the spec?

[hawkinsw]

Awesome!

I am only concerned because in CSS3 the image type is no longer limited to urls and can be other things like fragments and solid colors. See https://www.w3.org/TR/css3-images/. Because of that, speaking in the note only about urls as the type of multiple values seems misleading or limiting.

In light of that, do you have any thoughts on the content of the Note?

[npm1]

Yep, that seems like a different issue. I agree that we should be more specific about the fact that we only observe background images associated to a URL. I don't think the Note in 3.2 is inaccurate, as it says "we assume that there is one image request for each Element that a background-image property affects and for each URL that the background-image property specifies" which in particular implies that there is no image request when there is no URL, right?

But I do think that in https://wicg.github.io/element-timing/#sec-elements-exposed we need to add context about the fact that only background images with URL will be exposed. Does that make sense?

[hawkinsw]

Yes, thanks for agreeing! I am sorry that I was being inarticulate and made it harder than it needed to be! I'm not sure exactly how we should rectify the situation. If you're okay with it, I can specify some changes and put them in a PR and we can see what people think? Thanks again for working through this with me! Will

…

On Mon, Aug 19, 2019 at 11:46 AM npm1 ***@***.***> wrote: Yep, that seems like a different issue. I agree that we should be more specific about the fact that we only observe background images associated to a URL. I don't think the Note in 3.2 is inaccurate, as it says "we assume that there is one image request for each Element that a background-image property affects and for each URL that the background-image property specifies" which in particular implies that there is no image request when there is no URL, right? But I do think that in https://wicg.github.io/element-timing/#sec-elements-exposed we need to add context about the fact that only background images with URL will be exposed. Does that make sense? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#35>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACCP2CWXZMK7C3QNTYR3BPDQFK56FANCNFSM4IMETLTA> .

[npm1]

No, thank you for the feedback :) PR is welcome.