Repository navigation
Merge upstream quattro (kernel menu, fingerprint enrollment, first-run update) - #90
Merged
Merged
Conversation
The rerun after an interrupted change treated any one slot added since as the change's new key. If the change stopped before luksChangeKey and a recovery key was enrolled before the rerun, entering the old password removed the recovery key, and entering the recovery key removed the old password's slot. luksChangeKey picks the new slot itself, so the journal can't name it, and LUKS2 leaves both keys between its two header updates. With both slots on the disk the rerun now asks for the new password, and the old one when it has only the new, and removes a slot only once the new password opens the added slot and the old password the old one. Otherwise it removes nothing and says how to settle the disk by hand. Found by Greptile on omacom#14675. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…dvice The test split its answers on spaces, and the new password has one, so two of its three cases stopped before the confirm step and still passed on the first case's output. Each case now keeps its answers whole, starts from empty output and prompts, checks it reached the confirm step and says its own message. When the new password opens the added slot but the old one doesn't open the old slot, that may be a typo or a slot rekeyed since, so the message now asks to check the old password before removing the slot. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A failed enrollment on a Mac's Touch ID showed how setup's output misled: it told everyone to keep moving the finger around the sensor, which suits a swipe sensor, and printed fprintd's raw "Enroll result: enroll-stage-passed" lines. A press sensor needs a separate touch for each sample, fprintd sends one stage after its duplicate check before enrollment starts, and one stage can take several touches, so after a single touch enrollment looked under way and timed out a minute later. Setup now asks fprintd for the reader's scan type and, for a press sensor, says to touch, lift and touch again until it says Enrolled. It turns each result into an instruction: the duplicate check's stage prints nothing, each later stage is a numbered step followed by what to do next, retries and failures say what to do, and verify reads the same way, " (done)" and " (not done)" suffixes included. fprintd's routine lines are hidden, but anything else it prints, such as why it could not claim the device, is shown, and fprintd's exit status still decides each step. sudo is authenticated first, and named, because with fingerprint sudo set up its own touch read as the first sample; if it fails, setup says why and stops. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The first enroll-stage-passed was always hidden, on the assumption that
it is fprintd's duplicate check. fprintd runs that check only on readers
that identify, so on any other reader the first real stage was hidden
and every later step was numbered one low.
fprintd's D-Bus interface doesn't say whether a reader identifies, but
it gives the reader's name. Touch ID ("Apple secure enclave fingerprint
sensor") identifies, and its check needs no touch while nothing is
enrolled, so its first stage stays hidden. Every other reader's first
stage counts: on one that doesn't identify it is the first enrollment
stage, and a duplicate check that took a touch is shown as step 1.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A failed verify printed its message and fell off the end with status 0; it now exits 1 like a failed enrollment. The test now covers a failed verify after a successful enrollment: setup fails and leaves PAM and the lock screen alone. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Co-authored-by: Mark Groves <mark-groves@users.noreply.github.com>
When busctl couldn't read the reader's scan type, setup took the swipe branch and told a press reader to swipe its finger across the sensor to enroll and verify. A reader fprintd can't describe is now asked to touch or swipe. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
verify-disconnected and every other finished failure read the same "The check stopped", hiding what fprintd-verify reported. A disconnect now says so, as enrollment does, and any other failure names fprintd's result. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The expected strings spelled ✓ and ✗ as $'\u2713' and $'\u2717', which bash only turns into those characters under a UTF-8 locale. Under C or POSIX the enrollment check failed even though setup printed the right lines. The fixtures now hold the characters themselves. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The remove-and-retry line added the step's own "again" prompt after "then:", so a press reader read "Lift your finger off the sensor, then: Keep lifting your finger and touching the sensor." It now says "Lift your finger off the sensor, then try again." on every reader: libfprint sends this retry from swipe drivers too (elanspi, upekts, vfs101), so the line names no gesture. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* Switch the default kernel from Update > Kernel When more than one kernel is installed, Update > Kernel lists them with the boot default checked. Picking one runs omarchy-kernel set, which authenticates once through pkexec, puts the kernel first in BOOT_ORDER in /etc/default/limine, and rebuilds only that kernel's entry to reorder the boot menu. omarchy-kernel and omarchy-kernel list report the default and the installed kernels. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Only notify once the new kernel is set Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Restore the boot order when a kernel switch fails, widen the Kernel menu A rebuild that fails or never reaches the boot menu now puts the previous BOOT_ORDER back, so neither omarchy-kernel get nor the next kernel update acts on a rejected choice. A wildcard that ties several kernels no longer picks one alphabetically: their limine.conf order decides and only root can read it, so get reports the tie instead. Update > Kernel gets the wide layout Fonts uses so long kernel names fit. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Refuse kernel switches under Direct Boot, read commented BOOT_ORDER Direct Boot has the firmware load one UKI without Limine, so a reordered boot menu would change nothing; say how to turn it off instead. A BOOT_ORDER assignment with a trailing comment now counts, as it does for limine-entry-tool. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Refuse Direct Boot for root too, put the menu back after a late failure The Direct Boot check reads efibootmgr itself so sudo omarchy-kernel set is refused like the menu path. A rebuild that reorders the boot menu and then fails in a post hook now rebuilds the previous default under the restored order, so the menu and BOOT_ORDER agree again. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Title the kernel notification New default Kernel, with the kernel as body Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Use the Linux penguin for kernel notifications Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Kernel notification says Now using, and reboots when clicked Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Keep the kernel notification up for 30 seconds Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Lock kernel switches, stage the config write, roll back to the menu leader Overlapping switches from the detached menu action could let one that failed restore its backup over another that had succeeded, so a switch now holds a lock from reading the old state through rollback. The new BOOT_ORDER is staged beside /etc/default/limine and renamed over it, so a failed write cannot delete the saved order. Rollback rebuilds the kernel that led the boot menu before the switch, which root reads from the menu itself, so a wildcard tie no longer leaves nothing to restore. The kernel command gets its GROUP_DESCRIPTIONS entry. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Style the kernel failure notification like the success one, retry on click Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * Simplify omarchy-kernel Merge the BOOT_ORDER reader into one function, write the config in place inside the rollback instead of staging a separate file, and have get name a kernel only when the first match is an exact name rather than counting wildcard ties. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: David Heinemeier Hansson <david@hej.dk>
Only update the disabled field when enabling an output from the Display panel. Let the existing named monitor rule retain its mode, position, and scale, and cover command generation, output escaping, action launch, and the last-display guard. Co-authored-by: ZihaoMu <16487352+zihaomu@users.noreply.github.com>
…#16143) Notify workspace modes after float patterns or tag relationships change, before the first matching window opens. Clear the hook when helpers reload and cover late rules, tag ordering, multiple workspaces, and callback lifetime. Co-authored-by: ZihaoMu <16487352+zihaomu@users.noreply.github.com>
Co-authored-by: ZihaoMu <16487352+zihaomu@users.noreply.github.com>
…om#16046) Co-authored-by: ZihaoMu <16487352+zihaomu@users.noreply.github.com>
…acom#16211) * Offer the first-run update only when there is something to update First run sends a critical "Update System" toast as soon as the network is up, whatever the machine has pending. On a fresh 4.5.0rc1 install checkupdates reports nothing, the bar's update icon stays hidden because omarchy-update-available says Omarchy is up to date, and the toast leads to an update with nothing to do. Send it only when omarchy-update-available finds an update (Omarchy itself or a development checkout, the bar's signal) or checkupdates lists any package. checkupdates exits 2 when every package is current and 1 when it could not tell; only a definite "nothing" holds the toast back, so a failed check prompts as before. * Check for pending updates with one checkupdates run omarchy-update-available runs checkupdates to find Omarchy's package, and the first-run check then ran it again for the rest. Each run syncs the package databases over the network, so an install with nothing to update paid for that twice before skipping the toast. checkupdates alone covers every repository package, Omarchy's included. omarchy-update-available is still asked when OMARCHY_PATH is a development checkout, the one thing checkupdates cannot see. The test's stand-in for it now runs checkupdates as the real one does, and counts the runs. * Quote the path literal and spell out the update check The repository style quotes string literals in [[ ]] comparisons and prefers a full if/else over an early return that skips what follows. * Read OMARCHY_PATH as the session sets it AGENTS.md: runtime code relies on $OMARCHY_PATH and does not default it.
|
Important Review skippedBot user detected. To trigger a single review, invoke the ⚙️ Run configuration
You can disable this status message by setting the Use the checkbox below for a quick retry:
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
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
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.
