Skip to content

feat(cli): bluefin-cli runs the shell experience, uwelcome stays the greeting - #1370

Merged
hanthor merged 1 commit into
mainfrom
feat/bluefin-cli
Nov 20, 2026
Merged

hanthor merged 1 commit into
mainfrom
feat/bluefin-cli

Conversation

@hanthor

@hanthor hanthor commented Nov 20, 2026 •

Copy link
Copy Markdown
Member

bluefin-cli takes over Utah's shell experience (shell tools, prompt hooks and the command-line Homebrew bundles) from Common's ublue-bling and bling.sh/bling.fish. uwelcome stays the terminal greeting: bluefin-cli integrates it (tuna-os/bluefin-cli#527). Utah is the reference for projectbluefin/common#1719.

What changes

  • ujust bluefin-cli (60-custom.just) overrides Dakota's recipe (ublue-bling + cli.Brewfile): it runs bluefin-cli shell <shell> on and bluefin-cli install cli, or off when the rc file already has bluefin-cli's marker line. With no binary it says so instead of command not found.
  • bling.sh, bling.fish and ublue-bling become shims that run bluefin-cli, so old rc lines and muscle memory keep working. The desktop contract pins them.
  • configure-services.sh removes env.sh and bash-preexec-rearm.sh, which only Common's bling read; bluefin-cli keeps bash-preexec armed itself.
  • Docs: desktop-contract/references/bluefin-cli.md (old → new mapping), package-contract/references/ujust.md, README counts.

What stays

Package
bluefin-cli is in [unavailable] with projectbluefin/utah-packages#1235 (the factory recipe) as its tracker: until the pinned factory carries it, the shims do nothing and ujust bluefin-cli says it is missing. It moves to [parity] and [factory].parity once a factory pin bump with it lands. Unavailable count 9 → 10.

Tests

@mergeraptor
mergeraptor Bot requested review from bketelsen and daegalus November 20, 2026 06:13
@mergeraptor mergeraptor Bot added the approved Indicates a PR has been approved by an approver from all required OWNERS files. label Nov 20, 2026
@mergeraptor

mergeraptor Bot commented Nov 20, 2026 •

Copy link
Copy Markdown
Contributor

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: 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 /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@hivecommons-hive hivecommons-hive Bot added author/clanker A contributor's own hive authored this PR under their login no-dco A commit is missing Signed-off-by — run `git rebase --signoff` role/trusted Author is trusted for unattended merge (trusted-author role or registry/claim tier in require_tiers) stage/050-review-pending Review swarm requested; awaiting a verdict for the current head labels Nov 20, 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.

Summary: This change moves Utah's terminal setup (shell shortcuts, prompt and command-line tool bundles) to a newer tool called bluefin-cli. The old setup commands keep working, and they now call bluefin-cli for you. The standard welcome message when a terminal opens stays as it is. I found no blocking problems. One small safeguard is suggested below: it stops a possible endless loop if the old setup command ever comes back.


correctness (low, advisory)

Utah still ships the vendored Dakota recipe that calls the old command (system_files/shared/usr/share/ublue-os/just/system.just:16-18, bluefin-cli: → @ublue-bling). The new ublue-bling shim runs exec ujust bluefin-cli (system_files/shared/usr/bin/ublue-bling:4). Today the override in 60-custom.just:112-137 shadows the vendored recipe, and the just ≥ 1.56 floor in tests/test_ujust_overrides.py:19-23 keeps that true. If the override ever stops winning again, the two would call each other forever. #449 shows this has happened before. See the inline suggestion for a one-line guard.

What I checked: the recipe's rc paths ($HOME/.bashrc, $HOME/.zshrc, $HOME/.config/fish/config.fish) match bluefin-cli's own (internal/shell/shells.go:55-69). The marker it greps is bluefin-cli's shellMaker (internal/shell/shell.go:32). bluefin-cli shell … on|off removes the lines the old ublue-bling wrote (internal/shell/shell.go:80-89), so old rc files migrate cleanly. bluefin-cli's init script has its own load-once guard (internal/shell/resources/shell.sh:12), so the shim plus a new rc line cannot initialise twice. The Containerfile:233-238 copy order puts Utah's shims over Common's files. The contract run at this head passed. I did not boot an image, and bluefin-cli is still listed as unavailable, so the shims' runtime path is unverified.

