Commit graph

6 commits

Author SHA1 Message Date
dc8353a9e9
test: coverage round 4 (84.39% → 84.44%) + 6 new property tests
30 unit tests + 6 property tests across two parallel agents.

- ContactLive.Show + HrrrNativeClient + NexradClient: sort_observations
  + sort_soundings actual ordering per field, closest_observations
  capped-at-5 proximity, nil-pos2 half_dist=0 path, HrrrNativeClient
  scrambled-level density + surface finiteness, NexradClient cache
  population + zero-byte + year-boundary URL rounding. Properties:
  sort_observations preservation, haversine symmetry, build_native
  surface_temp_k finiteness.

- PathLive + Viewshed + GefsFetchWorker + SnmpClient: GPS source
  URL preservation, QRZ 404 surfacing, propagation_updated same-midpoint
  no-op; Viewshed effective_reach_km BLOCKED boundaries + find_reach_km
  zero max_range; GefsFetchWorker 502/400/410 + wind_u/wind_v aliases
  + nil profile; SnmpClient fully-qualified OID + double-dot drop +
  empty poll + unknown radio type. Properties: destination_point
  round-trip < 0.5 km, valid_time = run_time + fh*3600 invariant,
  parse_snmpget_output totality.
2026-04-24 10:32:05 -05:00
e11ebc9af8
test: coverage round 1 (80.45% → 82.05%) + 12 new property tests
Adds 67 test cases + 12 property tests across four surfaces — parallel
agents targeted the lowest-coverage files in the tree.

- ContactLive.Show (48.56% → 72.01%): 16 unit tests + 2 properties
  covering IEMRE/radar/ducting/propagation-analysis render branches,
  every terrain-verdict class, expanded HRRR/terrain sections, pending-
  edit banner, internal-network conn, and total-over-fields properties
  for obs + sounding sort handlers.

- HrrrNativeClient (33% → 58%): 8 unit tests + 4 properties covering
  partial-surface fallback, empty input, optional-var nils, bogus
  binary dispatch, and level-count + monotonic-height + URL-encoding
  + duct-subset invariants.

- NexradClient / HrrrClient / GefsClient: 13 unit tests + 4 properties
  covering 500/404/timeout paths, malformed idx bodies, parse_idx
  totality, byte-range invariants, and URL-builder round-trips.

- HrrrBackfill / HrrrNativeDeriveFields / ImportContestLogs: 7 unit
  tests + 1 property. Seeds contacts + native profiles, stubs HRRR
  with Req.Test, exercises the `--limit 0` and `filter_points_needing_
  backfill` paths, CSV import happy + malformed paths, and a
  band-string round-trip property.

2599 → 2666 tests, 170 → 182 properties, 0 failures.
2026-04-24 10:15:37 -05:00
5564b6703d
test: push line coverage from 78.67% → 80.45%
Adds 30 more test cases (2594 → 2624 tests) targeting the last
low-coverage hot spots.

Notable additions:
- ContactLive.Show fully-hydrated render path (seeds HrrrProfile,
  TerrainProfile, ContactCommonVolumeRadar, Sounding, SurfaceObservation,
  IemreObservation) exercises refresh_computed, build_propagation_analysis,
  build_data_sources, compute_elevation_profile, and the downstream
  render branches they unlock.
- handle_info :terrain_ready and :sounding_fetched paths for same-contact
  and unrelated-contact dispatch.
- Flagged-invalid badge + line-through header style.
- Owner-label button path (Edit vs Suggest Edit).
- NotifyListener init/1 subscribes on schedule; handle_info(:subscribe)
  succeeds against the live Repo.
- RecalibrateScorer runs cleanly against an empty corpus via the
  Recalibrator insufficient-data fallback.
- HrrrNativeClient.extract_native_profiles/2 tolerates malformed input
  without crashing.
2026-04-24 08:27:18 -05:00
764643bc62
test: expand coverage and refactor to idiomatic style
Changes under lib/ (source):
- iemre_fetch_worker: replace `if transient_failure?/1` boolean predicate
  + `if/else` with a classifier that pattern-matches the error reason
  directly, returning `:transient | :permanent`, and dispatch via
  `case`. Three-clause head is clearer than the boolean predicate.
- scores_file.extract_points and nexrad_client.extract_box: force `//1`
  step on the row/col comprehensions so an image box whose centre
  lands outside the raster collapses to an empty iteration instead
  of firing Elixir 1.19's ambiguous-range warning. Adds a regression
  test for the right-edge box.

