Meeting minutes
<krit> w3c/
Align type and media attributes from SVGStyleElement with HTMLStyleElement
<caribou> PR 1001
Karl: w3c/
Dirk: Change should be easy. Only alignment needs to be achieved.
Dirk: Why Chrome isn't aligned.
Karl: Luke created WPTs. Safari and Firefox are aligned but WPT later marked as obsolete by Luke.
Divyansh: Chromium doesn't support these properties. That's why WPTs are failing for Chromium.
Dirk: Let's agree of making 'type' attribute obsolete. 'media' attribute resolution deferred for now.
Nikolas: 'media' attribute is used for use-cases (like Braille) which are not relevant for SVG. Thus, we can probably drop that as well.
RESOLUTION: 'type' attribute to be deprecated. 'media' attribute resolution deferred for now.
Adding resolution of content dependent units used in sizing properties for inner SVG elements w3c/svgwg#999
Adding resolution of content dependent units used in sizing properties for inner SVG elements
<nzimmermann> New issue to track possible media attribute deprecation for SVGStyleElement - w3c/
Divyansh: We already got the resolution for this, but we are not sure if this needs to go in CSS spec or SVG spec. CSS Working Group suggested to put this in SVG spec.
Dirk: Create a PR for this to go in CSS spec.
RESOLUTION: Add this to SVG spec
CSS Gradients w3c/svgwg#1021
CSS Gradients w3c/svgwg#1021
<krit> https://
<krit> "SVG does not calculate gradients in pre-multiplied space, so 'transparent' really means transparent black"
Tav: Color gradient in Chrome is different from Firefox and Safari for color-interpolation. We need to align on this.
Tav: Chrome renders P3 colors as P3 colors. Firefox and Safari seem to map P3 colors to sRGB. Safari 18 uses the default rendering for all three values because it doesn't implement color-interpolation.
<dmangal> https://
Divyansh: There is a platform specific bug for Chromium on mac. https://
Tav: There are number of suggestions to fix this: 1) We can do this in the spec to define the color space for interpolation or 2) Allow additional color spaces.
Dirk: This might not be the purview of the SVG Working group as this might be more of the nature that how CSS gradients work.
Dirk: Ragvesh and team can check and get back on this.
Should getBBox returns values when the shape can't be drawn w3c/svgwg#1018
Karl: Should getBBox return values when the shape can't be drawn? Say, a negative value in width and/or height attribute, e.g. <rect id="rect1" x="1" y="2" width="-10" height="20"/>
Karl: In Chrome and Firefox, getBBox returns 0 if if the element can't be drawn. What should be the correct behavior here?
Dirk: We can check how this works with filter and get back on this with results of different cases.
Nikolas: In case width is set to Invalid value but height is set to a Valid value, returning non-zero values might bring in confusion and developer may need to do special handling for these cases. 0 still represents Zero area and thus we can infer that the shape isn't drawn.
Ragvesh: Need to go back and check.
<caribou> +1 to Nikolas' point as well
Dirk: WPTs will be updated. We will discuss this again next week.
Divyansh: Whatever we end up deciding for Bounding box, we might want to do the same for ClientRect as well.
::selection and other highlights in SVG w3c/svgwg#894
Karl: Firefox, Chromium and Safari are all inconsistent in how color is applied via SVG specific CSS properties and HTML like CSS properties on text while selected vs un-selected. w3c/
Karl: Comment on bug with your opinions. We will discuss this back again in upcoming meetings.