Use floret-kit's shared snack chip

The undo chip moves into the kit so Calendula's reschedule confirmation is
the same component. Re-pins the kit onto the branch that carries it.

.forgejo/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,24 @@
+# Kept enabled so anything that doesn't fit the four templates still has a way
+# in.
+blank_issues_enabled: true
+
+contact_links:
+  - name: Translate Agendula
+    url: https://weblate.dev.jeanlucmakiola.de/engage/agendula/
+    about: >-
+      Translations are managed on Weblate, not here — it owns every values-*
+      file, so a hand-edited translation gets overwritten on the next sync.
+      No coding needed: pick or request a language and translate in the browser.
+
+  - name: Contributing guide
+    url: https://codeberg.org/jlmakiola/agendula/src/branch/main/CONTRIBUTING.md
+    about: >-
+      Before opening a pull request: how to build (there's a submodule), where
+      code goes, and the one architectural rule a change is reviewed against.
+
+  - name: Sync sources and scope
+    url: https://codeberg.org/jlmakiola/agendula/src/branch/main/README.md
+    about: >-
+      Agendula is a front-end over the OpenTasks provider, so it works with
+      DAVx5, SmoothSync, DecSync and friends. Google Tasks and Microsoft To Do
+      are out of scope by design — check here before requesting a backend.

.forgejo/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,42 @@
+<!--
+Thanks for contributing to Agendula!
+
+Please skim CONTRIBUTING.md if you haven't:
+https://codeberg.org/jlmakiola/agendula/src/branch/main/CONTRIBUTING.md
+
+Two things it's easy to get wrong:
+  • The one architectural rule — provider column names, `TaskContract`,
+    `ContentResolver` and the authority string never leak above `data/tasks/`.
+  • Don't bump `versionName` / `versionCode`. That bump reaching `main` is what
+    cuts a release, so it belongs only in a release PR.
+-->
+
+### What this changes
+
+
+### Why
+
+<!-- Closes #123 — link the issue this implements or fixes. -->
+
+
+### How it was tested
+
+<!--
+Which of these ran green, and anything you exercised by hand. On-device notes
+are especially useful for UI changes, and for anything touching the provider
+read/write paths (OpenTasks / tasks.org installed).
+
+  ./gradlew lintDebug :app:testDebugUnitTest :app:assembleDebug
+  python3 scripts/check_translations.py
+-->
+
+
+### Checklist
+
+- [ ] `./gradlew lintDebug :app:testDebugUnitTest :app:assembleDebug` passes locally
+- [ ] New domain logic comes with JVM unit tests under `app/src/test/`
+- [ ] Provider details stay inside `data/tasks/`
+- [ ] No `values-*/strings.xml` touched (Weblate owns those; new English strings in `values/` are fine)
+- [ ] `CHANGELOG.md` updated under `## [Unreleased]`, if the change is user-visible
+- [ ] No `versionName` / `versionCode` bump
+- [ ] No planning or design documents committed

.forgejo/workflows/ci.yaml

@@ -37,14 +37,29 @@ jobs:
       - name: Reproducible-release invariant
         run: bash scripts/check_reproducible_release.sh
 
-      # Decide whether anything that affects the app build changed. Docs,
-      # F-Droid metadata and the licence don't, so those PRs skip the SDK +
-      # Gradle work below but still report a green `ci`.
+      # Decide whether anything that affects the app build changed. Docs, store
+      # metadata, licence texts and forge housekeeping don't, so those PRs skip
+      # the SDK + Gradle work below but still report a green `ci`.
       - name: Classify change scope
         id: scope
+        env:
+          # Deliberately a skip-list, not a build-list: a path nobody thought
+          # about defaults to building. Only paths the Gradle build provably
+          # never reads belong here — note that the workflows themselves, the
+          # `.gitmodules` submodule pointer and `scripts/` are *not* in it.
+          SKIP_RE: '(\.md$|^docs/|^fastlane/|^fdroid-metadata/|^design/|^\.(forgejo|gitea)/ISSUE_TEMPLATE/|^\.editorconfig$|^\.gitattributes$|^\.gitignore$|^LICENSE$)'
         run: |
           set -e
           BASE="${{ github.base_ref }}"
+          # Normally the bare branch name; tolerate a full ref, which would
+          # otherwise make the merge-base lookup fail and quietly degrade this
+          # guard into "always build".
+          BASE="${BASE#refs/heads/}"
+          if [ -z "$BASE" ]; then
+            echo "No base branch on this event — running the full build to be safe."
+            echo "code=true" >> "$GITHUB_OUTPUT"
+            exit 0
+          fi
           # Full (not --depth=1) base fetch so the merge-base is present even when
           # the PR branch forked several commits back; a shallow tip has no merge
           # base with a divergent branch and `git diff base...HEAD` aborts.
@@ -58,11 +73,15 @@ jobs:
           fi
           CHANGED=$(git diff --name-only "$MB" HEAD)
           echo "Changed files:"; echo "$CHANGED"
-          if echo "$CHANGED" | grep -vE '(\.md$|^docs/|^fdroid-metadata/|^fastlane/|^LICENSE$)' | grep -q .; then
+          RELEVANT=$(echo "$CHANGED" | grep -vE "$SKIP_RE" || true)
+          if [ -n "$RELEVANT" ]; then
+            # Naming them makes "why did my docs PR build for four minutes?"
+            # answerable from the log alone.
+            echo "Build-relevant changes:"; echo "$RELEVANT"
             echo "code=true" >> "$GITHUB_OUTPUT"
           else
-            echo "code=false" >> "$GITHUB_OUTPUT"
             echo "Docs/metadata-only change — skipping the Android build."
+            echo "code=false" >> "$GITHUB_OUTPUT"
           fi
 
       - name: Setup Java
@@ -72,9 +91,14 @@ jobs:
           distribution: 'zulu'
           java-version: '17'
 
+      # Fully qualified on purpose. Codeberg resolves bare `uses:` refs against
+      # data.forgejo.org, Forgejo's own action mirror — actions/checkout,
+      # setup-java and cache all exist there, but android-actions/setup-android
+      # does not, and the job dies with "repository not found". Gitea's instance
+      # defaults to GitHub, which is why this never surfaced before the split.
       - name: Setup Android SDK
         if: steps.scope.outputs.code == 'true'
-        uses: android-actions/setup-android@v3
+        uses: https://github.com/android-actions/setup-android@v3
         with:
           # Default ("tools platform-tools") drags in the Android Emulator
           # (~300 MB) which the build never uses.

.forgejo/workflows/translations.yaml

@@ -0,0 +1,39 @@
+name: Translations
+
+# Fast, SDK-free parity check for translation resources, so Weblate PRs (which
+# only touch values-*/strings.xml) get quick feedback without the full Android
+# build. The deeper checks still run in CI via lintDebug (ExtraTranslation).
+#
+# Runs on every PR (no path filter) so the required "Translations / check"
+# status is always reported — like the `ci` job. A path-filtered workflow is
+# skipped on unrelated PRs and never posts its status, which leaves that
+# required check pending forever and blocks the merge of any code-only PR into a
+# release/* branch. The check itself is cheap and simply passes when the
+# committed translations are consistent, so always running it costs nothing.
+on:
+  pull_request:
+
+concurrency:
+  group: translations-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  check:
+    runs-on: docker
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Ensure python3
+        run: |
+          if ! command -v python3 >/dev/null 2>&1; then
+            if command -v apt-get >/dev/null 2>&1; then
+              apt-get update && apt-get install -y python3
+            elif command -v apk >/dev/null 2>&1; then
+              apk add --no-cache python3
+            fi
+          fi
+          python3 --version
+
+      - name: Check translation parity
+        run: python3 scripts/check_translations.py

.gitea/workflows/release.yaml

@@ -3,12 +3,18 @@ name: Release — F-Droid repo + Gitea/Codeberg release
 # A release is cut by merging a release branch into main with a bumped
 # versionName (see docs/RELEASING.md). This workflow reads that versionName and,
 # if no matching tag exists yet, runs tests, builds + signs the APK, publishes
-# it to the F-Droid repo, creates the vX.Y.Z tag + Gitea release, and mirrors
-# that release to Codeberg with the signed APK + a SHA-256 checksum as a
+# it to the F-Droid repo, creates the vX.Y.Z tag + Gitea release, and publishes
+# the release on Codeberg with the signed APK + a SHA-256 checksum as a
 # direct-download channel — the tag is an output of the pipeline, not its
 # trigger. Ordinary merges (no version bump) fall through `detect` and do
 # nothing.
 #
+# This file lives in .gitea/workflows on purpose: Codeberg is canonical for git,
+# issues, PRs and releases, but every secret (app key, F-Droid repo key, Hetzner
+# credentials) lives on the self-hosted Gitea instance, and this is the only
+# directory Codeberg cannot see. Contributor-triggerable work lives in
+# .forgejo/workflows and references no secret. See docs/RELEASING.md.
+#
 # A manual workflow_dispatch (from a branch) runs the re-sign-only recovery
 # path: it re-signs the existing F-Droid index with the repo key and re-uploads,
 # without building an APK or creating a release. Used for key rotation / repo
@@ -27,6 +33,14 @@ jobs:
   # whether this push actually cuts a new release (no tag for it yet). Keeps the
   # heavy job from running on every merge to main.
   detect:
+    # Gitea only. The workflow directory split already keeps this file invisible
+    # to Codeberg — Forgejo's lookup is first-match-wins, and .forgejo/workflows
+    # exists — but that only holds while .forgejo/ is non-empty. Move the last
+    # file out of it and Codeberg would fall back to .gitea/workflows and start
+    # running the release pipeline on the contributor-facing runner, with no
+    # secrets. repository_owner differs between the two forges regardless of
+    # URL, proxy or instance rename, so this closes it permanently.
+    if: github.repository_owner == 'makiolaj'
     runs-on: docker
     outputs:
       is_release: ${{ steps.v.outputs.is_release }}
@@ -42,8 +56,16 @@ jobs:
       - name: Resolve version and whether it is a new release
         id: v
         env:
-          TOKEN: ${{ secrets.GITHUB_TOKEN }}
-          API: ${{ github.server_url }}/api/v1/repos/${{ github.repository }}
+          # Tags are read from Codeberg, which is canonical — deliberately NOT
+          # from the Gitea API this workflow runs on. The Codeberg -> Gitea sync
+          # is a push mirror, i.e. `git push --mirror`, which deletes refs the
+          # source does not have. A tag minted here on Gitea is therefore wiped
+          # by the next sync (Codeberg does not have it yet) and only reappears
+          # once the tag push at the end of this workflow propagates back.
+          # Asking Gitea inside that window would report "no tag" for a release
+          # that already shipped, and cut it a second time.
+          # Public repo, so this read needs no token.
+          TAG_API: https://codeberg.org/api/v1/repos/jlmakiola/agendula
         run: |
           set -e
           VERSION=$(grep -oP 'versionName\s*=\s*"\K[^"]+' app/build.gradle.kts)
@@ -65,15 +87,28 @@ jobs:
           fi
           # A tag for this version already existing means the release shipped on
           # an earlier push; do nothing. Absent => this merge cuts the release.
-          STATUS=$(curl -s -o /dev/null -w '%{http_code}' \
-            -H "Authorization: token $TOKEN" "$API/git/refs/tags/v$VERSION")
-          if [ "$STATUS" = "200" ]; then
-            echo "Tag v$VERSION already exists — nothing to release."
+          #
+          # Anything other than a clean 200/404 is treated as fatal rather than
+          # as "no tag". A Codeberg outage or a network blip would otherwise
+          # read as absent and re-cut a release that has already shipped —
+          # republishing to F-Droid. Failing here is recoverable; a duplicate
+          # release is not.
+          STATUS=$(curl -s -o /dev/null -w '%{http_code}' "$TAG_API/git/refs/tags/v$VERSION" || echo 000)
+          case "$STATUS" in
+            200)
+              echo "Tag v$VERSION already exists on Codeberg — nothing to release."
               echo "is_release=false" >> "$GITHUB_OUTPUT"
-          else
-            echo "No tag for v$VERSION yet — cutting the release."
+              ;;
+            404)
+              echo "No tag for v$VERSION on Codeberg yet — cutting the release."
               echo "is_release=true" >> "$GITHUB_OUTPUT"
-          fi
+              ;;
+            *)
+              echo "Codeberg tag lookup for v$VERSION returned HTTP $STATUS." >&2
+              echo "Refusing to guess: treating this as 'no tag' could re-cut a shipped release." >&2
+              exit 1
+              ;;
+          esac
 
