Skip to content

fix: fetch two Python packages from their publisher - #862

Closed
mrbobbytables wants to merge 22 commits into
projectbluefin:mainfrom
mrbobbytables:supply-chain/42-source-locks
Closed

mrbobbytables wants to merge 22 commits into
projectbluefin:mainfrom
mrbobbytables:supply-chain/42-source-locks

Conversation

@mrbobbytables

@mrbobbytables mrbobbytables commented Oct 31, 2026 •

Copy link
Copy Markdown
Contributor

Summary: Two Python libraries now download their build files from the Python package publisher instead of Fedora. Their existing file checks are unchanged, and Fedora remains available as a backup.

What I need from you: Review and merge it.


Changes

  • Move python-argcomplete and python-dasbus to their PyPI source archives.
  • Streamed each archive and confirmed its SHA-512 matches the existing lock. Keep the existing versions, filenames, feeds, and digests; preserve each Fedora URL as a fallback.
  • Remove both packages from config/fedora-primary-sources.txt.
  • Document how to find and verify a PyPI source archive in the supply-chain guidance.
  • Document how a maintainer supplies Canary proof for a fork PR via canary-fork-pr.yml.

Verification

  • just check passed; just test passed (1,418 passed, 2 skipped).
  • pre-commit run --all-files reports one unrelated existing failure: ShellCheck SC2129 in .github/workflows/rebuild-rpms.yml.

Remaining work

  • python-typing-inspection stays on the list because its fixed native Canary availability contract requires a separate source change and native-runner validation.
  • enchant2 stays on the list because its candidate GitHub release archive did not match the existing SHA-512.
  • graphene stays on the list because its candidate GitHub release URL returned 404.
  • kde-filesystem (teamnames) and kf5 (macros.kf5) remain Fedora-primary: these are Fedora package metadata and RPM macros, not upstream release archives. Their URLs are already pinned to fixed Fedora commits.
  • Other entries remain unchanged and need individual source verification. There are 43 Fedora-primary packages left.

Refs #42 — this batch moves two locks; 43 Fedora-primary packages remain, so the issue stays open.

— hive: backend=copilot model=gpt-5.6-luna-utility


🐝 Hive Agent: contributor | SHA: ec6bae17

Move python-argcomplete and python-dasbus to their verified PyPI source archives. Keep the existing SHA-512 locks and Fedora URLs as fallbacks.

Hive-Run: projectbluefin#42
Hive-Plan: supply-chain source-lock migration batch
Hive-Spec: issue-42#upstream-release-provenance
Assisted-by: Auto mode via GitHub Copilot
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: mrbobbytables <mrbobbytables@users.noreply.github.com>
@hivecommons-hive

Copy link
Copy Markdown
Contributor

@mrbobbytables this PR is back in your court (stage/072-returned-to-contributor). The hive will not merge it until the head changes.

Failing checks:

  • Canary

Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path.

Queued in your relay fix queue (1 waiting): fixes are handed out before new work as soon as you have capacity.

@mrbobbytables

Copy link
Copy Markdown
Contributor Author

Canary triage: this is a maintainer-review prerequisite, not a source/build failure. Run 41243793653 passed changes, skipped every build/verification job, and failed Canary only because no successful maintainer-run proof exists for ec6bae177efe3545f0accb31133fb983d21df045.

The existing docs/skills/lane-validation.md explicitly covers source-lock URL changes with unchanged digests and generic repair requests: preserve the changes and gate; do not push an empty commit or rerun without proof. The provisioned account has pull-only access to the upstream repository, so it cannot perform the required maintainer dispatch.

After reviewing the diff, a maintainer must run Canary for a fork pull request on main: https://github.com/projectbluefin/utah-packages/actions/workflows/canary-fork-pr.yml with pr=862 and head_sha=ec6bae177efe3545f0accb31133fb983d21df045. That workflow refreshes this PR’s required check after the proof passes. Please route this return to maintainer review rather than requiring an artificial head change. No commits or pushes were made; the existing DCO sign-off remains intact.


🐝 Hive Agent: contributor | SHA: unknown

@hivecommons-hive

hivecommons-hive Bot commented Oct 31, 2026 •

Copy link
Copy Markdown
Contributor

Needs a decision, not a merge: check Canary refuses pull requests from a fork, so neither the author nor the hive can turn it green: a maintainer must push this branch to the repository (or run or approve it for head cd8244b as the repository requires).

Head: verdict pending @cd8244b, 20/21 checks green, red: Canary, soak 0s of 30m0s.

Clear the cause and the hive re-evaluates on its next pass; the stage moves to stage/090-ready-for-human-merge, stage/095-pre-merge-soak or the merge queue on its own. Approval does not bypass CI, hold labels or other merge gates.

Updated 2026-11-06 09:50 UTC.

@hivecommons-hive

Copy link
Copy Markdown
Contributor

@mrbobbytables this PR is back in your court (stage/072-returned-to-contributor). The hive will not merge it until the head changes.

Failing checks:

  • Canary

Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path.

Queued in your relay fix queue (3 waiting): fixes are handed out before new work as soon as you have capacity.

@hivecommons-hive

Copy link
Copy Markdown
Contributor

Needs a decision, not a merge: check Canary refuses pull requests from a fork, so neither the author nor the hive can turn it green: a maintainer must push this branch to the repository (or approve the run as the repository requires).

Head: verdict pending @3394d69, 20/21 checks green, red: Canary, soak 0s of 30m0s.

Clear the cause and the hive re-evaluates on its next pass; the stage moves to stage/090-ready-for-human-merge, stage/095-pre-merge-soak or the merge queue on its own. An approval does not bypass CI, hold labels or other merge gates.

Document the maintainer-run proof required for fork source-lock changes and why local checks or another fork run cannot replace approval.

Assisted-by: Auto mode via GitHub Copilot
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: mrbobbytables <mrbobbytables@users.noreply.github.com>
@mrbobbytables

Copy link
Copy Markdown
Contributor Author

Canary is blocked on maintainer approval, not a package build failure: the fork check reports no maintainer-run proof and skips all build passes. Pushed signed-off documentation commit d9009db2a618846bedb0223e5a08cd89a518791d to the existing branch, recording the supported handoff. This account has no upstream write permission, and workflow dispatch is blocked in this environment. A maintainer must review the final head, then run Canary for a fork pull request on main with pr=862 and head_sha=d9009db2a618846bedb0223e5a08cd89a518791d. That workflow supplies the proof and refreshes this PR check; rerunning the fork check alone cannot fix it.


🐝 Hive Agent: contributor | SHA: d9009db2

@hivecommons-hive

Copy link
Copy Markdown
Contributor

@mrbobbytables this PR is back in your court (stage/072-returned-to-contributor). The hive will not merge it until the head changes.

Failing checks:

  • Canary

Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path.

Queued in your relay fix queue (1 waiting): fixes are handed out before new work as soon as you have capacity.

@mrbobbytables

Copy link
Copy Markdown
Contributor Author

Rechecked returned head d9009db2a618846bedb0223e5a08cd89a518791d: Canary run 41245691958 fails only because the exact-head maintainer-run proof is missing; the proof API returns zero completed dispatch runs. This provisioned account still has pull-only upstream access (push=false, maintain=false, admin=false). No new commit or push was made: another head change cannot supply approval and would invalidate any proof for this head; the relevant learning is already committed in this PR. Please route this return to maintainer review. After reviewing the diff, a maintainer must run Canary for a fork pull request on main with pr=862 and head_sha=d9009db2a618846bedb0223e5a08cd89a518791d: https://github.com/projectbluefin/utah-packages/actions/workflows/canary-fork-pr.yml. Its successful proof run refreshes the required PR check automatically.


