ie#656: cut 0.109.0 — PyPI 0.108.0 shipped without pgw#1133, so master's number is already spent - #654
Merged
Merged
Conversation
… number is spent `refs/tags/v0.108.0` points at `821090c3` (ie#655), cut BEFORE pgw#1133 merged as `5e2a74ae`. Its publish failed the tree-provenance gate, a `ci.yml` dispatch on `821090c3` went green, the re-run succeeded, and PyPI's `gen_worker-0.108.0-py3-none-any.whl` (uploaded 18:09:42Z) has NO `gen_worker/convert/dtype_pins.py` and a `clone.py` that never mentions the pins. So pgw#1133's close-out line "the eventual 0.108.0 artifact contains both changes" is not what shipped, and master could not be tagged again at 0.108.0: one artifact, one number, and that number is taken. 0.109.0 carries pgw#1133 alone. The 0.107.0/0.108.0 pair was tagged without a changelog cut, so this also assembles the four fragments that DID ship in 0.108.0 into their own section, dated and attributed to the tree that carries them — the CHANGELOG now says which artifact holds what instead of implying one release contains work that predates it.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
refs/tags/v0.108.0on origin points at821090c3— ie#655's commit, cut before pgw#1133 merged as5e2a74ae. That tag's publish failed the tree-provenance gate at 17:48Z; aci.ymldispatch on821090c3went green and the re-run published.Verified on the artifact, not inferred:
gen_worker-0.108.0-py3-none-any.whl(PyPI, uploaded 2026-08-11T18:09:42Z) contains nogen_worker/convert/dtype_pins.py, and itsgen_worker/convert/clone.pynever mentionsdtype_pinsorpinned.So pgw#1133 is unpublished, and master cannot be tagged at 0.108.0 again — one artifact, one number.
This PR:
pyproject.toml0.108.0 → 0.109.0;changelog.d/pgw1133.mdinto a 0.109.0 section;ie655,pgw1123,pgw1127,pgw1128) into their own 0.108.0 section — 0.107.0/0.108.0 were tagged without a cut, so the CHANGELOG had no section for them at all.Nothing else changes: no source, no tests, no locks.
Found by the ie#656 lane (the served bf16
AutoencoderKLWanontensorhub/wan22-ti2v-5b), which needed a conversion surface carrying the pgw#1133 fix and discovered the number was gone. That remediation did not wait on this — it repaired the tree with a singledtype: "source"VAE merge leg on the standing conversion endpoint.