prop/lib/microwaveprop/workers/propagation_grid_worker.ex
Graham McIntire cd7f2fc2b8
feat(propagation): Phase 3 Stream A cutover — Rust owns f00..f18
The hourly cron now only seeds grid_tasks. The chain step, native-duct
merge, NEXRAD merge, commercial-link merge, scoring, ProfilesFile
write, and band-score writes all moved to rust/prop_grid_rs.

Elixir changes:
- GridTaskEnqueuer.seed_with_analysis/1: inserts 1 kind='analysis' row
  (f00) + 18 kind='forecast' rows (f01..f18).
- PropagationGridWorker: stripped from 423 LOC to a thin seeder.
  perform(%{}) → GridTaskEnqueuer.seed_with_analysis.
  Deleted: process_forecast_hour, merge_native_duct_data,
  merge_nexrad_data, merge_commercial_link_data, compute_scores_*,
  persist_profiles, record_run_timing (Rust emits spans to Prometheus
  instead), apply_nexrad_observations, apply_duct_grid, timed helpers.
  Test rewritten for the new shape: 0 Oban fan-out jobs, 19 grid_tasks
  rows with the expected kind distribution.

HrrrNativeClient and NexradClient remain — they have other callers
(HrrrNativeGridWorker for per-QSO duct batch; NexradWorker and
CommonVolumeRadarWorker for per-contact radar). Only f00's direct
use moved.
2026-04-19 18:13:41 -05:00

51 lines
2 KiB
Elixir

defmodule Microwaveprop.Workers.PropagationGridWorker do
@moduledoc """
Hourly seed worker for the Rust `prop-grid-rs` chain.
Post-Phase-3-cutover, Elixir no longer runs any fetch/decode/score
work for the propagation grid. The hourly cron fires this worker
with empty args, and it inserts 19 `grid_tasks` rows (1 analysis
f00 + 18 forecast f01..f18) for Rust to drain. Rust owns everything
from there: HRRR fetch, wgrib2 decode, native-level duct merge,
NEXRAD composite, commercial-link degradation, band scoring, and
both the ProfilesFile (MessagePack) and the per-band score files.
The old chain-step perform/2 clauses and the merge/score helpers
moved to `rust/prop_grid_rs/src/pipeline.rs`. `Propagation.record_run_timing`
is still called from this module when the Rust worker reports a
chain step, through the PropagationNotifyListener path.
"""
use Oban.Worker,
queue: :propagation,
priority: 0,
max_attempts: 5,
# Deduplicate seed jobs in a 1-hour window. The hourly cron fires
# `%{}` args; unique protects against FreshnessMonitor re-fires
# during an outage stacking multiple chains for the same run.
unique: [period: 3600, states: [:available, :scheduled, :executing, :retryable]]
alias Microwaveprop.Propagation.GridTaskEnqueuer
alias Microwaveprop.Weather.HrrrClient
require Logger
@impl Oban.Worker
def perform(%Oban.Job{args: args}) when args == %{} do
seed_chain()
end
defp seed_chain do
# HRRR takes ~45min to publish after the hour. Use 2 hours ago to
# ensure availability.
two_hours_ago = DateTime.add(DateTime.utc_now(), -2, :hour)
run_time = two_hours_ago |> HrrrClient.nearest_hrrr_hour() |> DateTime.truncate(:second)
Logger.info("PropagationGrid: seeding chain run_time=#{run_time} (f00 analysis + f01-f18 forecasts via grid_tasks)")
case GridTaskEnqueuer.seed_with_analysis(run_time) do
{:ok, _count} -> :ok
{:error, reason} -> {:error, reason}
end
end
end