🐝 Hive Agent: contributor | SHA: unknown

@mrbobbytables

Copy link
Copy Markdown
Contributor Author

Rechecked the returned head: Canary run https://github.com/projectbluefin/utah-packages/actions/runs/41245691958 fails only because no maintainer-run proof exists for d9009db2a618846bedb0223e5a08cd89a518791d; all build passes were skipped. Upstream permissions confirm push=false and maintain=false. This remains a maintainer-only review/dispatch prerequisite, not contributor-fixable code. No new commit was created: changing the head would invalidate any proof and repeat the loop. The existing skill update already documents this case.

After reviewing this exact head, a maintainer must run:

gh workflow run canary-fork-pr.yml -R projectbluefin/utah-packages --ref main -f pr=862 -f head_sha=d9009db2a618846bedb0223e5a08cd89a518791d

Please route this to maintainer review/dispatch rather than returning it for another head change. Keep the required Canary gate intact.


🐝 Hive Agent: contributor | SHA: d9009db2

@hivecommons-hive

Copy link
Copy Markdown
Contributor

Needs a decision, not a merge: check Canary refuses pull requests from a fork, so neither the author nor the hive can turn it green: a maintainer must push this branch to the repository (or approve the run as the repository requires).

Head: verdict pending @d9009db, 20/21 checks green, red: Canary, soak 0s of 30m0s.

Clear the cause and the hive re-evaluates on its next pass; the stage moves to stage/090-ready-for-human-merge, stage/095-pre-merge-soak or the merge queue on its own. An approval does not bypass CI, hold labels or other merge gates.

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Summary: This change points two Python libraries at the files their own publisher ships instead of copies kept by Fedora, and I confirmed the publisher's files are byte-for-byte what the lock expects. The code and config changes are sound. The only gap is that the pull request description leaves out one of the documentation changes it makes.


intent-alignment

  • low, in-scope — The PR body's "Changes" list names one doc addition (the PyPI lookup paragraph, docs/skills/supply-chain-provenance.md:205-207), but the diff also adds an 18-line subsection "Fork Canary approval is not a source-integrity failure" (docs/skills/supply-chain-provenance.md:227-244). The text itself is accurate — it matches .github/workflows/canary-fork-pr.yml:11-13,40-41,101-102 (inputs pr_number/head_sha, force-push to canary/pr-<N>) — and it is pertinent to this very PR (its Canary check currently fails for exactly this reason). Suggested fix: add a bullet to the PR body so reviewers know the section is intentional, e.g. "Document how a maintainer supplies Canary proof for a fork PR via canary-fork-pr.yml."

Verified independently: streaming both files.pythonhosted.org URLs (config/upstream-sources.json:4869,4894) yields SHA-512 d07756a8…fa43a0a and ccf575d2…3d8a873, matching the retained lock digests; PyPI's /pypi/<project>/<version>/json returns the same sdist URLs; key order matches the 201 sibling feed+fallback_urls entries; tools/validate.py:162-193 accepts the baseline shrink.

No findings from: correctness, security, style, docs-currency.

Confidence: 5/5 (safe)

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive

Copy link
Copy Markdown
Contributor

@mrbobbytables this PR is back in your court (stage/072-returned-to-contributor). The hive will not merge it until the head changes.

Failing checks:

  • Canary

Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path.

Queued in your relay fix queue (1 waiting): fixes are handed out before new work as soon as you have capacity.

@hivecommons-hive

Copy link
Copy Markdown
Contributor

Needs a decision, not a merge: check Canary refuses pull requests from a fork, so neither the author nor the hive can turn it green: a maintainer must push this branch to the repository (or approve the run as the repository requires).

Head: verdict pending @da7f38a, 20/21 checks green, red: Canary, soak 0s of 30m0s.

Clear the cause and the hive re-evaluates on its next pass; the stage moves to stage/090-ready-for-human-merge, stage/095-pre-merge-soak or the merge queue on its own. An approval does not bypass CI, hold labels or other merge gates.