   # Releases: build + sign + publish, then mint the tag and Gitea release.
   # Also runs on manual dispatch, where it skips the build and just re-signs and
@@ -360,17 +395,21 @@ jobs:
             -F "attachment=@/tmp/$ASSET" \
             "$API/releases/$ID/assets?name=$ASSET" -o /dev/null -w "asset upload HTTP %{http_code}\n"
 
-      # Mirror the release to the Codeberg mirror as a direct-download channel
-      # for users who don't want F-Droid. Gitea already push-mirrors branches +
-      # tags to Codeberg, but releases aren't git objects so they don't sync —
-      # we create the release there over the API and attach the signed APK plus
-      # a SHA-256 checksum. The APK is identical to the F-Droid one (same app
-      # key), so this adds no trust surface. Best-effort: a Codeberg outage
-      # (it 504s under load) must never fail an already-published F-Droid
-      # release. Needs the CODEBERG_RELEASE_TOKEN secret; skips cleanly if unset.
+      # Publish the release on Codeberg, which is canonical for tags and
+      # releases (see docs/RELEASING.md). Codeberg push-mirrors branches + tags
+      # to Gitea, but releases aren't git objects and don't sync in either
+      # direction — so this step pushes the tag straight to Codeberg and creates
+      # the release there over the API, attaching the signed APK plus a SHA-256
+      # checksum as the direct-download channel for users who don't want
+      # F-Droid. The APK is identical to the F-Droid one (same app key), so this
+      # adds no trust surface. Needs the CODEBERG_RELEASE_TOKEN secret; skips
+      # cleanly if unset.
       - name: Publish release to Codeberg
         if: env.IS_RELEASE == 'true'
-        continue-on-error: true
+        # NOT continue-on-error: this step reported green through 0.2.1, 0.2.2,
+        # 0.3.0, 0.3.1 and 0.3.2 while never once publishing, which is how a
+        # crash-fix release reached F-Droid but not the Codeberg/Obtainium
+        # users who needed it. A broken mirror must fail the release loudly.
         env:
           TOKEN: ${{ secrets.CODEBERG_RELEASE_TOKEN }}
           API: https://codeberg.org/api/v1/repos/jlmakiola/agendula
@@ -399,29 +438,26 @@ jobs:
             sed -i -e '/./,$!d' release-notes.md
           fi
           [ -s release-notes.md ] || echo "_See CHANGELOG.md for ${VERSION}._" > release-notes.md
-          # Forgejo 500s on POST /releases when the tag ALREADY exists, and a
-          # bare tag is not a release — so GET /releases/tags then 404s and the
-          # upsert has no id to fall back on ("Could not resolve Codeberg
-          # release id", the 0.3.1 failure). Pushing the tag first therefore
-          # guarantees the 500 rather than avoiding it; 0.3.0 published because
-          # it POSTed while the tag was still absent. So: POST with the tag
-          # ABSENT and let the API mint tag + release together from
-          # target_commitish. Branches mirror reliably, so the commit is already
-          # on Codeberg — only a tag the mirror raced in ahead of us is in the
-          # way, and clearing it is safe precisely because no release owns it.
-          ID=$(curl -s -H "Authorization: token $TOKEN" "$API/releases/tags/$TAG" | jq -r '.id // empty')
-          if [ -z "$ID" ]; then
-            curl -s -o /dev/null -w "codeberg tag DELETE HTTP %{http_code}\n" -X DELETE \
-              -H "Authorization: token $TOKEN" "$API/tags/$TAG"
-          fi
-          python3 - "$TAG" "$SHA" "$PRERELEASE" <<'PY' > cb-payload.json
+          # Push the tag to Codeberg ourselves. Under Codeberg-canonical the
+          # mirror runs Codeberg -> Gitea, so waiting for a tag to arrive here
+          # from Gitea (what 0.3.2 did) would wait forever. The tag this
+          # pipeline minted on Gitea is in fact *deleted* by the next mirror
+          # sync until Codeberg has it — so pushing it here is what makes it
+          # durable on both forges.
+          #
+          # Pushing the ref first and attaching with NO target_commitish is
+          # deliberate: a release POST carrying a target_commitish for a commit
+          # or tag Codeberg hasn't received yet is what produced the
+          # empty-bodied 500s. Attaching to a ref that already exists doesn't
+          # need the API to write one.
+          git tag -f "$TAG" "$SHA"
+          git push -f "https://jlmakiola:${TOKEN}@codeberg.org/jlmakiola/agendula.git" \
+            "refs/tags/$TAG"
+          python3 - "$TAG" "$PRERELEASE" <<'PY' > cb-payload.json
           import json, sys
-          tag, sha, pre = sys.argv[1:4]
+          tag, pre = sys.argv[1:3]
           print(json.dumps({
               "tag_name": tag,
-              # Recreate the tag as part of the release; the commit is on
-              # Codeberg already via the branch mirror.
-              "target_commitish": sha,
               "name": tag,
               "body": open("release-notes.md").read(),
               "draft": False,
@@ -429,20 +465,31 @@ jobs:
               "prerelease": pre == "true",
           }))
           PY
-          # Upsert (re-run safe): a release that already exists is PATCHed in
-          # place — the delete above is skipped in that case, so re-running
-          # never disturbs a published release.
-          if [ -n "$ID" ]; then
+          # Create (or update) the release. Codeberg 500s on a POST/GET against a
+          # tag it has only just received — the release request outruns the
+          # indexing of the ref we pushed a moment ago — so a single attempt can
+          # fail even though the very same call succeeds seconds later. Retry
+          # with backoff, and PATCH in place if a release already exists (re-run
+          # safe, so re-running never disturbs a published release). A 5xx body
+          # still exits curl 0, so the loop, not `set -e`, controls the flow.
+          ID=""
+          for attempt in 1 2 3 4 5 6; do
+            EXIST=$(curl -s -H "Authorization: token $TOKEN" "$API/releases/tags/$TAG" | jq -r '.id // empty' 2>/dev/null || true)
+            if [ -n "$EXIST" ]; then
               curl -s -o /dev/null -w "release PATCH HTTP %{http_code}\n" -X PATCH \
                 -H "Authorization: token $TOKEN" -H "Content-Type: application/json" \
-              -d @cb-payload.json "$API/releases/$ID"
-          else
-            curl -s -o cb-response.json -w "release POST HTTP %{http_code}\n" -X POST \
-              -H "Authorization: token $TOKEN" -H "Content-Type: application/json" \
-              -d @cb-payload.json "$API/releases"
-            ID=$(jq -r '.id // empty' cb-response.json 2>/dev/null || true)
+                -d @cb-payload.json "$API/releases/$EXIST"
+              ID="$EXIST"; break
             fi
-          if [ -z "$ID" ]; then echo "Could not resolve Codeberg release id." >&2; exit 1; fi
+            CODE=$(curl -s -o cb-response.json -w "%{http_code}" -X POST \
+              -H "Authorization: token $TOKEN" -H "Content-Type: application/json" \
+              -d @cb-payload.json "$API/releases")
+            echo "release POST attempt $attempt HTTP $CODE"
+            ID=$(jq -r '.id // empty' cb-response.json 2>/dev/null || true)
+            [ -n "$ID" ] && break
+            sleep $((attempt * 10))
+          done
+          if [ -z "$ID" ]; then echo "Could not resolve Codeberg release id after retries." >&2; exit 1; fi
 
           # Attach APK + checksum, replacing any prior asset of the same name.
           for A in "$ASSET_APK" "$ASSET_SUM"; do

.gitea/workflows/renovate.yml

@@ -0,0 +1,61 @@
+name: Renovate
+
+on:
+  # Weekly sweep. Mondays 05:00 UTC — this cron owns the cadence; the repo's
+  # renovate.json5 deliberately has no internal schedule (avoids double-gating).
+  schedule:
+    - cron: '0 5 * * 1'
+  # Manual run for an on-demand sweep from the Actions tab.
+  workflow_dispatch:
+
+# Never let two Renovate runs touch the repo at once.
+concurrency:
+  group: renovate
+  cancel-in-progress: false
+
+jobs:
+  renovate:
+    # Gitea only — same guard, and the same reason, as release.yaml's `detect`:
+    # this file is invisible to Codeberg only while .forgejo/ is non-empty, and
+    # a repo-write token must never run on the contributor-facing runner.
+    if: github.repository_owner == 'makiolaj'
+    runs-on: docker
+    # Run the Renovate image *as* the job container and invoke the `renovate`
+    # binary directly. The renovatebot/github-action wrapper is a thin Node
+    # action that shells out to `docker run …` — it needs a Docker CLI + socket
+    # inside the job, which the Gitea runner's plain node container has not, so
+    # it died on "Unable to locate executable file: docker". Running the image
+    # directly drops the docker-in-docker requirement entirely.
+    # Full tag pinned; Renovate's github-actions manager keeps it bumped.
+    container:
+      image: ghcr.io/renovatebot/renovate:43.232.0
+    steps:
+      - name: Run Renovate
+        run: renovate
+        env:
+          # Renovate targets Codeberg (canonical) while still RUNNING on the
+          # Gitea runner. Moving the job to Codeberg would put a repo-write
+          # token on the contributor-facing runner, which is exactly what the
+          # .forgejo/ vs .gitea/ split exists to prevent — so the token stays
+          # where the other secrets live and only the API calls cross over.
+          #
+          # Platform is `forgejo`, not `gitea`: Codeberg runs Forgejo, and the
+          # pinned image ships a distinct forgejo platform module.
+          RENOVATE_PLATFORM: forgejo
+          RENOVATE_ENDPOINT: https://codeberg.org/api/v1
+          # Codeberg bot-account token (Gitea secret). Needs repo read/write +
+          # PR scope on jlmakiola/agendula.
+          RENOVATE_TOKEN: ${{ secrets.RENOVATE_TOKEN }}
+          # Scope to this repo only — no org-wide autodiscovery.
+          RENOVATE_AUTODISCOVER: 'false'
+          RENOVATE_REPOSITORIES: '["jlmakiola/agendula"]'
+          # Commits/PRs authored as the bot, not a real maintainer. This address
+          # must be a verified email on the Codeberg bot account, otherwise the
+          # commits show up unattributed there.
+          RENOVATE_GIT_AUTHOR: 'Renovate Bot <renovate@jeanlucmakiola.de>'
+          # Read-only github.com PAT (no scopes needed). Nearly every dependency
+          # is *released* on GitHub, and without this, changelog/release-note
+          # lookups hit the 60/h anonymous rate limit and PRs arrive with an
+          # empty "Release Notes" section.
+          RENOVATE_GITHUB_COM_TOKEN: ${{ secrets.GITHUB_COM_TOKEN }}
+          LOG_LEVEL: info

.gitignore

@@ -53,5 +53,15 @@ Thumbs.db
 # F-Droid local artifacts (the pipeline generates them in CI)
 /fdroid/
 
+# Release-pipeline scratch files. release.yaml writes these into the workspace
+# while cutting a release; a self-hosted runner reuses that workspace, so they
+# must never end up committed (release-notes.md did, through 0.3.2).
+/release-notes.md
+/payload.json
+/existing.json
+/response.json
+/cb-payload.json
+/cb-response.json
+
 # KSP
 .ksp/

.gitmodules

@@ -1,3 +1,3 @@
 [submodule "floret-kit"]
 	path = floret-kit
-	url = https://gitea.jeanlucmakiola.de/makiolaj/floret-kit.git
+	url = https://codeberg.org/jlmakiola/floret-kit.git

CHANGELOG.md

@@ -7,6 +7,18 @@ All notable changes to this project are documented here. The format follows
 
 ## [Unreleased]
 
+### Added
+- Agendula can now be translated. Pick or request a language on Weblate and
+  translate in the browser — the link sits at the top of the language picker in
+  **Settings → App language**. Partial translations are fine; anything
+  untranslated falls back to English.
+
+### Changed
+- Agendula's home is now **Codeberg** (`jlmakiola/agendula`) — that's where the
+  source, issues, pull requests and releases live. The Source and License links
+  in Settings, the issue-reporting link and the F-Droid metadata all point there
+  now. The self-hosted Gitea instance stays as build infrastructure.
+
 ## [0.3.2] - 2026-07-20
 
 ### Fixed

CONTRIBUTING.md

@@ -33,10 +33,33 @@ would expose provider details to a ViewModel or the UI, it's in the wrong layer.
 ./gradlew lintDebug                 # Android lint (CI runs this on every PR)
 ```
 
-CI (`.gitea/workflows/ci.yaml`) runs a reproducible-release invariant check,
+CI (`.forgejo/workflows/ci.yaml`, on Codeberg) runs a reproducible-release invariant check,
 then lint → unit tests → debug build on every pull request, so run these locally
 before opening a PR. Keep CI green.
 
+## Translations
+
+**Never edit a `values-*/strings.xml` file in a pull request.** Translations are
+owned by a self-hosted Weblate that writes to this repository directly, and a
+hand-edit is overwritten on the next sync.
+
+→ **[Translate Agendula on Weblate](https://weblate.dev.jeanlucmakiola.de/engage/agendula/)**
+
+Adding a *new* English string to `values/strings.xml` is normal PR work; Weblate
+picks it up and offers it to translators. Partial translations are expected and
+fine — missing keys are informational. Stale and orphaned keys are not, so run
+
+```sh
+python3 scripts/check_translations.py
+```
+
+before pushing. It reports those more clearly than lint's `MissingTranslation`
+does, and it's what the `Translations` check runs on every PR.
+
+A new language also needs one `<locale>` line in
+`app/src/main/res/xml/locales_config.xml` — that file is the single source of
+truth for both the in-app picker and the Android 13+ per-app language setting.
+
 ## Where to put code
 
 | Layer | Lives in | Rule of thumb |

README.md

@@ -6,6 +6,7 @@
 Reads, writes, and reminds — on top of an existing tasks provider, with no own
 sync stack.</p>
 
+<a href="https://codeberg.org/jlmakiola/agendula/actions"><img src="https://codeberg.org/jlmakiola/agendula/actions/workflows/ci.yaml/badge.svg?branch=main" alt="CI"></a>
 <img src="https://img.shields.io/badge/Android-10%2B-3DDC84?logo=android&logoColor=white" alt="Android 10+">
 <img src="https://img.shields.io/badge/Kotlin-Compose-7F52FF?logo=kotlin&logoColor=white" alt="Kotlin + Compose">
 <img src="https://img.shields.io/badge/Material%203-Expressive-4285F4" alt="Material 3 Expressive">
@@ -13,7 +14,7 @@ sync stack.</p>
 
 </div>
 
-Agendula is the task-list sibling to [Calendula](https://gitea.jeanlucmakiola.de/makiolaj/calendula).
+Agendula is the task-list sibling to [Calendula](https://codeberg.org/jlmakiola/calendula).
 Where Calendula is a pure front-end over Android's `CalendarContract`, Agendula is
 a pure front-end over the **OpenTasks `TaskContract` provider** — the store that
 DAVx5 (and SmoothSync, DecSync, …) syncs your CalDAV `VTODO` tasks into. No own
@@ -40,6 +41,18 @@ adapter — because it builds on the provider, not on any one sync app. Google
 Tasks / Microsoft To Do are out of scope by design (proprietary; they would mean
 owning a sync stack). Open standards — CalDAV / iCalendar / DecSync — are the lane.
 
+## Translations
+
+Agendula ships in English so far, and would like not to. Translations are
+managed on a self-hosted **Weblate**, and partial ones are fine — an
+untranslated string simply falls back to English.
+
+**→ [Help translate Agendula](https://weblate.dev.jeanlucmakiola.de/engage/agendula/)**
+
+No coding needed: register on the Weblate server, pick (or request) a language,
+and translate the strings in your browser. You can also reach this link in the
+app from the top of **Settings → App language**.
+
 ## License
 
 MIT — see [LICENSE](LICENSE).

app/build.gradle.kts

@@ -111,6 +111,18 @@ android {
         }
     }
 
+    lint {
+        // Community translations are expected to be partial — a missing string
+        // falls back to the English base at runtime — so don't fail the build on
+        // it. Likewise a translated <plurals> may not fill every CLDR quantity
+        // form its locale defines (e.g. Arabic needs "zero"); the missing form
+        // falls back to "other" at runtime, so MissingQuantity is informational
+        // too. Stale/extra keys (ExtraTranslation) stay fatal; scripts/
+        // check_translations.py guards the same invariants with clearer,
+        // translator-facing messages.
+        informational += listOf("MissingTranslation", "MissingQuantity")
+    }
+
     testOptions {
         unitTests {
             all { it.useJUnitPlatform() }

app/src/main/java/de/jeanlucmakiola/agendula/data/prefs/SettingsPrefs.kt

@@ -12,7 +12,7 @@ import de.jeanlucmakiola.agendula.domain.TaskFormField
 import de.jeanlucmakiola.floret.reminders.ReminderOverride
 import de.jeanlucmakiola.floret.reminders.ReminderOverrideCodec
 import de.jeanlucmakiola.floret.reminders.applyReminderOverride
-import de.jeanlucmakiola.floret.reminders.reminderLeadFor
+import de.jeanlucmakiola.floret.reminders.reminderLeadsFor
 import kotlinx.coroutines.flow.Flow
 import kotlinx.coroutines.flow.map
 import javax.inject.Inject
@@ -39,15 +39,17 @@ data class Settings(
     val bottomAddBar: Boolean = false,
     /**
      * Per-list overrides of [reminderLeadMinutes]: a list present in the map
-     * overrides the global default (a null value = no reminder); absent = inherit.
+     * overrides the global default (an empty list = no reminder); absent =
+     * inherit. Agendula offers a single reminder, so each override is a
+     * one-element (or empty) list.
      */
-    val perListReminderOverride: Map<Long, Int?> = emptyMap(),
+    val perListReminderOverride: Map<Long, List<Int>> = emptyMap(),
     /** Optional edit-form fields shown by default; the rest sit behind "More fields". */
     val defaultEditFields: Set<TaskFormField> = emptySet(),
 ) {
     /** The lead time for a task in [listId]: its override if set, else the global default. */
     fun reminderLeadFor(listId: Long): Int? =
-        perListReminderOverride.reminderLeadFor(listId, reminderLeadMinutes)
+        perListReminderOverride.reminderLeadsFor(listId, listOf(reminderLeadMinutes)).firstOrNull()
 }
 
 /** App preferences, backed by DataStore. Mirrors Calendula's prefs shape. */

app/src/main/java/de/jeanlucmakiola/agendula/ui/common/ReminderLeadPicker.kt

@@ -54,7 +54,9 @@ fun ReminderLeadPicker(
     onDismiss: () -> Unit,
     presets: List<Int> = REMINDER_PRESETS,
 ) {
-    val selectedMinutes = (selected as? ReminderOverride.Minutes)?.minutes
+    // Agendula is single-reminder: an override carries a one-element list, so
+    // take the single value for this single-select picker.
+    val selectedMinutes = (selected as? ReminderOverride.Minutes)?.minutes?.firstOrNull()
     val customSelected = selectedMinutes != null && selectedMinutes !in presets
     val seed = decomposeReminderMinutes(selectedMinutes?.takeIf { customSelected })
 
@@ -65,7 +67,7 @@ fun ReminderLeadPicker(
     val options = buildList {
         if (allowInherit) add(ReminderOverride.Inherit)
         if (allowNone) add(ReminderOverride.None)
-        presets.forEach { add(ReminderOverride.Minutes(it)) }
+        presets.forEach { add(ReminderOverride.Minutes(listOf(it))) }
     }
     val rowCount = options.size + 1 // + the custom row
 
@@ -105,7 +107,7 @@ fun ReminderLeadPicker(
                 unit = unit,
                 onUnitChange = { unit = it },
                 onConfirm = { minutes ->
-                    onSelect(ReminderOverride.Minutes(minutes))
+                    onSelect(ReminderOverride.Minutes(listOf(minutes)))
                     onDismiss()
                 },
             )
@@ -168,5 +170,5 @@ private fun CustomReminderEditor(
 private fun reminderOverrideLabel(override: ReminderOverride): String = when (override) {
     ReminderOverride.Inherit -> stringResource(R.string.reminder_use_default)
     ReminderOverride.None -> stringResource(R.string.reminder_none)
-    is ReminderOverride.Minutes -> reminderLeadTimeLabel(override.minutes)
+    is ReminderOverride.Minutes -> reminderLeadTimeLabel(override.minutes.first())
 }

app/src/main/java/de/jeanlucmakiola/agendula/ui/detail/TaskDetailScreen.kt

@@ -501,13 +501,17 @@ private fun taskWhenLines(task: Task): Pair<String, String?>? {
     val due = task.due
     return when {
         start != null && due != null -> {
-            val sameDay = start.formatDate() == due.formatDate()
-            val primary = if (sameDay) due.formatDate() else "${start.formatDate()} – ${due.formatDate()}"
-            val secondary = if (task.isAllDay) null else "${start.formatTime()} – ${due.formatTime()}"
+            val allDay = task.isAllDay
+            val sameDay = start.formatDate(allDay) == due.formatDate(allDay)
+            val primary =
+                if (sameDay) due.formatDate(allDay)
+                else "${start.formatDate(allDay)} – ${due.formatDate(allDay)}"
+            val secondary = if (allDay) null else "${start.formatTime()} – ${due.formatTime()}"
             primary to secondary
         }
-        due != null -> due.formatDate() to if (task.isAllDay) null else due.formatTime()
-        start != null -> start.formatDate() to if (task.isAllDay) null else start.formatTime()
+        due != null -> due.formatDate(task.isAllDay) to if (task.isAllDay) null else due.formatTime()
+        start != null ->
+            start.formatDate(task.isAllDay) to if (task.isAllDay) null else start.formatTime()
         else -> null
     }
 }

app/src/main/java/de/jeanlucmakiola/agendula/ui/edit/TaskEditScreen.kt

@@ -653,7 +653,7 @@ private fun ScheduleRow(
             )
         } else {
             Text(
-                text = value.formatDate(),
+                text = value.formatDate(allDay),
                 style = MaterialTheme.typography.titleMedium,
                 color = valueColor,
                 modifier = Modifier.clickable(onClick = onPick).padding(vertical = 8.dp, horizontal = 6.dp),
@@ -865,7 +865,7 @@ private fun ParentPickerSheet(
                     GroupedRow(
                         title = task.title.ifBlank { stringResource(R.string.task_untitled) },
                         position = positionOf(index, section.tasks.size),
-                        summary = task.due?.formatDate(),
+                        summary = task.due?.formatDate(task.isAllDay),
                         selected = task.taskId == selectedId,
                         minHeight = 56.dp,
                         onClick = { choose(task.taskId) },

app/src/main/java/de/jeanlucmakiola/agendula/ui/settings/SettingsScreen.kt

@@ -43,6 +43,7 @@ import androidx.compose.material.icons.filled.Gavel
 import androidx.compose.material.icons.filled.Language
 import androidx.compose.material.icons.filled.Notifications
 import androidx.compose.material.icons.filled.Palette
+import androidx.compose.material.icons.filled.Translate
 import androidx.compose.material.icons.filled.Tune
 import androidx.compose.material.icons.rounded.AccountTree
 import androidx.compose.material.icons.rounded.Circle
@@ -84,7 +85,6 @@ import de.jeanlucmakiola.agendula.domain.TaskFormField
 import de.jeanlucmakiola.floret.components.AboutCard
 import de.jeanlucmakiola.floret.components.AboutLink
 import de.jeanlucmakiola.floret.components.CollapsingScaffold
-import de.jeanlucmakiola.floret.components.LanguagePickerRow
 import de.jeanlucmakiola.floret.components.GroupedRow
 import de.jeanlucmakiola.floret.components.OptionPicker
 import de.jeanlucmakiola.floret.components.Position
@@ -92,8 +92,10 @@ import de.jeanlucmakiola.agendula.ui.common.ReminderLeadPicker
 import de.jeanlucmakiola.floret.components.pastelize
 import de.jeanlucmakiola.floret.components.positionOf
 import de.jeanlucmakiola.floret.identity.collapseExit
+import de.jeanlucmakiola.floret.locale.AppLanguage
 import de.jeanlucmakiola.floret.identity.expandEnter
 import de.jeanlucmakiola.floret.reminders.ReminderOverride
+import de.jeanlucmakiola.floret.reminders.reminderOverrideFor
 import de.jeanlucmakiola.agendula.ui.common.reminderLeadTimeLabel
 
 /** The settings sub-screens reached from the hub's category rows. */
@@ -210,19 +212,70 @@ private fun SettingsHub(
             leading = { CategoryIcon(Icons.Default.Notifications, ChipAccent.Primary) },
             onClick = { onOpenSection(SettingsSection.Reminders) },
         )
-        LanguagePickerRow(
-            position = Position.Middle,
-            title = stringResource(R.string.settings_language),
-            autoLabel = stringResource(R.string.settings_language_auto),
-            localesConfig = R.xml.locales_config,
-            leading = { CategoryIcon(Icons.Default.Language, ChipAccent.Neutral) },
-        )
+        LanguageRow(position = Position.Middle)
         ReportProblemRow(position = Position.Bottom)
 
         AppVersionText()
     }
 }
 
+/**
+ * The app-language row. Deliberately not floret-kit's `LanguagePickerRow`: the
+ * picker it opens carries a "Help translate" header, and inviting contributions
+ * right where a user goes looking for their language is app-specific framing,
+ * not a family primitive. Everything else matches that recipe.
+ */
+@Composable
+private fun LanguageRow(position: Position) {
+    val context = LocalContext.current
+    // Setting a locale recreates the activity; mirror the choice locally so the
+    // row updates instantly even before the recreation lands.
+    var current by remember { mutableStateOf(AppLanguage.currentTag()) }
+    var showDialog by remember { mutableStateOf(false) }
+
+    // null = follow the system; the rest are BCP-47 tags from locales_config.xml.
+    val options = remember { listOf<String?>(null) + AppLanguage.supportedTags(context, R.xml.locales_config) }
+
+    GroupedRow(
+        title = stringResource(R.string.settings_language),
+        summary = languageLabel(current),
+        position = position,
+        leading = { CategoryIcon(Icons.Default.Language, ChipAccent.Neutral) },
+        onClick = { showDialog = true },
+    )
+
+    if (showDialog) {
+        OptionPicker(
+            title = stringResource(R.string.settings_language),
+            predictiveBack = true,
+            options = options,
+            selected = current,
+            label = { languageLabel(it) },
+            onSelect = {
+                current = it
+                AppLanguage.apply(it)
+            },
+            onDismiss = { showDialog = false },
+            // Invite contributions right where users pick their language.
+            header = {
+                val translateUrl = stringResource(R.string.about_translate_url)
+                GroupedRow(
+                    title = stringResource(R.string.settings_translate),
+                    summary = stringResource(R.string.settings_translate_hint),
+                    position = Position.Alone,
+                    leading = { CategoryIcon(Icons.Default.Translate, ChipAccent.Neutral) },
+                    onClick = { openUrl(context, translateUrl) },
+                )
+                Spacer(Modifier.height(16.dp))
+            },
+        )
+    }
+}
+
+@Composable
+private fun languageLabel(tag: String?): String =
+    if (tag == null) stringResource(R.string.settings_language_auto) else AppLanguage.displayName(tag)
+
 /** Opens the project's issue tracker; no data leaves the device until submitted. */
 @Composable
 private fun ReportProblemRow(position: Position) {
@@ -483,10 +536,10 @@ private fun RemindersScreen(
     if (showOffset) {
         ReminderLeadPicker(
             title = stringResource(R.string.settings_default_reminder),
-            selected = ReminderOverride.Minutes(state.settings.reminderLeadMinutes),
+            selected = ReminderOverride.Minutes(listOf(state.settings.reminderLeadMinutes)),
             allowInherit = false,
             allowNone = false,
-            onSelect = { if (it is ReminderOverride.Minutes) viewModel.setReminderLeadMinutes(it.minutes) },
+            onSelect = { if (it is ReminderOverride.Minutes) viewModel.setReminderLeadMinutes(it.minutes.first()) },
             onDismiss = { showOffset = false },
         )
     }
@@ -504,14 +557,8 @@ private fun RemindersScreen(
 }
 
 /** The stored override for [listId], as a picker choice (absent → inherit). */
-private fun listOverrideChoice(state: SettingsUiState, listId: Long): ReminderOverride {
-    val map = state.settings.perListReminderOverride
-    return when {
-        !map.containsKey(listId) -> ReminderOverride.Inherit
-        map[listId] == null -> ReminderOverride.None
-        else -> ReminderOverride.Minutes(map.getValue(listId)!!)
-    }
-}
+private fun listOverrideChoice(state: SettingsUiState, listId: Long): ReminderOverride =
+    state.settings.perListReminderOverride.reminderOverrideFor(listId)
 
 /** Row summary for a list: its override, or the inherited global default. */
 @Composable
@@ -519,7 +566,7 @@ private fun listOverrideSummary(choice: ReminderOverride, globalDefault: Int): S
     ReminderOverride.Inherit ->
         stringResource(R.string.settings_list_reminder_inherits, reminderLeadTimeLabel(globalDefault))
     ReminderOverride.None -> stringResource(R.string.reminder_none)
-    is ReminderOverride.Minutes -> reminderLeadTimeLabel(choice.minutes)
+    is ReminderOverride.Minutes -> reminderLeadTimeLabel(choice.minutes.first())
 }
 
 // ---------------------------------------------------------------------------

app/src/main/java/de/jeanlucmakiola/agendula/ui/tasklist/TaskListScreen.kt

@@ -98,6 +98,9 @@ import de.jeanlucmakiola.agendula.domain.TaskSection
 import de.jeanlucmakiola.agendula.domain.TaskSections
 import de.jeanlucmakiola.agendula.ui.common.priorityAccent
 import de.jeanlucmakiola.floret.components.Position
+import de.jeanlucmakiola.floret.components.SnackChip
+import de.jeanlucmakiola.floret.components.SnackChipHeight
+import de.jeanlucmakiola.floret.components.SnackChipMargin
 import de.jeanlucmakiola.floret.time.formatDateTimeCompact
 import de.jeanlucmakiola.floret.components.pastelize
 import de.jeanlucmakiola.floret.components.positionOf
@@ -189,13 +192,18 @@ fun TaskListScreen(
             Box(
                 modifier = Modifier
                     .align(Alignment.BottomStart)
-                    .padding(start = 16.dp, bottom = inner.calculateBottomPadding() + 16.dp)
-                    .height(56.dp),
+                    .padding(
+                        start = SnackChipMargin,
+                        bottom = inner.calculateBottomPadding() + SnackChipMargin,
+                    )
+                    .height(SnackChipHeight),
                 contentAlignment = Alignment.CenterStart,
             ) {
-                UndoChip(
+                SnackChip(
                     visible = undoTarget != null,
-                    onUndo = {
+                    message = stringResource(R.string.task_deleted),
+                    actionLabel = stringResource(R.string.undo),
+                    onAction = {
                         undoTarget?.let { viewModel.undoDelete(it.taskId) }
                         undoTarget = null
                     },
@@ -205,47 +213,6 @@ fun TaskListScreen(
     }
 }
 
-/**
- * A compact floating "snackchip" for an undoable delete — a rounded pill (not a
- * full-width snackbar) sized to its content, sliding up from the bottom centre.
- */
-@Composable
-private fun UndoChip(visible: Boolean, onUndo: () -> Unit, modifier: Modifier = Modifier) {
-    AnimatedVisibility(
-        visible = visible,
-        enter = slideInVertically { it } + fadeIn(),
-        exit = slideOutVertically { it } + fadeOut(),
-        modifier = modifier,
-    ) {
-        Surface(
-            color = MaterialTheme.colorScheme.surfaceContainerHighest,
-            contentColor = MaterialTheme.colorScheme.onSurface,
-            shape = RoundedCornerShape(50),
-            shadowElevation = 6.dp,
-        ) {
-            Row(
-                verticalAlignment = Alignment.CenterVertically,
-                modifier = Modifier.padding(start = 20.dp, end = 8.dp, top = 6.dp, bottom = 6.dp),
-                horizontalArrangement = Arrangement.spacedBy(12.dp),
-            ) {
-                Text(
-                    text = stringResource(R.string.task_deleted),
-                    style = MaterialTheme.typography.bodyMedium,
-                )
-                TextButton(
-                    onClick = onUndo,
-                    contentPadding = PaddingValues(horizontal = 16.dp, vertical = 6.dp),
-                ) {
-                    Text(
-                        text = stringResource(R.string.undo),
-                        style = MaterialTheme.typography.labelLarge,
-                    )
-                }
-            }
-        }
-    }
-}
-
 @OptIn(ExperimentalMaterial3Api::class, ExperimentalFoundationApi::class)
 @Composable
 private fun TaskListBody(

app/src/main/res/values/strings.xml

@@ -190,13 +190,16 @@
     <string name="settings_about_logo_desc">Agendula app icon</string>
     <string name="settings_language">App language</string>
     <string name="settings_language_auto">System default</string>
+    <string name="settings_translate">Help translate</string>
+    <string name="settings_translate_hint">Add or improve a language on Weblate</string>
     <string name="settings_report_problem">Report a problem</string>
     <string name="settings_report_problem_hint">Open the issue tracker</string>
-    <string name="about_source_url" translatable="false">https://gitea.jeanlucmakiola.de/makiolaj/agendula</string>
+    <string name="about_source_url" translatable="false">https://codeberg.org/jlmakiola/agendula</string>
     <string name="crash_report_issue_title">Crash report</string>
-    <string name="report_issue_url" translatable="false">https://gitea.jeanlucmakiola.de/makiolaj/agendula/issues/new</string>
-    <string name="about_license_url" translatable="false">https://gitea.jeanlucmakiola.de/makiolaj/agendula/src/branch/main/LICENSE</string>
+    <string name="report_issue_url" translatable="false">https://codeberg.org/jlmakiola/agendula/issues/new</string>
+    <string name="about_license_url" translatable="false">https://codeberg.org/jlmakiola/agendula/src/branch/main/LICENSE</string>
     <string name="about_support_url" translatable="false">https://ko-fi.com/jeanlucmakiola</string>
+    <string name="about_translate_url" translatable="false">https://weblate.dev.jeanlucmakiola.de/engage/agendula/</string>
     <string name="settings_theme">Theme</string>
     <string name="settings_theme_system">Follow system</string>
     <string name="settings_theme_light">Light</string>

docs/ARCHITECTURE.md

@@ -12,7 +12,7 @@ Agendula is a Material 3 Expressive **front-end** over the OpenTasks
 `TaskContract` provider — it reads, writes, and reminds on top of a tasks store
 that some other app (DAVx5, SmoothSync, DecSync CC, tasks.org, …) syncs over
 CalDAV. **Agendula owns no database and no sync stack.** It is the task-list
-sibling to [Calendula](https://gitea.jeanlucmakiola.de/makiolaj/calendula),
+sibling to [Calendula](https://codeberg.org/jlmakiola/calendula),
 which does the same thing for `CalendarContract`.
 
 The whole design hangs off one rule:
@@ -244,8 +244,8 @@ the `@IoDispatcher`). `AgendulaApp` is the `@HiltAndroidApp` entry point;
 | Other | DataStore, kotlinx-datetime, kotlinx-coroutines |
 | Tests | JUnit5 (Jupiter) + Truth + Turbine + coroutines-test; the data source is the JVM-testable seam |
 | Versioning | committed `versionName` is the source of truth; a bump reaching `main` triggers the release and the pipeline mints the `vX.Y.Z` tag. `versionCode = MAJOR*10000 + MINOR*100 + PATCH`. See [`RELEASING.md`](RELEASING.md). |
-| CI | Gitea workflows (`.gitea/workflows/ci.yaml`, `release.yaml`) |
-| Distribution | F-Droid (`fdroid-metadata/`) |
+| CI | Split by forge: `.forgejo/workflows/ci.yaml` on Codeberg (canonical, no secrets), `.gitea/workflows/release.yaml` on Gitea (all secrets). See [`RELEASING.md`](RELEASING.md). |
+| Distribution | F-Droid (`fdroid-metadata/`) + Codeberg release APKs |
 
 ---
 

docs/README.md

@@ -3,7 +3,7 @@
 Agendula is a Material 3 Expressive **task** app for Android: a pure front-end over
 the OpenTasks `TaskContract` provider (synced by DAVx5 / SmoothSync / DecSync
 over CalDAV), with no own database or sync stack. Sibling to
-[Calendula](https://gitea.jeanlucmakiola.de/makiolaj/calendula). See the
+[Calendula](https://codeberg.org/jlmakiola/calendula). See the
 top-level [`../README.md`](../README.md) for the project pitch.
 
 ## Index

docs/RELEASING.md

@@ -1,15 +1,18 @@
 # Agendula — releasing
 
-Agendula is distributed through a **self-hosted F-Droid repo** (on Hetzner) with
-a human-readable **Gitea release** per version. Both are produced automatically
-by `.gitea/workflows/release.yaml` when a **bumped `versionName` reaches `main`**
-— the pipeline builds and publishes that version, then creates the matching
-`vX.Y.Z` tag and Gitea release itself. There are no APK assets on the Gitea
-release: distribution lives in the F-Droid repo; the release is the changelog of
-record.
+Agendula is distributed through a **self-hosted F-Droid repo** (on Hetzner) and a
+**Codeberg release** per version carrying the signed APK as a direct download.
+Both are produced automatically by `.gitea/workflows/release.yaml` when a
+**bumped `versionName` reaches `main`** — the pipeline builds and publishes that
+version, then creates the matching `vX.Y.Z` tag and the releases itself. The
+parallel **Gitea release** is the changelog of record on the build instance and
+carries no APK assets.
 
-While Agendula is pre-1.0 (`versionName` starts with `0.`), every Gitea release
-is flagged as a **pre-release**. This happens automatically and graduates to a
+Codeberg is the canonical forge; Gitea is build infrastructure. See
+[Two forges, one repo](#two-forges-one-repo) for how the two are wired.
+
+While Agendula is pre-1.0 (`versionName` starts with `0.`), every release is
+flagged as a **pre-release**. This happens automatically and graduates to a
 stable release at `1.0.0` — no manual toggling.
 
 ---
@@ -80,35 +83,49 @@ re-running the workflow safely retries).
 CI and release are split so a change is built once on its PR and only does
 release work when a merge actually cuts a release:
 
-- **`ci.yaml`** (on `pull_request`) — the reproducible-release invariant guard
+- **`ci.yaml`** (`.forgejo/workflows/`, on `pull_request`, **Codeberg**) — the
+  reproducible-release invariant guard
   (`scripts/check_reproducible_release.sh`), then lint + unit tests + a debug
   assemble and a Trivy scan, once per PR. Docs/metadata-only PRs skip the Android
   build but still report a green `CI` check.
-- **`release.yaml`** (on push to `main`, plus `workflow_dispatch`) — a cheap
-  `detect` job reads `versionName` and checks whether a tag for it already
-  exists. Only when it doesn't does the `release` job run: unit tests on the
-  merged commit, pin `versionCode`, build & sign the release APK with the **app
-  key**, copy it into the F-Droid repo, generate the per-version changelog from
-  the fastlane tree, re-sign the index with the **repo key**, upload `repo/` +
-  `metadata/`, then create the `vX.Y.Z` tag + Gitea release (CHANGELOG section as
-  notes, flagged pre-release while `MAJOR` is 0), attach the R8 `mapping.txt`,
-  and mirror the release to **Codeberg** with the signed APK + a SHA-256 checksum
-  (both best-effort). Ordinary merges with no version bump fall through `detect`
-  and do nothing.
+- **`translations.yaml`** (`.forgejo/workflows/`, on `pull_request`, **Codeberg**)
+  — an SDK-free parity check (`scripts/check_translations.py`) over
+  `values-*/strings.xml`, so Weblate PRs get fast feedback. Runs on every PR
+  without a path filter, so the required `Translations / check` status is always
+  reported.
+- **`renovate.yml`** (`.gitea/workflows/`, Mondays 05:00 UTC plus
+  `workflow_dispatch`, **Gitea**) — the dependency sweep. Runs the pinned
+  Renovate image as its job container and targets Codeberg's API; the cadence
+  lives here, not in `renovate.json5`, so the two don't double-gate.
+- **`release.yaml`** (`.gitea/workflows/`, on push to `main` plus
+  `workflow_dispatch`, **Gitea**) — a cheap `detect` job reads `versionName` and
+  checks **Codeberg** for a tag for it. Only when there isn't one does the
+  `release` job run: unit tests on the merged commit, pin `versionCode`, build &
+  sign the release APK with the **app key**, copy it into the F-Droid repo,
+  generate the per-version changelog from the fastlane tree, re-sign the index
+  with the **repo key**, upload `repo/` + `metadata/`, then create the `vX.Y.Z`
+  tag + Gitea release (CHANGELOG section as notes, flagged pre-release while
+  `MAJOR` is 0), attach the R8 `mapping.txt`, and publish the release on
+  **Codeberg** with the signed APK + a SHA-256 checksum. Ordinary merges with no
+  version bump fall through `detect` and do nothing.
 
 ### Codeberg direct-download channel
 
-Alongside F-Droid, each release is mirrored to the Codeberg repo
+Alongside F-Droid, each release is published on the Codeberg repo
 (`jlmakiola/agendula`) as a plain download for users who don't want F-Droid.
-Gitea already **push-mirrors** branches and tags to Codeberg, but releases
-aren't git objects and don't sync, so the pipeline creates the release over the
-Codeberg API and attaches `agendula_v<version>.apk` + its `.sha256`. It's the
-same APK the F-Droid repo serves (same **app key**), so it adds no trust surface.
-The step is best-effort: a Codeberg outage never fails an already-published
-F-Droid release, and it skips cleanly if `CODEBERG_RELEASE_TOKEN` is unset.
+Releases aren't git objects and don't sync with the push mirror in either
+direction, so the pipeline pushes the `vX.Y.Z` tag straight to Codeberg, creates
+the release over the Codeberg API, and attaches `agendula_v<version>.apk` + its
+`.sha256`. It's the same APK the F-Droid repo serves (same **app key**), so it
+adds no trust surface. It skips cleanly if `CODEBERG_RELEASE_TOKEN` is unset,
+but it is **not** `continue-on-error`: through 0.2.1–0.3.2 this step reported
+green while never once publishing, which is how a crash-fix release reached
+F-Droid but not the Codeberg/Obtainium users who needed it. A broken mirror
+fails the release loudly.
+
 One-time setup: the Codeberg repo's **Releases** unit must be enabled and a
-`CODEBERG_RELEASE_TOKEN` secret (Codeberg access token, `write:repository` scope)
-added to Gitea Actions.
+`CODEBERG_RELEASE_TOKEN` secret (Codeberg access token, `write:repository` scope
+— it pushes the tag as well as creating the release) added to Gitea Actions.
 
 ### Manual re-sign / recovery
 
@@ -120,6 +137,51 @@ rotation or repo recovery without publishing a new app version.
 
 ---
 
+## Two forges, one repo
+
+**Codeberg (`jlmakiola/agendula`) is canonical** — git, issues, PRs, tags and
+releases. The self-hosted Gitea instance is build infrastructure: it holds the
+signing key, publishes the F-Droid repo, and runs the release pipeline. Codeberg
+push-mirrors `main` and tags to Gitea, and a bumped `versionName` arriving there
+triggers `release.yaml` exactly as before.
+
+Workflows are separated by **directory**, not by conditionals. Forgejo looks in
+`.forgejo/workflows` → `.gitea/workflows` → `.github/workflows` and stops at the
+first that exists; Gitea doesn't know `.forgejo/` at all:
+
+| Directory | Runs on | Contains | Secrets |
+| --- | --- | --- | --- |
+| `.forgejo/workflows/` | Codeberg | `ci.yaml`, `translations.yaml` | **none** |
+| `.gitea/workflows/` | Gitea | `release.yaml`, `renovate.yml` | signing key, F-Droid, Hetzner, bot tokens |
+
+The line is drawn at **secrets, not at CI-vs-release**. That's what makes fork
+PRs safe: everything a contributor can trigger lives in `.forgejo/` and can
+reference no secret. Renovate stays on the Gitea runner *even though it opens
+PRs on Codeberg* — it talks to Codeberg's API rather than moving its token onto
+the contributor-facing runner. `detect` (and the Renovate job) additionally
+carries a
+`github.repository_owner == 'makiolaj'` guard, because the directory split only
+holds while `.forgejo/` is non-empty — empty it and Codeberg would fall back to
+`.gitea/` and run the release pipeline on the contributor-facing runner.
+
+Two consequences worth remembering:
+
+- **`detect` reads tags from Codeberg**, not from the Gitea instance it runs on.
+  Push mirroring is `git push --mirror`, so a tag minted on Gitea is deleted by
+  the next sync until the Codeberg tag push propagates back. Asking Gitea inside
+  that window would re-cut a shipped release.
+- **Any ref that exists only on Gitea gets deleted** by the mirror. That's
+  correct under Codeberg-canonical, but don't debug a "vanished" branch without
+  remembering it.
+
+`floret-kit` is a submodule of this repo and follows the same move: `.gitmodules`
+points at `https://codeberg.org/jlmakiola/floret-kit.git`, so a clone resolves
+without reaching the personal Gitea instance. The Gitea copy is **kept** — every
+existing tag records the old submodule URL, so rebuilds of past releases still
+resolve.
+
+---
+
 ## Secrets (Gitea → repo Settings → Actions → Secrets)
 
 The workflow fails loudly if the F-Droid ones are missing — it will **never**
@@ -133,7 +195,9 @@ user's pinned repo).
 | `FDROID_CONFIG_BASE64` | F-Droid `config.yml` (base64) — repo metadata + keystore passwords. |
 | `HETZNER_HOST`, `HETZNER_USER`, `HETZNER_PASS` | Upload target for the F-Droid repo. |
 | `GITHUB_TOKEN` | Provided by Gitea Actions; used to create the release + attach assets. |
-| `CODEBERG_RELEASE_TOKEN` | Codeberg access token (`write:repository` scope) — creates the mirrored Codeberg release + uploads the APK/checksum. Best-effort; if unset the Codeberg step skips. |
+| `CODEBERG_RELEASE_TOKEN` | Codeberg access token (`write:repository` scope) — pushes the tag to Codeberg, creates the release there and uploads the APK/checksum. If unset the step skips; if set and failing, the release fails. |
+| `RENOVATE_TOKEN` | Codeberg bot-account token — repo read/write + PR scope on `jlmakiola/agendula`. Used only by `renovate.yml`. |
+| `GITHUB_COM_TOKEN` | Read-only github.com PAT (no scopes). Without it Renovate's changelog lookups hit the 60/h anonymous rate limit and PRs arrive with empty release notes. |
 
 The app key signs APKs; the repo key signs the index (its fingerprint is what
 users pin). Neither key nor `config.yml` is ever uploaded to the server — they

docs/STORAGE-AND-SYNC.md

@@ -0,0 +1,357 @@
+# Agendula — storage and sync
+
+> Decided direction, captured 2026-08-01. Supersedes the earlier "Posture B =
+> bundle OpenTasks" working notes, which are withdrawn (see
+> [Dead ends](#dead-ends--do-not-revisit)). This is the detailed companion to
+> `ARCHITECTURE.md` §7 and the `ProviderResolver` comments, **and it redefines
+> what Posture B means** — those two need a follow-up edit.
+> `ROADMAP.md` / `PLAN.md` remain known-stale and are due a deliberate pass;
+> this document does not attempt it.
+
+## The plan, in short
+
+**What we're building**
+
+1. **Our own provider.** Vendor the Apache-2.0 dmfs task provider in-tree as
+   `:provider`, renamed to our own authority and permission namespace. Our
+   database, our namespace — coexists with everything, replaces nothing.
+2. **Our own sync.** An Agendula sync adapter, so remote storage never depends
+   on another app's roadmap.
+3. **The user picks the mode.** Local-only · Synced · External provider.
+4. **Least permission.** Ask only for what the chosen mode needs, when it needs
+   it.
+5. **Kit-first.** Anything that isn't task-domain goes to floret-kit.
+
+**In what order**
+
+| # | Step | Why now |
+|---|---|---|
+| 1 | Merge `fix/provider-interaction-review` | unmerged and rotting; touches the same permission flow as step 2 |
+| 2 | Vendor `:provider` under our own authority | the identity, done once — and it ships a complete local-first app |
+| 3 | Export / backup | our data now lives only in our app's private storage |
+| 4 | File the DAVx5 issue | cheap, non-blocking, serves F-Droid users |
+| 5 | Sync adapter | the 1.x arc; design discussion pending |
+
+Everything below is the reasoning behind those choices, the alternatives that
+were rejected, and the constraints they have to survive.
+
+---
+
+## The decision
+
+Agendula gets its **own identity all the way down** — its own task database
+under its own authority, its own sync, and a storage mode the user picks. It
+does not adopt, replace, or impersonate another project's provider.
+
+Four parts:
+
+1. **Own DB, bundled in-process.** Vendor the Apache-2.0 dmfs
+   `opentasks-provider` as an in-tree `:provider` Gradle module, renamed to
+   authority `de.jeanlucmakiola.agendula.tasks` with permissions
+   `de.jeanlucmakiola.agendula.permission.READ_TASKS` / `…WRITE_TASKS`. Not a
+   separate provider *app*; not a schema written from scratch. We keep the dmfs
+   `TaskContract` shape — it's proven, it's what our whole data layer already
+   speaks, and it's what every CalDAV engine already understands — we just own
+   the namespace it lives in.
+2. **Own sync adapter**, so remote storage never depends on another app's
+   roadmap. Protocol coverage is deliberately open — separate discussion.
+3. **The user chooses the backend**: local-only, synced, or an external provider
+   that's already on the device.
+4. **Ask for only what the chosen mode actually needs**, at the moment it needs
+   it.
+
+Plus a standing rule: **anything that isn't task-domain goes to floret-kit.**
+
+### The vocabulary, redefined
+
+`ARCHITECTURE.md` §7 and `ProviderResolver`'s KDoc still describe Posture B as
+"bundle OpenTasks and find `org.dmfs.tasks` first." Replace with:
+
+- **Posture A** — front-end over an *external* provider (OpenTasks, tasks.org).
+  Still fully supported; it stops being the default and becomes a **user
+  choice**.
+- **Posture B** — our own bundled provider under **our own** authority.
+  Coexists with everything; replaces nothing.
+
