Brian Hung
Brian Hung
> You can even use Chevrotain as HF does. I would use an incremental parser that's also error insensitive like [lezer](https://lezer.codemirror.net/). Syntax highlighting for parsing incomplete formulas is very similar...
@sequba That would also work!
> (in sheets other then sheetId) Think it might be useful to also include cells in `sheetId` if for some reason, a cell references another cell in the same sheet...
An alternative serialization method might be to just use `setAttribute('marks', node.marks)` to distinguish between `attrs` and `marks`. This would also allow `marks` to retain its expected properties of being an...
> I don't think `yDocToProsemirrorJSON` is the correct signature to convert a `Y.XmlFragment` to JSON. In my mind, a `Y.XmlFragment` is an array of `Node`s, not a `Node` with type...
> we should change the signature To clarify, this would mean in essence deprecating `yDocToProsemirrorJSON` in favor of the other two utility methods `YXmlFragmentToProsemirrorJSON` and `YXmlElementToProsemirrorJSON`? Because if we change...
If anyone wants to tackle this in a PR, two main functions to address are `createTypeFromElementNode` and `createNodeFromYElement`. Particularly, serialize the marks of a node as `Y.XmlElement` attributes using `setAttribute`...
Same thing happens with `createDraft` and `finishDraft`. ```ts let state = createDraft({}) state.x = 10 let patches, inversePatches, nextState nextState = finishDraft(state, (p, ip) => { patches = p inversePatches...
headless seems to be running into the same problem 😅 https://twitter.com/adamwathan/status/1530666459545931776