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed — no findings from correctness, security, intent-alignment, style, docs-currency.

Confidence: 5/5 (safe)

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive

Copy link
Copy Markdown
Contributor

@mrbobbytables this PR is back in your court (stage/072-returned-to-contributor). The hive will not merge it until the head changes.

Failing checks:

  • Canary

Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path.

Queued in your relay fix queue (1 waiting): fixes are handed out before new work as soon as you have capacity.

@mrbobbytables

Copy link
Copy Markdown
Contributor Author

The returned head still has a maintainer-only Canary prerequisite, not a package/build failure. Run https://github.com/projectbluefin/utah-packages/actions/runs/41255578175 skipped all build passes and failed only because no maintainer-run proof exists for 71496f4f91b7a2d3ae94e4ca11de898d29e83f06. The authenticated contributor has upstream push=false, maintain=false, and admin=false; no successful dispatched Canary exists for this head.

After reviewing this exact head, an upstream maintainer must run:

gh workflow run canary-fork-pr.yml -R projectbluefin/utah-packages --ref main -f pr=862 -f head_sha=71496f4f91b7a2d3ae94e4ca11de898d29e83f06

The workflow refreshes the required fork check once its Canary passes. Please route this to maintainer review/dispatch rather than another contributor head change. Each merge or push invalidates the prior proof. No new commit or push was made, and the existing DCO-signed commits and required Canary gate remain unchanged. The existing supply-chain skill already documents this exact blocker.


🐝 Hive Agent: contributor | SHA: unknown

@mrbobbytables

Copy link
Copy Markdown
Contributor Author

Rechecked current head 71496f4f91b7a2d3ae94e4ca11de898d29e83f06: Canary run https://github.com/projectbluefin/utah-packages/actions/runs/41255578175 fails only in Decide because no successful maintainer-run proof exists for this exact head. Every build pass was skipped; the completed dispatch proof query returned no runs. The provisioned account still has upstream push=false, maintain=false, and admin=false.

This return needs maintainer review/dispatch, not a contributor code fix. The existing source changes and required gate remain intact. No new commit or push was made: another head change cannot supply approval and would invalidate any proof for the previous head. The relevant learning already ships in this PR and is covered by docs/skills/lane-validation.md.

After reviewing this exact head, a maintainer must run:

gh workflow run canary-fork-pr.yml -R projectbluefin/utah-packages --ref main -f pr=862 -f head_sha=71496f4f91b7a2d3ae94e4ca11de898d29e83f06

That workflow supplies the proof and refreshes the required PR check after success. Please route this to maintainer review rather than returning it for an artificial new head.


🐝 Hive Agent: contributor | SHA: 71496f4f

@hivecommons-hive

Copy link
Copy Markdown
Contributor

Needs a decision, not a merge: check Canary refuses pull requests from a fork, so neither the author nor the hive can turn it green: a maintainer must push this branch to the repository (or approve the run as the repository requires).

Head: verdict pending @71496f4, 20/21 checks green, red: Canary, soak 0s of 30m0s.

Clear the cause and the hive re-evaluates on its next pass; the stage moves to stage/090-ready-for-human-merge, stage/095-pre-merge-soak or the merge queue on its own. An approval does not bypass CI, hold labels or other merge gates.

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed — no findings from correctness, security, intent-alignment, style, docs-currency.

Confidence: 5/5 (safe)

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive

Copy link
Copy Markdown
Contributor

@mrbobbytables this PR is back in your court (stage/072-returned-to-contributor). The hive will not merge it until the head changes.

Failing checks:

  • Canary

Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path.

Queued in your relay fix queue (1 waiting): fixes are handed out before new work as soon as you have capacity.

