Conversation
|
This is really cool, thanks a lot. I'm wondering if removing changes to math.gl references should warrant a breaking change for us, that we would bump the major version number. |
|
Thanks for the welcome here! I think it's a good point about bumping the version, as some users could still be on the typeless math.gl and luma.gl packages. But on that subject, I've got another PR waiting (I wanted to get these commits out of the way first) that adds Non-Iterable data support, type inference, extension props. It may make more sense, if we're thinking version bumps anyway, to go ahead and fold these commits in to it after all. I'll go ahead and make that new PR and we can figure out what to do with this one, based on what you think of it. |
|
Was thinking another way forward, maybe less intrusive, would be to just push the typings overlap fix for luma.gl and math.gl down the road and remove 508fba4. Would that be better? |
|
@jstaab seems like a sound idea - if it's not too much work for you. |
|
@danmarshall should be ready for a review |
|
Thanks! Published 4.9.20 |
A collection of small touch-ups I've collected over time. Just running down a commit summary:
Removes math.gl and luma.gl types because they now have their ownLayer<any, any>to better match all Layers.Matrix4for instances ofmodelMatrix