prop/lib/microwaveprop/weather/hrrr.ex

defmodule Microwaveprop.Weather.Hrrr do
  @moduledoc false

  import Ecto.Query

  alias Ecto.UUID
  alias Microwaveprop.Propagation.Grid
  alias Microwaveprop.Radio
  alias Microwaveprop.Radio.Contact
  alias Microwaveprop.Repo
  alias Microwaveprop.Weather.HrrrNativeProfile
  alias Microwaveprop.Weather.HrrrProfile
  alias Microwaveprop.Weather.Narr
  alias Microwaveprop.Weather.NarrProfile
  alias Microwaveprop.Weather.ProfileLookup
  alias Microwaveprop.Weather.SoundingParams

  # ── Delegates to ProfileLookup ──

  defdelegate find_nearest_hrrr(lat, lon, timestamp), to: ProfileLookup
  defdelegate hrrr_profiles_for_path(contact), to: ProfileLookup
  defdelegate hrrr_profiles_for_contacts(contacts), to: ProfileLookup
  defdelegate find_nearest_native_profile(lat, lon, timestamp), to: ProfileLookup
  defdelegate best_profile_for_contact(contact), to: ProfileLookup
  defdelegate hrrr_for_contact(contact), to: ProfileLookup
  defdelegate hrrr_data_fully_present?(contact), to: ProfileLookup
  defdelegate has_hrrr_profile?(lat, lon, valid_time), to: ProfileLookup
  defdelegate hrrr_points_present_batch(points), to: ProfileLookup
  defdelegate round_to_hrrr_grid(lat, lon), to: ProfileLookup
  defdelegate purge_grid_point_profiles(), to: ProfileLookup

  # ── HRRR profile persistence ──

  @spec upsert_hrrr_profile(map()) :: {:ok, HrrrProfile.t()} | {:error, Ecto.Changeset.t()}
  def upsert_hrrr_profile(attrs) do
    %HrrrProfile{}
    |> HrrrProfile.changeset(attrs)
    |> Repo.insert(
      on_conflict: :nothing,
      conflict_target: [:lat, :lon, :valid_time]
    )
  end

  @spec upsert_hrrr_profiles_batch([map()], keyword()) :: {non_neg_integer(), nil}
  def upsert_hrrr_profiles_batch(profiles, _opts \\ []) do
    Microwaveprop.Instrument.span(
      [:db, :upsert_hrrr_profiles],
      %{count: length(profiles)},
      fn -> do_upsert_hrrr_profiles_batch(profiles) end
    )
  end

  defp do_upsert_hrrr_profiles_batch(profiles) do
    now = DateTime.truncate(DateTime.utc_now(), :second)

    profiles
    |> Enum.chunk_every(500)
    |> Enum.reduce({0, nil}, fn chunk, {total_count, _} ->
      entries =
        Enum.map(chunk, fn attrs ->
          is_gp =
            rem(round(attrs.lat * 1000), 125) == 0 and
              rem(round(attrs.lon * 1000), 125) == 0

          Map.merge(attrs, %{
            id: UUID.generate(),
            is_grid_point: is_gp,
            inserted_at: now,
            updated_at: now
          })
        end)

      {count, rows} =
        Repo.insert_all(HrrrProfile, entries,
          on_conflict: :nothing,
          conflict_target: [:lat, :lon, :valid_time]
        )

      {total_count + count, rows}
    end)
  end

  # ── Native duct info ──

  @doc """
  Returns just `hrrr_native_profiles.best_duct_band_ghz` near the
  given cell. Convenience wrapper around `nearest_native_duct_info/3`
  for callers that don't need Richardson.
  """
  @spec nearest_native_duct_ghz(float(), float(), DateTime.t()) :: float() | nil
  def nearest_native_duct_ghz(lat, lon, %DateTime{} = timestamp) do
    case nearest_native_duct_info(lat, lon, timestamp) do
      {:ok, %{best_duct_band_ghz: ghz}} -> ghz
      _ -> nil
    end
  end

  @doc """
  Returns `{:ok, %{best_duct_band_ghz: ghz, bulk_richardson: r}}` for
  the nearest `hrrr_native_profiles` cell to (`lat`, `lon`) at
  `timestamp` within ±0.07° / ±1h, or `{:error, :not_found}`.

  Richardson gates the 1.15× refractivity boost in `Scorer` — a duct
  band reading under turbulent conditions (high Richardson) is likely
  to be mixed out before it supports the target path. The pair is
  cheap to fetch together, so callers that score a cell prefer this
  over the bare `nearest_native_duct_ghz/3`.
  """
  @spec nearest_native_duct_info(float(), float(), DateTime.t()) ::
          {:ok, %{best_duct_band_ghz: float() | nil, bulk_richardson: float() | nil}}
          | {:error, :not_found}
  def nearest_native_duct_info(lat, lon, %DateTime{} = timestamp) do
    dlat = 0.07
    dlon = 0.07
    time_start = DateTime.add(timestamp, -3600, :second)
    time_end = DateTime.add(timestamp, 3600, :second)

    from(h in HrrrNativeProfile,
      where:
        h.lat >= ^(lat - dlat) and h.lat <= ^(lat + dlat) and
          h.lon >= ^(lon - dlon) and h.lon <= ^(lon + dlon) and
          h.valid_time >= ^time_start and h.valid_time <= ^time_end,
      order_by:
        fragment(
          "ABS(? - ?) + ABS(? - ?) + ABS(EXTRACT(EPOCH FROM ? - ?))",
          h.lat,
          ^lat,
          h.lon,
          ^lon,
          h.valid_time,
          ^timestamp
        ),
      limit: 1,
      select: %{best_duct_band_ghz: h.best_duct_band_ghz, bulk_richardson: h.bulk_richardson}
    )
    |> Repo.one()
    |> case do
      nil -> {:error, :not_found}
      row -> {:ok, row}
    end
  end

  # ── Reconciliation ──

  @doc """
  Reconcile `contacts.hrrr_status` for every `:queued` contact:

    * flips to `:complete` when every path point has an hrrr_profiles
      row at the rounded HRRR hour;
    * flips to `:unavailable` when any path point falls outside the
      HRRR CONUS grid (pos2 or the great-circle midpoint can drift
      into Canada, the mid-Atlantic, or the Pacific even when pos1 is
      stateside — the Rust hrrr-point-worker returns no profile for
      those points and the contact would otherwise sit `:queued`
      forever).

  Returns `{:ok, n}` where `n` is the total number of contacts advanced
  (completed + oconus).
  """
  @spec reconcile_hrrr_statuses() :: {:ok, non_neg_integer()}
  def reconcile_hrrr_statuses do
    queued =
      Contact
      |> where([c], c.hrrr_status == :queued and not is_nil(c.pos1))
      |> Repo.all()

    {oconus, rest} = Enum.split_with(queued, &contact_has_oconus_path_point?/1)
    to_complete = Enum.filter(rest, &hrrr_data_fully_present?/1)

    _ =
      if oconus != [] do
        ids = Enum.map(oconus, & &1.id)
        Radio.set_enrichment_status!(ids, :hrrr_status, :unavailable)
      end

    _ =
      if to_complete != [] do
        ids = Enum.map(to_complete, & &1.id)
        Radio.set_enrichment_status!(ids, :hrrr_status, :complete)
      end

    {:ok, length(oconus) + length(to_complete)}
  end

  defp contact_has_oconus_path_point?(contact) do
    contact
    |> Radio.contact_path_points()
    |> Enum.any?(fn {lat, lon} ->
      not Grid.contains?(%{"lat" => lat, "lon" => lon})
    end)
  end

  # ── Backfill scalars ──

  @doc """
  Derive `surface_refractivity` / `min_refractivity_gradient` /
  `ducting_detected` for every `hrrr_profiles` row where those columns
  are NULL but the pressure-level `profile` JSONB is populated.
  Required one-off after the Rust hrrr_points worker shipped without
  deriving these scalars — existing rows have the raw levels but the
  DB scalars are NULL. Idempotent; rows with scalars already set are
  skipped by the WHERE clause.

  Batches by `limit` rows per pass to keep transactions short on the
  Turing Pi 2 Postgres node. Returns the total number of rows advanced.
  """
  @spec backfill_hrrr_scalars(keyword()) :: non_neg_integer()
  def backfill_hrrr_scalars(opts \\ []) do
    batch = Keyword.get(opts, :batch_size, 500)
    max_batches = Keyword.get(opts, :max_batches, 1_000)

    Enum.reduce_while(1..max_batches, 0, fn _, total ->
      case backfill_hrrr_batch(batch) do
        0 -> {:halt, total}
        n -> {:cont, total + n}
      end
    end)
  end

  defp backfill_hrrr_batch(limit) do
    rows =
      HrrrProfile
      |> where([h], is_nil(h.surface_refractivity))
      |> limit(^limit)
      |> select([h], %{id: h.id, profile: h.profile})
      |> Repo.all()

    updates =
      Enum.flat_map(rows, fn %{id: id, profile: profile} ->
        case SoundingParams.derive(profile || []) do
          %{} = derived ->
            [
              %{
                id: id,
                surface_refractivity: Map.get(derived, :surface_refractivity),
                min_refractivity_gradient: Map.get(derived, :min_refractivity_gradient),
                ducting_detected: Map.get(derived, :ducting_detected, false),
                duct_characteristics: Map.get(derived, :duct_characteristics)
              }
            ]

          _ ->
            []
        end
      end)

    now = DateTime.truncate(DateTime.utc_now(), :second)

    _ =
      if updates != [] do
        {values_sql, params} =
          updates
          |> Enum.with_index()
          |> Enum.reduce({"", []}, &build_values_row(&1, &2, now))

        Repo.query!(
          "UPDATE hrrr_profiles AS h " <>
            "SET surface_refractivity = v.surface_refractivity, " <>
            "min_refractivity_gradient = v.min_refractivity_gradient, " <>
            "ducting_detected = v.ducting_detected, " <>
            "duct_characteristics = v.duct_characteristics, " <>
            "updated_at = v.updated_at " <>
            "FROM (VALUES #{values_sql}) " <>
            "AS v(id, surface_refractivity, min_refractivity_gradient, " <>
            "ducting_detected, duct_characteristics, updated_at) " <>
            "WHERE h.id = v.id",
          params
        )
      end

    length(updates)
  end

  defp build_values_row({u, _idx}, {sql, params}, now) do
    base = Enum.count(params)
    frag = "($#{base + 1}::uuid, $#{base + 2}, $#{base + 3}, $#{base + 4}, $#{base + 5}, $#{base + 6})"
    sep = if sql == "", do: "", else: ", "

    params =
      params ++
        [
          UUID.dump!(u.id),
          u.surface_refractivity,
          u.min_refractivity_gradient,
          u.ducting_detected,
          u.duct_characteristics || [],
          now
        ]

    {sql <> sep <> frag, params}
  end

  # ── Path profiles ──

  @doc "Find all atmospheric profiles along a contact's path, from any source."
  @spec profiles_along_path(map()) :: [HrrrProfile.t() | NarrProfile.t()]
  def profiles_along_path(contact) do
    hrrr_path = hrrr_profiles_for_path(contact)

    if hrrr_path == [] do
      Narr.narr_profiles_for_path(contact)
    else
      hrrr_path
    end
  end
end