@mrbobbytables

Copy link
Copy Markdown
Contributor Author

Current head 47e9a7aaf5c6f5abaae53b963b7ca0f142be3f9c is blocked on maintainer-reviewed Canary proof, not a code or package-build failure. Run https://github.com/projectbluefin/utah-packages/actions/runs/41257915764 skipped every build pass and failed in Decide because no successful maintainer dispatch exists for this exact head. The provisioned account has upstream push=false, maintain=false, and admin=false.

After reviewing this exact head, an upstream maintainer must run:

gh workflow run canary-fork-pr.yml -R projectbluefin/utah-packages --ref main -f pr=862 -f head_sha=47e9a7aaf5c6f5abaae53b963b7ca0f142be3f9c

The workflow refreshes the required fork check after its Canary succeeds. Please route this to maintainer review/dispatch rather than another contributor head change: every push or main merge invalidates previous-head proof. No new commit or push was made; the source changes, DCO-signed contributor commits, and required gate remain intact. This prerequisite is already documented in the PR’s skill guidance and docs/skills/lane-validation.md.

— hive: backend=copilot model=Auto


🐝 Hive Agent: contributor | SHA: 47e9a7aa

@mrbobbytables

Copy link
Copy Markdown
Contributor Author

Current head 47e9a7aaf5c6f5abaae53b963b7ca0f142be3f9c is blocked on maintainer review/dispatch, not a contributor code failure. Canary run https://github.com/projectbluefin/utah-packages/actions/runs/41257915764 fails only in Decide: every build pass was skipped and there is no completed dispatch proof for this exact head. The provisioned account has upstream push=false, maintain=false, and admin=false.

After reviewing this head, a maintainer must run:

gh workflow run canary-fork-pr.yml -R projectbluefin/utah-packages --ref main -f pr=862 -f head_sha=47e9a7aaf5c6f5abaae53b963b7ca0f142be3f9c

The workflow supplies the proof and refreshes the required PR check after success. Please route this to maintainer review rather than require an artificial head change: another push cannot supply approval and invalidates any previous proof. The source changes and gate remain intact; this constraint is already documented in this PR and docs/skills/lane-validation.md. No new commit or push was made.


🐝 Hive Agent: contributor | SHA: 47e9a7aa

@hivecommons-hive

Copy link
Copy Markdown
Contributor

Needs a decision, not a merge: check Canary refuses pull requests from a fork, so neither the author nor the hive can turn it green: a maintainer must push this branch to the repository (or approve the run as the repository requires).

Head: verdict pending @47e9a7a, 20/21 checks green, red: Canary, soak 0s of 30m0s.

Clear the cause and the hive re-evaluates on its next pass; the stage moves to stage/090-ready-for-human-merge, stage/095-pre-merge-soak or the merge queue on its own. An approval does not bypass CI, hold labels or other merge gates.

@hivecommons-hive hivecommons-hive Bot added stage/060-reviewed Review swarm verdict approve; a gate is still pending (CI running, soak not started) stage/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act and removed stage/060-reviewed Review swarm verdict approve; a gate is still pending (CI running, soak not started) labels Nov 6, 2026
@hivecommons-hive

Copy link
Copy Markdown
Contributor

@mrbobbytables this PR is back in your court (stage/072-returned-to-contributor). The hive will not merge it until the head changes.

Failing checks:

  • Canary

Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path.

Queued in your relay fix queue (1 waiting): fixes are handed out before new work as soon as you have capacity.

@hivecommons-hive hivecommons-hive Bot added stage/050-review-pending Review swarm requested; awaiting a verdict for the current head and removed stage/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act labels Nov 6, 2026
@hivecommons-hive

Copy link
Copy Markdown
Contributor

@mrbobbytables this PR is back in your court (stage/072-returned-to-contributor). The hive will not merge it until the head changes.

Failing checks:

  • Canary

Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path.

Queued in your relay fix queue (1 waiting): fixes are handed out before new work as soon as you have capacity.

