Files

T

timothyandClaude Opus 4.8 7e6b9d2747 test(77): characterize clock-boundary schedule padding; docs

#77's core (pad/snap schedules to :00/:15/:30 via filler) already exists —
FillerMode.Pad + PadToNearestMinute (Classic), pad_to_next/pad_until
(Sequential); Block is inherently time-anchored. No production code change;
this locks the behaviour end-to-end and documents that it exists.

- PlayoutBuildGoldenTests.Classic_clock_padded: a PostRoll FillerMode.Pad(15)
  preset through the real PlayoutBuilder snaps content to :15 (golden +
  explicit quarter-hour assertion). Splits Verify -> CompareGolden for reuse.
- ChannelGuideProjectorClockPadTests: the guide projection coalesces trailing
  filler so programmes STOP on the padded boundary (XMLTV half).
- docs: decisions.md (2026-07-17 entry), domain-model.md (clock-boundary row),
  testing.md (test map + count 540->542).

Deferred (UI, blocked on #388): one-click per-channel/schedule clock-align
toggle + 60-min increment option.

Refs #77

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

2026-07-17 01:22:07 +02:00

5.2 KiB

Raw Blame History

Testing map

Purpose: authoritative map of what each test project/suite covers and how to run it. Read this before adding tests, not just docs/contributing.md §8 (which now just points here).

Test projects

Project	Covers	Notes
`ErsatzTV.Tests`	API controllers + MediatR handlers	In-memory SQLite fixture: a shared `SqliteConnection("Data Source=:memory:;Foreign Keys=False")` kept open + `EnsureCreatedAsync()` (not full migration replay) + `PRAGMA foreign_keys=OFF`, then seed; a tiny `IDbContextFactory` wraps `new TvContext(...)`. 828 tests currently.
`ErsatzTV.Core.Tests`	Domain logic, scheduling, IPTV/XMLTV generation	References `ErsatzTV.Application` directly — there is no separate `Application.Tests` project. 542 tests + 1 skipped under `TZ=UTC` (the Block playout golden additionally skips under a non-UTC `TZ`; see Golden-file nets).
`ErsatzTV.Architecture.Tests`	Layering rules via NetArchTest.eNhancedEdition	Core↛Infra/App/EF; FFmpeg↛all; App↛concrete providers. 5 tests. See `docs/contributing.md` §1.
`ErsatzTV.FFmpeg.Tests`	FFmpeg command construction	Build a pipeline, assert the exact rendered arg string (`PipelineBuilderBaseTests.cs`).
`web/` (vitest)	React SPA unit tests	330 tests; run alongside typecheck + build (see below).

Golden-file nets

Three golden-file suites guard the highest-value, most-subtle output:

M3U: ErsatzTV.Core.Tests/Iptv/ChannelPlaylistGoldenTests.cs (ersatztv#11) — env var ETV_UPDATE_GOLDENS
XMLTV: ChannelGuideGoldenTests (ersatztv#28) — env var ETV_UPDATE_GOLDENS
Playout build: ErsatzTV.Core.Tests/Scheduling/Goldens/PlayoutBuildGoldenTests.cs (ersatztv#163) — env var ETV_UPDATE_PLAYOUT_GOLDENS (deliberately separate from ETV_UPDATE_GOLDENS so regenerating one net can't silently rewrite the other). Snapshots the PlayoutItems each builder produces over a pinned build window. Covers the Classic (PlaybackOrder.Chronological) and Block builders; Sequential (YAML) + Scripted are tracked in ersatztv#381. A third case, Classic_clock_padded (ersatztv#77), locks clock-boundary padding: a FillerMode.Pad + PadToNearestMinute=15 PostRoll preset snaps content to :15, and the test both goldens the output and asserts every content item after the first starts on a quarter-hour. Its EPG counterpart is ErsatzTV.Core.Tests/Channels/ChannelGuideProjectorClockPadTests.cs (guide programmes stop on the padded boundary). The build reads no wall clock — time enters only via the caller-supplied start — so a pinned start is fully deterministic. Snapshots the raw PlayoutItem.Start/Finish (UTC), not the *Offset properties (those call .ToLocalTime() and would make the golden machine-TZ dependent). The Block case is TZ-sensitive by construction (BlockPlayoutBuilder maps template times via TimeZoneInfo.Local), so it is guarded with Assume.That(TimeZoneInfo.Local.BaseUtcOffset == Zero): it runs under TZ=UTC (CI) and reports inconclusive (a graceful skip, not a failure) under any other TZ. A real TZ seam for the block builder is ersatztv#380's scope.

All three locate their golden files via [CallerFilePath]. A missing golden is a hard fail, not a skip. Regenerate via ETV_UPDATE_GOLDENS=1 dotnet test ... (M3U/XMLTV) or ETV_UPDATE_PLAYOUT_GOLDENS=1 dotnet test ... (playout build).

Never set ETV_UPDATE_GOLDENS / ETV_UPDATE_PLAYOUT_GOLDENS in CI or from an agent. A golden diff during normal test runs means the code changed the output — regenerating to make the diff go away hides the change instead of surfacing it. Only a human who has confirmed the change is intentional should regenerate.

Timezone independence

The suite is timezone-independent (ersatztv#24). When constructing test PlayoutItems, always set a real Start (e.g. startState.CurrentTime.UtcDateTime) — never rely on the default DateTime.MinValue, which underflows DateTimeOffset.MinValue once a non-UTC local offset is applied (StartOffset calls ToLocalTime()). CI runs in UTC; local runs may not.

Running tests

Full .NET gate:

dotnet build ErsatzTV.sln
TZ=UTC dotnet test

CI adds --blame-hang-timeout 2m to catch hangs.

Fast subsets:

# single project
dotnet test ErsatzTV.Tests

# filtered
dotnet test ErsatzTV.Core.Tests --filter FullyQualifiedName~ChannelPlaylistGoldenTests

Web (web/):

npm test           # vitest
npm run typecheck  # tsc -b --pretty false
npm run lint       # eslint .
npm run build      # tsc -b && vite build

Per-PR verification gate

Before opening a PR: build the solution, run both .NET test projects (plus ErsatzTV.Architecture.Tests and ErsatzTV.FFmpeg.Tests if touched), and run the web test/lint/ typecheck/build steps above. All must be green. A golden-file diff or an architecture-test failure is a hard stop — fix the code, don't regenerate/relax the test.

5.2 KiB Raw Blame History