+The A/B seam itself is unchanged and still earns its keep: `ProviderResolver` is
+the only thing that knows an authority, `AndroidTasksDataSource` the only thing
+that touches a resolver. UI, ViewModels, domain and repository are untouched by
+all of this.
+
+---
+
+## Why not squat `org.dmfs.tasks`
+
+The rejected plan was to bundle the provider under dmfs's own authority so
+DAVx5 would sync into it unwittingly. Reasons it's out, in order of how badly
+each one bites:
+
+1. **It is a structural identity mismatch, and the resulting bug is invisible to
+   both sides.** Content-provider *authorities* are how a sync engine finds a
+   provider, but Android **account visibility is keyed by package name**: since
+   API 26 an app only sees accounts whose authenticator has made them visible to
+   *its package*, and `GET_ACCOUNTS` alone no longer suffices. A sync engine's
+   allowlist would name the package `org.dmfs.tasks`, not
+   `de.jeanlucmakiola.agendula`. So the bundled provider could find **zero**
+   accounts — and the dmfs provider uses account enumeration to prune task lists
+   whose account has gone away. The failure mode isn't "no sync", it's "our
+   provider quietly purges synced lists." *(Reasoned from the platform rules,
+   not from having read DAVx5's source — but the class of bug is structural, and
+   every future place anything keys on package rather than authority is a fresh
+   instance of it.)* An explicit integration under our own name makes this bug
+   impossible by construction.
+2. **Play Store install-time landmine.** Two apps cannot declare the same
+   authority (`INSTALL_FAILED_CONFLICTING_PROVIDER`) or the same `<permission>`
+   name (`INSTALL_FAILED_DUPLICATE_PERMISSION`, waived only for identical
+   signing certs). Anyone with OpenTasks installed gets a failed install,
+   surfacing as one-star reviews we can't usefully answer.
+3. **Migration data loss.** "Uninstall OpenTasks first" takes its DB with it.
+   CalDAV-synced tasks reconcile back; local-only tasks are simply gone, and
+   OpenTasks has no export (dmfs/opentasks #170, #204, #71 — years-old,
+   unimplemented; their wiki punts to a desktop client).
+4. **Squatting another project's namespace doesn't scale.** At low install
+   counts nobody notices. At scale we'd be generating issues on dmfs's tracker
+   that aren't dmfs's fault, and silently maintaining a schema fork under their
+   name.
+
+---
+
+## The `:provider` module
+
+**Source.** dmfs `opentasks-provider`, Apache-2.0. Target the **1.4.2** source
+(DB version 23 — the version that actually carries `is_recurring`; tasks.org's
+fork is DB 22 and lacks it, which is why `TaskMapper.task` reads `rrule`/`rdate`
+for recurrence detection rather than trusting the column).
+
+**Layout: in-tree module, not a git submodule.** floret-kit is a submodule
+because we co-develop it. This is a fork we will sync from upstream
+approximately never, so in-tree is simpler for both stores and honest about what
+it is. Ship a `provider/PROVENANCE.md`: upstream commit, and every change we
+made.
+
+**License hygiene, on day one.** The app is MIT, the provider is Apache-2.0 —
+permissive into permissive, fine — but the module keeps its Apache-2.0 headers,
+`LICENSE`, and `NOTICE`. Ten minutes now; embarrassing to retrofit once it's in
+two store listings.
+
+**What we change:**
+
+- Authority → `de.jeanlucmakiola.agendula.tasks` (it's already a string
+  resource, `opentasks_authority`).
+- Permission names → `de.jeanlucmakiola.agendula.permission.*`. These are
+  **hardcoded in the AAR manifest**, which is the single clearest reason
+  vendoring is mandatory rather than merely preferable — you cannot rename them
+  in a prebuilt artifact without `tools:` node surgery we'd rather not ship.
+- Drop `<uses-permission android:name="android.permission.GET_ACCOUNTS" />`.
+  We own our own accounts, so we don't need it — but note the provider's
+  account-cleanup path is written *assuming* it, so this is a review-and-rework
+  item, not a free deletion. **Verify** the provider's local-list/local-account
+  path works with no account present at all; that's the entire local-only mode.
+- Drop the exported `BOOT_COMPLETED` / `TIME_SET` / `TIMEZONE_CHANGED` receiver,
+  or keep it deliberately and give it an explicit `android:exported`. The AAR is
+  from the `targetSdk 29` era; AGP hard-errors on a merged manifest with an
+  intent-filtered component and no explicit `exported` once targetSdk ≥ 31, and
+  we're on 36. Fixed at source instead of patched around.
+- Modernize the build: it ships `minSdk 21` / `targetSdk 29`, Robolectric 3.5.1,
+  JUnit 4.12. We're `minSdk 29` / `targetSdk 36` and Play raises its target-API
+  floor annually, so this isn't optional upkeep.
+
+**Our data now lives in our app's private storage.** Uninstall means deletion.
+That single fact is what promotes export/backup from "nice to have" to a v1
+feature — see [Storage modes](#storage-modes--the-users-choice).
+
+---
+
+## Storage modes — the user's choice
+
+| Mode | Backing store | Sync | Needs |
+|---|---|---|---|
+| **Local** | our bundled provider | none | no permissions at all — same-uid provider access needs no grant |
+| **Synced** | our bundled provider | our sync adapter | network + an account the user configures |
+| **External** | OpenTasks / tasks.org | whatever that provider's engine does (DAVx5 …) | that provider's `READ`/`WRITE_TASKS`, granted at runtime |
+
+Local and Synced are the same store — Synced is Local with an account attached,
+so switching on sync is not a migration.
+
+**Resolver ordering needs deciding.** Today `ProviderResolver.CANDIDATES` is a
+fixed priority list and the first hit wins. Once we bundle our own provider,
+"first hit" is the wrong rule: someone who used Agendula locally and *later*
+installs DAVx5 + OpenTasks would see an external candidate outrank the provider
+that actually holds their data. Options: rank ours first whenever it's
+non-empty, or make the mode an explicit Settings choice (it's user-visible
+either way, so probably both — auto-pick a sane default, let Settings override).
+
+**Export/backup is a v1 feature.** Not, as previously framed, a migration safety
+net for "uninstall OpenTasks" — that scenario no longer exists. It's data
+portability for Local-mode users, whose tasks otherwise exist in exactly one
+place with no second copy. On Play, where most users won't have a sync engine,
+that's the majority.
+
+---
+
+## Sync — own adapter
+
+**Decided:** Agendula ships its own sync. Not because DAVx5 is bad, but because
+depending on it makes one external maintainer's roadmap the gate on our core
+feature — the same shape of dependency the whole identity decision exists to
+escape.
+
+The two precedents diverge and the choice between them is the whole point:
+tasks.org has its own authority **and its own sync** (sovereign); jtx Board has
+its own authority and **depends on DAVx5** (and got added, though a working
+relationship with bitfire is part of that story). We're taking the tasks.org
+shape.
+
+**Play sharpens this.** DAVx5 is a paid app on Google Play and free only on
+F-Droid — *worth confirming, since it's load-bearing* — which means most Play
+users will never have it. Lobbying bitfire is therefore an **F-Droid-audience
+feature, not a sync strategy**.
+
+**Still file the DAVx5 issue** — cheap, non-blocking, real value for F-Droid
+users. And make it the strongest possible version of the ask: our provider *is*
+the dmfs provider with renamed strings, so it's byte-identical contract
+compliance and a small enum-shaped addition with near-zero ongoing maintenance
+for them. Say that explicitly. "Here's a change that can't break anything" lands
+very differently from "please support my app."
+
+**Open — the next discussion.** Protocol coverage ("support as much as
+possible"), the account model, conflict resolution, and where the DAV/iCalendar
+work lives. One constraint to settle early: we're MIT; `dav4jvm` is Apache-2.0
+and fine, but **verify `ical4android`'s license** before assuming it's usable.
+
+---
+
+## Permissions — only what the mode needs
+
+The manifest is static, so "request only what we need" is really two different
+mechanisms, and conflating them is how apps end up over-permissioned:
+
+- **Runtime (dangerous) permissions** — genuinely stageable. Ask at the moment
+  the feature is used, never up front.
+- **Install-time (normal) permissions** — declared unconditionally; the only
+  lever is **not declaring them until the feature ships**, and not letting a
+  bundled dependency drag in ones we don't use.
+
+| Permission | When |
+|---|---|
+| *(none)* for our own provider | same-uid access needs no grant — `ProviderStatus.NEEDS_PERMISSION` must never fire in Local/Synced mode |
+| `org.dmfs.permission.*`, `org.tasks.permission.*` | requested **only** when the user selects External mode; declared always (static manifest) |
+| `POST_NOTIFICATIONS` | when reminders are first enabled |
+| `USE_EXACT_ALARM` / `SCHEDULE_EXACT_ALARM` | when exact due-time reminders are used. Note Play reviews `USE_EXACT_ALARM` and requires the app to be a calendar/alarm/task app — we qualify, but it needs a justification in the listing |
+| `INTERNET` | **don't declare it until sync ships** |
+| `GET_ACCOUNTS` | never — stripped from the vendored provider; our own account type doesn't need it to see its own accounts |
+
+**Work item:** the permission gate in `RootScreen` / `PermissionViewModel` /
+`ProviderResolver.hasPermission` currently assumes an external provider always
+needs a grant. It needs a bypass for our own provider. Modest, but it's the
+exact flow `fix/provider-interaction-review` just touched — merge that first.
+
+---
+
+## What lands in floret-kit
+
+Standing rule, matching the kit's own thesis (*share the mechanics, keep the
+look — the kit never knows about a specific app's domain*): if it isn't
+task-domain, it goes to the kit.
+
+| Candidate | Kit module | Note |
+|---|---|---|
+| ContentProvider seam — `ColumnReader`, failures, observer→Flow | `core-provider` | **already on the kit's deferred list**, blocked on migrating Calendula to the name-based reader. Bundling our own provider is the forcing function that makes this worth doing. |
+| Runtime-permission staging — request/state machine, rationale plumbing, "ask at point of use" | new, e.g. `core-permissions` | pure mechanics, and Calendula has the identical problem |
+| DAV client + iCalendar parse/serialize | new, e.g. `core-dav` | **the big one.** Calendula is a calendar app; it needs the same primitives. Worth designing for two consumers from the start rather than extracting later |
+| Export/backup plumbing — SAF, file writing, share-out | kit | the *serialization* of tasks is domain; the plumbing isn't |
+| Sync-adapter/account scaffolding | kit, probably | the `AbstractThreadedSyncAdapter` + authenticator boilerplate is identical everywhere; the delta logic is domain |
+
+**Stays app-local:** the vendored `:provider` module (task-specific, and
+Apache-2.0 against the kit's MIT), `TaskContract` and the mappers, domain models
+and smart lists, all screens, and the reminder *scheduler* (per the kit's
+existing "not shared" call — Agendula pulls, Calendula pushes).
+
+---
+
+## Distribution — F-Droid and Play from day one
+
+Not on either yet; both are targets, so build *for* them rather than retrofitting.
+
+- **F-Droid** requires from-source. The in-tree `:provider` module satisfies it;
+  a JitPack artifact would not. floret-kit's composite build already keeps the
+  kit from-source, and the reproducibility guard (`vcsInfo { include = false }`
+  in `app/build.gradle.kts`) is already in place.
+- **Play** requires a rising target-API floor, a data-safety declaration, and
+  justification for `USE_EXACT_ALARM`. It also means dangerous permissions we
+  don't use are a liability, not just clutter — which is most of why
+  `GET_ACCOUNTS` and the stray receiver come out of the vendored provider.
+- **Parked, not solved:** dual-distribution signing. F-Droid reproducible builds
+  verify against *our* signed APK; Play App Signing re-signs with Google's key.
+  Both can coexist, but it needs a deliberate pass before the first Play upload.
+  See `RELEASING.md`.
+
+---
+
+## Dead ends — do not revisit
+
+- **One APK that detects at install time and adapts.** Impossible.
+  `<provider>` authorities and `<permission>` declarations are frozen at build
+  time and read by the OS at install; there is no install-time hook where our
+  code runs. And unlike a component, a `<permission>` cannot be runtime-toggled
+  — no `setComponentEnabledSetting` equivalent.
+- **Dynamic feature modules** to deliver the provider conditionally. Conditions
+  are limited to hardware features / SDK / country — there is no "only if app X
+  is absent" — and they require Play, so they're dead for F-Droid regardless.
+- **`frontend` / `standalone` build flavors.** Two flavors means two
+  `applicationId`s (two listings, two signing lines, and switching costs a user
+  their local data), or one `applicationId` and they can't coexist in a repo
+  anyway. Obsolete now that our provider coexists with everything instead of
+  replacing anything.
+- **Maven Central for the provider.** `org.dmfs:opentasks-provider` *is* there —
+  but only up to `1.1.8.1` (2016, `jar` packaging, 3 versions). No DB 23.
+  Verified.
+- **JitPack (`com.github.dmfs.opentasks:opentasks-provider:1.4.2`).** Has the
+  right version, but it's a prebuilt artifact (fails F-Droid from-source), it
+  can't have its hardcoded permission names renamed, and adding JitPack widens
+  the dependency trust surface — `settings.gradle.kts` is currently `google()` +
+  `mavenCentral()` only, under `FAIL_ON_PROJECT_REPOS`. *Still usable for a
+  throwaway spike* (a library string resource can be overridden from the app
+  module, so the authority rename works), but not for anything we ship.
+
+---
+
+## Sequencing
+
+1. **Merge `fix/provider-interaction-review`** (`47cf99a`, currently unmerged
+   into `main`). It's blocking nothing and rotting, and it touches the exact
+   permission flow step 2 changes.
+2. **Vendor `:provider`** under our own authority and permission namespace, with
+   the permission-gate bypass. This is the identity, done once, done right —
+   and it ships a complete local-first app to both stores.
+3. **Export/backup.** Now a v1 feature, not a migration hack.
+4. **File the DAVx5 issue.** Non-blocking, cheap, serves F-Droid users.
+5. **Sync adapter.** The 1.x arc; design discussion pending.
+
+**Scope honesty:** the withdrawn notes costed this at "2–4 days shippable, +1
+week for F-Droid." Steps 2–5 are a substantially larger program than that, and
+the roadmap should say so rather than inheriting the old estimate.
+
+---
+
+## Open questions
+
+1. **Sync protocol coverage**, account model, conflict resolution — the next
+   discussion.
+2. **Resolver ordering / mode selection UX** once our provider coexists with
+   external ones (see [Storage modes](#storage-modes--the-users-choice)).
+3. **Does the vendored provider work with no account at all?** Local-only mode
+   depends on it entirely. First thing the vendoring work should prove.
+4. **`ical4android` licensing** vs our MIT.
+5. **jtx Board** as an additional External-mode candidate — richer contract,
+   later. (`PLAN.md` decision #3, still open.)
+
+---
+
+## Related
+
+The provider-interaction review on `fix/provider-interaction-review` fixed,
+among others: recurrence-aware editing (routes through the instances URI),
+all-day UTC handling, the `DUE`/`DURATION` collision, per-task reminders (Alarm
+property rows), and flow-recovery robustness. That makes `ROADMAP.md` open
+decision #5 ("recurrence-aware editing out of scope for v1") **stale**.

fdroid-metadata/de.jeanlucmakiola.agendula.yml

@@ -6,5 +6,5 @@ Summary: A modern Material 3 Expressive task app for Android.
 Categories:
   - Time
 
-SourceCode: https://gitea.jeanlucmakiola.de/makiolaj/agendula
-IssueTracker: https://gitea.jeanlucmakiola.de/makiolaj/agendula/issues
+SourceCode: https://codeberg.org/jlmakiola/agendula
+IssueTracker: https://codeberg.org/jlmakiola/agendula/issues

renovate.json5

@@ -0,0 +1,122 @@
+{
+  $schema: "https://docs.renovatebot.com/renovate-schema.json",
+
+  extends: [
+    "config:recommended",
+    // chore(deps): … — match the repo's conventional-commit style.
+    ":semanticCommits",
+  ],
+
+  // `config:recommended` brings in mergeConfidence:age-confidence-badges, whose
+  // Age column is a Mend badge. Mend's Merge Confidence index only covers Maven
+  // Central: org.jetbrains.kotlin, junit, truth, turbine et al resolve, but
+  // every androidx/compose artifact lives on Google's Maven repo and comes back
+  // as a grey UNKNOWN — i.e. most of this project. Renovate already knows the
+  // real answer, since it derives release timestamps itself for the
+  // minimumReleaseAge rules below (Google Maven serves `last-modified` on its
+  // POMs), so take the age from there and leave Mend to the Confidence column,
+  // which still carries signal for the Maven Central half.
+  prBodyDefinitions: {
+    Age: "{{#if releaseTimestamp}}{{{newVersionAgeInDays}}} d{{else}}unknown{{/if}}",
+  },
+  // Default heading links to the Merge Confidence docs; this column is ours now.
+  prBodyHeadingDefinitions: {
+    Age: "Age",
+  },
+
+  // No automerge: a dependency bump goes through the same review (and, for
+  // anything touching the build, the same on-device check) as a feature
+  // before it can ride a release — see docs/RELEASING.md and the mandatory
+  // `scripts/verify-release.sh` gate.
+  automerge: false,
+
+  // One reviewable surface; the dashboard issue lists everything pending.
+  dependencyDashboard: true,
+
+  // The cooling-off periods below are advisory, not a gate. "flexible" still
+  // prefers a version that has cleared its window, but when every candidate is
+  // too young it opens the PR at the newest one anyway, so merging early stays
+  // a judgement call. (The default, "strict", would suppress the PR entirely
+  // until a release aged in.) A still-young branch carries a yellow
+  // `renovate/stability-days` check so it's visible which side of the line
+  // it's on; with automerge off, nothing acts on that check by itself.
+  //
+  // NOT "none": that short-circuits the candidate loop in filter-checks.ts, and
+  // that loop is what calls postprocessRelease — the only thing that fetches a
+  // Maven artifact's Last-Modified header. Skipping it leaves releaseTimestamp
+  // unset, which empties the Age column and quietly makes minimumReleaseAge and
+  // the stability check no-ops, since both need that timestamp to compare.
+  internalChecksFilter: "flexible",
+
+  labels: ["dependencies"],
+  prConcurrentLimit: 5,
+  prHourlyLimit: 0,
+
+  // Cadence is owned by the Gitea Actions cron (.gitea/workflows/renovate.yml,
+  // Mondays) — no internal `schedule` here, so the two don't double-gate and
+  // silently skip a run.
+
+  // Workflows are split by forge and neither directory is .github: CI lives in
+  // .forgejo/workflows (Codeberg) and the release/renovate jobs in
+  // .gitea/workflows (Gitea). Extend the github-actions manager (same syntax)
+  // to watch both — otherwise the pinned Renovate image tag and the action
+  // versions in either file would never get bumped. See docs/RELEASING.md.
+  // `fileMatch` is deprecated; the replacement takes the regex delimited, and
+  // Renovate's config migration was already rewriting this on every run.
+  "github-actions": {
+    managerFilePatterns: ["/^\\.(gitea|forgejo)/workflows/[^/]+\\.ya?ml$/"],
+  },
+
+  packageRules: [
+    // Cooling-off period, scaled by blast radius: how long a release should
+    // have been out (and un-yanked, un-hotfixed) before it's considered
+    // settled. Advisory only — see `internalChecksFilter` above.
+    {
+      matchUpdateTypes: ["major"],
+      minimumReleaseAge: "30 days",
+    },
+    {
+      matchUpdateTypes: ["minor"],
+      minimumReleaseAge: "20 days",
+    },
+    {
+      matchUpdateTypes: ["patch", "digest", "pin", "rollback"],
+      minimumReleaseAge: "10 days",
+    },
+    // material3 is deliberately pinned to the 1.5 *alpha* line for the
+    // Expressive APIs (see gradle/libs.versions.toml). Follow the alpha train
+    // but keep it in its own PR, reviewed in isolation; revisit the pin when
+    // 1.5.0 stable lands.
+    {
+      matchPackageNames: ["androidx.compose.material3:material3"],
+      ignoreUnstable: false,
+      groupName: "material3 (alpha)",
+    },
+    // Test-only deps: group into one low-noise PR.
+    {
+      matchPackageNames: [
+        "org.junit.jupiter:**",
+        "org.junit.platform:**",
+        "com.google.truth:**",
+        "app.cash.turbine:**",
+        "androidx.test:**",
+        "androidx.test.espresso:**",
+        "androidx.test.ext:**",
+      ],
+      groupName: "test dependencies",
+    },
+    // Last word on the PR table. The merge-confidence preset sets prBodyColumns
+    // from inside a packageRule of its own, and only for the datasources Mend
+    // supports — so a plain top-level prBodyColumns would lose to it for maven
+    // deps, and the Gradle wrapper / Actions / container bumps would keep the
+    // default columns and show no age at all. A rule declared after it wins,
+    // and gives every PR the same table.
+    // "Pending" earns its place under a flexible filter: when the bump lands on
+    // a version that has cleared its window but a newer one hasn't, that newer
+    // version is named here rather than silently withheld.
+    {
+      matchPackageNames: ["*"],
+      prBodyColumns: ["Package", "Type", "Change", "Age", "Pending", "Confidence"],
+    },
+  ],
+}

scripts/check_translations.py

@@ -0,0 +1,94 @@
+#!/usr/bin/env python3
+"""Validate Android translation resources against the base strings.xml.
+
+Community translations live in ``app/src/main/res/values-<locale>/strings.xml``
+and are produced via Weblate. This guard keeps incoming translation PRs honest:
+
+  * every translation file must be well-formed XML;
+  * a translation must not define keys absent from the base — those are stale
+    keys left behind after a rename/removal upstream;
+  * a translation must not translate strings marked ``translatable="false"`` in
+    the base (URLs, IDs and the like).
+
+Missing keys are *allowed* and only reported as coverage: a missing string
+falls back to the English base at runtime, so partial translations are fine
+(this mirrors the lint config, which downgrades ``MissingTranslation``).
+
+Exits non-zero if any error is found. Errors are emitted as Gitea/GitHub
+Actions ``::error`` annotations so they surface inline on the PR.
+"""
+from __future__ import annotations
+
+import sys
+import xml.etree.ElementTree as ET
+from pathlib import Path
+
+RES_DIR = Path("app/src/main/res")
+BASE = RES_DIR / "values" / "strings.xml"
+RESOURCE_TAGS = ("string", "plurals", "string-array")
+
+
+def entries(path: Path) -> dict[str, bool]:
+    """Map resource name -> is-translatable for every entry in ``path``."""
+    root = ET.parse(path).getroot()
+    return {
+        el.attrib["name"]: el.attrib.get("translatable", "true") != "false"
+        for el in root
+        if el.tag in RESOURCE_TAGS and "name" in el.attrib
+    }
+
+
+def main() -> int:
+    if not BASE.exists():
+        print(f"::error::base resource file {BASE} not found", file=sys.stderr)
+        return 1
+
+    base = entries(BASE)
+    base_keys = set(base)
+    nontranslatable = {name for name, ok in base.items() if not ok}
+    translatable_total = len(base_keys - nontranslatable)
+
+    files = sorted(RES_DIR.glob("values-*/strings.xml"))
+    if not files:
+        print("No translation files found (values-*/strings.xml).")
+        return 0
+
+    errors = 0
+    for path in files:
+        locale = path.parent.name[len("values-"):]
+        try:
+            translated = entries(path)
+        except ET.ParseError as exc:
+            print(f"::error file={path}::{locale}: malformed XML: {exc}")
+            errors += 1
+            continue
+
+        keys = set(translated)
+        stale = sorted(keys - base_keys)
+        translated_fixed = sorted(keys & nontranslatable)
+        missing = base_keys - nontranslatable - keys
+
+        for name in stale:
+            print(f"::error file={path}::{locale}: stale key '{name}' is not in the base strings.xml")
+            errors += 1
+        for name in translated_fixed:
+            print(
+                f"::error file={path}::{locale}: key '{name}' is translatable=\"false\" "
+                "in the base and must not be translated"
+            )
+            errors += 1
+
+        covered = translatable_total - len(missing)
+        pct = covered * 100 // translatable_total if translatable_total else 100
+        verdict = "OK" if not (stale or translated_fixed) else "FAIL"
+        print(f"{locale:<10} {covered}/{translatable_total} keys ({pct}%) — {verdict}")
+
+    if errors:
+        print(f"\n{errors} translation error(s) found.", file=sys.stderr)
+        return 1
+    print("\nAll translation files are consistent with the base.")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())