Closed marlitas closed 3 months ago
This change affected the center-and-variability
phet-io API, but did not require any changes in the migration wrapper since we are adding a Property tandem path. It did not change the location or remove any other Properties that would require a migration rule. I tested by running the migration wrapper and found no errors.
Ready for review by @amanda-phet, and tagging @zepumph for code review to confirm my migration diagnosis from above.
I have absolutely no clue how this happened, but somehow I linked the commits to a buoyancy issue I've never heard of? https://github.com/phetsims/buoyancy/issues/64
Really not sure. Here are the commits: https://github.com/phetsims/soccer-common/commit/96b2d224504bc1af552400d9224a664e5ac3dfff https://github.com/phetsims/phet-io-sim-specific/commit/21cf4fb7c774c4812182c67d17d511cd5a4b5a33
Looks perfect thanks. There are multiple reasons why migration doesn't matter here:
Rather thank link this seven times (under each ball - meanShareAndBalance.balancePointScreen.view.sceneView.soccerBallNodes.soccerBallNode*
), I thought it would be linked once under soccerBallNodes
.
If that's hard to do I'm fine leaving this as it is, but I think it would be cleaner to have this property at the same level as soccerBallNode*
.
I was able to make the above request happen by customizing the tandem. I added documentation to make that clear in the code, and documentation in PhetioObject
makes it seem like this is a behavior that is not recommended but not unprecedented either:
If tandem is specified in the options (not recommended), it will use that tandem and nest it anywhere that tandem exists. Use this option with caution since it allows you to nest the tandem anywhere in the tree.
Over to @amanda-phet to confirm and close if all looks well.
Looks good, thanks!
Discussion Is this a change that would affect soccer-common? Check with MK if this changes the migration (MS: I don't think it will)