Robert Frampton

Makes sense, we've talked about doing this in the past. I think the reason we haven't done it already is dealing with Metal components that provide code snippets, because even...

Cool. That approach should work on Electric as well.

CC @apollonian @causztic. Let me know if you have any questions or if there is anything you'd like to change!

Right now I have a branch with the following behavior. It captures any incremental dom errors and waits until the tree of components is done trying to render, then it...

Yep, I'll send a PR later this morning. Edit: ran into some issues, might have uncovered an unrelated problem.

If we make this change now it will break a lot of tests for the other product teams that are using Jest, including Clay.

Thanks for reporting @matuzalemsteles, Unfortunately this is probably an issue in [closure-template](https://github.com/google/closure-templates)'s `SoyToIncrementalDomSrcCompiler` that is implemented by [metal-tools-soy](https://www.npmjs.com/package/metal-tools-soy). I doubt this is something we'll be able to easily fix ourselves....

When rendering children, Metal still calls [incrementalDom.elementVoid](https://github.com/metal/metal.js/blob/master/packages/metal-jsx/src/JSXRenderer.js#L95) and passes it a function that increments the number of children of the current component. From what I remember this is done to...

Wouldn't this be a breaking change that would require us to update all metal components? Even if we had some kind of backwards compatibility, how would it work if one...

Hey @matuzalemsteles This one is a bit tricky, because for the case of `ClayDropdownBase`, the `ClayPortal` component isn't actually providing the markup, rather the markup is being passed from `ClayDropdownBase`...