Settings -> App -> "Install from USB stick" (car build only). It looks
in the top folder of every mounted volume (found from the app's own
folder on it, since the head unit has no file picker) for
NilsMG4-unstable-<version>.apk, takes the newest one newer than the
installed version, checks it against its .sha256 sidecar and the
package name, and installs it with the same PackageInstaller session
as the over-the-air update. The stable build gets a stub and no button.
Tested on the emulator: 2.18.0 installed a 2.18.1 from the storage's
top folder without a confirmation, as the system user, so the OTA
install step should not need one either.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QLJfZQZeAKAwSqViTmLzki
The car updates itself again, from this fork's own releases, while the
repository stays private:
- publish-car (flake) creates the pre-release car-<version> on the
forge and attaches the APK and its .sha256. It needs a forge token
that may write releases, from ~/.config/nilsmg4/forge-token or
$FORGE_TOKEN; it is sent in a header and never printed.
- The server relays releases and attachments (/v1/app/releases,
/v1/app/assets/...), read with a read-only FORGE_TOKEN from its
Secret, GitHub-shaped with download URLs back at itself. Redirects
away from the forge are not followed; off (404) without a token.
- The app asks its server with its own paired token, allows exactly
the server's host, downloads the APK itself instead of through the
DownloadManager (so the token stays in the app), verifies SHA-256 and
package name, and tries a PackageInstaller session (it may install
silently as the system user) before falling back to Downloads.
OTA.md is rewritten for this flow; the k8s and server READMEs explain
the new token.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QLJfZQZeAKAwSqViTmLzki
mise.toml, the comments in app/build.gradle (corrected where they no
longer described the build: the car build is signed by build-car, and
the updater reads the version from build-car's file name), OTA.md's
message table (now the updater's current English messages), the Turkish
messages quoted in CHANGELOG.md and OTA.md (glossed in English), two
test assertion messages, an icon comment, and the ignition-state comment
in EVHardware.kt. The vendored EVHardware library still has about 120
half-French comment lines; noted in SPECS.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QLJfZQZeAKAwSqViTmLzki
OTA.md, mise.toml and docs/MG4_PLATFORM_INSTALL.md now use
com.stinnesbeck.nilsmg4(.unstable) and NilsMG4-*.apk. OTA.md says the
updater is switched off and that its publishing steps are upstream's
GitHub process; it also drops the old four-part version example.
CHANGELOG.md keeps the old ID as upstream history.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QLJfZQZeAKAwSqViTmLzki
Two byte-identical copies of the platform key live on this machine, under
LUMINA/DriveHub_Kamera/tools and MG/DriveHub_Kamera/tools. OTA.md named the
LUMINA one; the MG one sits inside this workspace and is the copy actually
reached for.
Both work today only because someone keeps them in sync by hand. A private
signing key duplicated across two directories, with a checked-in document
pointing at one of them, is worth collapsing to a single location.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The refresh button found 2.2.3, downloaded it, and reported
"Kurulum reddedildi (imza uyuşmuyor?)" for a build signed with exactly the same
platform certificate as the one already running. The signature was never the
problem.
install() shelled out to `pm install -r` with a path inside the app's private
cache. pm does not read the file — it hands the path to the package manager
service, which opens it from its own process and SELinux context, where that
directory is not readable. The install failed for a reason nobody could see:
install() returned a bare boolean, and UpdateHook mapped every false to the one
message it had. The pm output that would have named the real error was captured
into a StringBuilder and discarded.
A PackageInstaller session takes no path. We open the archive ourselves, as the
app, and stream it into the session, so no second process ever has to reach our
cache. Approach taken from merthankaraman/DriveHub_Dort, which hit the same wall
on the same head unit.
Consequences worth knowing when reading a future failure:
- Committing a session is asynchronous, so the refresh button can only say an
install started. OtaInstallResultReceiver has the verdict, and surfaces the
package manager's own INSTALL_FAILED_* text rather than guessing.
- STATUS_PENDING_USER_ACTION is not an error. The receiver launches the
confirmation activity the platform supplies.
- The PendingIntent must be MUTABLE or it arrives with no status extras.
install() now returns an enum, so a certificate mismatch, an unparseable or
wrong-package archive, and a failed session each say what they are.
Also checks the archive's package name against ours, which was a genuine gap:
the platform key signs every app on this unit, so a certificate match alone
never meant the APK was this app.
The `pm install -r` exit-code test goes with the shell-out it covered. Nothing
in the session path is exercisable off-device.
REQUEST_INSTALL_PACKAGES and the receiver are declared in the unstable manifest
only — the stable channel has no updater at all, and its merged manifest was
checked to confirm it gains neither.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The signing key and the pre-release tick are the two things that silently
make an update invisible, and neither is guessable from the code.
Co-Authored-By: Claude <noreply@anthropic.com>
nilsandClaude Opus 5.5