Test coverage added:
- Feature wrappers in Backtest.Features (theta_e_jump, shear_at_top,
  duct_thickness, best_duct_freq, duct_usable_for_band,
  distance_to_front / parallel_to_front stubs, all_features/0).
- NotifyListener handle_info tolerance for malformed payloads +
  unknown messages, plus the PubSub broadcast side effect.
- Viewshed.effective_reach_km across every verdict / diffraction
  tier / ducting-score tier combination.
- SnmpClient parse_af60_output unit conversions + unknown-column
  handling; parse_snmpget_output OID normalisation and junk-line
  tolerance.
- HrrrNativeClient native_level_count, native_variables,
  native_messages shape, duct_messages / duct_byte_ranges.
- Changeset coverage for GeomagneticObservation, SolarFluxObservation,
  SolarXrayObservation, Ionosphere.Observation, HrrrClimatology, and
  Metar5minObservation — lifted each from ~50% / 0% to 100%.
- Property tests: GefsClient.dewpoint_from_rh invariants (Td ≤ T,
  monotonic in RH, finite across the operating envelope) + idempotent
  nearest_run. NexradClient.pixel_to_dbz monotonicity + bounded
  output; latlon_to_pixel round-trip for every grid cell.

Also cleared several unrelated compiler/linter warnings:
- Three private test helpers (create_contact, insert_contact,
  current_hour) had `\\ %{}` / `\\ 0` defaults that no caller used.
- profiles_file_test bound `dir` in a pattern match without using it.

Suite: 2,359 tests + 146 properties pass (was 2,300 / 139); coverage
70.38% → 70.89%; credo strict clean.
2026-04-23 13:28:42 -05:00
e7a7ae073d Phase 9.3, 9.4, and Phase 3 NEXRAD pipeline
Task 9.3 - Weight recalibration via gradient descent:
- Recalibrator module fits logistic regression weights using Nx
- Trains on QSO positives vs random baseline negatives
- Cross-validates by month, normalizes weights to sum to 1.0
- Mix task: mix recalibrate_scorer --sample 5000 --epochs 2000

Task 9.4 - Side-by-side scorer comparison:
- ScorerDiff.compare/3 re-scores grid with old vs new weights
- Reports mean diff, regressions, improvements, per-band breakdown
- Mix task: mix scorer_diff --new-weights '{...}'

Phase 3 - NEXRAD ingestion pipeline:
- NexradClient fetches IEM n0q composite PNGs, extracts per-point
  box statistics (mean/max dBZ, texture variance)
- NexradObservation schema with unique (lat, lon, observed_at)
- NexradWorker on :nexrad queue for background processing
- nexrad_texture backtest feature in Features module
- mix nexrad_backfill --limit 200

All tasks added to AdminTaskWorker and Release for production use.
1116 tests, 0 failures.
2026-04-10 12:48:36 -05:00
900685aa06 Phase 1 tasks 1.1-1.5: HRRR native hybrid-sigma ingestion
- Spike docs at docs/research/hrrr_native_levels.md confirming files
  are on AWS S3 for 5+ years, 50 hybrid levels, and include TKE and
  SPFH needed for Phase 2 turbulence features. Architectural finding:
  per-point on-demand fetching is impractical (~530 MB/file), so
  the ingestion worker batches per (date, hour) instead.
- hrrr_native_profiles schema: arrays per level plus cached surface
  scalars and placeholder columns for Phase 2/4 derived fields.
  Strictly additive — the existing hrrr_profiles table is untouched.
- HrrrNativeClient: pure URL/message-list helpers, build_native_profile/1
  that turns a parsed wgrib2 map into the schema shape (TDD'd).
- Exposed HrrrClient.download_grib_ranges/2 so the native client
  reuses the existing parallel byte-range download + disk cache.
- HrrrNativeGridWorker: Oban worker keyed on {year, month, day, hour},
  unique at :infinity, pulls distinct (lat, lon) points from contacts
  in the ±30 min window, downloads the native grib2, extracts per
  point, bulk-upserts.
- mix hrrr_native_backfill --limit N enqueues the top-N hours by
  contact count.

Phase 1 gate still pending Task 1.6 (sanity-check backtest after
live data lands).
2026-04-09 16:23:51 -05:00