[iwoplaza]

Stacked on #3331.

pnpm run changes printed every changed file (400+ lines for this release) and every package that had any file touched. Deciding what to bump still took manual filtering: private packages, test-only changes, and packages already bumped.

New packages/tgpu-dev-cli/release-plan.mjs with getReleasePlan({ base, head }) sorts each changed package into one of these statuses. base defaults to the latest release tag (see #3331), so "unchanged since base" means "unchanged since it was last published". Commits that went straight to release and were merged back into main still count as unreleased.

status	meaning
needs-release	public, has source changes, version unchanged since base. Comes with a suggested nextVersion (patch bump; prereleases are left to bump by hand).
new	public, but not in the latest release yet, so its current version is the first one to publish
already-bumped	version differs from the one in the latest release
non-source-only	only tests, docs, *.md, vitest configs, tsconfig.json, or devDependencies/version in package.json changed
private	not published

pnpm run changes now prints a summary. Use --verbose to also list the files, with the ones that aren't shipped greyed out:

ℹ Packages to release:
  @typegpu/cli                0.12.3 → 0.12.4   (15 source files)
  @typegpu/color              0.12.1 → 0.12.2   (2 source files)
  ...
  typegpu                     0.12.9 → 0.12.10  (69 source files)

ℹ Changed, but no release needed:
  @typegpu/algorithm          private
  @typegpu/geometry           private
  typegpu-testing-utility     private

Tested (comparing against v0.12.9) on:

• origin/main: the list above.
• The bump PR bump: typegpu 0.12.10 (+more) #3328 (--head origin/patch-release-workflow): 7 packages "already bumped", and react/three still to release.
• test: Depth textures can be viewed as unfilterable-float #3322 (test-only commit): "tests, docs or configs only".
• fix: Make the '~unstable' export compatible with Babel's CommonJS transform #3306: "needs release".

Limitation: changes that only remove unused code (like #3087 in @typegpu/react and @typegpu/three this cycle) still count as source changes, since a file diff can't tell them apart.

Next in the stack: tgpu-dev-cli bump, which applies these versions.

[github-actions]

pkg.pr.new

packages
Ready to be installed by your favorite package manager ⬇️

https://pkg.pr.new/software-mansion/TypeGPU/tgpu-gen@a66b662bb4ce5f2f1ab36b15d2c56cae70c24fee

https://pkg.pr.new/software-mansion/TypeGPU/typegpu@a66b662bb4ce5f2f1ab36b15d2c56cae70c24fee

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/cli@a66b662bb4ce5f2f1ab36b15d2c56cae70c24fee

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/color@a66b662bb4ce5f2f1ab36b15d2c56cae70c24fee

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/gl@a66b662bb4ce5f2f1ab36b15d2c56cae70c24fee

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/radiance-cascades@a66b662bb4ce5f2f1ab36b15d2c56cae70c24fee

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/react@a66b662bb4ce5f2f1ab36b15d2c56cae70c24fee

https://pkg.pr.new/software-mansion/TypeGPU/@typegpu/three@a66b662bb4ce5f2f1ab36b15d2c56cae70c24fee

https://pkg.pr.new/software-mansion/TypeGPU/unplugin-typegpu@a66b662bb4ce5f2f1ab36b15d2c56cae70c24fee

benchmark
view benchmark

commit
view commit

[github-actions]

Resolution Time Benchmark

---
config:
  themeVariables:
    xyChart:
      plotColorPalette: "#E63946, #3B82F6, #059669"
---
xychart
  title "Random Branching (🔴 PR | 🔵 main | 🟢 release)"
  x-axis "max depth" [1, 2, 3, 4, 5, 6, 7, 8]
  y-axis "time (ms)"
  line [0.75, 1.63, 3.47, 4.94, 6.07, 10.26, 21.13, 23.41]
  line [0.78, 1.52, 3.13, 5.45, 6.00, 10.21, 19.25, 24.22]
  line [1.11, 2.14, 4.97, 6.71, 8.13, 14.01, 26.15, 25.69]

Loading

---
config:
  themeVariables:
    xyChart:
      plotColorPalette: "#E63946, #3B82F6, #059669"
---
xychart
  title "Linear Recursion (🔴 PR | 🔵 main | 🟢 release)"
  x-axis "max depth" [1, 2, 3, 4, 5, 6, 7, 8]
  y-axis "time (ms)"
  line [0.18, 0.35, 0.50, 0.66, 0.87, 0.90, 1.05, 1.17]
  line [0.24, 0.41, 0.54, 0.59, 0.78, 0.84, 1.01, 1.28]
  line [0.27, 0.61, 0.79, 0.89, 1.22, 1.31, 1.59, 1.75]

Loading

---
config:
  themeVariables:
    xyChart:
      plotColorPalette: "#E63946, #3B82F6, #059669"
---
xychart
  title "Full Tree (🔴 PR | 🔵 main | 🟢 release)"
  x-axis "max depth" [1, 2, 3, 4, 5, 6, 7, 8]
  y-axis "time (ms)"
  line [0.83, 1.89, 3.19, 6.02, 10.61, 22.22, 46.64, 97.21]
  line [0.84, 1.97, 3.37, 5.88, 10.29, 21.45, 46.40, 94.66]
  line [0.90, 2.31, 4.15, 7.70, 15.67, 28.14, 59.29, 122.82]

Loading

[github-actions]

Bundle size comparison (import * as ... in PR vs import * as ... in target):

🟢 Decreased	➖ Unchanged	🔴 Increased	❔ Unknown
0	356	0	0

No notable changes.

If you wish to run a comparison for other, slower bundlers, run the 'Tree-shake test' from the GitHub Actions menu.