No findings from: security, intent-alignment, style, docs-currency.

Confidence: 5/5 (safe)

-- hive: agent=reviewer backend=claude model=claude-opus-5-5 effort=medium claude=2.1.327

#!/usr/bin/bash
# Compatibility shim: on Utah, bluefin-cli replaces Universal Blue's
# ublue-bling. The old command still turns the terminal experience on or off.
exec ujust bluefin-cli

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.

correctness (low): The vendored system.just:16-18 recipe bluefin-cli still calls ublue-bling. If the 60-custom.just override ever stops shadowing it (the precedence regression in #449), this shim and that recipe would call each other forever. A guard turns that into a clear error:

Suggested change
exec ujust bluefin-cli
if [[ -n "${UTAH_BLING_SHIM:-}" ]]; then
echo 'error: ujust bluefin-cli resolved to the old ublue-bling recipe' >&2
exit 1
fi
UTAH_BLING_SHIM=1 exec ujust bluefin-cli

@hivecommons-hive

Copy link
Copy Markdown
Contributor

Hive status — updated in place; edits do not notify. Full history is in the Activity Monitor.

Review findings from reviewer-2 (folded here because this head already carries its hive review):

Looks correct to me, with one consequence worth confirming before merge.

What I verified against the tree at ba34a58:

  • The override resolves. Utah's vendored system.just:16 still defines bluefin-cli: ublue-bling; 00-entry.just imports 00-common.just then 60-custom.just, and tests/test_ujust_overrides.py:123-134 builds that same layout with allow-duplicate-recipes, so the new recipe tests would fail if Dakota's recipe won.
  • The shims land over Common's files: Containerfile:110-114 stages Common's system_files/shared before Utah's /tmp/utah-local, which is copied last.
  • The removed helpers have no other reader. In projectbluefin/common, bling/env.sh is read only by usr/bin/ublue-bling:39 and bash-preexec-rearm.sh only by bling/bling.sh:75. Both are replaced here, and no Utah contract or test pins either file.
  • bling.sh detects the shell with the same readlink /proc/$$/exe Common's bling.sh uses, so bash and zsh behave as before.

Interim regression (declared in the body, stating the concrete effect): bluefin-cli is in [unavailable] and its factory recipe, utah-packages#1235, is still open with changes requested. From the image that carries this PR until a factory pin with bluefin-cli lands:

  • Anyone whose ~/.bashrc/~/.zshrc sources bling.sh loses bling's aliases, starship, zoxide, direnv and mise hooks on update, with no message (bling.sh:13 is a silent no-op).
  • ujust bluefin-cli and ublue-bling, which work today, exit 1 with "bluefin-cli is not installed".

The brew install tuna-os/tap/bluefin-cli workaround in packages/utah.toml restores both, but nothing in the shell tells users about it. If that window is acceptable, this is ready. If not, holding the merge until utah-packages#1235 publishes and the pin moves avoids it with no code change.

Minor, not blocking: Common's bling.sh had a BLING_SOURCED guard against double sourcing. The shim drops it, so an rc setup that sources it twice runs bluefin-cli init twice. Whether that matters depends on bluefin-cli's init being idempotent, which I did not verify.

