Repository navigation
feat: add the prefix name shifter to the long-tail backlog (#505) - #819
Conversation
…uefin#819) * feat: add the prefixdevname tool that shifts Windows-style prefix names Import the prefixdevname Rust crate from the GitHub tag v0.2.0 archive, unmodified, and lock that archive sha512 since upstream publishes no release asset. Register the build in Packit, refresh the source-lock and the architecture counts, and move prefixdevname from the long-tail backlog area to [resolved] and regenerate the backlog report. No hosted build, publication or runtime closure is claimed. Refs projectbluefin#505 Hive-Run: projectbluefin#505 Hive-Plan: issue-505 implementation step 3 (prefixdevname source family) Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Claude Opus 5.5 via Claude Code Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com> * docs: link the prefixdevname closing record to its pull request Fill in the pr field on the prefixdevname [resolved] record now that the import PR has a number. Refs projectbluefin#505 Hive-Run: projectbluefin#505 Hive-Plan: issue-505 implementation step 5 (closing record) Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Claude Opus 5.5 via Claude Code Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com>
719ac78 to
28d1257
Compare
|
@kylerankin this PR is back in your court ( Review asks:
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. |
…uefin#819) * feat: add the prefixdevname tool that shifts Windows-style prefix names Import the prefixdevname Rust crate from the GitHub tag v0.2.0 archive, unmodified, and lock that archive sha512 since upstream publishes no release asset. Register the build in Packit, refresh the source-lock and the architecture counts, and move prefixdevname from the long-tail backlog area to [resolved] and regenerate the backlog report. No hosted build, publication or runtime closure is claimed. Refs projectbluefin#505 Hive-Run: projectbluefin#505 Hive-Plan: issue-505 implementation step 3 (prefixdevname source family) Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Claude Opus 5.5 via Claude Code Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com> * docs: link the prefixdevname closing record to its pull request Fill in the pr field on the prefixdevname [resolved] record now that the import PR has a number. Refs projectbluefin#505 Hive-Run: projectbluefin#505 Hive-Plan: issue-505 implementation step 5 (closing record) Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Claude Opus 5.5 via Claude Code Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com>
28d1257 to
dfdcbc6
Compare
|
@kylerankin this PR is back in your court ( Review asks:
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. |
There was a problem hiding this comment.
Summary: This change adds a small network-card naming helper to the package set, with a fully pinned and reproducible source bundle (I regenerated it and the checksum matches exactly). The source bundle and bookkeeping are fine, but the build recipe itself cannot succeed as written: three of its four install commands use a form that fails outright, and every file would land in the wrong directory, so the system's device manager would never see the rule or find the helper. Both problems have a two-line fix below.
What I need from you: Apply the suggested %install/%files fix.
correctness
-
%installfails:install -Dwith a trailing-slash destination is "Not a directory" —packages/prefixdevname/prefixdevname.spec:34-36. GNU coreutilsinstall -D src dir/treatsdir/as the final file path and aborts (install: cannot create regular file '…/rules.d/': Not a directory, reproduced with coreutils 9.7). Line 33 (the binary) is the only one that names the destination file. The spec will never reach%files. -
%{_libdir}puts udev/dracut files under/usr/lib64, where nothing reads them —packages/prefixdevname/prefixdevname.spec:33-36,42-47. udev reads/usr/lib/udev/rules.d, dracut reads/usr/lib/dracut/modules.d, and the upstream rule itself hardcodesPROGRAM="/usr/lib/udev/prefixdevname"(rules/71-prefixdevname.rulesin the v0.2.0 tag). Upstream'sMakefile installandredhat/prefixdevname.spec:37-42both use/usr/lib(%{_prefix}/lib). The repo's other udev packagers use%{_udevrulesdir}/%{_prefix}/lib/udev(e.g.packages/bolt/bolt.spec:73,packages/alsa-utils/alsa-utils.spec:131). Both issues fixed together:
