Return the main state of the sensor (e.g., '10:30' or '10:30 +5'). Calculates the state from the first entry in the filtered data.

Return the current watchdog state. Shows the name and delay of the previous station for the next train.

Return the number of minutes until the user must leave. Returns 'Leave now!' if the walk time has already passed.

Return the calculated punctuality percentage.

Return current timezone-aware datetime.

Parse ISO datetime string to timezone-aware datetime.

Patch DataUpdateCoordinator to prevent background tasks and simplify tests.

Return a minimal real coordinator with departure_history for _update_history tests.

Return the paths to sensor files.

Set up frame helper for the test session.

JSON serializer for objects not serializable by default json code.

Return the path to strings.json.

Test that all translation keys used in code are present in strings.json.

Test direct station match (200 OK) via JSON API.

Test that station names ending with a dot are encoded as %2E.

Test graceful failure when all lookups fail.

Test fallback to HTML parsing when JSON fails or returns 300 with HTML.

Test successful station candidate fetching via JSON API.

Test downloading stations when not cached.

Test loading stations from persistent cache.

Test calendar event generation with duration, walk time, filtering and links.

Test that all keys used in config_flow.py are in strings.json.

Test that the config flow details schema contains the paused option.

Test that 429 responses throttle the next local update tick.

Test that 429 exception (aiohttp.ClientResponseError) throttled updates.

Test alternative connections are generated for same-destination trains.

Test that data source mapping works.

Test excluding cancelled trains.

Test that departures are filtered by favorite trains.

Test that the coordinator handles API errors gracefully without crashing.

Test handling of invalid JSON response.

Test that throttled local ticks do not fetch again without raw cache data.

Test parsing occupancy data.

Test that options are correctly correctly added to the URL.

Test that updates can be paused in the coordinator.

Test platform change detection.

Test QoS parsing and facilities extraction.

Test retry logic on transient failures.

Test that it returns cached data after max retries fail.

Test route details parsing.

Test trip ID parsing.

Test updating data.

Test correctly encoding of station and via parameters.

Test wagon order and sector extraction.

Test that next_departures_text correctly formats with custom template.

Test deduplication with custom key for KVV (as requested in #116).

Test deduplication with default key (DB IRIS style).

Test deduplication fallback when key resolving yields empty string.

Test that the elevator binary sensor publishes defective facilities.

Main test entry point for link validation.

Test consistency for extra translation sections like repairs, train_types.

Test the matching logic (exact, starts-with, contains, fuzzy).

Test that validating the user input works and creates an entry (multi-step).

Test that validating advanced user input works, including excluded_directions.

Test the flow when multiple station matches are found.

Test manual entry override when no matches found.

Test error keys returned when official or FaserF servers are unreachable.

Test we get the form.

Test a full lifecycle of editing multiple option pages.

Test Hass.io discovery flow when addon is already installed.

Test Hass.io discovery flow when addon is not installed.

Test that special characters in station names don't crash or cause injection-like issues.

Test handling of very large responses (DoS protection simulation).

Test that the leave now sensor returns a numeric value when not yet time to leave.

Test 0 state when time is up.

Test sensor with no data.

Test that the leave now sensor returns a numeric value (0) instead of a string.

Test the calculation logic of the Leave Now sensor.

Test binary sensors attributes with live data.

Test that calendar event generation matches live coordinator configuration.

Test that the live data contains required fields and station messages.

Test various filtering features against the live API response.

Test that sensor attributes and states map correctly with live data.

Test manual entry path.

Verify that markdown files have proper spacing between elements.

Test that next_departures_text correctly picks up time from diverse keys.

Test options flow menu.

Test that the options flow general options schema contains the paused option.

Test extraction of station links from various HTML patterns including query-based and provider-specific ones.

Test parsing with irrelevant content.

Test that departures in the past are filtered out.

Verify that different platforms result in unique fetch URLs for caching.

Test the statistics calculation logic.

Test statistics when history is empty.