zhixing-system/zhixing-server/src/zhixing_server/modules/selection/application/run.py

"""Application orchestration for persisted whole-universe B1 runs."""

from __future__ import annotations

import logging
import time
from collections.abc import Callable, Sequence
from concurrent.futures import ThreadPoolExecutor
from dataclasses import dataclass
from datetime import date
from typing import Literal, Protocol, cast

from ..domain.models import SelectionEvaluation, StockHistory
from ..domain.pattern_scoring import (
    PatternCase,
    PatternCaseLibraryLoader,
    PatternScore,
    PatternScorer,
)
from ..domain.runs import (
    BatchSelectionRunStore,
    BatchSelectionUniverseReader,
    SelectionExecutionSource,
    SelectionRerunRequired,
    SelectionResultQuery,
    SelectionRun,
    SelectionRunInProgress,
    SelectionRunItem,
    SelectionRunStatus,
    SelectionRunStore,
    SelectionStock,
    SelectionUniverseReader,
)
from .evaluate import EvaluateZhixingB1

logger = logging.getLogger(__name__)
StrategyName = Literal["zhixing_b1"]
_FAILURE_STATUSES = {"insufficient_history", "missing_target_bar", "data_error"}


class SelectionEvaluator(Protocol):
    """Minimal single-stock evaluator required by the batch orchestrator."""

    def execute(self, ts_code: str, target_trade_date: date) -> SelectionEvaluation: ...


@dataclass(frozen=True, slots=True)
class PreparedSelectionRun:
    """A claimed run and its immutable market-data source snapshot."""

    run: SelectionRun
    source: SelectionExecutionSource


class RunZhixingB1:
    """Prepare, execute, and query persisted Zhixing B1 result batches."""

    def __init__(
        self,
        reader: SelectionUniverseReader,
        store: SelectionRunStore,
        evaluator: SelectionEvaluator | None = None,
        pattern_case_loader: PatternCaseLibraryLoader | None = None,
        pattern_scorer: PatternScorer | None = None,
        *,
        pattern_scoring_enabled: bool = False,
        max_workers: int = 4,
        batch_size: int = 200,
    ) -> None:
        """Inject storage ports and configure bounded chunk execution."""

        if max_workers < 1:
            raise ValueError("max_workers must be at least 1")
        if batch_size < 1:
            raise ValueError("batch_size must be at least 1")
        self.reader = reader
        self.store = store
        self.evaluator = evaluator or EvaluateZhixingB1(reader)
        self.pattern_case_loader = pattern_case_loader
        self.pattern_scorer = pattern_scorer
        self.pattern_scoring_enabled = pattern_scoring_enabled
        self.max_workers = max_workers
        self.batch_size = batch_size

    def prepare(
        self,
        strategy: StrategyName,
        target_trade_date: date,
        *,
        rerun: bool,
    ) -> PreparedSelectionRun:
        """Validate source eligibility before claiming the rerunnable key."""

        source = self.reader.load_execution_source(strategy, target_trade_date)
        run = self.store.prepare_run(
            strategy,
            target_trade_date,
            source,
            rerun=rerun,
        )
        return PreparedSelectionRun(run=run, source=source)

    def execute(self, prepared: PreparedSelectionRun) -> None:
        """Evaluate every eligible stock and converge the persisted run status.

        This method is the boundary used by FastAPI's in-process background
        task.  An unexpected batch-level error is recorded before the worker
        returns so the UI never mistakes a lost worker exception for success.
        """

        stocks = _unique_stocks(prepared.source.stocks)
        evaluated_count = 0
        selected_stock_count = 0
        signal_count = 0
        failed_count = 0
        history_rows = 0
        batch_count = _chunk_count(len(stocks), self.batch_size)
        read_seconds = 0.0
        evaluate_seconds = 0.0
        persist_seconds = 0.0
        scoring_seconds = 0.0
        try:
            pattern_cases, pattern_library_error = self._prepare_pattern_cases(prepared.run.id)
            with ThreadPoolExecutor(max_workers=self.max_workers) as executor:
                for batch_stocks in _chunks(stocks, self.batch_size):
                    read_started = time.perf_counter()
                    histories = self._load_histories(
                        batch_stocks,
                        prepared.source.target_trade_date,
                    )
                    read_seconds += time.perf_counter() - read_started
                    history_rows += sum(
                        len(history.bars) for history in histories if history is not None
                    )

                    evaluate_started = time.perf_counter()
                    evaluations = tuple(
                        executor.map(
                            self._evaluate_stock,
                            batch_stocks,
                            histories,
                            [prepared.source.target_trade_date] * len(batch_stocks),
                        )
                    )
                    evaluate_seconds += time.perf_counter() - evaluate_started

                    scoring_started = time.perf_counter()
                    items = tuple(
                        _to_item(
                            stock.ts_code,
                            stock.name,
                            evaluation,
                            pattern_score=self._score_stock(
                                prepared.run.id,
                                stock,
                                history,
                                evaluation,
                                pattern_cases,
                                pattern_library_error,
                            ),
                        )
                        for stock, history, evaluation in zip(
                            batch_stocks,
                            histories,
                            evaluations,
                            strict=True,
                        )
                    )
                    scoring_seconds += time.perf_counter() - scoring_started

                    evaluated_count += len(items)
                    selected_stock_count += sum(item.status == "selected" for item in items)
                    signal_count += sum(item.signal_count for item in items)
                    failed_count += sum(item.status in _FAILURE_STATUSES for item in items)

                    persist_started = time.perf_counter()
                    self._record_items(prepared.run.id, items)
                    persist_seconds += time.perf_counter() - persist_started

            status = _run_status(evaluated_count, failed_count)
            self.store.finish_run(
                prepared.run.id,
                status,
                evaluated_count=evaluated_count,
                selected_stock_count=selected_stock_count,
                signal_count=signal_count,
                failed_count=failed_count,
            )
        except Exception as exc:  # noqa: BLE001 - worker boundary must persist failure state
            logger.error(
                "selection_run_failed run_id=%s error_type=%s reason=%s",
                prepared.run.id,
                exc.__class__.__name__,
                _safe_item_error(exc),
            )
            try:
                self.store.finish_run(
                    prepared.run.id,
                    "failed",
                    evaluated_count=evaluated_count,
                    selected_stock_count=selected_stock_count,
                    signal_count=signal_count,
                    failed_count=max(failed_count, 1),
                    error_type="batch_error",
                    error_message=str(exc),
                )
            except Exception:  # noqa: BLE001 - preserve the original worker failure
                logger.error(
                    "selection_run_failure_persist_failed run_id=%s",
                    prepared.run.id,
                )
        finally:
            logger.info(
                "selection_run_summary run_id=%s stock_count=%d history_rows=%d "
                "batch_count=%d worker_count=%d read_seconds=%.3f "
                "evaluate_seconds=%.3f scoring_seconds=%.3f persist_seconds=%.3f",
                prepared.run.id,
                len(stocks),
                history_rows,
                batch_count,
                self.max_workers,
                read_seconds,
                evaluate_seconds,
                scoring_seconds,
                persist_seconds,
            )

    def _prepare_pattern_cases(
        self,
        run_id: str,
    ) -> tuple[tuple[PatternCase, ...] | None, str | None]:
        """Load the complete case library once without failing selection."""

        if not self.pattern_scoring_enabled:
            return None, None
        if self.pattern_case_loader is None or self.pattern_scorer is None:
            reason = "pattern scoring is enabled but not configured"
            logger.error("selection_pattern_library_failed run_id=%s reason=%s", run_id, reason)
            return None, reason
        try:
            return self.pattern_case_loader.load(), None
        except Exception as exc:  # noqa: BLE001 - scoring enrichment must not fail selection
            reason = _safe_item_error(exc)
            logger.warning(
                "selection_pattern_library_failed run_id=%s error_type=%s reason=%s",
                run_id,
                exc.__class__.__name__,
                reason,
            )
            return None, reason

    def _score_stock(
        self,
        run_id: str,
        stock: SelectionStock,
        history: StockHistory | None,
        evaluation: SelectionEvaluation,
        cases: tuple[PatternCase, ...] | None,
        library_error: str | None,
    ) -> PatternScore:
        """Score one selected stock once and isolate enrichment failures."""

        if not self.pattern_scoring_enabled or evaluation.status != "selected":
            return PatternScore()
        if library_error is not None:
            return PatternScore.failed(library_error)
        if history is None or cases is None or self.pattern_scorer is None:
            return PatternScore.failed("pattern scoring history or case library is unavailable")
        try:
            return self.pattern_scorer.score(history, cases)
        except Exception as exc:  # noqa: BLE001 - one score must not fail the selection run
            reason = _safe_item_error(exc)
            logger.warning(
                "selection_pattern_score_failed run_id=%s ts_code=%s error_type=%s reason=%s",
                run_id,
                stock.ts_code,
                exc.__class__.__name__,
                reason,
            )
            return PatternScore.failed(reason)

    def _load_histories(
        self,
        stocks: Sequence[SelectionStock],
        target_trade_date: date,
    ) -> tuple[StockHistory | None, ...]:
        """Load one chunk when the reader supports it, with old-path fallback."""

        typed_stocks = tuple(stocks)
        loader = getattr(self.reader, "load_histories", None)
        if callable(loader):
            batch_reader = cast(BatchSelectionUniverseReader, self.reader)
            loaded = batch_reader.load_histories(typed_stocks, target_trade_date)
            histories_by_code = {history.ts_code: history for history in loaded}
            return tuple(
                histories_by_code.get(
                    stock.ts_code,
                    StockHistory(ts_code=stock.ts_code, name=stock.name),
                )
                for stock in typed_stocks
            )

        if isinstance(self.evaluator, EvaluateZhixingB1):
            return tuple(
                self.reader.load_history(stock.ts_code, target_trade_date) for stock in typed_stocks
            )
        return (None,) * len(typed_stocks)

    def _evaluate_stock(
        self,
        stock: SelectionStock,
        history: StockHistory | None,
        target_trade_date: date,
    ) -> SelectionEvaluation:
        """Evaluate one stock inside a worker and isolate its exception."""

        ts_code = stock.ts_code
        try:
            execute_history: Callable[[StockHistory, date], SelectionEvaluation] | None = getattr(
                self.evaluator,
                "execute_history",
                None,
            )
            if history is not None and execute_history is not None:
                return execute_history(history, target_trade_date)
            return self.evaluator.execute(ts_code, target_trade_date)
        except Exception as exc:  # noqa: BLE001 - isolate one stock from the batch
            logger.warning(
                "selection_item_failed ts_code=%s error_type=%s reason=%s",
                ts_code,
                exc.__class__.__name__,
                _safe_item_error(exc),
            )
            return SelectionEvaluation(
                ts_code=ts_code,
                target_trade_date=target_trade_date,
                status="data_error",
                reason=_safe_item_error(exc),
            )

    def _record_items(self, run_id: str, items: Sequence[SelectionRunItem]) -> None:
        """Use batch persistence while retaining the old single-item seam."""

        record_items = getattr(self.store, "record_items", None)
        if callable(record_items):
            batch_store = cast(BatchSelectionRunStore, self.store)
            batch_store.record_items(run_id, tuple(items))
            return
        for item in items:
            self.store.record_item(run_id, item)

    def get_run(
        self,
        run_id: str,
        *,
        query: SelectionResultQuery | None = None,
    ) -> SelectionRun | None:
        """Read one persisted run for polling."""

        return self.store.get_run(run_id, query=query)

    def get_latest(
        self,
        strategy: StrategyName,
        target_trade_date: date | None = None,
        *,
        query: SelectionResultQuery | None = None,
    ) -> SelectionRun | None:
        """Read the current result by date or the latest result for a strategy."""

        return self.store.get_latest_run(strategy, target_trade_date, query=query)


def _to_item(
    ts_code: str,
    name: str,
    evaluation: SelectionEvaluation,
    *,
    pattern_score: PatternScore | None = None,
) -> SelectionRunItem:
    """Translate a single-stock domain result into a stored item."""

    return SelectionRunItem(
        ts_code=ts_code,
        name=name or (evaluation.signals[0].name if evaluation.signals else ""),
        status=evaluation.status,
        signal_count=len(evaluation.signals),
        reason=evaluation.reason,
        pattern_score=pattern_score or PatternScore(),
        signals=evaluation.signals,
    )


def _run_status(evaluated_count: int, failed_count: int) -> SelectionRunStatus:
    """Map per-stock outcomes into a visible batch status."""

    if failed_count == 0:
        return "success"
    if evaluated_count == 0 or failed_count >= evaluated_count:
        return "failed"
    return "partial_success"


def _safe_item_error(error: Exception) -> str:
    """Keep per-stock failure context readable without persisting tracebacks."""

    return " ".join(str(error).split())[:500] or error.__class__.__name__


def _chunks(
    values: Sequence[SelectionStock],
    size: int,
) -> tuple[tuple[SelectionStock, ...], ...]:
    """Split a stable stock sequence into bounded immutable chunks."""

    return tuple(tuple(values[index : index + size]) for index in range(0, len(values), size))


def _chunk_count(value_count: int, size: int) -> int:
    """Return the number of chunks without materializing empty chunks."""

    return (value_count + size - 1) // size


def _unique_stocks(stocks: Sequence[SelectionStock]) -> tuple[SelectionStock, ...]:
    """Keep the first source row for each stock so it is evaluated once."""

    seen: set[str] = set()
    unique: list[SelectionStock] = []
    for stock in stocks:
        ts_code = stock.ts_code
        if ts_code in seen:
            continue
        seen.add(ts_code)
        unique.append(stock)
    return tuple(unique)


__all__ = [
    "PreparedSelectionRun",
    "RunZhixingB1",
    "SelectionRerunRequired",
    "SelectionRunInProgress",
]