InsightsRepository in :core:model (SpendTrends, SpendForecast and parts); :data DTOs for E-34/E-35
(round-trip the contract examples, fixing ExampleCoverageTest), InsightsRepositoryImpl through KvittApi and
OnlineGate with category/account names from Room (deleted rows included); FakeInsightsRepository with
scenarios (loaded, empty household, account without history, offline, unreachable, long); the Insights
section below S-11's charts: household trend and per-category trends on SpendLineChart with average and
moving average, a forecast card per account, "Needs connection" / "Could not reach the server" over the last
data of the session. Strings sv/en; 41 goldens.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TQd3pEP9UeN2uifbHknQ7B
A drill-down opened from a breakdown row now shows its own period chip and
Month/Week toggle (initialised from S-11, S-11 keeps its own filter), the
header total, a line chart of the group's spend per bucket with the dashed
range average, the trailing 3-bucket moving average and a legend, and a
caption comparing the latest full bucket with the average; the expense rows
follow. Bucket drill-downs are unchanged. The series is the group's column of
AnalyticsState.buckets from spendSeries, so no contract change.
:core:design chart: SpendLineChart and LineLegend beside StackedBarChart,
shared axis/slot helpers (bar chart pixels unchanged), AnalyticsFilterBar
measure and group-by toggles optional.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TQd3pEP9UeN2uifbHknQ7B
Eight modules each had a copy of the same configuration-context harness (ScreenSnapshots/Snapshots); M12f had to fix
the same capture bug in two of them. :core:design's test fixtures (src/testFixtures, package
se.hagstedt.kvitt.design.testing) now hold Variant, TestClock, TestEnvironment, ScreenSnapshots with its three
capture modes (Root; Screen, with the hand-advanced frame clock; ScreenWhenIdle), MainDispatcherRule and withMain.
Test fixtures rather than a :core:testing module: :core:design's own tests can use them without a dependency on
itself, no module is added to the §6.2 graph, and lint treats them as test sources. The architecture check rejects
test fixtures outside test configurations. Modules keep only their own frame (analytics household, settle size and
inline overlays, settings avatar pictures). Every golden is unchanged.
kvitt.android.screenshots applies Roborazzi with outputDir src/test/screenshots and adds the screenshot test
libraries; the feature plugin applies it and adds the test fixtures, so the seven feature build scripts lose their
copied blocks. Removed: roborazzi-junit-rule and roborazzi-compose (never imported), the explicit
lifecycle-viewmodel-compose (feature plugin, :app) and hilt-lifecycle-viewmodel-compose in :app (both still arrive
through the features; release and debug runtime classpaths are identical), and the unused catalog entries
androidx-hilt-navigation-compose and androidx-lifecycle-viewmodel-compose. @Before/@After setMain pairs in nine test
classes use MainDispatcherRule.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TQd3pEP9UeN2uifbHknQ7B
The task class and its wiring were duplicated verbatim in both build scripts; bundleCategoryKeywords(resourcePath)
registers it once per module. Resource paths and inputs are unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TQd3pEP9UeN2uifbHknQ7B
Drop the en_dark variant from every screenshot helper and render every
golden at mdpi (Robolectric qualifier w411dp-h<height>dp-mdpi instead of
xhdpi/xxhdpi), so a 411 x 891 dp screen is 411 x 891 px. All goldens
deleted and re-recorded; 1143 -> 865 PNGs. README describes the layout.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TQd3pEP9UeN2uifbHknQ7B
Server: E-3 invite_kind; E-12 link ?from=<admin name>, E-4 passes it into kvitt://invite;
X-Kvitt-App-Version-Name stored in devices.app_version_name (migration 0003, schema 3);
OpenAPI and examples follow. Phone: :core:model, :core:fakes, :data, features and :app
updated together; goldens re-recorded for S-20 (version name) and new S-1/S-22 states.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TQd3pEP9UeN2uifbHknQ7B
Bugs: viewerPercentOf reads basis points (Samples in bp); FullSplitPicker and the
split labels use editor_split_full_on_you ("Allt på dig"); AmountInput.of restores
money with both decimals (AmountInput.percent for ratio fields), replacing M7c's
moneyInput; quickRowCategories does the D-060 fill itself; KvittScaffold resolves
resource-backed snackbars (showKvittSnackbar) while shown.
Gaps: ConfirmDialog optional and styled body; KvittButton loading form (M7a uses it,
ProgressBottomButton removed); InviteCodeField wraps after a group, placeholder,
imeAction; EmptyState icon tile and actionEnabled; DateFormatter.monthName and
dayRangeLong; AnalyticsFilterBar periodOwnRow; segmentedItem builder; screen tokens
in KvittDimens (OnboardingDimens removed); common_ok, common_undo; module-proposed
keys recorded in strings.csv; quick-row labels end in an ellipsis on line two.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TQd3pEP9UeN2uifbHknQ7B
Gradle project in android/ with every module of SPEC §6.2 (build-logic convention plugins, version catalog,
architecture check in check), :core:model types and repository interfaces of §14.2.4 and §14.5, :domain
implementing §14.5.2 and §14.5.4 (125 vector cases), scenario-driven fakes with the empty scenario, Room 3
entities and KvittDatabase v1 with exported schema, :app shell with Hilt, en/sv strings with MissingTranslation
as a lint error, Roborazzi goldens of the empty screen in light and dark.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
jacobh2