Fixes four of the five open bug reports, plus #97 (mailbox.org sign-in, reported by email).
- **#92 widgets reverting / "Can't load content"**: R8's horizontal class merging folded `TodayWidget`, `UpNextWidget`, `ListsWidget` and `WeekWidget` into one class. Glance maps providers to receivers by class name, so `TodayWidget().updateAll()` also targeted every Up Next, Lists and Week slot. Any slot without a running session got Today's composition, which is why widgets turned into the Today ring. In a 4x1 slot the ring size goes negative and Glance shows its error layout. A keep rule for `GlanceAppWidget` subclasses stops the merge. I checked the fresh `fullReleaseTest` mapping: all five classes are separate and none has a `$r8$classId`.
- **#70 crash in ru-RU**: fixed in floret-kit (jlmakiola/floret-kit#20). Its year-stripping regex broke `d MMM y 'г'.` into an unterminated quote. The new version also keeps the dot that closes the day-month part (de-DE `05.03.`). That PR is merged; the submodule points at its tip, which is on floret-kit main.
- **#86 widget padding**: when the Tasks widget drops its redundant section label (e.g. the Today list), a spacer above the list now puts the first row 8dp under the header (one `widget_tasks_header_gap` dimen shared with the preview), as in the preview, instead of 2dp.
- **#76 "Starts" clipped**: the label and values in the task form's Starts/Due rows share one line when they fit. Otherwise the values move to their own end-aligned line, using a `FlowRow` rather than a custom layout. The time can wrap under the date, and the clear button always keeps its 40dp.
- **#97 mailbox.org sign-in ("isn't a CalDAV server")**: the SRV/TXT lookup never reached the network. dnsjava was never set up for Android, so every lookup came back empty. mailbox.org's server is only published by SRV (`dav.mailbox.org`), so discovery fell back to `mailbox.org/.well-known/caldav`, which redirects to `auto.mailbox.org` and answers 405. Posteo (SRV-only, port 8443) hit the same dead end. Lookups now go through `android.net.DnsResolver.rawQuery` (minSdk is 29), so they follow the current network and honour Private DNS. SRV and TXT are queried at once under one 10s deadline, and a failure reads as no records. The wait stops as soon as the sign-in is cancelled, and answered lookups are kept for a minute so the signed-in pass does not ask again. `:caldav` only parses the raw answers (`DnsAnswers`, which replaces `DnsJavaResolver`). PRIVACY.md now says so.
- If an email address still finds no server, the address step switches straight to a server-address field with its own explanation, rather than ending at an error. That also happens when nothing answered for the domain (a mail-only domain), with the error left visible in case it was a typo or no network, and after a sign-in on the email domain. The email (without a `mailto:` prefix) is kept as the username for the sign-in step, and Back returns to the email.
Not included: **#75** (ticking in the 2x2 widget does nothing). The widget is most likely Today, which was one of the merged classes, so this may fix it too. I couldn't show the link from the code, so the issue stays open until it's checked on a device.
Closes #70, closes #76, closes #86, closes #92, closes #97
Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de>
Reviewed-on: https://codeberg.org/jlmakiola/agendula/pulls/96
Merging #88 left two `@Update fun update(alarm: TaskAlarmEntity)` declarations in `TaskAlarmDao`: one returning `Int` from the release branch, and one returning `Unit` from the feature branch. `release/v1.2.0` doesn't compile with both. This keeps the `Int` one, which `ProviderWriter` uses.
Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de>
Reviewed-on: https://codeberg.org/jlmakiola/agendula/pulls/89
Reminders can count from a task's start as well as its due date (#72), and each reminder on an all-day task can have its own time (#78).
## Changes
- A reminder counts from the start or the due date. Tasks with only a start date get reminders. Defaults in Settings can mix both, globally and per list.
- Each reminder on an all-day task can have its own time of day. Reminders without one ring at the all-day time set in Settings. 00:00 can't be stored, so it becomes 00:01.
- The editor shows the defaults a new task, or a newly added date, will get, and keeps them in step with dates, list and all-day until you change a reminder. What it shows is what gets saved. Removing the shown defaults now leaves the task without a reminder (Room store only; OpenTasks mode still falls back to the default).
- Switching between timed and all-day converts the reminders right away in the editor. Switching straight back restores them.
- Labels name the date when a task has both: "15 minutes before the start", "On the due date at 09:00". New reminders preselect the start.
- Defaults that have already passed don't ring when the task is saved.
- Sync: reminders rounded to whole days by builds before 1.2 pick up their real time of day on the next download.
## Deviations from the issues
- The start/due anchor lives in floret-kit (`ReminderLead`, floret-kit#19) rather than being packed into minute values. That PR is additive, so Calendula is unaffected.
- `error_reminder_without_due` was renamed to `error_reminder_without_date`, and the "(before start)/(before due)" strings were replaced by full sentences. The stale keys were deleted from the translation files so lint passes, and Weblate needs to offer the new strings to translators.
- The dead single-reminder API (`setAlarm`, `alarms()`, `editable()`, `reminderFor`) was removed.
- `AlarmLift` is back to its frozen 2→3 migration state.
## Release branch
`release/v1.2.0` is merged into this branch. Through it, this PR also brings main's commits the release branch was missing, including the reminder sync from #81 that this work builds on. I resolved conflicts in the data source interface, `TaskAlarmDao`, CHANGELOG, PRIVACY.md and the floret-kit pointer by keeping both sides.
Closes #72
Closes #78
Co-authored-by: Weblate <noreply-mt-weblate@weblate.org>
Co-authored-by: Sup99999 <alelmnour1o234@proton.me>
Co-authored-by: Robin Schanbacher <robin@schanbros.de>
Co-authored-by: Knee8811 <mk-weby@protonmail.com>
Co-authored-by: Sonja Kraus <klein.felix@pm.me>
Co-authored-by: Weblate <noreply@weblate.org>
Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de>
Reviewed-on: https://codeberg.org/jlmakiola/agendula/pulls/88
Brings Agendula's screen transitions and settings in line with Calendula.
**Motion**
- Screens slide in on the M3 Expressive spring and leave through floret's `predictiveBackExit`, everywhere: main navigation and every settings sub-screen.
- Settings sub-screens get Calendula's back-swipe preview (shrink, edge shift, rounded corners). The central settings BackHandler is gone; each screen handles its own back.
- Main navigation keeps NavHost's own predictive back, which seeks the new exit over the previous screen.
- `enableOnBackInvokedCallback` is set, so Android 14/15 show the preview too.
**Settings**
- Reminders → Notifications, Task form → New task form, with Calendula's subtitles and labels inside Notifications.
- Export and import move out of Storage into a new Backup & restore section; Storage keeps only the task store.
- The language row uses floret-kit's `LanguagePickerRow` with the translate header.
**Deviations from the issue**
- The main-navigation swipe preview scales and fades but doesn't shift or round its corners like Calendula's. Matching that exactly would mean replacing navigation-compose with Calendula-style overlays. Settings match exactly.
- The AOSP/GrapheneOS Settings transition suggested in the issue isn't adopted; the family stays on its M3 Expressive motion.
- Renamed labels keep their string keys, so existing translations stay until Weblate re-reviews them rather than falling back to English.
Depends on jlmakiola/floret-kit#17 (the submodule points at it).
Closes #66
Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de>
Reviewed-on: https://codeberg.org/jlmakiola/agendula/pulls/69
Builds on the read-only provider (#62) and makes it writable for other apps, as long as Agendula uses its own store.
- `tasks`: insert, update and delete. Writes follow the same rules as the app's own edits. Status, progress and completion time stay consistent, a subtree moves and deletes as one, and a write that changes nothing doesn't mark the task for sync.
- `instances`: completion columns complete, cancel or reopen that one occurrence. Other content columns edit the whole series. Changing a time column is rejected, so a series is moved through `tasks`.
- `properties`: alarms and categories can be inserted, updated and deleted.
- `tasklists` stays read-only. Batches run in one transaction.
- A client can write back a row it read. Columns it carries unchanged are ignored, and so are rdate/exdate values in the shape they were synced in.
- A task inserted with a time but no `tz` is floating, as in OpenTasks.
- Deleting a task now takes its whole subtree in the app as well, including "this and following" from the first occurrence and in External mode.
