fix: 修复补跑误含 cninfo 三步与时区不一致 (P1-2/P1-3)

- P1-2: 提取 DEFAULT_NEWS_STEPS(全链路去 report+cninfo),定时/补跑/默认三处统一;
  启动补跑不再误执行 cninfo 公告管道
- P1-3: 新增 scheduler/timeutil.py(调度时区统一入口,SCHEDULE_TZ 可覆盖,
  默认 Asia/Shanghai);run_scheduler 定时/补跑/--once、crawler 补跑保护、
  reporter 兜底日期全部改用调度时区,避免系统时区非上海时日期错位
- 新增 6 个测试;全量 282 passed

continuation.md

@@ -4,6 +4,24 @@
 
 ---
 
+## 本次完成 (2026-08-23) — P1-2/P1-3 补跑步骤与时区一致性修复
+
+**P1-2(守护进程补跑误含 cninfo 三步)**:
+- 原因:`scripts/run_scheduler.py` 启动补跑逻辑 `steps = [k for k in STEP_COMMANDS if k != "report"]` 只排除 report,**漏掉 cninfo_crawl/cninfo_extract/cninfo_pdf**;错过定时任务重启补跑时会额外执行整套公告管道(且 cninfo_crawl 无 --date,补跑历史日期静默空转)
+- 修复:提取共享常量 `scheduler.pipeline.DEFAULT_NEWS_STEPS`(= 全链路去除 report 与 cninfo 三步),定时任务、补跑、run_pipeline 默认三处统一引用
+
+**P1-3(调度时区与 date.today() 不一致)**:
+- 原因:cron 触发器显式用 `Asia/Shanghai`,但 `date.today()`/`datetime.now()` 取**系统时区**;若系统时区非上海(如容器 UTC),07:00 上海(=前一日 23:00 UTC)触发时日期会错一天,整条 pipeline 落错日目录
+- 修复:新增 `scheduler/timeutil.py`(schedule_tz/today_str/now,env `SCHEDULE_TZ` 可覆盖,默认 Asia/Shanghai);`run_scheduler` 定时/补跑/`--once`、`pipeline.run_step` crawler 补跑保护、`reporter.generate_report` 兜底日期全部改用它
+
+**测试**:
+- 新增 6 个(DEFAULT_NEWS_STEPS 不含 cninfo/run_pipeline 默认步骤行为/补跑源码引用/时区两例/--once 缺省日期),并入 `tests/test_incremental.py`
+- 全量 **282 passed**,3 个 crawler 基线失败与本次无关;ruff 干净(5 个 reporter 既有问题未动)
+
+**验证**:DEFAULT_NEWS_STEPS = [crawler, xwlb, extractor, dedup, llm, embedding, qdrant] ✓;today_str = 20260823(Asia/Shanghai)✓
+
+---
+
 ## 本次完成 (2026-08-22) — 多源新闻记录链路修复(方案 A + B)
 
 **用户需求**:一条新闻有多个来源时,全部来源都要记录并可见;此前"找不到多源"。

scheduler/__init__.py

@@ -7,6 +7,7 @@
 """
 
 from .pipeline import (
+    DEFAULT_NEWS_STEPS,
     STEP_COMMANDS,
     STEP_TIMEOUTS,
     PipelineResult,
@@ -16,6 +17,7 @@ from .pipeline import (
 )
 
 __all__ = [
+    "DEFAULT_NEWS_STEPS",
     "STEP_COMMANDS",
     "STEP_TIMEOUTS",
     "PipelineResult",

scheduler/pipeline.py

@@ -16,11 +16,13 @@ import os
 import subprocess
 import time
 from dataclasses import dataclass, field
-from datetime import date, datetime
+from datetime import datetime
 from pathlib import Path
 
 from loguru import logger
 
+from .timeutil import today_str
+
 # 断点状态文件(按日期隔离,记录每步骤结果)
 DEFAULT_STATE_PATH = Path("data/pipeline/state.json")
 
