Re: [editing] ContentEditable with UserSelect=None needs to be documented (#20)

So in other words: I argue for given direct access to the placement info in terms of pixels, type getClientRects. If Firefox draws more than one caret, the pixel placement of both should be given. But this is in part also because I cannot think of how an abstraction could handle bi-di and vertical text automatically and still give enough information to the JS editor to make effective choices of how to move the caret.

@rniwa What type of abstraction were you thinking of?

---
Reply to this email directly or view it on GitHub:
https://github.com/w3c/editing/issues/20#issuecomment-111059105