Deviation: tag property ids stay stable, with one rare exception. If two tags on one task have hash-colliding spellings, removing one can shift the other's id. Avoiding that would mean storing the ids, which didn't seem worth it at about a 1 in 2³¹ chance.
Closes #63
Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de>
Reviewed-on: https://codeberg.org/jlmakiola/agendula/pulls/64
In Own storage mode, Agendula now publishes its Room store as a ContentProvider that follows the OpenTasks contract. It serves `tasklists`, `tasks`, `instances` and `properties` (reminders and tags) at `<package>.tasks`, behind a dangerous-level `<package>.permission.READ_TASKS` that the user grants per app. An app that already reads OpenTasks or tasks.org only has to change the authority and the permission.
On a query, the provider lays the Room store out in OpenTasks' columns and loads that into an in-memory SQLite database, so a client's selection, arguments and sort order run as real SQL. The snapshot is rebuilt after a write, after a storage mode switch, or once it is an hour old, because instance ids and `distance_from_current` move with the clock. Observers on the authority get one notification per burst of changes. In External mode every table is empty.
Recurrence expansion moved out of `RoomTasksDataSource` into shared functions, so the app and the provider expand a series the same way.
`docs/CONTENT_PROVIDER.md` describes the contract for client developers, with the authority and permission for each build. README, CHANGELOG and the privacy policy mention the new permission.
Where this differs from the issue: the issue asked for read and write permissions, and this PR is read-only. Insert, update and delete throw. Accepting writes from other apps means deciding what happens to the fields Agendula keeps verbatim for CalDAV round-tripping, which was the issue's second open question, and that needs its own design. That is tracked in #63.
Closes #43
Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de>
Reviewed-on: https://codeberg.org/jlmakiola/agendula/pulls/62
Agendula 1.1.0. Merging this triggers the release pipeline (F-Droid, Codeberg, Play) and mints the v1.1.0 tag.
- Tags, synced as CATEGORIES (#35)
- Saved filters, also usable in the Tasks widget (#47)
- Offline flavor without sync or network, published to Codeberg only (#39)
- New Today, Up next, Lists and Week widgets; widgets can show several lists
- Synced lists can be renamed and deleted
- All-day tasks no longer land a day off west of UTC in smart lists and filters
versionName 1.1.0 / versionCode 1010099. CHANGELOG has the 1.1.0 section, and What's New is written for en-US, en-GB, de-DE, pt-BR, plus new fr-FR and ar.
After the tag lands, fdroiddata!49998 needs a 1.1.0 build entry with gradle: full.
Closes #35, closes #39, closes #47
Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de>
Reviewed-on: https://codeberg.org/jlmakiola/agendula/pulls/51
### What this changes
Adds beta releases. Pushing a `release/*` branch whose committed `versionName` is `X.Y.Z-beta.N` makes the new `.gitea/workflows/beta.yaml` run the unit tests, build and sign the APK with the app key, and publish it as a **Codeberg pre-release** (APK + `.sha256`), plus a Gitea pre-release with the R8 mapping. F-Droid (self-hosted and official) and Play never get a beta; Obtainium only offers it with *Include prereleases* on.
- **`scripts/version_info.sh`** is the single source for `versionName` → `versionCode`, used by `release.yaml`, `beta.yaml`, the changelog sync and the store-listing check. From 1.1.0: `X*1000000 + Y*10000 + Z*100 + N` for betas (N = 1–98), `+ 99` for stable, so `1.1.0-beta.1` → `1010001`, `1.1.0` → `1010099`. All 1.0.x versions keep the legacy formula, so `release/v1.0.1` (code `10001`) stays valid.
- **Shared publish scripts:** `scripts/publish_codeberg_release.sh`, `scripts/publish_gitea_release.sh` and `scripts/release_notes.sh`, moved out of `release.yaml`. The stable path behaves as before.
- **Guards:**
- CI fails a PR whose committed `versionCode` doesn't match its `versionName`.
- CI fails a PR into `main` that carries a beta version.
- `release.yaml`'s `detect` refuses a beta on `main` as a backstop.
- `beta.yaml` refuses a beta of a version that has already shipped as stable.
- Betas get no store What's New file.
- **Docs:** "Cutting a beta" and the versionCode table in `docs/RELEASING.md`; a note on beta tags in `docs/fdroid-official/README.md`; how to opt in to betas in the README.