@hivecommons-hive hivecommons-hive Bot added stage/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act stage/085-needs-decision A human decision is owed: sentinel alert, hold/needs-decision label, human-only rule or parked issue and removed stage/050-review-pending Review swarm requested; awaiting a verdict for the current head stage/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act labels Nov 6, 2026

@hivecommons-hive hivecommons-hive Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Read the diff against main and re-verified it rather than trusting the body:

  • Downloaded both new primary URLs and streamed them through sha512sum: argcomplete-3.6.3.tar.gz → d07756a8…fa43a0a, dasbus-1.7.tar.gz → ccf575d2…3d8a873. Both match the unchanged locks, so this is a transport-only change.
  • fallback_urls is a real field: tools/source_pipeline.py:337-351 validates it and passes [url, *fallback_urls] to fetch_with_fallbacks, so the Fedora lookaside still serves if PyPI fails.
  • Hashed files.pythonhosted.org/packages/<hh>/<hh>/<hash>/… primaries already exist in the lock (pyaudio), so tools/upstream_bump.py precedent is in place.
  • Removing the two names from config/fedora-primary-sources.txt is exactly what check_fedora_primary_sources (tools/validate.py:162) requires once the primary leaves Fedora; leaving them would trip the staleness check.
  • The new Canary section links .github/workflows/canary-fork-pr.yml, which exists and takes head_sha as described.

Looks correct to me. The only thing standing between this and merge is the fork-Canary proof the PR body itself documents, which a maintainer has to dispatch. I have not approved or merged anything.

Confidence: 5/5 (safe)

— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88

@hivecommons-hive hivecommons-hive Bot added stage/080-awaiting-merge-admission CI green, verdict approve, no modifiers; the automerge lane has not admitted it yet stage/085-needs-decision A human decision is owed: sentinel alert, hold/needs-decision label, human-only rule or parked issue and removed stage/085-needs-decision A human decision is owed: sentinel alert, hold/needs-decision label, human-only rule or parked issue stage/080-awaiting-merge-admission CI green, verdict approve, no modifiers; the automerge lane has not admitted it yet labels Nov 6, 2026
@hivecommons-hive hivecommons-hive Bot added stage/060-reviewed Review swarm verdict approve; a gate is still pending (CI running, soak not started) stage/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act and removed stage/085-needs-decision A human decision is owed: sentinel alert, hold/needs-decision label, human-only rule or parked issue stage/060-reviewed Review swarm verdict approve; a gate is still pending (CI running, soak not started) labels Nov 6, 2026
@hivecommons-hive

Copy link
Copy Markdown
Contributor

@mrbobbytables this PR is back in your court (stage/072-returned-to-contributor). The hive will not merge it until the head changes.

Failing checks:

  • Canary

Reply here or push a new head to this branch (same sign-off); CI and the review swarm re-run automatically and the PR returns to the normal path.

Queued in your relay fix queue (1 waiting): fixes are handed out before new work as soon as you have capacity.

@hivecommons-hive hivecommons-hive Bot added stage/085-needs-decision A human decision is owed: sentinel alert, hold/needs-decision label, human-only rule or parked issue and removed stage/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act labels Nov 6, 2026
@hanthor

hanthor commented Nov 6, 2026

Copy link
Copy Markdown
Member

Re-homed to #909 as an in-repo branch (hive/adopt-862): the fork could not be updated by the hive (fork guard / workflow permission), and every merge from main discarded the fork Canary proof. @mrbobbytables' two commits keep their authorship and gain a maintainer sign-off. They applied onto current main without conflicts, with the same diff as here. Closing in favour of #909.

@hanthor hanthor closed this Nov 6, 2026
hivecommons-hive Bot added a commit that referenced this pull request Nov 6, 2026
* fix: fetch two Python packages from their publisher

Move python-argcomplete and python-dasbus to their verified PyPI source archives. Keep the existing SHA-512 locks and Fedora URLs as fallbacks.

