sophhub 0.4.19 → 0.4.21

package/skills/sophnet-schedule/src/services/table_parser.py

@@ -1,286 +1,286 @@
-"""
-Markdown 表格解析器。
-
-从 markitdown 或 OCR 输出的 Markdown 内容中提取结构化日程事件。
-移植自 skills/schedule-reminder/scripts/parse_schedule.py 的列名识别逻辑，
-输出格式与 add_event.py / import_events.py 兼容。
-
-用法：
-    from services.table_parser import parse_events_from_markdown
-    events = parse_events_from_markdown(md_content)
-    # events: [{"summary": "...", "date": "YYYY-MM-DD", "start": "HH:MM", ...}, ...]
-"""
-
-from __future__ import annotations
-
-import re
-import sys
-from datetime import date, datetime
-from pathlib import Path
-from typing import Optional
-
-sys.path.insert(0, str(Path(__file__).parent.parent))
-
-from services.datetime_utils import current_year
-
-# ─────────────────────────── 列名别名表 ──────────────────────────
-# 各字段的候选列名（中英文、简写、Excel 常见变体）
-
-COLUMN_ALIASES: dict[str, set[str]] = {
-    "date": {
-        "日期", "date", "Date", "DATE", "日", "活动日期", "会议日期",
-        "事件日期", "日程日期", "计划日期", "时间",
-    },
-    "start": {
-        "开始时间", "开始", "start", "Start", "START", "开始时刻",
-        "起始时间", "from", "From", "时间开始", "起始", "开始时间段",
-    },
-    "end": {
-        "结束时间", "结束", "end", "End", "END", "结束时刻",
-        "to", "To", "时间结束", "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",
-    },
-    "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", "参会人", "参会人员",
-    },
-    "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", "remark", "Remark", "comments", "附注", "补充",
-    },
-}
-
-_DEFAULT_DURATION_MINUTES = 60
-_DEFAULT_ADVANCE_MINUTES  = 15
-
-
-# ─────────────────────────── 列名检测 ────────────────────────────
-
-def detect_column_mapping(
-    header_cells: list[str],
-    user_map: dict[str, str] | None = None,
-) -> dict[str, int]:
-    """
-    根据表头行自动推断字段→列索引映射。
-
-    优先级：用户自定义 > 精确别名匹配 > 子串匹配。
-
-    Args:
-        header_cells: Markdown 表头各列文字列表
-        user_map:     自定义映射 {"自定义列名": "字段名", ...}
-
-    Returns:
-        {"title": 0, "date": 1, "start": 2, ...}
-    """
-    mapping: dict[str, int] = {}
-    used_indices: set[int]  = set()
-
-    # 用户自定义（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:
-                    mapping[field] = i
-                    used_indices.add(i)
-
-    # 精确匹配
-    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
-            if cell.strip() in aliases:
-                mapping[field] = i
-                used_indices.add(i)
-                break
-
-    # 子串匹配（兜底）
-    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_lower = cell.strip().lower()
-            if any(a.lower() in cell_lower or cell_lower in a.lower()
-                   for a in aliases):
-                mapping[field] = i
-                used_indices.add(i)
-                break
-
-    return mapping
-
-
-# ─────────────────────────── 时间规范化 ──────────────────────────
-
-def normalize_date(raw: str) -> Optional[str]:
-    """
-    将各种日期格式规范化为 YYYY-MM-DD，失败返回 None。
-    支持：2026-04-01 / 2026/04/01 / 04-01 / 4/1（补全年份）
-    """
-    raw = raw.strip()
-    for fmt in ("%Y-%m-%d", "%Y/%m/%d"):
-        try:
-            return datetime.strptime(raw, fmt).strftime("%Y-%m-%d")
-        except ValueError:
-            pass
-    for fmt in ("%m-%d", "%m/%d"):
-        try:
-            d = datetime.strptime(raw, fmt)
-            return d.replace(year=current_year()).strftime("%Y-%m-%d")
-        except ValueError:
-            pass
-    return None
-
-
-def normalize_time(raw: str) -> Optional[str]:
-    """
-    将时间字符串规范化为 HH:MM，失败返回 None。
-    支持：14:30 / 14:30:00 / 9:05
-    """
-    raw = raw.strip().replace("：", ":")
-    m = re.match(r"(\d{1,2}):(\d{2})", raw)
-    if m:
-        return f"{int(m.group(1)):02d}:{int(m.group(2)):02d}"
-    return None
-
-
-def add_minutes_to_time(time_str: str, minutes: int) -> str:
-    """HH:MM 加 N 分钟，返回 HH:MM（不跨日边界处理）。"""
-    h, m = map(int, time_str.split(":"))
-    total = h * 60 + m + minutes
-    return f"{(total // 60) % 24:02d}:{total % 60:02d}"
