wtclaude
aa055469a8
The defect Phase 14b found in `MyEventsViewModel` and flagged next door: the
notifications surface has the identical shape, and it leaks the same way.
A drawer route's view model outlives a sign-out. `navigateTopLevel` uses
`popUpTo(HOME) { saveState = true }` with `restoreState = true`, so the
`NavBackStackEntry` keeps its `ViewModelStore` and a view model that loaded only
in `init` never runs again. Signing out and back in as somebody else showed the
second account the FIRST account's inbox — titles and body text written for
another person — with no request made at all, while the badge above the list
showed the new account's real unread count, because the shell refreshes that on
every session change.
`InboxCache` was never the hole: it is keyed by (base URL, user id) and a snapshot
has never crossed an account. The hole was the in-memory state, which nothing
invalidated.
Both view models now key on the signed-in account id, so a resume revalidation
that returns the same user does not refetch. The inbox resets its state *before*
loading rather than after, because `load()` paints the cache only when there is no
`Success` on screen — otherwise the previous account's rows stay up for the whole
round trip.
The settings screen behind the inbox's gear is fixed with it, and there the stale
render is worse than disclosure: those controls are written from, so a screen
still showing the previous account's preferences would send this account's PUT
built out of them.
Walked on the emulator against a local website, before and after: two accounts
with deliberately different inboxes, signed out and in within one process. Before,
the second account saw the first's rows and the server logged no inbox fetch;
after, it logs the fetch and shows its own.
572 tests, 0 failures.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016wDDVXWMDz82WqE1i969r4
Runic Gateway — Android app
A native Android client for a Runic Gateway shard's public site + player self-service. It is
purely an API client of the website backend — it never talks to the link/ sidecar or the game
shard directly, and it ships none of the shard/sidecar wiring. It surfaces the same content and
player features as the website's browser client, minus every administrative/management console.
The authoritative design contract is docs/android/PLAN.md
in the RunicGateway/docs repo. The authoritative API reference is the committed OpenAPI spec at
website/server/swagger/swagger-output.json.
Status
M0 — repo scaffold. Gradle + Compose + Hilt skeleton with CI (lint + unit test + debug build). The functional Kotlin pass (M1–M4) and the design pass (M5) follow — see the plan's milestones (§9).
Stack
| Concern | Choice |
|---|---|
| Language / UI | Kotlin + Jetpack Compose (Material 3) |
| Navigation | Navigation-Compose, single-activity |
| HTTP | Retrofit + OkHttp, kotlinx.serialization |
| Async | Coroutines + Flow |
| DI | Hilt |
| Prefs / base URL | Jetpack DataStore |
| Tokens at rest | EncryptedSharedPreferences |
| Images | Coil |
| Min SDK | Android 10 (API 29) |
| Target / compile SDK | 35 |
Dependency and plugin versions are pinned in gradle/libs.versions.toml.
Build
Requires JDK 17 and the Android SDK (ANDROID_HOME / local.properties).
./gradlew assembleDebug # build a debug APK -> app/build/outputs/apk/debug/
./gradlew test # JVM unit tests
./gradlew lint # Android lint
./gradlew installDebug # install on a connected device/emulator
The app self-configures its server URL on first run (PLAN.md §3), so a single build works against any shard's website — there is no compiled-in API host.
CI
.gitea/workflows/pr-checks.yml gates PRs into main and edge with
./gradlew lint test assembleDebug on the org's self-hosted runner (JDK 17 + Android SDK). Debug
builds are auto-signed, so the gate needs no secrets. This pipeline is verified green end-to-end on
the runner (M0). A signed release APK attached to a Gitea release comes at M6.
edge is in the trigger deliberately: a workstream that lands its phases on a working branch
before one cutover PR into main otherwise gets no CI at all until the cutover — which is what
happened to all nine M12 phase PRs (docs/website/ENGAGEMENT.md §7.1 Q8). sonarqube.yml is
unaffected: it is a push-on-main analysis, not a PR gate.
The workflow carries a few runner-specific accommodations (each explained in comments in the file), because this self-hosted runner differs from a stock GitHub runner:
- JDK 17 is installed via
apt(notactions/setup-java) — the runner can't resolveapi.adoptium.net, while the Ubuntu mirrors are reachable. - SDK packages are installed explicitly via
sdkmanager, withset +o pipefailsoyesdying ofSIGPIPEdoesn't fail the step. gradlewischmod +x'd in the run step — the runner's checkout does not preserve the git executable bit, so./gradlewalone fails with "Permission denied".
Contributing
See CONTRIBUTING.md. AI-assisted contributions must be disclosed (org
policy): tick the PR box naming the tool and add a Co-Authored-By trailer to AI-authored commits.
Licensed GPL-3.0-or-later.