Repository navigation
Conversation
Dakota ships fish from GNOME OS and no zsh. fish leaves [unavailable] (the factory builds it; it was a deliberate exclusion, #419) so base.toml's entry installs, and zsh leaves the base manifest. Common's /etc/zsh files stay, unused, exactly as on Dakota. The repo gate keeps its deliberate-exclusion mechanism with an empty set; its tests use a stand-in. Closes #1230 Assisted-by: Claude Opus via GitHub Copilot Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: danathar, hanthor The full list of commands accepted by this bot can be found here. Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
Withdrawn: this PR is not back in your court after all. The failing checks ( |
Danathar
left a comment
There was a problem hiding this comment.
Stand-in review (the hive's reviewers are out of quota). Approving. fish ships and zsh goes, matching Dakota and the #1230 owner decision. No zsh use remains in system_files, ujust recipes, profile.d, contracts or scripts at head. Common's /etc/zsh/* files stay but only zsh reads them, and Dakota ships the same files. Dakota ships /usr/bin/fish and no zsh. In the testsuite only common_shell.feature mentions zsh, and both scenarios are @requires_brew; the latest Utah e2e run skipped them ("Homebrew not present"). Small follow-ups: docs/bluefin-package-gaps.md (lines 29, 93, 97, 106) still says fish is blocked and zsh wanted. Those two zsh scenarios would fail on Utah if CI ever gets Homebrew, so they're worth scoping to Bluefin-classic. baselines/triage.toml still lists fish as unavailable. Users with a zsh login shell lose it, which is worth one line in the release text. Merge note: this conflicts with the #1287–#1289 chain (README, contract, packages/utah.toml, presets, counts). Both lower "genuinely unavailable" 7→6, so the second to land needs a rebase and a count regeneration (expected 5).
|
Folded into the part-2 landing PR (fold/dakota-sq) with this change unchanged apart from rebase and conflict resolution. Closing here; review continues there. |
Stacked on #1286, which is stacked on #1284.
What
fishleavespackages/utah.toml[unavailable], along with its deliberate-exclusion block (Standing: fish is deliberately excluded from the Utah image (D3) #419).packages/base.tomlalready lists it, so it installs. It is published in the pinned factory image: I checked thesha256:523507…repodata.zshis removed frompackages/base.toml. It was not in the Hummingbird base (rpm -q zshsays not installed), so dropping it from the manifest drops it from the image, along with the zsh RPM's/etc/zshrc,/etc/zshenvand/etc/skel/.zshrc./etc/zsh/{zlogin,zprofile,zshenv,zshrc}stay. Dakota ships the same files byte for byte without zsh installed (same hashes in the Dakota and Utah image scans), so they are inert.scripts/check-repo-availability.py:DELIBERATELY_EXCLUDEDis now empty. The gate tests patch in a stand-in, so the "deliberate exclusion stops resolving / leaves the overlay" paths are still exercised.docs/bluefin-package-gaps.mdD3 is marked superseded by parity(dakota): shells — drop zsh, revisit fish #1230 (bash stays the default shell). Site data and doc counts were regenerated (6 genuinely unavailable).Verification
pytest tests/test_repo_availability_gate.py tests/test_package_resolution.py tests/test_check_doc_counts.py tests/test_site.py: 103 passed.just check: not completed locally. Itsgit submodule updatehung for 10+ minutes cloningtilingshellfrom GitHub on this host, so I killed it. CI runs the full gate.Closes #1230