%install
-install -D -m 0755 target/release/prefixdevname %{buildroot}%{_libdir}/udev/prefixdevname
-install -D -m 0644 rules/71-prefixdevname.rules %{buildroot}%{_libdir}/udev/rules.d/
-install -D -m 0755 dracut/71prefixdevname/module-setup.sh %{buildroot}%{_libdir}/dracut/modules.d/71%{name}/
-install -D -m 0755 dracut/71prefixdevname-tools/module-setup.sh %{buildroot}%{_libdir}/dracut/modules.d/71%{name}-tools/
+install -D -p -m 0755 target/release/prefixdevname %{buildroot}%{_udevdir}/prefixdevname
+install -D -p -m 0644 rules/71-prefixdevname.rules %{buildroot}%{_udevrulesdir}/71-prefixdevname.rules
+install -D -p -m 0755 dracut/71prefixdevname/module-setup.sh %{buildroot}%{_prefix}/lib/dracut/modules.d/71%{name}/module-setup.sh
+install -D -p -m 0755 dracut/71prefixdevname-tools/module-setup.sh %{buildroot}%{_prefix}/lib/dracut/modules.d/71%{name}-tools/module-setup.sh
%files
%license LICENSE
%doc README.md
-%{_libdir}/udev/prefixdevname
-%{_libdir}/udev/rules.d/71-prefixdevname.rules
-%dir %{_libdir}/dracut/modules.d/71%{name}
-%{_libdir}/dracut/modules.d/71%{name}/*
-%dir %{_libdir}/dracut/modules.d/71%{name}-tools
-%{_libdir}/dracut/modules.d/71%{name}-tools/*
+%{_udevdir}/prefixdevname
+%{_udevrulesdir}/71-prefixdevname.rules
+%dir %{_prefix}/lib/dracut/modules.d/71%{name}
+%{_prefix}/lib/dracut/modules.d/71%{name}/*
+%dir %{_prefix}/lib/dracut/modules.d/71%{name}-tools
+%{_prefix}/lib/dracut/modules.d/71%{name}-tools/*(%{_udevdir}/%{_udevrulesdir} come from systemd-rpm-macros, pulled in by the existing BuildRequires: systemd-devel.)
Verified clean: tools/generated_sources.py:980-1065 mirrors _pydantic_core_* (lines 760-825); regenerating prefixdevname-0.2.0-vendored.tar.xz from the pinned tag archive reproduces sha512 223679b0… exactly as locked in packages/prefixdevname/sources:1 and config/upstream-sources.json; tests/test_generated_sources*.py, test_architecture_counts.py, test_factory_build_backlog.py (125 passed), tools/factory_build_backlog.py --check and tools/validate.py (651 / 645 rawhide / 6 upstream) all agree with the diff.
No findings from: security, intent-alignment, style, docs-currency.
Confidence: 1/5 (needs attention) — 2 high findings
— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
…uefin#819) * feat: add the prefixdevname tool that shifts Windows-style prefix names Import the prefixdevname Rust crate from the GitHub tag v0.2.0 archive, unmodified, and lock that archive sha512 since upstream publishes no release asset. Register the build in Packit, refresh the source-lock and the architecture counts, and move prefixdevname from the long-tail backlog area to [resolved] and regenerate the backlog report. No hosted build, publication or runtime closure is claimed. Refs projectbluefin#505 Hive-Run: projectbluefin#505 Hive-Plan: issue-505 implementation step 3 (prefixdevname source family) Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Claude Opus 5.5 via Claude Code Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com> * docs: link the prefixdevname closing record to its pull request Fill in the pr field on the prefixdevname [resolved] record now that the import PR has a number. Refs projectbluefin#505 Hive-Run: projectbluefin#505 Hive-Plan: issue-505 implementation step 5 (closing record) Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Claude Opus 5.5 via Claude Code Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com>
dfdcbc6 to
57b8755
Compare
|
@kylerankin this PR is back in your court ( Failing checks:
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. |
|
Needs a decision, not a merge: check Head: verdict pending @57b8755, 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 |
There was a problem hiding this comment.
Summary: This change adds a small network-card naming helper to the package set and, since the last review, now builds a reproducible source bundle that includes all of its Rust dependencies. The bundle and its checksum verify correctly, but the two problems from the previous review are still there: the recipe's install step uses a command form that fails outright, and it places files under a directory where the system never looks for them. The package still cannot build or work as written.
What I need from you: Fix the two install issues in the spec (patch below).
