- From: Tab Atkins Jr. via GitHub <sysbot+gh@w3.org>
- Date: Mon, 26 Feb 2018 21:28:24 +0000
- To: public-houdini-archive@w3.org
Okay, I think I've hit on the necessary design. I need a CSSImageValue, subclass of CSSStyleValue, representing all images. In the next level we'll add some subtypes for this, like CSSLinearGradientFunctionValue or something. I also need a CSSURLValue, subclass of CSSStyleValue, representing the `url()` syntax. *All* `url()`s are reified as this, because in some cases we can't tell what type a `url()` should resolve to until after we fetch it, and this behavior might spread to *all* `<image>` cases (if we ever allow, for example, pointing to an SVG `<linearGradient>` as an `<image>`, which seems reasonable). We'll add a CSSImageish typedef that accepts CSSImageValue and CSSURLValue, and things that want to take a CSSImageValue should take the typedef instead; if a CSSURLValue is used and doesn't end up pointing to an image, it'll just be an invalid image, no problem. In the next level we'll add a `CSSURLImageValue`, which you can obtain from a `CSSURLValue` (either thru a constructor, or a promise-returning method, or something), which exposes the width/height/etc stuff that you expect from an image. This will *never* be reified directly from an underlying value, you must *always* manually obtain it via the converter. -- GitHub Notification of comment by tabatkins Please view or discuss this issue at https://github.com/w3c/css-houdini-drafts/issues/716#issuecomment-368656793 using your GitHub account
Received on Monday, 26 February 2018 21:28:28 UTC