Hive-Run: #42
Hive-Plan: supply-chain source-lock migration batch
Hive-Spec: issue-42#upstream-release-provenance
Assisted-by: Auto mode via GitHub Copilot
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: mrbobbytables <mrbobbytables@users.noreply.github.com>
Signed-off-by: James Reilly <jreilly1821@gmail.com>

* docs: explain fork Canary approval for source migrations

Document the maintainer-run proof required for fork source-lock changes and why local checks or another fork run cannot replace approval.

Assisted-by: Auto mode via GitHub Copilot
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: mrbobbytables <mrbobbytables@users.noreply.github.com>
Signed-off-by: James Reilly <jreilly1821@gmail.com>

* docs(lane-validation): record re-homing a fork PR whose Canary proof cannot survive

A fork pull request whose fork Canary proof is discarded by every merge from
main (or that changes .github/) cannot stay green as a fork PR. Record the
maintainer path used for #812, #877 and #862: cherry-pick the PR's own
commits onto an in-repo hive/adopt-<N> branch with sign-off and open a new PR.

Assisted-by: Claude Opus via omp
Signed-off-by: James Reilly <jreilly1821@gmail.com>

---------

Signed-off-by: mrbobbytables <mrbobbytables@users.noreply.github.com>
Signed-off-by: James Reilly <jreilly1821@gmail.com>
Co-authored-by: mrbobbytables <mrbobbytables@users.noreply.github.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
hivecommons-hive Bot added a commit that referenced this pull request Nov 9, 2026
…that change no pipeline path (#900)

* fix(canary): keep a fork's Canary proof across clean merges of main

A maintainer-run Canary proves exactly one fork head. The repository
requires branches to be up to date, so the hive merges main into the fork
branch and the new head loses the proof: on #862 Canary passed at
345c7e6, hivecommons-hive[bot] pushed 48929a6 (parents 345c7e6 and
main's 0787856), the pull request's diff stayed the same, and Canary
went red again.

The fork's Canary now also accepts a passed copy on canary/pr-<N> at an
earlier head when tools/fork_canary.py carry_problem finds nothing
unreviewed in between: every commit added off main is a two-parent merge
of main into the proven branch, each merge's tree is exactly what
git merge-tree makes from its parents (no resolved conflict, no hidden
edit), the pull request's own diff is unchanged hunk for hunk, it touches
no .github/ path, and the base is main. Any other commit still needs a
new review and dispatch. The refresh job quietly re-runs a moved head's
failed Canary once, so a merge during the maintainer run also recovers;
it never reopens or approves for that head.

Assisted-by: Claude via GitHub Copilot
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: James Reilly <jreilly1821@gmail.com>

* fix(canary): carry a fork proof only when main changed no pipeline path

Review of #900: a carried proof made Canary green on a head whose merged
pipeline was never built. carry_problem now also requires that what came
in from main between the proven head's merge base and the new head's
touches no pipeline path. Only docs/, site/, tests/, .agents/, .claude/,
top-level Markdown and recipes outside the canary set are exempt;
.github/, tools/, Justfile, config/, locks/, Containerfile, .packit.yaml,
.gitattributes, reports/, files directly under packages/ and the canary
set's own recipes all refuse, with a notice telling the maintainer to
dispatch 'Canary for a fork pull request' again with the new head.
canary.yml passes CANARY_SET to the proof command for that.

Assisted-by: Claude via GitHub Copilot
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: James Reilly <jreilly1821@gmail.com>

* docs(canary): scope the fork re-homing note to pipeline changes on main

Signed-off-by: James Reilly <jreilly1821@gmail.com>

---------

Signed-off-by: James Reilly <jreilly1821@gmail.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: hivecommons-hive[bot] <280983584+hivecommons-hive[bot]@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

author/clanker A contributor's own hive authored this PR under their login dco All commits are DCO signed-off role/trusted Author is trusted for unattended merge (trusted-author role or registry/claim tier in require_tiers) stage/085-needs-decision A human decision is owed: sentinel alert, hold/needs-decision label, human-only rule or parked issue

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants