I don't think inside InterpolationTypes is the right place for what PropertyInterpolationTypesMapping does. InterpolationTypes should only care about value types, not the properties that use them. CSSProperties.in would be a more appropriate place to declaratively encode the mapping.

Currently the switch/if-else code gives us on-demand initialisation of InterpolationTypes for each CSS property/SVG attribute. Without it you'd be forcing everything to initialise when most pages would only use a handful out of multiple hundreds.

I'll take this to give it some thought.

BTW static create() methods SGTM, the other ideas brought up are orthogonal.

TODO: investigate whether the discussion of MakeUnique() and WrapUnique() affects this.
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/Bo_4JqYXaiE

I'm not likely to work on this soon, so making it Available. I'm actively working on the broader Objective (issue 654309), however, and happy to brief someone if they want to take it up before I get back to it.

Could you clarify whether the switch/if-else thing is part of this bug? The title suggests something orthogonal.

Hmm, no, it shouldn't be. This bug only specifically deals with the static ::create functions. My intuition at the time was that if the inheritance hierarchy is approached correctly (for some as-yet-unclear value of 'correct'), then switch/if-else and property map logic would not be required. I think there are connected design problems (well, question marks) here, but let's leave this bug to be only about the static create methods, and address the switch/if-else logic separately if it becomes appropriate.

This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue.

Sorry for the inconvenience if the bug really should have been left as Available.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot