Add scale block support and fix transform chain replay logic#331
Open
tracygardner wants to merge 2 commits intomainfrom
Open
Add scale block support and fix transform chain replay logic#331tracygardner wants to merge 2 commits intomainfrom
tracygardner wants to merge 2 commits intomainfrom
Conversation
The anchor compensation in resize() uses world-space AABB coordinates. When a mesh is rotated, the world AABB is larger than the actual mesh bounds, so applying a subsequent resize produced a different position offset in the live preview than at runtime (where resize runs before rotation in the sequential chain). Fix: when any transform block (resize, rotate_to, scale) in a DO chain has a value changed or is moved, reset the mesh to its initial creation state (identity rotation, unit scaling, base position) and replay the entire DO chain in order. This exactly replicates the runtime execution sequence so the live preview always matches what "play" produces. Also extends scale block support to getMeshFromBlock/getMeshesFromBlock and the contextBlock detection in updateMeshFromBlock, and adds BLOCK_MOVE replay so reordering DO-chain blocks updates the preview. https://claude.ai/code/session_011dYQCXitV5ZhAgoT7537XF
…r walk Two bugs in the rotation gizmo onDragEndObservable caused shapes to jump vertically after being rotated with the gizmo: 1. Wrong while-loop condition: `!mesh.parent.physics` checked the PARENT's physics instead of the current node's, stopping one level too early for compound meshes. The subsequent `if (!mesh?.physics) return` then always fired early, skipping the entire block update. 2. Non-physics meshes were completely bypassed by `if (!mesh?.physics) return`, so the rotate_to block was never created/updated. The next time applyDoChainToMeshes ran (e.g. from another block change event), it would replay the DO chain without any rotate_to block, resetting the mesh rotation (and thus the Y position compensation) to zero — causing the visible jump. Fix: separate physicsMesh walk (corrected condition checks current node), remove the early-exit guard, use findParentWithBlockId for reliable mesh→block lookup on both compound and simple meshes. https://claude.ai/code/session_011dYQCXitV5ZhAgoT7537XF
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
This PR adds support for the
scaleblock type and refactors transform block handling to properly replay the entire DO-chain when any transform block is modified. This ensures the live preview correctly matches runtime behavior, especially when multiple transforms are chained together (e.g., resize → rotate_to → resize).Key Changes
Added
scaleblock support: ExtendedgetMeshFromBlock()andgetMeshesFromBlock()to recognize and handle thescaleblock type alongsiderotate_toandresize.Introduced DO-chain replay logic: Created two new helper functions:
getContainerBlock(): Walks up the block hierarchy to find the containing create_/load_ block that owns the DO chainapplyTransformBlockToMeshes(): Applies a single transform block's effect to a list of meshesapplyDoChainToMeshes(): Resets meshes to their creation state, restores base position, then replays all enabled transform blocks in orderFixed transform block updates: Modified
updateMeshFromBlock()to replay the entire DO chain when transform blocks are moved or their fields change, rather than applying transforms in isolation. This ensures anchor compensation inresize()always operates on an unrotated mesh, matching runtime behavior.Added BLOCK_MOVE event handling: Transform blocks now properly handle block reordering by replaying the full chain, ensuring the preview reflects the new block order.
Updated block change handlers: Modified
blocks/transform.jsto trigger DO-chain replay when a transform block itself is moved/connected, and addedscaleto the list of recognized transform block types.Notable Implementation Details
resize()to re-capture geometry dimensions from the reset state.https://claude.ai/code/session_011dYQCXitV5ZhAgoT7537XF