Update Chrome/Opera data for SVG element APIs using mdn-bcd-collector #7197
Conversation
github-actions
bot
added
the
data:api 🐇
api/SVGGraphicsElement.json
Outdated
| @@ -5,10 +5,10 @@ | |||
| "mdn_url": "https://developer.mozilla.org/docs/Web/API/SVGGraphicsElement", | |||
| "support": { | |||
| "chrome": { | |||
| "version_added": true | |||
| "version_added": "30" | |||
There was a problem hiding this comment.
http://mdn-bcd-collector.appspot.com/tests/api/SVGGraphicsElement doesn't have an instance because it's a base class for other things. It was introduced in https://chromium.googlesource.com/chromium/src/+/1c0121db5bdec1c757a78a9eb7b2888985809cd8 and probably moved some attributes around on the prototype chain. As a result the version numbers for subfeatures are nuanced. We may need partial_implementation.
Can you check if there are other interfaces updated in this PR for which we also don't test any instance? It might be good to break those out into its own PR. This PR is big, but until I got to this point seemed pretty straightforward.
There was a problem hiding this comment.
It looks like we don't have a custom test implemented for this API, oops. I've reverted the changes to this and will add a custom test so we can test this properly. (This was the only one we missed, it looks like.)
There was a problem hiding this comment.
Great, thank you!
It was I who removed the custom test for SVGComponentTransferFunctionElement + SVGGraphicsElement + SVGTextContentElement + SVGTextPositioningElement in foolip/mdn-bcd-collector#643.
SVGTextPositioningElement isn't updated in this PR, but the remaining two are, and those tests also don't use any instances:
http://mdn-bcd-collector.appspot.com/tests/api/SVGComponentTransferFunctionElement
http://mdn-bcd-collector.appspot.com/tests/api/SVGTextContentElement
| @@ -5,10 +5,10 @@ | |||
| "mdn_url": "https://developer.mozilla.org/docs/Web/API/SVGComponentTransferFunctionElement", | |||
| "support": { | |||
| "chrome": { | |||
| "version_added": true | |||
| "version_added": "5" | |||
There was a problem hiding this comment.
This is one of the interfaces without instances in the tests mentioned in #7197 (comment), but this data looks plenty plausible anyway, and based on https://caniuse.com/svg-filters it couldn't really have been earlier than this.
api/SVGTextContentElement.json
Outdated
| @@ -427,10 +427,10 @@ | |||
| "__compat": { | |||
| "support": { | |||
| "chrome": { | |||
| "version_added": true | |||
| "version_added": "43" | |||
There was a problem hiding this comment.
This seems wrong. http://mdn-bcd-collector.appspot.com/tests/api/SVGTextContentElement doesn't test any instances and I think this is just a matter of the interface object not being exposed until Chrome 43, or something like that. The IDL file for this goes back to 2005: https://trac.webkit.org/changeset/10577/webkit
I'll revert the changes to this file to get this PR landed.
Data from (confirmed by) mdn-bcd-collector: - Chrome ≤15 - Edge 12 / 79 - Firefox ≤4 - Firefox Android ≤86 - IE 10 - Opera ≤12.1 / 15 - Safari 6 - Safari iOS ≤10.3 Chrome ≤15 is assumed to be 5 in most cases where it was 5 here: mdn#7197 Firefox ≤4 is assumed to be 3 based on: https://developer.mozilla.org/en-US/docs/Mozilla/Firefox/Releases/3 Mirrored data: - Firefox Android 4 - Opera Android ≤12.1 / 14 - Safari iOS 6 - Samsung Internet 1.0 - WebView Android ≤37 Most of the data is very regular. Notable exceptions: - SVGFEConvolveMatrixElement entries are assumed to have shipped in Chrome 6, like api.SVGFEConvolveMatrixElement and its subfeatures. - SVGFEDropShadowElement was shipped in Chrome 13 / Firefox 30, and seemingly never supported in IE, EdgeHTML or Presto. - In Firefox, something strange happened with 3 of the interfaces. In Firefox 21, almost all subfeatures of these tests pass: http://mdn-bcd-collector.appspot.com/tests/api/SVGFEComponentTransferElement http://mdn-bcd-collector.appspot.com/tests/api/SVGFEFloodElement http://mdn-bcd-collector.appspot.com/tests/api/SVGFEGaussianBlurElement (except edgeMode) In Firefox 22, the height/result/width/x/y fail. They're back in Firefox 29/35. This looks like a mishap, but is captured in the data anyway.
Data from (confirmed by) mdn-bcd-collector: - Chrome ≤15 - Edge 12 / 79 - Firefox ≤4 - Firefox Android ≤86 - IE 10 - Opera ≤12.1 / 15 - Safari 6 - Safari iOS ≤10.3 Chrome ≤15 is assumed to be 5 in most cases where it was 5 here: mdn#7197 Firefox ≤4 is assumed to be 3 based on: https://developer.mozilla.org/en-US/docs/Mozilla/Firefox/Releases/3 Mirrored data: - Firefox Android 4 - Opera Android ≤12.1 / 14 - Safari iOS 6 - Samsung Internet 1.0 - WebView Android ≤37 Most of the data is very regular. Notable exceptions: - SVGFEConvolveMatrixElement entries are assumed to have shipped in Chrome 6, like api.SVGFEConvolveMatrixElement and its subfeatures. - SVGFEDropShadowElement was shipped in Chrome 13 / Firefox 30, and seemingly never supported in IE, EdgeHTML or Presto. - In Firefox, something strange happened with 3 of the interfaces. In Firefox 21, almost all subfeatures of these tests pass: http://mdn-bcd-collector.appspot.com/tests/api/SVGFEComponentTransferElement http://mdn-bcd-collector.appspot.com/tests/api/SVGFEFloodElement http://mdn-bcd-collector.appspot.com/tests/api/SVGFEGaussianBlurElement (except edgeMode) In Firefox 22, the height/result/width/x/y fail. They're back in Firefox 29/35. This looks like a mishap, but is captured in the data anyway.
Data from (confirmed by) mdn-bcd-collector: - Chrome ≤15 - Edge 12 / 79 - Firefox ≤4 - Firefox Android ≤86 - IE 10 - Opera ≤12.1 / 15 - Safari 6 - Safari iOS ≤10.3 Chrome ≤15 is assumed to be 5 in most cases where it was 5 here: mdn#7197 Firefox ≤4 is assumed to be 3 based on: https://developer.mozilla.org/en-US/docs/Mozilla/Firefox/Releases/3 Mirrored data: - Firefox Android 4 - Opera Android ≤12.1 / 14 - Safari iOS 6 - Samsung Internet 1.0 - WebView Android ≤37 Most of the data is very regular. Notable exceptions: - SVGFEConvolveMatrixElement entries are assumed to have shipped in Chrome 6, like api.SVGFEConvolveMatrixElement and its subfeatures. - SVGFEDropShadowElement was shipped in Chrome 13 / Firefox 30, and seemingly never supported in IE, EdgeHTML or Presto. - In Firefox, something strange happened with 3 of the interfaces. In Firefox 21, almost all subfeatures of these tests pass: http://mdn-bcd-collector.appspot.com/tests/api/SVGFEComponentTransferElement http://mdn-bcd-collector.appspot.com/tests/api/SVGFEFloodElement http://mdn-bcd-collector.appspot.com/tests/api/SVGFEGaussianBlurElement (except edgeMode) In Firefox 22, the height/result/width/x/y fail. They're back in Firefox 29/35. This looks like a mishap, but is captured in the data anyway.
Data from (confirmed by) mdn-bcd-collector: - Chrome ≤15 - Edge 12 / 79 - Firefox ≤4 - Firefox Android ≤86 - IE 10 - Opera ≤12.1 / 15 - Safari 6 - Safari iOS ≤10.3 Chrome ≤15 is assumed to be 5 in most cases where it was 5 here: #7197 Firefox ≤4 is assumed to be 3 based on: https://developer.mozilla.org/en-US/docs/Mozilla/Firefox/Releases/3 Mirrored data: - Firefox Android 4 - Opera Android ≤12.1 / 14 - Safari iOS 6 - Samsung Internet 1.0 - WebView Android ≤37 Most of the data is very regular. Notable exceptions: - SVGFEConvolveMatrixElement entries are assumed to have shipped in Chrome 6, like api.SVGFEConvolveMatrixElement and its subfeatures. - SVGFEDropShadowElement was shipped in Chrome 13 / Firefox 30, and seemingly never supported in IE, EdgeHTML or Presto. - In Firefox, something strange happened with 3 of the interfaces. In Firefox 21, almost all subfeatures of these tests pass: http://mdn-bcd-collector.appspot.com/tests/api/SVGFEComponentTransferElement http://mdn-bcd-collector.appspot.com/tests/api/SVGFEFloodElement http://mdn-bcd-collector.appspot.com/tests/api/SVGFEGaussianBlurElement (except edgeMode) In Firefox 22, the height/result/width/x/y fail. They're back in Firefox 29/35. This looks like a mishap, but is captured in the data anyway.
This PR sets results (from true/null to a real value) for Chrome and Opera for various SVG element APIs based upon results from the mdn-bcd-collector project, followed by mirroring to Chrome Android, Opera Android, Samsung Internet, and WebView. Data comes from the results from Chrome 1-85, as well as Opera 12.14, 15, and 70, on both macOS and Windows.
Notes:
falseto1(and one entry from3to1). This is because anything that is in Chrome 1 is guaranteed to be in WebView for Android 1 (as stated by Joe Medley).