suggestion for non-core transform syntax

slate-edit-table is an example of a Slate plugin that exports its own transforms.

The problem is that the transform syntax is rather ugly when using non-core transforms.

Here is an example to illustrate:

const {insertRow, moveSelectionBy} = slateEditTableInstance.transforms;
let transform = state.transform();

insertRow(transform);
transform.moveTo(state.selection);
moveSelectionBy(transform, 1, 0);
return transform.apply();

What I don’t like about the syntax is that

• because core transforms are chained, it helps to show that they are all part of the same transform
• non-core transforms lose this clarity

I would like to add a method, maybe something like call that is just syntactic sugar for calling a non-core transform with the first argument being the transform object and the remaining arguments being being the remaining arguments in the call.

The code below would then be equivalent to the code above but more readable.

const {insertRow, moveSelectionBy} = slateEditTableInstance.transforms;

return state.transform()
  .call(insertRow)
  .moveTo(state.selection)
  .call(moveSelectionBy, 1, 0)
  .apply();

What do you (a) think of the idea and (b) the name of the method call. I can create a pull request for this if you think this is a good idea.

[thesunny]

Worked on this last night including a unit test. Just need to clean up a little and will post a pull request.

[Soreine]

That’s a good idea, and regarding the different raises issues with extending the Transform object, I think the initial .call solution is quite clean and simple to do. 👍