-
-
-# ─────────────────────────── 表格解析 ────────────────────────────
-
-def _split_md_row(line: str) -> list[str]:
-    """将 Markdown 表格行拆分为单元格列表（去掉首尾 |）。"""
-    cells = line.strip().strip("|").split("|")
-    return [c.strip() for c in cells]
-
-
-def _is_separator_row(line: str) -> bool:
-    """判断是否为 Markdown 表格分隔行（如 |---|---|）。"""
-    stripped = line.strip()
-    # 去掉首尾 | 后，只含 -、:、|、空格
-    inner = stripped.strip("|")
-    return bool(inner) and bool(re.match(r"^[\s\-:|]+$", inner))
-
-
-def parse_events_from_markdown(
-    content: str,
-    user_map: dict[str, str] | None = None,
-    default_advance: int = _DEFAULT_ADVANCE_MINUTES,
-    default_duration: int = _DEFAULT_DURATION_MINUTES,
-) -> list[dict]:
-    """
-    从 Markdown 内容中提取所有日程事件。
-
-    每个事件输出结构：
-        {
-            "summary":  str,           # 必填
-            "date":     "YYYY-MM-DD",  # 必填
-            "start":    "HH:MM",       # 必填
-            "end":      "HH:MM",       # 可选
-            "location": str,           # 可选
-            "description": str,        # 备注
-            "remind":   int,           # 提醒分钟数（默认 15）
-        }
-
-    跳过：title/date/start 任一缺失或解析失败的行。
-    """
-    events: list[dict] = []
-    lines   = content.splitlines()
-    i       = 0
-
-    while i < len(lines):
-        line = lines[i]
-
-        # 找到一个表格的表头行
-        if not line.strip().startswith("|"):
-            i += 1
-            continue
-
-        header_cells = _split_md_row(line)
-
-        # 下一行应为分隔行
-        if i + 1 >= len(lines) or not _is_separator_row(lines[i + 1]):
-            i += 1
-            continue
-
-        col_map = detect_column_mapping(header_cells, user_map)
-
-        # title / date / start 是必需字段，至少要有 title
-        if "title" not in col_map:
-            i += 2
-            continue
-
-        # 读数据行
-        i += 2
-        while i < len(lines) and lines[i].strip().startswith("|"):
-            cells = _split_md_row(lines[i])
-            i += 1
-
-            def _cell(field: str) -> str:
-                idx = col_map.get(field)
-                if idx is None or idx >= len(cells):
-                    return ""
-                return cells[idx].strip()
-
-            title    = _cell("title")
-            date_raw = _cell("date")
-            start_raw= _cell("start")
-
-            if not title or title in ("-", ""):
-                continue
-
-            date_str  = normalize_date(date_raw)  if date_raw  else None
-            start_str = normalize_time(start_raw) if start_raw else None
-
-            if not date_str or not start_str:
-                continue
-
-            end_raw = _cell("end")
-            end_str = normalize_time(end_raw) if end_raw else None
-            if not end_str:
-                end_str = add_minutes_to_time(start_str, default_duration)
-
-            advance_raw = _cell("advance")
-            try:
-                remind = int(advance_raw) if advance_raw else default_advance
-            except ValueError:
-                remind = default_advance
-
-            note     = _cell("note")
-            location = _cell("location")
-            parts_raw = _cell("participants")
-            description = " | ".join(filter(None, [note, parts_raw]))
-
-            events.append({
-                "summary":     title,
-                "date":        date_str,
-                "start":       start_str,
-                "end":         end_str,
-                "location":    location,
-                "description": description,
-                "remind":      remind,
-            })
-
-    return events
+"""
+Markdown 表格解析器。
+
+从 markitdown 或 OCR 输出的 Markdown 内容中提取结构化日程事件。
+移植自 skills/schedule-reminder/scripts/parse_schedule.py 的列名识别逻辑，
+输出格式与 add_event.py / import_events.py 兼容。
+
+用法：
+    from services.table_parser import parse_events_from_markdown
+    events = parse_events_from_markdown(md_content)
+    # events: [{"summary": "...", "date": "YYYY-MM-DD", "start": "HH:MM", ...}, ...]
+"""
+
+from __future__ import annotations
+
+import re
+import sys
+from datetime import date, datetime
+from pathlib import Path
+from typing import Optional
+
+sys.path.insert(0, str(Path(__file__).parent.parent))
+
+from services.datetime_utils import current_year
+
+# ─────────────────────────── 列名别名表 ──────────────────────────
+# 各字段的候选列名（中英文、简写、Excel 常见变体）
+
+COLUMN_ALIASES: dict[str, set[str]] = {
+    "date": {
+        "日期", "date", "Date", "DATE", "日", "活动日期", "会议日期",
+        "事件日期", "日程日期", "计划日期", "时间",
+    },
+    "start": {
+        "开始时间", "开始", "start", "Start", "START", "开始时刻",
+        "起始时间", "from", "From", "时间开始", "起始", "开始时间段",
+    },
+    "end": {
+        "结束时间", "结束", "end", "End", "END", "结束时刻",
+        "to", "To", "时间结束", "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",
+    },
+    "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", "参会人", "参会人员",
+    },
+    "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", "remark", "Remark", "comments", "附注", "补充",
+    },
+}
+
+_DEFAULT_DURATION_MINUTES = 60
+_DEFAULT_ADVANCE_MINUTES  = 15
+
+
+# ─────────────────────────── 列名检测 ────────────────────────────
+
+def detect_column_mapping(
+    header_cells: list[str],
+    user_map: dict[str, str] | None = None,
+) -> dict[str, int]:
+    """
+    根据表头行自动推断字段→列索引映射。
+
+    优先级：用户自定义 > 精确别名匹配 > 子串匹配。
+
+    Args:
+        header_cells: Markdown 表头各列文字列表
+        user_map:     自定义映射 {"自定义列名": "字段名", ...}
+
+    Returns:
+        {"title": 0, "date": 1, "start": 2, ...}
+    """
+    mapping: dict[str, int] = {}
+    used_indices: set[int]  = set()
+
+    # 用户自定义（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:
+                    mapping[field] = i
+                    used_indices.add(i)
+
+    # 精确匹配
+    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
+            if cell.strip() in aliases:
+                mapping[field] = i
+                used_indices.add(i)
+                break
+
+    # 子串匹配（兜底）
+    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_lower = cell.strip().lower()
+            if any(a.lower() in cell_lower or cell_lower in a.lower()
+                   for a in aliases):
+                mapping[field] = i
+                used_indices.add(i)
+                break
+
+    return mapping
+
+
+# ─────────────────────────── 时间规范化 ──────────────────────────
+
+def normalize_date(raw: str) -> Optional[str]:
+    """
+    将各种日期格式规范化为 YYYY-MM-DD，失败返回 None。
+    支持：2026-04-01 / 2026/04/01 / 04-01 / 4/1（补全年份）
+    """
+    raw = raw.strip()
+    for fmt in ("%Y-%m-%d", "%Y/%m/%d"):
+        try:
+            return datetime.strptime(raw, fmt).strftime("%Y-%m-%d")
+        except ValueError:
+            pass
+    for fmt in ("%m-%d", "%m/%d"):
+        try:
+            d = datetime.strptime(raw, fmt)
+            return d.replace(year=current_year()).strftime("%Y-%m-%d")
+        except ValueError:
+            pass
+    return None
+
+
+def normalize_time(raw: str) -> Optional[str]:
+    """
+    将时间字符串规范化为 HH:MM，失败返回 None。
+    支持：14:30 / 14:30:00 / 9:05
+    """
+    raw = raw.strip().replace("：", ":")
+    m = re.match(r"(\d{1,2}):(\d{2})", raw)
+    if m:
+        return f"{int(m.group(1)):02d}:{int(m.group(2)):02d}"
+    return None
+
+
+def add_minutes_to_time(time_str: str, minutes: int) -> str:
+    """HH:MM 加 N 分钟，返回 HH:MM（不跨日边界处理）。"""
+    h, m = map(int, time_str.split(":"))
+    total = h * 60 + m + minutes
+    return f"{(total // 60) % 24:02d}:{total % 60:02d}"
+
+
+# ─────────────────────────── 表格解析 ────────────────────────────