### Why
To ship a test build of an upcoming version (e.g. 1.1.0) to opted-in testers before the stable release, without it reaching F-Droid or Play users.
### How it was tested
- `scripts/version_info.sh` against stable, beta, legacy and invalid version names.
- Both publish scripts against a mock forge API: create, re-run (PATCH plus asset replacement), Codeberg's 500-then-retry path, and the skip when no token is set.
- A scratch copy with `1.1.0-beta.1` committed: the version check passes, the changelog sync and `check_store_listing.py --complete` pass without a What's New, the PR-into-main guard trips, and a wrong `versionCode` is rejected.
- `sync_changelog_to_fastlane.sh` and `check_store_listing.py` (with and without `--complete`) still pass on the current `1.0.0`.
- All three workflow files parse as YAML.
Not run on the real runners yet. The first beta push is the live test of `beta.yaml`, which assumes a mirrored branch push starts a workflow on Gitea, the same way pushes to `main` already do.
### Checklist
- [x] No `versionName` / `versionCode` bump
- [x] No `values-*/strings.xml` touched
- [x] `CHANGELOG.md` not updated: this is release infrastructure, not a user-visible change
Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de>
Reviewed-on: https://codeberg.org/jlmakiola/agendula/pulls/38
First stable release. Merging this bumps versionName to 1.0.0 and triggers the release pipeline (F-Droid, Codeberg, Play).
**App**
- CalDAV sync built in, with Agendula's own task store; OpenTasks / tasks.org stay available and can be copied over in Settings → Storage
- repeating tasks, several reminders per task, lists managed in the app, iCalendar import/export, widget and Quick Settings tile
- a list can be kept out of the smart lists (#18) and gets its own notification channel (#17)
- duplicate a task with its subtasks (#16)
- HTML descriptions shown as plain text (#15)
- relative day words in reminder notifications (#14)
- asks for exact-alarm access instead of claiming USE_EXACT_ALARM, and re-arms reminders when that access changes
**Release plumbing**
- floret-kit bumped to v0.4.0; the old pin was never pushed, so a clean clone couldn't check out the submodule. 0.4.0 drops CrashConfig.issueTitle (crash issues are always filed in English)
- prebuilt .so files ship unstripped, so the build no longer depends on whether an NDK is installed; now checked by check_reproducible_release.sh
- official F-Droid recipe in docs/fdroid-official/, to submit to fdroiddata once v1.0.0 is tagged
- Google Play: fastlane uploads the AAB and every locale's What's New after the F-Droid release; a separate listing lane pushes text and graphics from the fastlane tree, which CI now checks against Play's limits
- store listing: title "Agendula: Tasks" in every locale, icon, feature graphic, screenshots and 1.0.0 changelogs in en-US, en-GB, de-DE and pt-BR
crash_report_issue_title is now unused but stays until Weblate removes the translated copies.
Closes #14, closes #15, closes #16, closes #17, closes #18
Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de>
Reviewed-on: https://codeberg.org/jlmakiola/agendula/pulls/20
Ships the first community translations: German (98%) and Brazilian
Portuguese (100%), both from Weblate, plus the translation infrastructure
and the Codeberg forge move that had accumulated under Unreleased.
A minor bump rather than a patch: two new user-facing languages is a
feature, the way 0.3.0 was cut for Reminders.
versionCode is the derived 400 (MAJOR*10000 + MINOR*100 + PATCH), and
scripts/sync_changelog_to_fastlane.sh generated changelogs/400.txt so the
official F-Droid listing shows this version's "What's New". At 908 chars
it exceeds the script's 500-char advisory and may be truncated in-client
— consistent with past feature releases (0.2.0 ran 1811, 0.3.0 692).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0146mhMLZ5AqJSZkADDc92Jz
The Weblate merge (#8) landed values-de and values-pt-rBR, but
locales_config.xml still listed only `en`. That file is the single source
of truth for which languages the app ships: AppLanguage parses it for the
in-app picker (Settings -> App language) and android:localeConfig points
the system per-app language settings at it. So both translations were
compiled into the APK and reachable by nothing — you could not pick either
one, and only a device already set to de/pt-BR would ever have seen them.