@@ -74,6 +76,14 @@ STEP_COMMANDS: dict[str, list[str]] = {
     "cninfo_pdf":     ["uv", "run", "a-share", "cninfo", "--enrich-pdf", "--pdf-limit", "100"],
 }
 
+# 新闻链路默认步骤:全链路除去 report(单独追加)与 cninfo 独立管道(P1-2:
+# 补跑/定时任务若误含 cninfo 三步,会重复执行整套公告管道,且 cninfo_crawl
+# 无 --date 参数,补跑历史日期时会静默空转)。
+DEFAULT_NEWS_STEPS: list[str] = [
+    k for k in STEP_COMMANDS
+    if k not in ("report", "cninfo_crawl", "cninfo_extract", "cninfo_pdf")
+]
+
 
 @dataclass
 class StepResult:
@@ -223,7 +233,8 @@ def run_step(name: str, date_str: str) -> StepResult:
     # 补跑保护:抓取类步骤只能产生当天数据(网站首页只含当前内容,历史文章已滚走),
     # 补跑历史日期时跳过抓取并告警,复用已有 data/raw/*/{date_str} 数据。
     # 注意:仅保护 crawler; xwlb 走 API 支持任意历史日期,无需跳过。
-    if name == "crawler" and date_str != date.today().strftime("%Y%m%d"):
+    # "今天"按调度时区判定(P1-3),避免系统时区与 cron 时区不一致时错判。
+    if name == "crawler" and date_str != today_str():
         started = datetime.now()
         logger.warning(
             "补跑模式:首页只含当天内容,无法补抓 {};跳过抓取,复用已有 data/raw/*/{}",
@@ -330,7 +341,7 @@ def run_pipeline(
                   记录,跳过连续成功的步骤,从第一个失败/未执行步骤继续。
         state_path: 断点状态文件路径(测试可注入)。
     """
-    names = steps or [k for k in STEP_COMMANDS if k not in ("report", "cninfo_crawl", "cninfo_extract", "cninfo_pdf")]
+    names = steps or list(DEFAULT_NEWS_STEPS)
     result = PipelineResult(started_at=datetime.now())
 
     state = _load_pipeline_state(state_path)

scheduler/reporter.py

@@ -1056,7 +1056,10 @@ def generate_report(day_str: str | None = None, *, upload: bool = True) -> int |
     `upload` 参数保留以兼容 scheduler/pipeline.py 调用，已无实际作用。
     返回 report_id（成功）或 None（无数据/失败）。
     """
-    day_str = day_str or date.today().strftime("%Y%m%d")
+    # 兜底日期按调度时区取(P1-3),与 pipeline 传入的 date_str 语义一致
+    if not day_str:
+        from .timeutil import today_str
+        day_str = today_str()
     logger.info("生成日报: {}", day_str)
 
     # 收集数据

scheduler/timeutil.py

@@ -0,0 +1,36 @@
+"""调度时区工具。
+
+背景(P1-3):
+    定时任务 cron 触发时区与"今天"的判定必须一致。此前 cron 用
+    Asia/Shanghai,而 `date.today()` / `datetime.now()` 取系统时区——
+    若系统时区不是 Asia/Shanghai(如 UTC),07:00 上海(=前一日 23:00 UTC)
+    触发时 `date.today()` 会返回错误日期,整条 pipeline 落在错日目录。
+
+统一入口:
+    schedule_tz(): 调度时区(env SCHEDULE_TZ 可覆盖,默认 Asia/Shanghai)
+    today_str():   按调度时区返回 YYYYMMDD
+"""
+
+from __future__ import annotations
+
+import os
+from datetime import datetime
+from zoneinfo import ZoneInfo
+
+# 默认调度时区(与 .env 的 SCHEDULE_TIMES 语义一致)
+DEFAULT_SCHEDULE_TZ = "Asia/Shanghai"
+
+
+def schedule_tz() -> ZoneInfo:
+    """调度时区(env SCHEDULE_TZ 可覆盖,默认 Asia/Shanghai)。"""
+    return ZoneInfo(os.environ.get("SCHEDULE_TZ") or DEFAULT_SCHEDULE_TZ)
+
+
+def now() -> datetime:
+    """当前时刻(调度时区,aware)。"""
+    return datetime.now(schedule_tz())
+
+
+def today_str() -> str:
+    """按调度时区返回今天的 YYYYMMDD。"""
+    return now().strftime("%Y%m%d")

scripts/run_scheduler.py

@@ -16,15 +16,16 @@ from __future__ import annotations
 import argparse
 import signal
 import sys
-from datetime import date, datetime
 from pathlib import Path
 from typing import Any
 
 from dotenv import load_dotenv
 from loguru import logger
 
-from scheduler import STEP_COMMANDS, run_pipeline
+from scheduler import DEFAULT_NEWS_STEPS, run_pipeline
 from scheduler.stock_reporter import generate_all_stock_reports
+from scheduler.timeutil import now as tz_now
+from scheduler.timeutil import schedule_tz, today_str
 
 
 def _setup_logger(level: str) -> None:
@@ -69,7 +70,7 @@ def _once(args: argparse.Namespace) -> int:
     if args.resume and args.steps:
         logger.error("--resume 与 --steps 不能同时使用(断点续跑针对全链路)")
         return 2
-    run_pipeline(args.date, steps=steps, resume=args.resume)
+    run_pipeline(args.date or today_str(), steps=steps, resume=args.resume)
     return 0
 
 
@@ -92,17 +93,18 @@ def _daemon(args: argparse.Namespace) -> int:
 
     scheduler = BackgroundScheduler()
 
-    # 包装函数:每次触发时重新计算日期,避免 date.today() 在注册时冻结。
+    # 包装函数:每次触发时按调度时区重新计算日期(P1-3),
+    # 避免 date.today() 在注册时冻结或与 cron 时区不一致。
     def _scheduled_pipeline(steps: list[str] | None = None) -> None:
-        run_pipeline(date.today().strftime("%Y%m%d"), steps=steps)
+        run_pipeline(today_str(), steps=steps)
 
     for hour, minute in times:
-        trigger = CronTrigger(hour=hour, minute=minute, timezone="Asia/Shanghai")
+        trigger = CronTrigger(hour=hour, minute=minute, timezone=str(schedule_tz()))
         is_first = (hour == first_hour and minute == first_minute)
         job_kwargs: dict | None = None
         if is_first:
             job_kwargs = {
-                "steps": [k for k in STEP_COMMANDS if k not in ("report", "cninfo_crawl", "cninfo_extract", "cninfo_pdf")] + ["report"]
+                "steps": list(DEFAULT_NEWS_STEPS) + ["report"]
             }
         scheduler.add_job(
             _scheduled_pipeline,
@@ -118,7 +120,7 @@ def _daemon(args: argparse.Namespace) -> int:
     cninfo_raw = os.environ.get("CNINFO_SCHEDULE_TIME", "06:30")
     cninfo_parts = cninfo_raw.split(":")
     cninfo_h, cninfo_m = int(cninfo_parts[0]), int(cninfo_parts[1]) if len(cninfo_parts) > 1 else 0
-    cninfo_trigger = CronTrigger(hour=cninfo_h, minute=cninfo_m, timezone="Asia/Shanghai")
+    cninfo_trigger = CronTrigger(hour=cninfo_h, minute=cninfo_m, timezone=str(schedule_tz()))
     cninfo_steps = ["cninfo_crawl", "cninfo_extract", "cninfo_pdf",
                     "dedup", "llm", "embedding", "qdrant"]
     scheduler.add_job(
@@ -135,7 +137,7 @@ def _daemon(args: argparse.Namespace) -> int:
     if stock_raw:
         stock_parts = stock_raw.split(":")
         stock_h, stock_m = int(stock_parts[0]), int(stock_parts[1]) if len(stock_parts) > 1 else 0
-        stock_trigger = CronTrigger(hour=stock_h, minute=stock_m, timezone="Asia/Shanghai")
+        stock_trigger = CronTrigger(hour=stock_h, minute=stock_m, timezone=str(schedule_tz()))
         scheduler.add_job(
             generate_all_stock_reports,
             trigger=stock_trigger,
@@ -159,7 +161,7 @@ def _daemon(args: argparse.Namespace) -> int:
     logger.info("调度器已启动,等待触发... (按 Ctrl+C 退出)")
 
     # 启动时检查是否有因重启/宕机错过的定时任务,30 分钟内补跑
-    now = datetime.now()
+    now = tz_now()
     for hour, minute in times:
         scheduled = now.replace(hour=hour, minute=minute, second=0, microsecond=0)
         missed_minutes = (now - scheduled).total_seconds() / 60
@@ -168,10 +170,12 @@ def _daemon(args: argparse.Namespace) -> int:
                 "检测到错过的定时任务 {:02d}:{:02d} ({} 分钟前),立即补跑一次",
                 hour, minute, int(missed_minutes),
             )
-            steps = [k for k in STEP_COMMANDS if k != "report"]
+            # P1-2:补跑只跑新闻链路(DEFAULT_NEWS_STEPS 不含 report 与 cninfo 三步),
+            # cninfo 公告管道由其自身定时任务负责,避免重复执行整套公告管道。
+            steps = list(DEFAULT_NEWS_STEPS)
             if (hour, minute) == sorted_times[0]:
                 steps.append("report")
-            run_pipeline(date.today().strftime("%Y%m%d"), steps=steps)
+            run_pipeline(today_str(), steps=steps)
 
     import contextlib
 
@@ -186,8 +190,8 @@ def main() -> int:
     parser = argparse.ArgumentParser(description="A 股新闻定时任务 (M7)")
     parser.add_argument("--once", action="store_true", help="立即执行一次全链路")
     parser.add_argument(
-        "--date", default=date.today().strftime("%Y%m%d"),
+        "--date", default=None,
-        help="日期 YYYYMMDD (仅 --once 模式)",
+        help="日期 YYYYMMDD (仅 --once 模式,默认按调度时区取今天)",
     )
     parser.add_argument("--steps", default=None,
                         help="仅执行指定步骤,逗号分隔 (如 crawler,extractor)")

tests/test_incremental.py

@@ -8,6 +8,7 @@
 from __future__ import annotations
 
 import json
+from datetime import UTC
 from pathlib import Path
 from types import SimpleNamespace
 
@@ -306,8 +307,7 @@ def test_crawler_today_runs_normally(monkeypatch: pytest.MonkeyPatch) -> None:
 
 def test_report_step_uses_date_str(monkeypatch: pytest.MonkeyPatch) -> None:
     """report 步骤必须使用传入的 date_str,而非 date.today()(P0-4)。"""
-    from scheduler import pipeline
+    from scheduler import pipeline, reporter
-    from scheduler import reporter
 
     received: list[str] = []
 
@@ -346,3 +346,106 @@ def test_pipeline_backfill_skips_crawler_keeps_rest(
     assert all(s.success for s in result.steps)
     state = pipeline._load_pipeline_state(state_path)
     assert state["20260101"]["crawler"]["status"] == "ok"
+
+
+# --------------------------------------------------------------------------- #
+# P1-2: 默认步骤不含 cninfo 独立管道 / P1-3: 调度时区统一
+# --------------------------------------------------------------------------- #
+
+def test_default_news_steps_excludes_cninfo_and_report() -> None:
+    """新闻链路默认步骤不含 report 与 cninfo 三步(P1-2)。"""
+    from scheduler import DEFAULT_NEWS_STEPS
+
+    assert "report" not in DEFAULT_NEWS_STEPS
+    assert "cninfo_crawl" not in DEFAULT_NEWS_STEPS
+    assert "cninfo_extract" not in DEFAULT_NEWS_STEPS
+    assert "cninfo_pdf" not in DEFAULT_NEWS_STEPS
+    # 新闻链路核心步骤齐全
+    assert {"crawler", "xwlb", "extractor", "dedup", "llm",
+            "embedding", "qdrant"} <= set(DEFAULT_NEWS_STEPS)
+
+
+def test_run_pipeline_default_steps_excludes_cninfo(
+    tmp_path: Path, monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    """run_pipeline 默认 steps 不含 cninfo(只跑新闻链路)。"""
+    from scheduler import pipeline
+    from scheduler.timeutil import today_str
+
+    calls: list[str] = []
+
+    def _fake_run(cmd, timeout=None):  # noqa: ARG001
+        name = next(c.split(".")[-1] for c in cmd if "scripts.run_" in c)
+        calls.append(name)
+        return SimpleNamespace(returncode=0)
+
+    monkeypatch.setattr(pipeline.subprocess, "run", _fake_run)
+    state_path = tmp_path / "state.json"
+    today = today_str()  # 今天:避免 crawler 补跑保护跳过
+    result = pipeline.run_pipeline(
+        today, state_path=state_path,
+    )
+    assert result.all_success is True
+    assert calls == ["run_crawler", "run_xwlb", "run_extractor", "run_dedup",
+                     "run_event_extraction", "run_embedding", "run_qdrant_ingest"]
+    assert "cninfo" not in " ".join(calls)
+
+
+def test_scheduler_daemon_backfill_uses_default_news_steps(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    """守护进程补跑 steps 使用 DEFAULT_NEWS_STEPS(不含 cninfo 三步)。"""
+    import inspect
+
+    import scripts.run_scheduler as rs
+    from scheduler import DEFAULT_NEWS_STEPS
+
+    # 验证补跑逻辑引用的常量:通过源码断言 + 常量内容双重保证
+    src = inspect.getsource(rs._daemon)
+    assert "list(DEFAULT_NEWS_STEPS)" in src
+    assert "cninfo_crawl" not in DEFAULT_NEWS_STEPS
+
+
+def test_today_str_matches_schedule_timezone(monkeypatch: pytest.MonkeyPatch) -> None:
+    """today_str 按调度时区返回 YYYYMMDD(P1-3)。"""
+    from datetime import datetime
+    from zoneinfo import ZoneInfo
+
+    from scheduler.timeutil import today_str
+
+    s = today_str()
+    assert len(s) == 8 and s.isdigit()
+    # 与调度时区(默认 Asia/Shanghai)当前日期一致
+    expect = datetime.now(ZoneInfo("Asia/Shanghai")).strftime("%Y%m%d")
+    assert s == expect
+
+
+def test_today_str_respects_env_override(monkeypatch: pytest.MonkeyPatch) -> None:
+    """SCHEDULE_TZ 环境变量可覆盖调度时区(P1-3)。"""
+    from datetime import datetime
+
+    from scheduler.timeutil import today_str
+
+    # 覆盖为 UTC 后,today_str 应返回 UTC 日期(而非默认 Asia/Shanghai)
+    monkeypatch.setenv("SCHEDULE_TZ", "UTC")
+    s = today_str()
+    expect = datetime.now(UTC).strftime("%Y%m%d")
+    assert s == expect
+
+
+def test_once_uses_today_str_when_date_missing(monkeypatch: pytest.MonkeyPatch) -> None:
+    """--once 不带 --date 时按调度时区取今天(P1-3)。"""
+    import scripts.run_scheduler as rs
+
+    captured: dict[str, str] = {}
+
+    def _fake_pipeline(date_str, steps=None, resume=False):  # noqa: ARG001
+        captured["date"] = date_str
+
+    monkeypatch.setattr(rs, "run_pipeline", _fake_pipeline)
+    from types import SimpleNamespace
+
+    args = SimpleNamespace(steps=None, resume=False, date=None)
+    rc = rs._once(args)
+    assert rc == 0
+    assert captured["date"] == rs.today_str()