+
+def _split_md_row(line: str) -> list[str]:
+    """将 Markdown 表格行拆分为单元格列表（去掉首尾 |）。"""
+    cells = line.strip().strip("|").split("|")
+    return [c.strip() for c in cells]
+
+
+def _is_separator_row(line: str) -> bool:
+    """判断是否为 Markdown 表格分隔行（如 |---|---|）。"""
+    stripped = line.strip()
+    # 去掉首尾 | 后，只含 -、:、|、空格
+    inner = stripped.strip("|")
+    return bool(inner) and bool(re.match(r"^[\s\-:|]+$", inner))
+
+
+def parse_events_from_markdown(
+    content: str,
+    user_map: dict[str, str] | None = None,
+    default_advance: int = _DEFAULT_ADVANCE_MINUTES,
+    default_duration: int = _DEFAULT_DURATION_MINUTES,
+) -> list[dict]:
+    """
+    从 Markdown 内容中提取所有日程事件。
+
+    每个事件输出结构：
+        {
+            "summary":  str,           # 必填
+            "date":     "YYYY-MM-DD",  # 必填
+            "start":    "HH:MM",       # 必填
+            "end":      "HH:MM",       # 可选
+            "location": str,           # 可选
+            "description": str,        # 备注
+            "remind":   int,           # 提醒分钟数（默认 15）
+        }
+
+    跳过：title/date/start 任一缺失或解析失败的行。
+    """
+    events: list[dict] = []
+    lines   = content.splitlines()
+    i       = 0
+
+    while i < len(lines):
+        line = lines[i]
+
+        # 找到一个表格的表头行
+        if not line.strip().startswith("|"):
+            i += 1
+            continue
+
+        header_cells = _split_md_row(line)
+
+        # 下一行应为分隔行
+        if i + 1 >= len(lines) or not _is_separator_row(lines[i + 1]):
+            i += 1
+            continue
+
+        col_map = detect_column_mapping(header_cells, user_map)
+
+        # title / date / start 是必需字段，至少要有 title
+        if "title" not in col_map:
+            i += 2
+            continue
+
+        # 读数据行
+        i += 2
+        while i < len(lines) and lines[i].strip().startswith("|"):
+            cells = _split_md_row(lines[i])
+            i += 1
+
+            def _cell(field: str) -> str:
+                idx = col_map.get(field)
+                if idx is None or idx >= len(cells):
+                    return ""
+                return cells[idx].strip()
+
+            title    = _cell("title")
+            date_raw = _cell("date")
+            start_raw= _cell("start")
+
+            if not title or title in ("-", ""):
+                continue
+
+            date_str  = normalize_date(date_raw)  if date_raw  else None
+            start_str = normalize_time(start_raw) if start_raw else None
+
+            if not date_str or not start_str:
+                continue
+
+            end_raw = _cell("end")
+            end_str = normalize_time(end_raw) if end_raw else None
+            if not end_str:
+                end_str = add_minutes_to_time(start_str, default_duration)
+
+            advance_raw = _cell("advance")
+            try:
+                remind = int(advance_raw) if advance_raw else default_advance
+            except ValueError:
+                remind = default_advance
+
+            note     = _cell("note")
+            location = _cell("location")
+            parts_raw = _cell("participants")
+            description = " | ".join(filter(None, [note, parts_raw]))
+
+            events.append({
+                "summary":     title,
+                "date":        date_str,
+                "start":       start_str,
+                "end":         end_str,
+                "location":    location,
+                "description": description,
+                "remind":      remind,
+            })
+
+    return events