Adds the two BCP-47 tags. Note the tags are `de` and `pt-BR` while the
resource folders are values-de and values-pt-rBR; the `r` is an Android
resource-qualifier artefact and must not appear here.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0146mhMLZ5AqJSZkADDc92Jz
Makes Codeberg canonical for git, issues, PRs, tags and releases. The self-hosted Gitea instance stays build infrastructure: signing key, F-Droid publishing, release pipeline. Ports the setup Calendula already runs on, adapted where Agendula genuinely differs.
**This PR is its own test.** It is the first PR opened on Codeberg, so a green `CI` check proves the new runner works *and* that the submodule resolves from its new home.
### 1 · Workflows split by directory
Forgejo's lookup is first-match-wins across `.forgejo/` → `.gitea/` → `.github/`, and Gitea cannot see `.forgejo/` at all. So each forge sees exactly one set, with no duplicated files and no expression to keep in sync:
| Directory | Runs on | Contains | Secrets |
| --- | --- | --- | --- |
| `.forgejo/workflows/` | Codeberg | `ci.yaml`, `translations.yaml` | **none** |
| `.gitea/workflows/` | Gitea | `release.yaml`, `renovate.yml` | all of them |
The line is drawn at **secrets, not CI-vs-release** — that is what makes fork PRs safe. Renovate deliberately does *not* move despite opening PRs here; it keeps running where its token already lives and merely talks to Codeberg's API.
CI also gains three fixes: an explicit `SKIP_RE` skip-list that names the build-relevant files in the log, base-ref normalisation, and a fully-qualified `android-actions/setup-android` — Codeberg resolves bare `uses:` refs against `data.forgejo.org`, which does not carry that action.
### 2 · Three release-pipeline safety changes
- `detect` and the Renovate job get an explicit `repository_owner` guard. The directory split only holds while `.forgejo/` is non-empty; empty it and Codeberg would fall back to `.gitea/` and start running these on the contributor-facing runner, without secrets.
- `detect` now 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 reports "no tag" for an already-shipped release and would cut it twice. It also now fails on any status other than 200/404 rather than reading a transient error as "no tag": a failed job is recoverable, a duplicate release is not.
- **The Codeberg publish step pushes the tag itself** instead of waiting for it to arrive by mirror. That wait was correct while Gitea mirrored *to* Codeberg; under Codeberg-canonical the mirror runs the other way and it would never resolve. Attaching the release to an already-pushed ref (no `target_commitish`) is what avoids the empty-bodied 500s, and the create call retries with backoff because Codeberg 500s on a tag it has only just received. The step stays **fail-loud**, not `continue-on-error` — it reported green through 0.2.1–0.3.2 while never once publishing, and that must not be possible again.
### 3 · Renovate
`renovate.json5` plus a Gitea-side job targeting Codeberg's API. `managerFilePatterns` covers **both** workflow directories, so the pinned Renovate image tag and the action versions in either file keep getting bumped.
Needs two new Gitea secrets: `RENOVATE_TOKEN` (Codeberg bot, repo read/write + PR scope) and `GITHUB_COM_TOKEN` (read-only github.com PAT, for changelog lookups).
### 4 · Weblate
A parity check (`scripts/check_translations.py`) runs on every PR without a path filter, so the required `Translations` status is always reported. Partial translations are expected, so `MissingTranslation` and `MissingQuantity` become informational — `ExtraTranslation` stays fatal. Agendula had no `lint` block at all, so the first locale to land would otherwise have failed the build.
**Settings → App language** now opens a picker carrying a "Help translate" header. That is why it drops floret-kit's `LanguagePickerRow` for a local row: the shared recipe has no `header` slot, and the framing is app-specific rather than a family primitive.
### 5 · Links repointed
In-app Source / License / report-issue URLs, F-Droid metadata, README (now with a Codeberg CI badge), and the docs.
`floret-kit` follows suit — `.gitmodules` points at `codeberg.org/jlmakiola/floret-kit`, so a clone no longer needs to reach the personal Gitea instance to resolve it. The Gitea copy is **kept**: every existing tag records the old submodule URL, so rebuilds of past releases still resolve.
### 6 · Housekeeping
Drops `release-notes.md` — a release-pipeline scratch file that got committed — and gitignores the five others the release job writes into the workspace.
