Conversation
zyn080302
force-pushed
the
fix/studio-release-locked-wheels-20260917
branch
from
September 17, 2026 03:01
d5fb434 to
7cbcff8
Compare
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Studio thin releases fail when cold-start precompilation rewrites third-party wheels:
agentkit-sdk-pythonno longer matches the original bytes recorded inuv.lock. Preserve a separate wheel snapshot before optimization. Full bundles retain precompilation; thin bundles use the locked third-party wheels and precompile local VeADK. Each bundle validates its own dependency hashes, and public artifact provenance remains strict.Release verification now calls the Release Server publisher to build full and thin bundles. Smoke tests explicitly reuse the selected full archive for both providers, so directory ordering cannot select a provider-specific thin archive. Regressions exercise actual precompilation/provenance checks, reject tampered wheels, and execute the workflow extraction logic with the thin archive ordered first. They run in the mandatory release gates.
The Sidecar gate installs its locked dev environment with cached uv and runs all four check groups from that environment. This removes pip resolution and installation overhead that exhausted the three-minute job budget. Lockfile changes trigger the gate and invalidate its dependency cache; the time limit, tests, coverage threshold, and failure propagation remain enforced.
Validation: RED/GREEN regressions cover wheel provenance, smoke archive selection, and the workflow's frozen environment and failure propagation. The full backend gate passed locally (271 tests, 96% Sidecar coverage). Pre-commit, changed-file Pyright, workflow shell syntax, and diff checks passed. CI covers full Python 3.10/3.12 tests, full/thin builds, and both provider smoke tests.
Rollout order: merge this PR, update the Release Server, then start a new release. Its builder executes the installed publisher, so merging this PR alone does not activate the fix.