Incremental advances collide with same-version artifacts that reached the
destination through another lineage — the stable -> rc bootstrap seed and
native rc builds are not byte-identical to edge's builds of the same
version. A collision means the package has not moved in the source since,
so keep the destination's published copy and continue; the hard failure
also aborted before the db rebuild and sync, leaving the advance half-done.
Also stop advancing pinned packages into rc: eligibility deferred to
package_builds_for_mirror, whose OMARCHY_RC_PINS gate is never set during
an advance, so edge's omarchy/omarchy-settings looked movable — and could
have overtaken an in-flight RC pin under a fresh filename.
1.7.0 makes OpenSSL and libzstd hard CMake requirements; both are in
every build chroot transitively, so the sync's version-only bump in
auto/sync-upstream would have built and shipped them as undeclared
linkage rather than failing. Declare them, and pin the new optional
Wayland idle-inhibit feature on by adding wayland-protocols to
makedepends instead of leaving it to what the chroot happens to carry.
Checksum matches the release's SHA256SUMS manifest and an independent
download.
Released 06:46 UTC today, inside the quarantine window; shipped by hand
per BYPASS_MIN_RELEASE_AGE's intent. The feed's asset name matches the
one the PKGBUILD builds, the download's size and SHA-512 match the
electron-builder feed exactly (byte-identical to what the app's own
updater installs), and the SHA-256 here is from that verified download.
0.11.2 is 14 hours old, inside the package's 24h min_release_age, so the
upstream sync would not pick it up until tomorrow; ship it now by hand.
Tarball commit b000b79 verified against the v0.11.2 tag, checksum from
an independent download. The new Cargo dependencies (ashpd, zbus,
gdk4-wayland) are vendored crates needing no new system libraries.
0.11.2 also ships FileManager1/portal integration files. Stage the
FileManager1 service under /usr/share/strata the way upstream's own
PKGBUILDs do — the app copies it per-user on opt-in. The portal files
stay unpackaged, matching upstream: enablement is per-user and
consent-gated in the app.
The package carries a deliberate commit pin (59d557a) so every
architecture packages the same tree. The AUR sync kept trying to revert
that to the unpinned branch-tip recipe, recording a version downgrade in
the process. Mark it local so exactly one source owns the recipe; it is
Omarchy-maintained until it moves to an upstream release feed.
makepkg always exports CARCH, so PKGBUILDs may branch on it at file
scope. Every place the tooling sourced a PKGBUILD did so without CARCH,
taking the wrong branch or aborting partway, and check-versions turned
the resulting empty pkgver/pkgrel into the version '-', which never
matches a published version and queues an endless rebuild.
package_pkgbuild_var reads one variable the way makepkg would see it,
for the architecture being checked, and reports whether the source
succeeded. check-versions, sync-rebuilds and omarchy-pkgs use it.
check-versions now warns and skips a package whose pkgver or pkgrel is
empty instead of comparing a partial version. A self-test covers a
PKGBUILD that branches on CARCH before assigning its version.
The file-scope case on CARCH ended in 'return 1', which aborted any
'source PKGBUILD' that did not export CARCH before pkgver and pkgrel
were assigned. check-versions reads PKGBUILDs that way, saw an empty
version, queued 8.12.34-35 on every tick, and promotion then refused to
overwrite the already-published artifact. Every channel has been in
backoff since 06:41 UTC today because of it.
The environment-theme patches shipped as a forked AppImage while
upstream lacked them. Keeping it current means a hand-built artifact
per release, and it has fallen three behind — 0.0.35 against 0.0.38.
t3code-bin tracks the upstream feed on its own, so drop the fork.
Nothing else in the repo referenced the package.
The AUR caught up: asusctl 6.4.0-1 (b0ec6ca) repoints source at the
GitHub repo upstream, which is what our patch existed to do. The new
PKGBUILD uses a release tarball instead of git+, so the patch context
no longer matches and every scheduled Sync AUR Packages run failed
applying it.
Remove the patch and resync from the AUR. With no .omarchy/patches
left, the package is no longer customized, so it tracks 6.4.0-1
without the .1 pkgrel suffix.
Package Link Studio 1.0.2 with its AUR-only MediaPipe and sounddevice dependencies. Wire Link Studio to GitHub release checksums, keep all three packages on the fast ring, and make MediaPipe's Bazel bootstrap a checksummed makepkg source.