### Not in this PR
The Codeberg → Gitea push mirror, the Weblate component, and the Codeberg bot account (all browser-side). Until the mirror is flipped, merging this does **not** reach the Gitea runner.
Co-authored-by: Jean-Luc Makiola <business@jeanlucmakiola.de>
Reviewed-on: https://codeberg.org/jlmakiola/agendula/pulls/2
No app-code change from 0.3.1: this exists to re-run the release pipeline
now that the Codeberg publish step is fixed. 0.3.1's APK reached F-Droid but
its Codeberg release 500'd, so anyone installing from Codeberg or Obtainium
is still on the crashing 0.3.0 — cutting 0.3.2 gets them the fix and proves
the workflow fix in the only way that counts, a real release.
Chosen over hand-patching the 0.3.1 release onto Codeberg: same outcome for
users, minus a manual APK upload, and it actually exercises the pipeline.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
R8 kept androidx.work.impl.WorkDatabase_Impl but pruned its no-arg
constructor: nothing calls it directly, Room only reaches it reflectively.
Room then threw InstantiationException, surfacing as "Failed to create an
instance of androidx.work.impl.WorkDatabase". WorkManager builds that
database from a startup ContentProvider, so 0.3.0 died before any of our
code ran — every install, every launch.
We don't depend on WorkManager directly; it arrives via Glance. AGP 9's
stricter R8 is what tipped this over, which is why 0.3.0 was the first
release to hit it.
Keep the Room no-arg constructor, and the ListenableWorker constructor
alongside it — same pruning hazard on the path WorkManager uses to
instantiate workers by name, which would have bitten once a Glance widget
update actually ran.
Reproduced the reporter's stack trace frame-for-frame on a releaseTest
build, then confirmed it launches clean afterwards. Fixes #1.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Ships the M5 reminders onboarding + Settings screen alongside the reworked
overview (Today progress ring, live Upcoming preview, unfurling search) and
the real launcher mark.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The press animation turned the glyph; put it on the scalloped cookie
container instead and make it stronger (scale 1→0.82, rotate 0→40°). The
glyph holds upright via a counter-rotation, so the shape spins while a
magnifier or list icon stays readable. New spinIcon opts a glyph into its own
quarter turn — the settings gear uses it, so it reads as a gear cranking.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Rework the top bar into a custom row with no app title (the launcher icon
already names the app). Settings stays pinned at the right and the search
action sits just left of it; neither moves. Tapping search unfurls a pill
leftward from the magnifier (expandHorizontally anchored at the end) holding
the auto-focused query field and a clear button, with the search icon
remaining as the bar's fixed trailing icon. Results render over the home
content as you type; a blank query leaves the home screen visible. Tapping
the icon again or system back closes search; the FAB hides while searching.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The first pass shifted the mark off-centre. Scale 0.66 about the canvas
centre, and centre the task CARD (not the card+bloom bounding box, which the
overhanging bloom badge drags low): pivot the Y-scale at the card's centre
(y=242.76) and translate +13.24 so the card sits dead-centre, horizontally
and vertically, with the bloom badging out to the lower-right.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The inline search bar was permanently visible. Replace it with a search
action button (a 6-sided cookie shape, sibling to the settings cookie) to
the left of settings: the search bar is absent until tapped, then opens
expanded and auto-focused, covering the home content with live results.
Back arrow or system back closes it; the FAB hides only while searching.
Pass windowInsets = 0 so the bar, already below the app bar, does not
re-apply the status-bar inset and float with a large top gap.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace the placeholder check-mark foreground with the finished agendula
mark (converted from design/icon/agendula_icon.svg): a rounded line-art task
card with a check, plus a small Calendula bloom badge in the open
bottom-right corner — the sibling of Calendula's calendar mark.
Strokes render in Calendula's off-white (#FAF6F0) over agendula's existing
plum background (#7A5C6B, the hue-rotated counterpart of Calendula's slate),
so the two apps read as a family while staying distinct. Scaled 0.66 to
match Calendula's footprint and ~2.8dp stroke weight; reused as the
<monochrome> slot for themed icons.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
makiolaj
Jean-Luc Makiolaandmakiolaj