Upstream
quattromoved fromce39c5e9(already on the fork via #89) to7075b61c(2026-10-21 21:47 UTC). 8 first-parent commits, 16 of them non-merges, cherry-picked onto forkquattro. The 2 merge commits (omacom#14874 and omacom#15512) were not replayed; their changes are in the commits they merged. Cursor stays on the official tarball, not a mise tool. Face, Grok Bot, and Cursor menu entries stay. The power-profile menu provider stays out. Fingerprint setup still goes throughomarchy-pam-pair-addandomarchy-apply-polkit-pam. Factory reset still scrubs inherited USBGuard and Thunderbolt trust.A resumed drive password change confirms the new slot (omacom#14874)
An interrupted
luksChangeKeycan leave both the old password and the new one on the disk, and LUKS2 does not tell the journal which slot is the new one. The rerun used to treat any slot added since as the change's new key, so a recovery key enrolled in between could be removed, or the old password's slot could be removed if the recovery key was typed. With both keys present, the rerun now asks for the new password, and for the old one when only the new key was added. It removes a slot only after the new password opens the added slot and the old password opens the old one. Otherwise it removes nothing. When the new password opens the added slot but the old password does not open the old slot, the message asks to check the old password before removing it.Fingerprint enrollment is guided (omacom#15512)
Setup asks fprintd for the reader's scan type. A press sensor is told to touch, lift, and touch again until it says Enrolled. A swipe sensor keeps the moving-finger instruction. When fprintd cannot say which, the prompts name neither. Each enroll and verify result becomes an instruction, and fprintd's routine lines stay hidden. Touch ID's first stage is its duplicate check and stays hidden; every other reader's first stage counts. The Touch ID name is the driver's
full_namefrom the aurora patch in omarchy-pkgs. sudo is authenticated, and named, before the first sample, so a fingerprint sudo touch is not taken as enrollment. A failed verify now exits 1, the same as a failed enrollment, and leaves PAM and the lock screen alone. A disconnected reader is named, and any other finished failure names fprintd's result. "Lift your finger off the sensor, then try again." is one line on every reader.Update > Kernel switches the default boot kernel (omacom#15966)
When more than one kernel is installed, Update > Kernel lists them with the boot default checked. Picking one runs
omarchy-kernel set, which authenticates once through pkexec, puts that kernel first inBOOT_ORDERin/etc/default/limine, and rebuilds that kernel's entry. A failed rebuild restores the previous order and the kernel that led the boot menu. Direct Boot refuses the switch, including for root, and says to turn it off under Setup > Direct Boot. A wildcard that ties several kernels is reported rather than guessed. The row uses the wide layout Fonts uses. In the terminal:omarchy kernel,omarchy kernel list, andomarchy kernel set <kernel>. The newkernelsmenu provider is in; the power-profile provider stays out.Re-enabling a display keeps its layout (omacom#16147)
Turning a display back on from the Display panel only clears
disabled. The existing named monitor rule keeps its mode, position, and scale instead of being reset to preferred, auto, and auto.Late personal float rules reach floating workspaces (omacom#16143)
Personal float and tag rules can load after saved workspace modes. Helpers now tell workspace modes when those patterns change, so the floating-workspace exclusions are current before the first matching window opens. Reloading helpers drops the previous config's hook.
Translation catalogs load before the first lookup (omacom#16059)
Shell and plugin catalogs are read through
FileViewbindings withblockLoading, so the firstt()sees the catalogs. Language, number format, and direction come from those same bindings. A key the catalogs do not carry still uses the caller's English fallback.Lutris and Battle.net install GPU libraries first (omacom#16046)
omarchy-install-gaming-gpu-lib32now runs beforeumu-launcherfor both Lutris and Battle.net. That satisfiesumu-launcher'slib32-vulkan-driverdependency before--noconfirmcan select an NVIDIA provider on an AMD or Intel machine.First run offers an update only when there is one (omacom#16211)
The critical "Update System" toast used to fire as soon as the network came up. It now waits until
checkupdateslists a package, or until a development checkout is behind its upstream.checkupdatesexits 2 when every package is current and 1 when it could not tell, so only a definite "nothing" holds the toast back.checkupdatesruns once;omarchy-update-availableis asked only about a dev checkout, whichcheckupdatescannot see.