Re-review of delta dfdcbc6…57b8755. New generator verified: running tools/generated_sources.py prefixdevname reproduced the locked sha512 223679b0…c13f1 byte-for-byte (config/upstream-sources.json:3089, packages/prefixdevname/sources:1).
correctness
-
Still present (high) —
packages/prefixdevname/prefixdevname.spec:33-35:install -D … <file> …/dir/with a trailing-slash destination. GNU install treats it as the destination file; reproduced with coreutils 9.7:install: cannot create regular file '…/rules.d/': Not a directory.%installaborts before%files. -
Still present (high) —
prefixdevname.spec:32-35,40-45install under%{_libdir}(/usr/lib64on x86_64/aarch64). udev reads/usr/lib/udev/rules.d, dracut reads/usr/lib/dracut/modules.d, and upstream's own rule (rules/71-prefixdevname.rules:9) and dracut module (dracut/71prefixdevname-tools/module-setup.sh,inst /usr/lib/udev/prefixdevname) hardcode/usr/lib/udev/prefixdevname; upstreamMakefile:19-25installs to/usr/lib. Sibling specs use%{_udevrulesdir}(packages/bolt/bolt.spec:73).Fix for both:
%install -install -D -m 0755 target/release/prefixdevname %{buildroot}%{_libdir}/udev/prefixdevname -install -D -m 0644 rules/71-prefixdevname.rules %{buildroot}%{_libdir}/udev/rules.d/ -install -D -m 0755 dracut/71prefixdevname/module-setup.sh %{buildroot}%{_libdir}/dracut/modules.d/71%{name}/ -install -D -m 0755 dracut/71prefixdevname-tools/module-setup.sh %{buildroot}%{_libdir}/dracut/modules.d/71%{name}-tools/ +install -D -p -m 0755 target/release/prefixdevname %{buildroot}%{_udevdir}/prefixdevname +install -D -p -m 0644 rules/71-prefixdevname.rules %{buildroot}%{_udevrulesdir}/71-prefixdevname.rules +install -D -p -m 0755 dracut/71prefixdevname/module-setup.sh %{buildroot}%{_prefix}/lib/dracut/modules.d/71%{name}/module-setup.sh +install -D -p -m 0755 dracut/71prefixdevname-tools/module-setup.sh %{buildroot}%{_prefix}/lib/dracut/modules.d/71%{name}-tools/module-setup.sh %files %license LICENSE %doc README.md -%{_libdir}/udev/prefixdevname -%{_libdir}/udev/rules.d/71-prefixdevname.rules -%dir %{_libdir}/dracut/modules.d/71%{name} -%{_libdir}/dracut/modules.d/71%{name}/* -%dir %{_libdir}/dracut/modules.d/71%{name}-tools -%{_libdir}/dracut/modules.d/71%{name}-tools/* +%{_udevdir}/prefixdevname +%{_udevrulesdir}/71-prefixdevname.rules +%dir %{_prefix}/lib/dracut/modules.d/71%{name} +%{_prefix}/lib/dracut/modules.d/71%{name}/module-setup.sh +%dir %{_prefix}/lib/dracut/modules.d/71%{name}-tools +%{_prefix}/lib/dracut/modules.d/71%{name}-tools/module-setup.sh
-
New (low) —
tools/generated_sources.py:993-1062adds_prefixdevname_transform/_prefixdevname_generatewith no test, while every other generator has one (tests/test_generated_sources.py:518PydanticCoreVendorTests,tests/test_generated_sources_generate.py:185,219). APrefixdevnameVendorTestsmirroring the pydantic one (fake tag archive + tinycrate, determinism, outside-top-dir rejection) would close it.
intent-alignment (advisory, low) — PR body says "the GitHub tag v0.2.0 archive is locked by sha512", but the lock is now the generated prefixdevname-0.2.0-vendored.tar.xz (config/upstream-sources.json:3086-3090); the tag archive is the generator's pinned input (tools/generated_sources.py:984-986). Update the body to say so.
docs-currency (advisory, low) — tools/generated_sources.py:4-29 module docstring enumerates each generated recipe and its reason; prefixdevname is not added (pydantic_core is also missing, pre-existing).
No findings from: security, style.
Confidence: 1/5 (needs attention) — 2 high findings
— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
@kylerankin the hive has adopted this PR ( |
|
1 of 2 commits on |
hanthor
left a comment
There was a problem hiding this comment.
Blocking
- CI has never run, so the new generator in
tools/generated_sources.pyis untested. - The branch needs a rebase.
Nit: find Cargo.lock by its exact f"{top}/Cargo.lock" path, as the existing generator does, rather than with endswith.
…uefin#819) * feat: add the prefixdevname tool that shifts Windows-style prefix names Import the prefixdevname Rust crate from the GitHub tag v0.2.0 archive, unmodified, and lock that archive sha512 since upstream publishes no release asset. Register the build in Packit, refresh the source-lock and the architecture counts, and move prefixdevname from the long-tail backlog area to [resolved] and regenerate the backlog report. No hosted build, publication or runtime closure is claimed. Refs projectbluefin#505 Hive-Run: projectbluefin#505 Hive-Plan: issue-505 implementation step 3 (prefixdevname source family) Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Claude Opus 5.5 via Claude Code Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com> * docs: link the prefixdevname closing record to its pull request Fill in the pr field on the prefixdevname [resolved] record now that the import PR has a number. Refs projectbluefin#505 Hive-Run: projectbluefin#505 Hive-Plan: issue-505 implementation step 5 (closing record) Hive-Spec: factory-build-backlog#closing-a-backlog-gap Assisted-by: Claude Opus 5.5 via Claude Code Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: Danathar <Danathar@users.noreply.github.com>
ba16578 to
340d915
Compare
There was a problem hiding this comment.
Looks correct to me; I reproduced the generated source lock independently.
What I checked (head fc9262d)
- Ran
python3 tools/generated_sources.py prefixdevname packages/prefixdevname <out>from a clean checkout of this head, on a different host, with network: outputprefixdevname-0.2.0-vendored.tar.xzis 4,017,312 bytes with SHA-512223679b0…c13f1— byte-identical to the lock inconfig/upstream-sources.jsonandpackages/prefixdevname/sources. The determinism claim (fixed mtime, zeroed ownership, cleared PAX headers, sorted members —tools/generated_sources.py,_prefixdevname_transform) holds in practice. - Fetched
https://github.com/msekletar/prefixdevname/archive/refs/tags/v0.2.0.tar.gzand computed SHA-512 myself: matchesPREFIXDEVNAME_INPUT_SHA512["0.2.0"], so_prefixdevname_generatereally does pin the input tree rather than trusting whatever GitHub serves. - Crate bytes are verified before vendoring: the pre-existing
_vendored_crate_members(tools/generated_sources.py:714-716) raises on a sha256 mismatch againstCargo.lock, so crates.io cannot substitute content. The new code reuses that path instead of re-implementing it. .hummingbird-upstream.jsonwith"branch": "upstream"and emptycommit/treehas precedent (bluefin-artwork-logos,gnome-rounded-blur,libfreeaptx);generateentries withstagelikewise (10 existing)..packit.yamlposition,[resolved]entry shape and backlog JSON deltas are consistent.
Canary is red for the usual fork reason — "a fork pull request cannot publish the canary image, and no maintainer-run Canary has passed" — not this diff; a maintainer needs to run the fork-Canary workflow against fc9262d59558b57d3e5469f788ca9959a97d5120.
Not checked: I did not compare prefixdevname.spec against Fedora's spec, since the provenance record declares this an upstream-branch recipe rather than a rawhide import.
Merge-order note: #823, #844, #834, #836 also bump the docs/architecture.md package-count table and edit factory-build-backlog.toml/.json; whichever lands after this one will need a count rebase.
Confidence: 5/5 (safe)
— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
Refresh architecture and backlog state from the current inventory. Add one inventory regeneration command, repair invalid imported YAML whitespace, and preserve workflow output values with grouped redirection. Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: James Reilly <jreilly1821@gmail.com>
Regenerate combined 680-recipe architecture and backlog state after projectbluefin#847 landed on main, and remove the prohibited historical DELTA session artifact carrying committed conflict markers. Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: James Reilly <jreilly1821@gmail.com>
Preserve prefixdevname recipe, source lock and Rust vendor generator/tests while structuring generated_sources cleanly alongside sibling generators. Regenerate combined 681-recipe inventory. Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: James Reilly <jreilly1821@gmail.com>
Merge current main (a590886) preserving all PR recipes and source locks, and regenerate combined inventory with just regenerate-inventory. Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: James Reilly <jreilly1821@gmail.com>
|
Merged current |
|
@kylerankin this PR is back in your court ( Failing checks:
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 (2 waiting): fixes are handed out before new work as soon as you have capacity. |
Drop the style-only grouped GITHUB_OUTPUT edit carried in from the interim baseline branch; main intentionally kept the original workflow so recipe PRs do not touch pipeline paths. Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: James Reilly <jreilly1821@gmail.com>
|
Follow-up |
|
@kylerankin this PR is back in your court ( Review asks:
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 (4 waiting): fixes are handed out before new work as soon as you have capacity. |
Merge current main and regenerate derived inventory with just regenerate-inventory. Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: James Reilly <jreilly1821@gmail.com>
|
@kylerankin this PR is back in your court ( Review asks:
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 (2 waiting): fixes are handed out before new work as soon as you have capacity. |
|
Canary was run on this pull request at |
|
@kylerankin the hive has adopted this PR ( |
|
stage 080-awaiting-merge-admission → 074-adopted-by-hive: conflicting |
Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: James Reilly <jreilly1821@gmail.com>
Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: James Reilly <jreilly1821@gmail.com>
Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Assisted-by: GPT-6.1 via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
|
Canary was run on this pull request at |
|
@kylerankin this PR has a new head without confirmed hive push evidence, so the hive has handed this PR back to you ( |
|
@kylerankin this PR is back in your court ( Review asks:
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 (2 waiting): fixes are handed out before new work as soon as you have capacity. |
There was a problem hiding this comment.
Summary: Re-review of the follow-up changes since the last approval. The new code now reads the one Cargo lock file at the top of the upstream release instead of the first file whose name happens to end in Cargo.lock, so a stray nested lock cannot change which third-party crates get bundled; that fix is tested and sound. One leftover: the pull request description still says it refreshes the recipe counts in the architecture document, but that document no longer carries counts and this PR no longer changes it.
What I need from you: Drop the architecture-counts bullet from the PR description.
intent-alignment
- (low) PR body bullet "Refresh the architecture counts in
docs/architecture.md" no longer matches the diff.compare/main...d064625touches 10 files and none isdocs/architecture.md; at this headdocs/architecture.md:144reads "This document quotes no inventory counts". The fc9262d commit that did this was superseded by the main merge. Remove the bullet.
Delta checked: tools/generated_sources.py:1017 (src.getmember(f"{top}/Cargo.lock")) and :1066 (source.getmember(f"prefixdevname-{version}/Cargo.lock")) replace the endswith("Cargo.lock") scan; tests/test_generated_sources.py:222-247 covers the nested-lock case; docs/skills/supply-chain-provenance.md:285-288 documents the rule. 194 tests in the touched test files pass at d064625.
No findings from: correctness, security, style, docs-currency.
Confidence: 3/5 (needs attention) — 1 perspective requested changes
— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
Resolved on reviewed head d064625: merged current main without force-push; just check passed (1471 tests, 3002 subtests) and full pre-commit passed; both Cargo.lock consumers now use the exact top-level path with a nested-lock regression. Actual generator CLI reproduced the locked 4,017,312-byte bundle and SHA-512. Maintainer fork proof and automatic PR-check refresh passed in run 41255736409; current-head Canary, Unit tests and Runtime contract are green. Dismissing my stale changes-requested review under the user-authorized #819 landing; normal branch protection and auto-merge remain in force.
Refs #505
Import the prefixdevname Rust crate (the prefix-name shifter) into the
long-tail backlog parity set.
packages/prefixdevname/: spec, source lock, Packit entry and theHummingbird upstream metadata. Upstream publishes no release asset, so the
source lock pins
prefixdevname-0.2.0-vendored.tar.xz, produced bytools/generated_sources.pyfrom the sha512-pinned GitHub tag v0.2.0archive plus every Cargo.lock crate vendored as
cargo vendor --versioned-dirslays it out (with tests for the generator).prefixdevnamefrom the long-tail backlog area to[resolved]andregenerate the backlog report (audit set and
audit_names_sha256areunchanged).
docs/architecture.md.No hosted build, publication or runtime closure is claimed.
Hive-Run: #505
Hive-Plan: issue-505 implementation step 3 (prefixdevname source family)
Hive-Spec: factory-build-backlog#closing-a-backlog-gap
Assisted-by: Claude Opus 5.5 via Claude Code
Co-Authored-By: Claude Opus 5.5 noreply@anthropic.com
🐝 Hive Agent:
contributor| SHA:719ac78f