@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/050-review-pending Review swarm requested; awaiting a verdict for the current head stage/060-reviewed Review swarm verdict approve; a gate is still pending (CI running, soak not started) labels Nov 20, 2026
@hivecommons-hive

hivecommons-hive Bot commented Nov 20, 2026 •

Copy link
Copy Markdown
Contributor

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

Review asks:

  • the branch conflicts with the base and needs a rebase

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 (8 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 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/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act stage/050-review-pending Review swarm requested; awaiting a verdict for the current head stage/060-reviewed Review swarm verdict approve; a gate is still pending (CI running, soak not started) labels Nov 20, 2026
@hivecommons-hive hivecommons-hive Bot added stage/050-review-pending Review swarm requested; awaiting a verdict for the current head stage/060-reviewed Review swarm verdict approve; a gate is still pending (CI running, soak not started) and removed stage/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act stage/050-review-pending Review swarm requested; awaiting a verdict for the current head labels Nov 20, 2026
@hivecommons-hive hivecommons-hive Bot added 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 and removed stage/060-reviewed Review swarm verdict approve; a gate is still pending (CI running, soak not started) stage/050-review-pending Review swarm requested; awaiting a verdict for the current head labels Nov 20, 2026
…greeting

bluefin-cli (tuna-os/bluefin-cli) replaces Common's ublue-bling and
bling.sh/bling.fish (projectbluefin/common#1719):

- `ujust bluefin-cli` (60-custom.just) turns bluefin-cli's shell experience
  on, installing the cli bundle, or off when the rc file carries
  bluefin-cli's line. It explains a missing binary.
- bling.sh, bling.fish and ublue-bling become shims, so rc files and muscle
  memory that still call bling get bluefin-cli. The contract pins them.
- configure-services.sh removes env.sh and bash-preexec-rearm.sh, which only
  Common's bling read; bluefin-cli keeps bash-preexec armed itself.

uwelcome stays the terminal greeting, with `ujust toggle-user-motd`, Utah's
interactive-shell guard and Dakota's bluefin-shell-defaults.sh unchanged.
bluefin-cli works through uwelcome where it is installed
(tuna-os/bluefin-cli#527).

bluefin-cli is [unavailable] until the factory publishes it
(projectbluefin/utah-packages#1235); the shims then do nothing.

Assisted-by: Claude Opus via GitHub Copilot
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
@hivecommons-hive hivecommons-hive Bot added stage/050-review-pending Review swarm requested; awaiting a verdict for the current head stage/060-reviewed Review swarm verdict approve; a gate is still pending (CI running, soak not started) stage/095-pre-merge-soak Lane admitted; the pre-merge soak window is running before the merge is attempted and removed stage/072-returned-to-contributor Contributor PR: CI red, changes requested or conflicting; diagnosis posted, author must act stage/050-review-pending Review swarm requested; awaiting a verdict for the current head stage/060-reviewed Review swarm verdict approve; a gate is still pending (CI running, soak not started) labels Nov 20, 2026
@hanthor
hanthor merged commit f72c1d4 into main Nov 20, 2026
44 checks passed
@hanthor
hanthor deleted the feat/bluefin-cli branch November 20, 2026 09:31
hanthor added a commit that referenced this pull request Nov 20, 2026
Dakota ships open-iscsi and open-isns, so Utah installs
iscsi-initiator-utils and isns-utils from the factory, with every unit
off. 85-utah-desktop.preset already carries Dakota's 50-iscsi.preset.
Fedora's package adds iscsi-starter.service and iscsi-onboot.service,
and Hummingbird's 90-default.preset enables both, so the preset now
disables them as well. The contract's disabled list asserts all nine
units. clean-stage strips /var, so tmpfiles.d/utah-iscsi.conf recreates
the node database directories with the RPMs' modes and policy labels.

nicstat is now published by the pinned factory and installed.

umotd is waived: uwelcome is Utah's terminal greeting, and bluefin-cli
works through it (#1370). erofs-utils stays planned, blocked on libqpl
(projectbluefin/utah-packages#1238).

Closes #1360.

Assisted-by: Claude Opus via GitHub Copilot
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

approved Indicates a PR has been approved by an approver from all required OWNERS files. author/clanker A contributor's own hive authored this PR under their login no-dco A commit is missing Signed-off-by — run `git rebase --signoff` role/trusted Author is trusted for unattended merge (trusted-author role or registry/claim tier in require_tiers) stage/095-pre-merge-soak Lane admitted; the pre-merge soak window is running before the merge is attempted

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant