sophhub 0.4.19 → 0.4.21

package/skills/schedule-reminder/src/scripts/parse_schedule.py

@@ -1,704 +1,704 @@
-#!/usr/bin/env python3
-"""
-parse_schedule.py - 解析 Markdown 日程表，输出今日/近期事项、冲突检测结果
-
-用法：
-  python3 parse_schedule.py --file /home/node/.openclaw/workspace/日程.md [--mode today|tomorrow|week|all]
-  python3 parse_schedule.py --stdin [--mode today]  # 从 stdin 读取 Markdown 内容
-
-输出：JSON，包含 events / upcoming / conflicts / recurring / tasks
-"""
-from __future__ import annotations
-
-import sys
-import re
-import json
-import argparse
-from datetime import datetime, date, timedelta
-from typing import Optional, List, Dict
-
-
-PRIORITY_ORDER = {"高": 0, "中": 1, "低": 2, "": 1}
-
-TASK_KEYWORDS = {"提交", "截止", "deadline", "交付", "汇报", "发布", "上线"}
-
-
-def parse_args():
-    p = argparse.ArgumentParser(description="解析 Markdown 日程表")
-    p.add_argument("--file", help="Markdown 日程文件路径")
-    p.add_argument("--stdin", action="store_true", help="从 stdin 读取内容")
-    p.add_argument(
-        "--mode",
-        choices=["today", "tomorrow", "week", "all"],
-        default="today",
-        help="筛选范围（默认 today）",
-    )
-    p.add_argument("--now", help="覆盖当前时间（格式 YYYY-MM-DD HH:MM，测试用）")
-    p.add_argument(
-        "--column-map",
-        help=(
-            "自定义列名映射，格式：'自定义列名=字段名,...'\n"
-            "字段名可选：date/start/end/title/location/participants/priority/advance/note\n"
-            "示例：--column-map '会期=date,主旨=title,时间段开始=start'"
-        ),
-    )
-    return p.parse_args()
-
-
-def read_content(args) -> str:
-    if args.stdin:
-        return sys.stdin.read()
-    if args.file:
-        path = args.file.replace("~", __import__("os").path.expanduser("~"))
-        with open(path, encoding="utf-8") as f:
-            return f.read()
-    print("错误：请指定 --file 或 --stdin", file=sys.stderr)
-    sys.exit(1)
-
-
-def parse_global_config(content: str) -> dict:
-    """从文件头部注释和引用块提取全局配置"""
-    cfg = {
-        "channel": "dingtalk",
-        "default_advance_minutes": 15,
-        "timezone": "Asia/Shanghai",
-    }
-    for line in content.splitlines():
-        line = line.strip()
-        m = re.match(r">\s*\*?\*?提醒渠道\*?\*?[：:]\s*(\S+)", line)
-        if m:
-            cfg["channel"] = m.group(1).strip()
-        m = re.match(r">\s*\*?\*?默认提前提醒\*?\*?[：:]\s*(\d+)", line)
-        if m:
-            cfg["default_advance_minutes"] = int(m.group(1))
-        m = re.match(r">\s*\*?\*?时区\*?\*?[：:]\s*(\S+)", line)
-        if m:
-            cfg["timezone"] = m.group(1)
-    return cfg
-
-
-def normalize_date(raw: str) -> Optional[date]:
-    raw = raw.strip()
-    for fmt in ("%Y-%m-%d", "%Y/%m/%d", "%m-%d", "%m/%d"):
-        try:
-            d = datetime.strptime(raw, fmt)
-            if d.year == 1900:
-                d = d.replace(year=date.today().year)
-            return d.date()
-        except ValueError:
-            pass
-    return None
-
-
-def normalize_time(raw: str) -> Optional[int]:
-    """返回分钟数（从当天0点起）"""
-    raw = raw.strip().replace("：", ":")
-    m = re.match(r"(\d{1,2}):(\d{2})", raw)
-    if m:
-        return int(m.group(1)) * 60 + int(m.group(2))
-    return None
-
-
-# ─────────────────────────────────────────────────────────
-# 列名自动映射（宽松模式，支持不标准的 Excel/CSV 转换结果）
-# ─────────────────────────────────────────────────────────
-
-# 各字段的候选列名（支持中英文、简写、常见变体）
-COLUMN_ALIASES = {
-    "date":         {"日期", "date", "Date", "DATE", "时间", "日", "活动日期", "会议日期", "事件日期",
-                     "日程日期", "计划日期"},
-    "start":        {"开始时间", "开始", "start", "Start", "START", "开始时刻", "起始时间", "from", "From",
-                     "开始时间段", "时间开始", "起始"},
-    "end":          {"结束时间", "结束", "end", "End", "END", "结束时刻", "to", "To",
-                     "时间结束", "截止", "终止"},
-    "title":        {"事项", "标题", "title", "Title", "TITLE", "内容", "会议", "活动", "事件",
-                     "subject", "Subject", "name", "Name", "任务", "描述", "会议名称", "活动名称",
-                     "事项名称", "议题", "主题", "项目", "meeting", "Meeting", "event", "Event",
-                     "topic", "Topic", "item", "Item"},
-    "location":     {"地点", "location", "Location", "LOCATION", "地址", "会议室", "venue", "Venue",
-                     "place", "Place", "room", "Room", "场所", "场地"},
-    "participants": {"参与者", "人员", "participants", "Participants", "与会者", "成员", "attendees",
-                     "Attendees", "负责人", "owner", "Owner", "参会人", "参会人员", "出席人员",
-                     "相关人员", "联系人"},
-    "priority":     {"优先级", "priority", "Priority", "PRIORITY", "重要程度", "紧急度", "级别",
-                     "importance", "Importance", "urgency", "Urgency"},
-    "advance":      {"提前(分钟)", "提前提醒", "提前", "advance", "Advance", "reminder", "Reminder",
-                     "提前时间", "提醒时间", "提前分钟"},
-    "note":         {"备注", "note", "Note", "NOTE", "notes", "Notes", "说明", "描述", "remark",
-                     "Remark", "comments", "Comments", "附注", "补充", "其他"},
-}
-
-
-def parse_user_column_map(raw: str) -> dict:
-    """
-    解析用户自定义列名映射字符串。
-    输入格式：'会期=date,主旨=title,时间段开始=start'
-    返回：{'会期': 'date', '主旨': 'title', '时间段开始': 'start'}
-    """
-    VALID_FIELDS = {"date", "start", "end", "title", "location",
-                    "participants", "priority", "advance", "note"}
-    result = {}
-    for part in raw.split(","):
-        part = part.strip()
-        if "=" not in part:
-            continue
-        custom_name, _, field = part.partition("=")
-        custom_name = custom_name.strip()
-        field = field.strip()
-        if field in VALID_FIELDS and custom_name:
-            result[custom_name] = field
-    return result
-
-
-def detect_column_mapping(header_cells: List[str],
-                           user_map: Optional[dict] = None) -> dict:
-    """
-    根据表头自动推断列索引映射。
-    优先级：用户自定义映射 > 精确别名匹配 > 子串匹配。
-    返回 {字段名: 列索引}
-    """
-    mapping: dict = {}
-    used_indices: set = set()
-
-    # 第零轮：用户自定义映射（最高优先级）
-    if user_map:
-        for i, cell in enumerate(header_cells):
-            cell_clean = cell.strip()
-            if cell_clean in user_map:
-                field = user_map[cell_clean]
-                if field not in mapping:  # 同一字段只取第一个
-                    mapping[field] = i
-                    used_indices.add(i)
-
-    # 第一轮：精确匹配
-    for field, aliases in COLUMN_ALIASES.items():
-        for i, cell in enumerate(header_cells):
-            if i in used_indices:
-                continue
-            if cell.strip() in aliases:
-                mapping[field] = i
-                used_indices.add(i)
-                break
-
-    # 第二轮：子串匹配（仅对未匹配的字段和未使用的列）
-    for field, aliases in COLUMN_ALIASES.items():
-        if field in mapping:
-            continue
-        for i, cell in enumerate(header_cells):
-            if i in used_indices:
-                continue
-            cell_clean = cell.strip()
-            matched = False
-            for alias in aliases:
-                if not alias:
-                    continue
-                # 列名包含别名（如"会议名称"包含"名称"→title）
-                # 但要避免短词误匹配（别名长度至少2）
-                if len(alias) >= 2 and (alias in cell_clean or cell_clean in alias):
-                    mapping[field] = i
-                    used_indices.add(i)
-                    matched = True
-                    break
-            if matched:
-                break
-
-    return mapping
-
-
-def report_unresolved_columns(header_cells: List[str], mapping: dict) -> List[str]:
-    """返回未能映射到任何已知字段的列名列表"""
-    mapped_indices = set(mapping.values())
-    return [
-        header_cells[i].strip()
-        for i in range(len(header_cells))
-        if i not in mapped_indices and header_cells[i].strip()
-    ]
-
-
-def check_required_fields(mapping: dict) -> List[str]:
-    """检查必要字段是否都已映射，返回缺失字段名"""
-    required = ["date", "title"]
-    return [f for f in required if f not in mapping]
-
-
-def is_standard_header(cells: List[str]) -> bool:
-    """判断是否是标准格式表头（首列精确为日期/Date/date）"""
-    SCHEDULE_HEADERS = {"日期", "Date", "date"}
-    return bool(cells) and cells[0].strip() in SCHEDULE_HEADERS
-
-
-def is_any_schedule_header(cells: List[str]) -> bool:
-    """宽松判断：表头中是否含有日期相关列"""
-    date_aliases = COLUMN_ALIASES["date"]
-    title_aliases = COLUMN_ALIASES["title"]
-    has_date = any(c.strip() in date_aliases for c in cells)
-    has_title = any(c.strip() in title_aliases for c in cells)
-    return has_date and has_title
-
-
-def extract_event_by_mapping(cells: List[str], mapping: dict, default_advance: int,
-                              section_name: str) -> Optional[dict]:
-    """根据列映射从数据行提取事项"""
-    def get(field):
-        idx = mapping.get(field)
-        if idx is not None and idx < len(cells):
-            return cells[idx].strip()
-        return ""
-
-    date_val = normalize_date(get("date"))
-    if not date_val:
-        return None
-
-    title = get("title")
-    if not title:
-        return None
-
-    start_str = get("start")
-    end_str = get("end")
-    start_min = normalize_time(start_str) if start_str else None
-    end_min = normalize_time(end_str) if end_str else None
-
-    priority_raw = get("priority")
-    priority = priority_raw if priority_raw in ("高", "中", "低") else "中"
-
-    advance_raw = get("advance")
-    try:
-        advance_min = int(advance_raw) if advance_raw else default_advance
-    except ValueError:
-        advance_min = default_advance
-
-    return {
-        "date": date_val.isoformat(),
-        "start_min": start_min,
-        "end_min": end_min,
-        "start_str": start_str,
-        "end_str": end_str,
-        "title": title,
-        "location": get("location"),
-        "participants": get("participants"),
-        "priority": priority,
-        "advance_min": advance_min,
-        "note": get("note"),
-        "section": section_name,
-        "is_task": any(k in title + get("note") for k in TASK_KEYWORDS),
-    }
-
-
-def parse_table_section(lines: List[str], section_name: str,
-                         default_advance: int,
-                         user_map: Optional[dict] = None) -> dict:
-    """
-    解析日程表格。返回 dict：
-      events: 解析出的事项列表
-      unresolved_columns: 无法映射的列名列表（宽松模式下）
-      missing_required: 缺失的必要字段（date/title）
-      mode: 'standard' | 'loose' | 'failed'
-    """
-    in_table = False
-    header_found = False
-    column_mapping: Optional[dict] = None
-    loose_mode = False
-    events = []
-    unresolved_columns: List[str] = []
-    missing_required: List[str] = []
-    parse_mode = "failed"
-
-    TASK_HEADERS = {"截止日期", "DeadLine", "deadline", "Deadline"}
-
-    for line in lines:
-        stripped = line.strip()
-        if not stripped.startswith("|"):
-            if in_table:
-                in_table = False
-                header_found = False
-                column_mapping = None
-                loose_mode = False
-            continue
-
-        cells = [c.strip() for c in stripped.strip("|").split("|")]
-        if len(cells) < 2:
-            continue
-
-        # 跳过任务表头，不被主日程解析器误抓
-        if cells[0].strip() in TASK_HEADERS:
-            continue
-
-        # 检测表头
-        if not in_table:
-            if is_standard_header(cells):
-                START_HEADERS = {"开始时间", "开始", "start", "Start", "START", "起始时间"}
-                second_col = cells[1].strip() if len(cells) > 1 else ""
-                if second_col in START_HEADERS:
-                    # 标准模式
-                    in_table = True
-                    header_found = True
-                    loose_mode = False
-                    column_mapping = None
-                    parse_mode = "standard"
-                    continue
-                else:
-                    # 首列是"日期"但列序不标准，降级为宽松模式
-                    mapping = detect_column_mapping(cells, user_map)
-                    missing = check_required_fields(mapping)
-                    if not missing:
-                        in_table = True
-                        header_found = True
-                        loose_mode = True
-                        column_mapping = mapping
-                        parse_mode = "loose"
-                        unresolved_columns = report_unresolved_columns(cells, mapping)
-                        missing_required = []
-                    else:
-                        missing_required = missing
-                        parse_mode = "failed"
-                    continue
-            elif is_any_schedule_header(cells) or user_map:
-                # 宽松模式：自动映射 + 用户自定义映射
-                mapping = detect_column_mapping(cells, user_map)
-                missing = check_required_fields(mapping)
-                if not missing:
-                    in_table = True
-                    header_found = True
-                    loose_mode = True
-                    column_mapping = mapping
-                    parse_mode = "loose"
-                    unresolved_columns = report_unresolved_columns(cells, mapping)
-                    missing_required = []
-                else:
-                    missing_required = missing
-                    parse_mode = "failed"
-                continue
-            continue
-            continue
-
-        # 跳过分隔行
-        if header_found and all(re.match(r"^[-:]+$", c) for c in cells if c):
-            continue
-
-        if not in_table:
-            continue
-
-        if loose_mode and column_mapping:
-            # 宽松模式：用列映射提取
-            ev = extract_event_by_mapping(cells, column_mapping, default_advance, section_name)
-            if ev:
-                events.append(ev)
-        else:
-            # 标准模式：按固定列序提取
-            if len(cells) < 4:
-                continue
-            date_val = normalize_date(cells[0]) if cells[0] else None
-            if not date_val:
-                continue
-            start_min = normalize_time(cells[1]) if len(cells) > 1 else None
-            end_min = normalize_time(cells[2]) if len(cells) > 2 else None
-            title = cells[3].strip() if len(cells) > 3 else ""
-            location = cells[4].strip() if len(cells) > 4 else ""
-            participants = cells[5].strip() if len(cells) > 5 else ""
-            priority = cells[6].strip() if len(cells) > 6 else "中"
-            advance_raw = cells[7].strip() if len(cells) > 7 else ""
-            note = cells[8].strip() if len(cells) > 8 else ""
-            if not title:
-                continue
-            try:
-                advance_min = int(advance_raw) if advance_raw else default_advance
-            except ValueError:
-                advance_min = default_advance
-            events.append({
-                "date": date_val.isoformat(),
-                "start_min": start_min,
-                "end_min": end_min,
-                "start_str": cells[1].strip() if len(cells) > 1 else "",
-                "end_str": cells[2].strip() if len(cells) > 2 else "",
-                "title": title,
-                "location": location,
-                "participants": participants,
-                "priority": priority if priority in ("高", "中", "低") else "中",
-                "advance_min": advance_min,
-                "note": note,
-                "section": section_name,
-                "is_task": any(k in title + note for k in TASK_KEYWORDS),
-            })
-
-    return {
-        "events": events,
-        "mode": parse_mode,
-        "unresolved_columns": unresolved_columns,
-        "missing_required": missing_required,
-    }
-
-
-def parse_task_section(lines: List[str], default_advance: int) -> list:
-    """解析待提醒任务表（截止日期形式）"""
-    in_table = False
-    header_found = False
-    tasks = []
-
-    TASK_HEADERS = {"截止日期", "DeadLine", "deadline", "Deadline"}
-
-    for line in lines:
-        stripped = line.strip()
-        if not stripped.startswith("|"):
-            if in_table:
-                in_table = False
-                header_found = False
-            continue
-
-        cells = [c.strip() for c in stripped.strip("|").split("|")]
-        if not cells:
-            continue
-
-        if cells[0] in TASK_HEADERS:
-            in_table = True
-            header_found = True
-            continue
-
-        if header_found and all(re.match(r"^[-:]+$", c) for c in cells if c):
-            continue
-
-        if not in_table:
-            continue
-
-        if len(cells) < 2:
-            continue
-
-        date_val = normalize_date(cells[0]) if cells[0] else None
-        if not date_val:
-            continue
-
-        title = cells[1].strip() if len(cells) > 1 else ""
-        priority = cells[3].strip() if len(cells) > 3 else "中"
-        advance_raw = cells[4].strip() if len(cells) > 4 else ""
-        note = cells[5].strip() if len(cells) > 5 else ""
-
-        # 解析"提前1天"、"提前2天"形式
-        advance_min = default_advance
-        m = re.search(r"提前\s*(\d+)\s*天", advance_raw)
-        if m:
-            advance_min = int(m.group(1)) * 24 * 60
-        else:
-            m = re.search(r"(\d+)", advance_raw)
-            if m:
-                advance_min = int(m.group(1))
-
-        tasks.append({
-            "date": date_val.isoformat(),
-            "start_min": 9 * 60,  # 默认早上9点触发
-            "end_min": 9 * 60 + 30,
-            "start_str": "09:00",
-            "end_str": "09:30",
-            "title": f"📋 {title}",
-            "location": "",
-            "participants": cells[2].strip() if len(cells) > 2 else "",
-            "priority": priority if priority in ("高", "中", "低") else "中",
-            "advance_min": advance_min,
-            "note": note,
-            "section": "tasks",
-            "is_task": True,
-        })
-
-    return tasks
-
-
-def detect_conflicts(events: list[dict]) -> list[dict]:
-    """检测同日时间冲突"""
-    by_date: dict[str, list[dict]] = {}
-    for e in events:
-        if e["start_min"] is None or e["end_min"] is None:
-            continue
-        by_date.setdefault(e["date"], []).append(e)
-
-    conflicts = []
-    for day, day_events in by_date.items():
-        sorted_events = sorted(day_events, key=lambda x: x["start_min"])
-        for i in range(len(sorted_events)):
-            for j in range(i + 1, len(sorted_events)):
-                a, b = sorted_events[i], sorted_events[j]
-                if a["end_min"] <= b["start_min"]:
-                    # 不重叠，但检查是否过于紧凑（< 5分钟间隔）
-                    gap = b["start_min"] - a["end_min"]
-                    if 0 < gap < 5:
-                        conflicts.append({
-                            "type": "tight",
-                            "date": day,
-                            "event_a": a["title"],
-                            "event_b": b["title"],
-                            "a_time": f"{a['start_str']}–{a['end_str']}",
-                            "b_time": f"{b['start_str']}–{b['end_str']}",
-                            "gap_min": gap,
-                            "message": (
-                                f"⚠️ 紧凑安排：【{a['title']}】{a['start_str']}–{a['end_str']} "
-                                f"与【{b['title']}】{b['start_str']} 之间仅 {gap} 分钟，几乎没有缓冲时间。"
-                            ),
-                        })
-                else:
-                    # 重叠
-                    overlap = a["end_min"] - b["start_min"]
-                    conflicts.append({
-                        "type": "overlap",
-                        "date": day,
-                        "event_a": a["title"],
-                        "event_b": b["title"],
-                        "a_time": f"{a['start_str']}–{a['end_str']}",
-                        "b_time": f"{b['start_str']}–{b['end_str']}",
-                        "overlap_min": overlap,
-                        "message": (
-                            f"⚠️ 时间冲突：【{a['title']}】{a['start_str']}–{a['end_str']} "
-                            f"与【{b['title']}】{b['start_str']}–{b['end_str']} "
-                            f"重叠 {overlap} 分钟。"
-                        ),
-                    })
-    return conflicts
-
-
-def filter_upcoming(events: list, mode: str, now: datetime) -> list:
-    """按 mode 筛选需要提醒的事项"""
-    today = now.date()
-    tomorrow = today + timedelta(days=1)
-    week_end = today + timedelta(days=7)
-    now_min = now.hour * 60 + now.minute
-
-    result = []
-    for e in events:
-        try:
-            ev_date = date.fromisoformat(e["date"])
-        except ValueError:
-            continue
-
-        # 已过期跳过
-        if ev_date < today:
-            continue
-        if ev_date == today and e["end_min"] is not None and e["end_min"] < now_min:
-            continue
-
-        # 低优先级且非任务，默认跳过
-        if e["priority"] == "低" and not e["is_task"]:
-            continue
-
-        include = False
-        if mode == "today" and ev_date == today:
-            include = True
-        elif mode == "tomorrow" and ev_date == tomorrow:
-            include = True
-        elif mode == "week" and today <= ev_date <= week_end:
-            include = True
-        elif mode == "all":
-            include = True
-
-        # today 模式额外：明日高优先级也纳入
-        if mode == "today" and ev_date == tomorrow and e["priority"] == "高":
-            include = True
-
-        if include:
-            result.append(e)
-
-    result.sort(key=lambda x: (x["date"], x["start_min"] or 0))
-    return result
-
-
-def build_reminder_time(ev: dict, now: datetime) -> Optional[str]:
-    """计算提醒时刻的 cron 表达式（单次）"""
-    if ev["start_min"] is None:
-        return None
-    try:
-        ev_date = date.fromisoformat(ev["date"])
-    except ValueError:
-        return None
-
-    remind_total_min = ev["start_min"] - ev["advance_min"]
-    if remind_total_min < 0:
-        remind_total_min = 0
-
-    r_hour = remind_total_min // 60
-    r_min = remind_total_min % 60
-
-    # 单次 cron：分 时 日 月 *
-    return f"{r_min} {r_hour} {ev_date.day} {ev_date.month} *"
-
-
-def build_reminder_time_str(ev: dict) -> str:
-    """人类可读的提醒时刻"""
-    if ev["start_min"] is None:
-        return ""
-    remind_total_min = max(0, ev["start_min"] - ev["advance_min"])
-    r_hour = remind_total_min // 60
-    r_min = remind_total_min % 60
-    return f"{r_hour:02d}:{r_min:02d}"
-
-
-def main():
-    args = parse_args()
-    content = read_content(args)
-    cfg = parse_global_config(content)
-    now = datetime.now()
-    if args.now:
-        try:
-            now = datetime.strptime(args.now, "%Y-%m-%d %H:%M")
-        except ValueError:
-            pass
-
-    lines = content.splitlines()
-
-    # 解析用户自定义列名映射
-    user_map = parse_user_column_map(args.column_map) if args.column_map else None
-
-    # 解析主日程表
-    schedule_result = parse_table_section(
-        lines, "schedule", cfg["default_advance_minutes"], user_map
-    )
-    events = schedule_result["events"]
-    # 解析任务表
-    tasks = parse_task_section(lines, cfg["default_advance_minutes"])
-    all_events = events + tasks
-
-    # 冲突检测（仅对 schedule 类事项）
-    conflicts = detect_conflicts(events)
-
-    # 筛选近期/今日事项
-    upcoming = filter_upcoming(all_events, args.mode, now)
-
-    # 为每个事项计算提醒时刻
-    for ev in upcoming:
-        ev["remind_cron"] = build_reminder_time(ev, now)
-        ev["remind_time_str"] = build_reminder_time_str(ev)
-
-    output = {
-        "meta": {
-            "mode": args.mode,
-            "now": now.strftime("%Y-%m-%d %H:%M"),
-            "config": cfg,
-            "parse_mode": schedule_result["mode"],
-        },
-        "summary": {
-            "total_events": len(all_events),
-            "upcoming_count": len(upcoming),
-            "conflict_count": len(conflicts),
-        },
-        "upcoming": upcoming,
-        "conflicts": conflicts,
-    }
-
-    # 列名问题提示（宽松模式下，让 agent 告知用户）
-    if schedule_result["unresolved_columns"]:
-        output["warnings"] = {
-            "unresolved_columns": schedule_result["unresolved_columns"],
-            "message": (
-                f"⚠️ 以下列名无法自动识别，其内容已忽略："
-                f"{', '.join(schedule_result['unresolved_columns'])}。"
-                f"如需记录这些列，请告诉我它们对应的字段（如：'会期' = 日期，'主旨' = 事项）。"
-            ),
-        }
-    if schedule_result["missing_required"]:
-        output["errors"] = {
-            "missing_required": schedule_result["missing_required"],
-            "message": (
-                f"❌ 表格缺少必要字段：{', '.join(schedule_result['missing_required'])}，"
-                f"无法解析。请确认表格中包含日期列和事项/标题列。"
-            ),
-        }
-
-    print(json.dumps(output, ensure_ascii=False, indent=2))
-
-
-if __name__ == "__main__":
-    main()
+#!/usr/bin/env python3
+"""
+parse_schedule.py - 解析 Markdown 日程表，输出今日/近期事项、冲突检测结果
+
+用法：
+  python3 parse_schedule.py --file /home/node/.openclaw/workspace/日程.md [--mode today|tomorrow|week|all]
+  python3 parse_schedule.py --stdin [--mode today]  # 从 stdin 读取 Markdown 内容
+
+输出：JSON，包含 events / upcoming / conflicts / recurring / tasks
+"""
+from __future__ import annotations
+
+import sys
+import re
+import json
+import argparse
+from datetime import datetime, date, timedelta
+from typing import Optional, List, Dict
+
+
+PRIORITY_ORDER = {"高": 0, "中": 1, "低": 2, "": 1}
+
+TASK_KEYWORDS = {"提交", "截止", "deadline", "交付", "汇报", "发布", "上线"}
+
+
+def parse_args():
+    p = argparse.ArgumentParser(description="解析 Markdown 日程表")
+    p.add_argument("--file", help="Markdown 日程文件路径")
+    p.add_argument("--stdin", action="store_true", help="从 stdin 读取内容")
+    p.add_argument(
+        "--mode",
+        choices=["today", "tomorrow", "week", "all"],
+        default="today",
+        help="筛选范围（默认 today）",
+    )
+    p.add_argument("--now", help="覆盖当前时间（格式 YYYY-MM-DD HH:MM，测试用）")
+    p.add_argument(
+        "--column-map",
+        help=(
+            "自定义列名映射，格式：'自定义列名=字段名,...'\n"
+            "字段名可选：date/start/end/title/location/participants/priority/advance/note\n"
+            "示例：--column-map '会期=date,主旨=title,时间段开始=start'"
+        ),
+    )
+    return p.parse_args()
+
+
+def read_content(args) -> str:
+    if args.stdin:
+        return sys.stdin.read()
+    if args.file:
+        path = args.file.replace("~", __import__("os").path.expanduser("~"))
+        with open(path, encoding="utf-8") as f:
+            return f.read()
+    print("错误：请指定 --file 或 --stdin", file=sys.stderr)
+    sys.exit(1)
+
+
+def parse_global_config(content: str) -> dict:
+    """从文件头部注释和引用块提取全局配置"""
+    cfg = {
+        "channel": "dingtalk",
+        "default_advance_minutes": 15,
+        "timezone": "Asia/Shanghai",
+    }
+    for line in content.splitlines():
+        line = line.strip()
+        m = re.match(r">\s*\*?\*?提醒渠道\*?\*?[：:]\s*(\S+)", line)
+        if m:
+            cfg["channel"] = m.group(1).strip()
+        m = re.match(r">\s*\*?\*?默认提前提醒\*?\*?[：:]\s*(\d+)", line)
+        if m:
+            cfg["default_advance_minutes"] = int(m.group(1))
+        m = re.match(r">\s*\*?\*?时区\*?\*?[：:]\s*(\S+)", line)
+        if m:
+            cfg["timezone"] = m.group(1)
+    return cfg
+
+
+def normalize_date(raw: str) -> Optional[date]:
+    raw = raw.strip()
+    for fmt in ("%Y-%m-%d", "%Y/%m/%d", "%m-%d", "%m/%d"):
+        try:
+            d = datetime.strptime(raw, fmt)
+            if d.year == 1900:
+                d = d.replace(year=date.today().year)
+            return d.date()
+        except ValueError:
+            pass
+    return None
+
+
+def normalize_time(raw: str) -> Optional[int]:
+    """返回分钟数（从当天0点起）"""
+    raw = raw.strip().replace("：", ":")
+    m = re.match(r"(\d{1,2}):(\d{2})", raw)
+    if m:
+        return int(m.group(1)) * 60 + int(m.group(2))
+    return None
+
+
+# ─────────────────────────────────────────────────────────
+# 列名自动映射（宽松模式，支持不标准的 Excel/CSV 转换结果）
+# ─────────────────────────────────────────────────────────
+
+# 各字段的候选列名（支持中英文、简写、常见变体）
+COLUMN_ALIASES = {
+    "date":         {"日期", "date", "Date", "DATE", "时间", "日", "活动日期", "会议日期", "事件日期",
+                     "日程日期", "计划日期"},
+    "start":        {"开始时间", "开始", "start", "Start", "START", "开始时刻", "起始时间", "from", "From",
+                     "开始时间段", "时间开始", "起始"},
+    "end":          {"结束时间", "结束", "end", "End", "END", "结束时刻", "to", "To",
+                     "时间结束", "截止", "终止"},
+    "title":        {"事项", "标题", "title", "Title", "TITLE", "内容", "会议", "活动", "事件",
+                     "subject", "Subject", "name", "Name", "任务", "描述", "会议名称", "活动名称",
+                     "事项名称", "议题", "主题", "项目", "meeting", "Meeting", "event", "Event",
+                     "topic", "Topic", "item", "Item"},
+    "location":     {"地点", "location", "Location", "LOCATION", "地址", "会议室", "venue", "Venue",
+                     "place", "Place", "room", "Room", "场所", "场地"},
+    "participants": {"参与者", "人员", "participants", "Participants", "与会者", "成员", "attendees",
+                     "Attendees", "负责人", "owner", "Owner", "参会人", "参会人员", "出席人员",
+                     "相关人员", "联系人"},
+    "priority":     {"优先级", "priority", "Priority", "PRIORITY", "重要程度", "紧急度", "级别",
+                     "importance", "Importance", "urgency", "Urgency"},
+    "advance":      {"提前(分钟)", "提前提醒", "提前", "advance", "Advance", "reminder", "Reminder",
+                     "提前时间", "提醒时间", "提前分钟"},
+    "note":         {"备注", "note", "Note", "NOTE", "notes", "Notes", "说明", "描述", "remark",
+                     "Remark", "comments", "Comments", "附注", "补充", "其他"},
+}
+
+
+def parse_user_column_map(raw: str) -> dict:
+    """
+    解析用户自定义列名映射字符串。
+    输入格式：'会期=date,主旨=title,时间段开始=start'
+    返回：{'会期': 'date', '主旨': 'title', '时间段开始': 'start'}
+    """
+    VALID_FIELDS = {"date", "start", "end", "title", "location",
+                    "participants", "priority", "advance", "note"}
+    result = {}
+    for part in raw.split(","):
+        part = part.strip()
+        if "=" not in part:
+            continue
+        custom_name, _, field = part.partition("=")
+        custom_name = custom_name.strip()
+        field = field.strip()
+        if field in VALID_FIELDS and custom_name:
+            result[custom_name] = field
+    return result
+
+
+def detect_column_mapping(header_cells: List[str],
+                           user_map: Optional[dict] = None) -> dict:
+    """
+    根据表头自动推断列索引映射。
+    优先级：用户自定义映射 > 精确别名匹配 > 子串匹配。
+    返回 {字段名: 列索引}
+    """
+    mapping: dict = {}
+    used_indices: set = set()
+
+    # 第零轮：用户自定义映射（最高优先级）
+    if user_map:
+        for i, cell in enumerate(header_cells):
+            cell_clean = cell.strip()
+            if cell_clean in user_map:
+                field = user_map[cell_clean]
+                if field not in mapping:  # 同一字段只取第一个
+                    mapping[field] = i
+                    used_indices.add(i)
+
+    # 第一轮：精确匹配
+    for field, aliases in COLUMN_ALIASES.items():
+        for i, cell in enumerate(header_cells):
+            if i in used_indices:
+                continue
+            if cell.strip() in aliases:
+                mapping[field] = i
+                used_indices.add(i)
+                break
+
+    # 第二轮：子串匹配（仅对未匹配的字段和未使用的列）
+    for field, aliases in COLUMN_ALIASES.items():
+        if field in mapping:
+            continue
+        for i, cell in enumerate(header_cells):
+            if i in used_indices:
+                continue
+            cell_clean = cell.strip()
+            matched = False
+            for alias in aliases:
+                if not alias:
+                    continue
+                # 列名包含别名（如"会议名称"包含"名称"→title）
+                # 但要避免短词误匹配（别名长度至少2）
+                if len(alias) >= 2 and (alias in cell_clean or cell_clean in alias):
+                    mapping[field] = i
+                    used_indices.add(i)
+                    matched = True
+                    break
+            if matched:
+                break
+
+    return mapping
+
+
+def report_unresolved_columns(header_cells: List[str], mapping: dict) -> List[str]:
+    """返回未能映射到任何已知字段的列名列表"""
+    mapped_indices = set(mapping.values())
+    return [
+        header_cells[i].strip()
+        for i in range(len(header_cells))
+        if i not in mapped_indices and header_cells[i].strip()
+    ]
+
+
+def check_required_fields(mapping: dict) -> List[str]:
+    """检查必要字段是否都已映射，返回缺失字段名"""
+    required = ["date", "title"]
+    return [f for f in required if f not in mapping]
+
+
+def is_standard_header(cells: List[str]) -> bool:
+    """判断是否是标准格式表头（首列精确为日期/Date/date）"""
+    SCHEDULE_HEADERS = {"日期", "Date", "date"}
+    return bool(cells) and cells[0].strip() in SCHEDULE_HEADERS
+
+
+def is_any_schedule_header(cells: List[str]) -> bool:
+    """宽松判断：表头中是否含有日期相关列"""
+    date_aliases = COLUMN_ALIASES["date"]
+    title_aliases = COLUMN_ALIASES["title"]
+    has_date = any(c.strip() in date_aliases for c in cells)
+    has_title = any(c.strip() in title_aliases for c in cells)
+    return has_date and has_title
+
+
+def extract_event_by_mapping(cells: List[str], mapping: dict, default_advance: int,
+                              section_name: str) -> Optional[dict]:
+    """根据列映射从数据行提取事项"""
+    def get(field):
+        idx = mapping.get(field)
+        if idx is not None and idx < len(cells):
+            return cells[idx].strip()
+        return ""
+
+    date_val = normalize_date(get("date"))
+    if not date_val:
+        return None
+
+    title = get("title")
+    if not title:
+        return None
+
+    start_str = get("start")
+    end_str = get("end")
+    start_min = normalize_time(start_str) if start_str else None
+    end_min = normalize_time(end_str) if end_str else None
+
+    priority_raw = get("priority")
+    priority = priority_raw if priority_raw in ("高", "中", "低") else "中"
+
+    advance_raw = get("advance")
+    try:
+        advance_min = int(advance_raw) if advance_raw else default_advance
+    except ValueError:
+        advance_min = default_advance
+
+    return {
+        "date": date_val.isoformat(),
+        "start_min": start_min,
+        "end_min": end_min,
+        "start_str": start_str,
+        "end_str": end_str,
+        "title": title,
+        "location": get("location"),
+        "participants": get("participants"),
+        "priority": priority,
+        "advance_min": advance_min,
+        "note": get("note"),
+        "section": section_name,
+        "is_task": any(k in title + get("note") for k in TASK_KEYWORDS),
+    }
+
+
+def parse_table_section(lines: List[str], section_name: str,
+                         default_advance: int,
+                         user_map: Optional[dict] = None) -> dict:
+    """
+    解析日程表格。返回 dict：
+      events: 解析出的事项列表
+      unresolved_columns: 无法映射的列名列表（宽松模式下）
+      missing_required: 缺失的必要字段（date/title）
+      mode: 'standard' | 'loose' | 'failed'
+    """
+    in_table = False
+    header_found = False
+    column_mapping: Optional[dict] = None
+    loose_mode = False
+    events = []
+    unresolved_columns: List[str] = []
+    missing_required: List[str] = []
+    parse_mode = "failed"
+
+    TASK_HEADERS = {"截止日期", "DeadLine", "deadline", "Deadline"}
+
+    for line in lines:
+        stripped = line.strip()
+        if not stripped.startswith("|"):
+            if in_table:
+                in_table = False
+                header_found = False
+                column_mapping = None
+                loose_mode = False
+            continue
+
+        cells = [c.strip() for c in stripped.strip("|").split("|")]
+        if len(cells) < 2:
+            continue
+
+        # 跳过任务表头，不被主日程解析器误抓
+        if cells[0].strip() in TASK_HEADERS:
+            continue
+
+        # 检测表头
+        if not in_table:
+            if is_standard_header(cells):
+                START_HEADERS = {"开始时间", "开始", "start", "Start", "START", "起始时间"}
+                second_col = cells[1].strip() if len(cells) > 1 else ""
+                if second_col in START_HEADERS:
+                    # 标准模式
+                    in_table = True
+                    header_found = True
+                    loose_mode = False
+                    column_mapping = None
+                    parse_mode = "standard"
+                    continue
+                else:
+                    # 首列是"日期"但列序不标准，降级为宽松模式
+                    mapping = detect_column_mapping(cells, user_map)
+                    missing = check_required_fields(mapping)
+                    if not missing:
+                        in_table = True
+                        header_found = True
+                        loose_mode = True
+                        column_mapping = mapping
+                        parse_mode = "loose"
+                        unresolved_columns = report_unresolved_columns(cells, mapping)
+                        missing_required = []
+                    else:
+                        missing_required = missing
+                        parse_mode = "failed"
+                    continue
+            elif is_any_schedule_header(cells) or user_map:
+                # 宽松模式：自动映射 + 用户自定义映射
+                mapping = detect_column_mapping(cells, user_map)
+                missing = check_required_fields(mapping)
+                if not missing:
+                    in_table = True
+                    header_found = True
+                    loose_mode = True
+                    column_mapping = mapping
+                    parse_mode = "loose"
+                    unresolved_columns = report_unresolved_columns(cells, mapping)
+                    missing_required = []
+                else:
+                    missing_required = missing
+                    parse_mode = "failed"
+                continue
+            continue
+            continue
+
+        # 跳过分隔行
+        if header_found and all(re.match(r"^[-:]+$", c) for c in cells if c):
+            continue
+
+        if not in_table:
+            continue
+
+        if loose_mode and column_mapping:
+            # 宽松模式：用列映射提取
+            ev = extract_event_by_mapping(cells, column_mapping, default_advance, section_name)
+            if ev:
+                events.append(ev)
+        else:
+            # 标准模式：按固定列序提取
+            if len(cells) < 4:
+                continue
+            date_val = normalize_date(cells[0]) if cells[0] else None
+            if not date_val:
+                continue
+            start_min = normalize_time(cells[1]) if len(cells) > 1 else None
+            end_min = normalize_time(cells[2]) if len(cells) > 2 else None
+            title = cells[3].strip() if len(cells) > 3 else ""
+            location = cells[4].strip() if len(cells) > 4 else ""
+            participants = cells[5].strip() if len(cells) > 5 else ""
+            priority = cells[6].strip() if len(cells) > 6 else "中"
+            advance_raw = cells[7].strip() if len(cells) > 7 else ""
+            note = cells[8].strip() if len(cells) > 8 else ""
+            if not title:
+                continue
+            try:
+                advance_min = int(advance_raw) if advance_raw else default_advance
+            except ValueError:
+                advance_min = default_advance
+            events.append({
+                "date": date_val.isoformat(),
+                "start_min": start_min,
+                "end_min": end_min,
+                "start_str": cells[1].strip() if len(cells) > 1 else "",
+                "end_str": cells[2].strip() if len(cells) > 2 else "",
+                "title": title,
+                "location": location,
+                "participants": participants,
+                "priority": priority if priority in ("高", "中", "低") else "中",
+                "advance_min": advance_min,
+                "note": note,
+                "section": section_name,
+                "is_task": any(k in title + note for k in TASK_KEYWORDS),
+            })
+
+    return {
+        "events": events,
+        "mode": parse_mode,
+        "unresolved_columns": unresolved_columns,
+        "missing_required": missing_required,
+    }
+
+
+def parse_task_section(lines: List[str], default_advance: int) -> list:
+    """解析待提醒任务表（截止日期形式）"""
+    in_table = False
+    header_found = False
+    tasks = []
+
+    TASK_HEADERS = {"截止日期", "DeadLine", "deadline", "Deadline"}
+
+    for line in lines:
+        stripped = line.strip()
+        if not stripped.startswith("|"):
+            if in_table:
+                in_table = False
+                header_found = False
+            continue
+
+        cells = [c.strip() for c in stripped.strip("|").split("|")]
+        if not cells:
+            continue
+
+        if cells[0] in TASK_HEADERS:
+            in_table = True
+            header_found = True
+            continue
+
+        if header_found and all(re.match(r"^[-:]+$", c) for c in cells if c):
+            continue
+
+        if not in_table:
+            continue
+
+        if len(cells) < 2:
+            continue
+
+        date_val = normalize_date(cells[0]) if cells[0] else None
+        if not date_val:
+            continue
+
+        title = cells[1].strip() if len(cells) > 1 else ""
+        priority = cells[3].strip() if len(cells) > 3 else "中"
+        advance_raw = cells[4].strip() if len(cells) > 4 else ""
+        note = cells[5].strip() if len(cells) > 5 else ""
+
+        # 解析"提前1天"、"提前2天"形式
+        advance_min = default_advance
+        m = re.search(r"提前\s*(\d+)\s*天", advance_raw)
+        if m:
+            advance_min = int(m.group(1)) * 24 * 60
+        else:
+            m = re.search(r"(\d+)", advance_raw)
+            if m:
+                advance_min = int(m.group(1))
+
+        tasks.append({
+            "date": date_val.isoformat(),
+            "start_min": 9 * 60,  # 默认早上9点触发
+            "end_min": 9 * 60 + 30,
+            "start_str": "09:00",
+            "end_str": "09:30",
+            "title": f"📋 {title}",
+            "location": "",
+            "participants": cells[2].strip() if len(cells) > 2 else "",
+            "priority": priority if priority in ("高", "中", "低") else "中",
+            "advance_min": advance_min,
+            "note": note,
+            "section": "tasks",
+            "is_task": True,
+        })
+
+    return tasks
+
+
+def detect_conflicts(events: list[dict]) -> list[dict]:
+    """检测同日时间冲突"""
+    by_date: dict[str, list[dict]] = {}
+    for e in events:
+        if e["start_min"] is None or e["end_min"] is None:
+            continue
+        by_date.setdefault(e["date"], []).append(e)
+
+    conflicts = []
+    for day, day_events in by_date.items():
+        sorted_events = sorted(day_events, key=lambda x: x["start_min"])
+        for i in range(len(sorted_events)):
+            for j in range(i + 1, len(sorted_events)):
+                a, b = sorted_events[i], sorted_events[j]
+                if a["end_min"] <= b["start_min"]:
+                    # 不重叠，但检查是否过于紧凑（< 5分钟间隔）
+                    gap = b["start_min"] - a["end_min"]
+                    if 0 < gap < 5:
+                        conflicts.append({
+                            "type": "tight",
+                            "date": day,
+                            "event_a": a["title"],
+                            "event_b": b["title"],
+                            "a_time": f"{a['start_str']}–{a['end_str']}",
+                            "b_time": f"{b['start_str']}–{b['end_str']}",
+                            "gap_min": gap,
+                            "message": (
+                                f"⚠️ 紧凑安排：【{a['title']}】{a['start_str']}–{a['end_str']} "
+                                f"与【{b['title']}】{b['start_str']} 之间仅 {gap} 分钟，几乎没有缓冲时间。"
+                            ),
+                        })
+                else:
+                    # 重叠
+                    overlap = a["end_min"] - b["start_min"]
+                    conflicts.append({
+                        "type": "overlap",
+                        "date": day,
+                        "event_a": a["title"],
+                        "event_b": b["title"],
+                        "a_time": f"{a['start_str']}–{a['end_str']}",
+                        "b_time": f"{b['start_str']}–{b['end_str']}",
+                        "overlap_min": overlap,
+                        "message": (
+                            f"⚠️ 时间冲突：【{a['title']}】{a['start_str']}–{a['end_str']} "
+                            f"与【{b['title']}】{b['start_str']}–{b['end_str']} "
+                            f"重叠 {overlap} 分钟。"
+                        ),
+                    })
+    return conflicts
+
+
+def filter_upcoming(events: list, mode: str, now: datetime) -> list:
+    """按 mode 筛选需要提醒的事项"""
+    today = now.date()
+    tomorrow = today + timedelta(days=1)
+    week_end = today + timedelta(days=7)
+    now_min = now.hour * 60 + now.minute
+
+    result = []
+    for e in events:
+        try:
+            ev_date = date.fromisoformat(e["date"])
+        except ValueError:
+            continue
+
+        # 已过期跳过
+        if ev_date < today:
+            continue
+        if ev_date == today and e["end_min"] is not None and e["end_min"] < now_min:
+            continue
+
+        # 低优先级且非任务，默认跳过
+        if e["priority"] == "低" and not e["is_task"]:
+            continue
+
+        include = False
+        if mode == "today" and ev_date == today:
+            include = True
+        elif mode == "tomorrow" and ev_date == tomorrow:
+            include = True
+        elif mode == "week" and today <= ev_date <= week_end:
+            include = True
+        elif mode == "all":
+            include = True
+
+        # today 模式额外：明日高优先级也纳入
+        if mode == "today" and ev_date == tomorrow and e["priority"] == "高":
+            include = True
+
+        if include:
+            result.append(e)
+
+    result.sort(key=lambda x: (x["date"], x["start_min"] or 0))
+    return result
+
+
+def build_reminder_time(ev: dict, now: datetime) -> Optional[str]:
+    """计算提醒时刻的 cron 表达式（单次）"""
+    if ev["start_min"] is None:
+        return None
+    try:
+        ev_date = date.fromisoformat(ev["date"])
+    except ValueError:
+        return None
+
+    remind_total_min = ev["start_min"] - ev["advance_min"]
+    if remind_total_min < 0:
+        remind_total_min = 0
+
+    r_hour = remind_total_min // 60
+    r_min = remind_total_min % 60
+
+    # 单次 cron：分 时 日 月 *
+    return f"{r_min} {r_hour} {ev_date.day} {ev_date.month} *"
+
+
+def build_reminder_time_str(ev: dict) -> str:
+    """人类可读的提醒时刻"""
+    if ev["start_min"] is None:
+        return ""
+    remind_total_min = max(0, ev["start_min"] - ev["advance_min"])
+    r_hour = remind_total_min // 60
+    r_min = remind_total_min % 60
+    return f"{r_hour:02d}:{r_min:02d}"
+
+
+def main():
+    args = parse_args()
+    content = read_content(args)
+    cfg = parse_global_config(content)
+    now = datetime.now()
+    if args.now:
+        try:
+            now = datetime.strptime(args.now, "%Y-%m-%d %H:%M")
+        except ValueError:
+            pass
+
+    lines = content.splitlines()
+
+    # 解析用户自定义列名映射
+    user_map = parse_user_column_map(args.column_map) if args.column_map else None
+
+    # 解析主日程表
+    schedule_result = parse_table_section(
+        lines, "schedule", cfg["default_advance_minutes"], user_map
+    )
+    events = schedule_result["events"]
+    # 解析任务表
+    tasks = parse_task_section(lines, cfg["default_advance_minutes"])
+    all_events = events + tasks
+
+    # 冲突检测（仅对 schedule 类事项）
+    conflicts = detect_conflicts(events)
+
+    # 筛选近期/今日事项
+    upcoming = filter_upcoming(all_events, args.mode, now)
+
+    # 为每个事项计算提醒时刻
+    for ev in upcoming:
+        ev["remind_cron"] = build_reminder_time(ev, now)
+        ev["remind_time_str"] = build_reminder_time_str(ev)
+
+    output = {
+        "meta": {
+            "mode": args.mode,
+            "now": now.strftime("%Y-%m-%d %H:%M"),
+            "config": cfg,
+            "parse_mode": schedule_result["mode"],
+        },
+        "summary": {
+            "total_events": len(all_events),
+            "upcoming_count": len(upcoming),
+            "conflict_count": len(conflicts),
+        },
+        "upcoming": upcoming,
+        "conflicts": conflicts,
+    }
+
+    # 列名问题提示（宽松模式下，让 agent 告知用户）
+    if schedule_result["unresolved_columns"]:
+        output["warnings"] = {
+            "unresolved_columns": schedule_result["unresolved_columns"],
+            "message": (
+                f"⚠️ 以下列名无法自动识别，其内容已忽略："
+                f"{', '.join(schedule_result['unresolved_columns'])}。"
+                f"如需记录这些列，请告诉我它们对应的字段（如：'会期' = 日期，'主旨' = 事项）。"
+            ),
+        }
+    if schedule_result["missing_required"]:
+        output["errors"] = {
+            "missing_required": schedule_result["missing_required"],
+            "message": (
+                f"❌ 表格缺少必要字段：{', '.join(schedule_result['missing_required'])}，"
+                f"无法解析。请确认表格中包含日期列和事项/标题列。"
+            ),
+        }
+
+    print(json.dumps(output, ensure_ascii=False, indent=2))
+
+
+if __name__ == "__main__":
+    main()