Default Electron to the gnome-libsecret password store so Hermes can use GNOME Keyring for secure remote tokens. Declare libsecret as a runtime dependency.
cmd_rc handed maybe_iso the bare train version, so omarchy-iso-release
named every candidate omarchy-X.Y.Z-rc.iso — rc2 uploaded over rc1's
URL. Pass the pinned rcN version instead; paired with omarchy-iso's
f9be60b, the artifact becomes omarchy-4.0.2rc2.iso. The ship path
keeps passing the final version.
Edge is the channel the 4.0.2 RC ISOs point pacman at until the rc-channel
migration lands, but edge only carried omarchy 4.0.1 — older than the
installed 4.0.2rc1, so those installs never saw an upgrade. Bumping master's
pin lets the normal edge pipeline build and publish 4.0.2rc2, whose migration
repoints [omarchy] at the rc channel.
Stable is unaffected: pinned packages never build for stable
(package_builds_for_mirror), it only receives them via the rc->stable advance
at ship time.
(cherry picked from rc-branch pin commit 11767d6)
pkgs.omarchy.org sits behind Cloudflare, and right after a sync the
plain db URL keeps serving the previous file from the edge cache
(observed: a cache HIT with age 900s+ returning the pre-release db).
That made status report the old version, wait_for_published poll a
frozen file for its full timeout, and would let RC auto-numbering
count from last release's versions. A unique query string per fetch
skips the cached entry.
package_builds_for_mirror gates pinned packages' rc builds on
OMARCHY_RC_PINS, but that check runs inside the container via
build.sh, and docker run only forwarded ARCH/MIRROR/PACKAGES —
so the rc trigger's pinned builds were always skipped as 'not
configured for direct rc builds'.
The rc trigger passes '--package omarchy omarchy-settings', and bin/build
already consumes every name up to the next --option, but bin/release only
took one — the second name fell through to 'Unknown option' and the
release exited before building anything. Parse greedily, like bin/build.
Ship the least-privilege rule with asdcontrol so authorization follows the package lifecycle. Install it after older broad rules for safe staggered upgrades.
Same upstream release, so only pkgrel moves. Built from v0.0.35-omarchy.4,
which carries the fixes from nine rounds of review on the upstream PR: the
CLI no longer overwrites an unreadable settings.json, setting a
single-appearance theme actually switches the client, reconnects no longer
flash published themes back to stock, and already-shipped clients keep
their config subscription instead of losing it to an unknown event.
`gh pr list` orders by creation date, so `--limit 30` cut the candidate set by
when PRs were opened, not when they merged. The `sort_by(.mergedAt)` that
followed only reordered whatever survived that cut. A PR opened before the
window but merged inside it — exactly the kind a release branch still needs —
never reached the already-on-branch check at all. #7649 and #7709 were both
missing from v4-0-2's list for this reason.
The window is now bounded by the branch point instead of a count: pull a wide
page and keep what merged after the merge-base's commit date, since anything
merged into the dev branch before the release branch left it is already there
by ancestry. v4-0-2 went from 11 candidates to 31.
Widening it surfaced a second gap. A change re-applied by hand carries neither
a PR number nor a cherry-pick trailer, so already_on_branch could not see it
and offered it again (#6939, applied as 33d7363c). It now also compares the PR
title against the branch's subjects, with any trailing "(#N)" stripped.
Comment-only. The .env requirement and the fact that only a PKGBUILD
version change triggers the build server both bit us once each; the
recipe now lives where the next rebuild starts.
The r2 artifact was built without the public Connect configuration official
artifacts carry (relay URL, Clerk publishable key, CLI OAuth client id), so
Connect prompted for login with no way to log in. Same code, rebuilt with
the configuration recovered from the published npm t3 package.
Rebuilt from v0.0.35-omarchy.2 of the environment-theme branch: the stream
event is capability-gated so old clients keep their config subscription,
republish dedupe is structural, the connect stream is gap-free, default
adoption is scoped per environment and keyed on the set generation, and
reserved theme ids cannot be published over. Joins the fast ring like
t3code-bin so patch rebuilds reach machines quickly.
Built from the environment-theme branch of ryanrhughes/t3code (tag
v0.0.35-omarchy.1) until upstream ships it: the app follows themes the
machine publishes into its state directory -- Omarchy retints it live --
and t3 theme set works. Same packaging as t3code-bin, conflicts with it,
and the next upstream release supersedes it after a rebase.
The desktop bundle already contains the full server -- the same entrypoint
npm's t3 package runs -- so a launcher that starts it through the bundled
Electron as Node puts the CLI on PATH without a second runtime. Upstream
only distributes the CLI via npm, which Omarchy's install scripting cannot
assume.