weflow-cli 1.15.1 → 1.17.0

package/scripts/promises.py

@@ -1,185 +1,185 @@
-"""欠账清单：我说过要给谁做什么，后来做了没有。
-
-这是微信永远回答不了的问题 —— 它知道"我明天发你"，但不知道"兑现"是什么。
-
-**默认不调模型**：先用规则把候选捞出来（我发的 + 承诺句式），再在同一个会话之后的对话里
-找"完成信号 / 对方催问 / 对方确认"三类证据。规则会误报，所以每条都带着原话和后续，
-让读者自己判 —— 而不是给一个"AI 认为你欠了 7 件事"的黑箱。
-
-用法：
-    python scripts/promises.py --days 60
-    python scripts/promises.py --days 90 --json
-    python scripts/promises.py --days 60 --talker 老表
-"""
-from __future__ import annotations
-
-import argparse
-import json
-import pathlib
-import re
-import sys
-from collections import defaultdict
-from datetime import datetime, timezone, timedelta
-
-SCRIPT_DIR = pathlib.Path(__file__).resolve().parent
-sys.path.insert(0, str(SCRIPT_DIR))
-
-from _utils import load_config, decrypt_lock  # noqa: E402
-import nt_decrypt as nd  # noqa: E402
-
-TZ = timezone(timedelta(hours=8))
-
-# 承诺句式：我 + 将来/稍后 + 一个交付动作。宁可多收，后面还有证据过滤。
-PROMISE_RE = re.compile(
-    r'(我|这边)[^。！？\n]{0,12}'
-    r'(明天|后天|下周|下周一|晚点|稍后|回头|等会|等会儿|下午|晚上|今天之内|这周)[^。！？\n]{0,12}'
-    r'(发|给|寄|做|弄|整理|传|送|安排|联系|确认|回复|处理)')
-DELIVER_RE = re.compile(r'(发了|发你了|发您了|已发|寄了|弄好了|做好了|整理好了|搞定了|已上传|传了|安排了|回复了|处理好了)')
-CHASE_RE = re.compile(r'(发了吗|寄了吗|好了吗|弄好了吗|什么时候|还没|催一下|怎么还没|在吗.{0,6}(发|给))')
-ACK_RE = re.compile(r'^(收到|好的|好滴|谢谢|多谢|ok|OK|嗯嗯|好嘞)')
-_NOT_PROMISE = re.compile(r'(我明天要|我下周要|我想|我觉得|我昨天|我上次|我已经)')
-
-
-def _clean(text):
-    text = ' '.join(str(text or '').split())
-    return re.sub(r'^[A-Za-z0-9_\-]+@?(chatroom|openim)?\s*:\s*', '', text)
-
-
-def same_account(name, wxid):
-    """配置里的 wxid 与 `Name2Id` 里的账号名**可能是同一个人的两种写法**。
-
-    实测本机：配置是 `wxid_账号A_后缀`，而 `Name2Id` 里只有 `wxid_账号A`（没有后缀）——
-    精确比较一个都匹配不上，于是"我发的消息"一条也捞不到（第一版就是这么得到 0 条承诺的）。
-    """
-    n, w = (name or '').strip(), (wxid or '').strip()
-    if not n or not w:
-        return False
-    return n == w or n.startswith(w + '_') or w.startswith(n + '_')
-
-
-def load_messages(conns, days, my_wxid):
-    """{talker: [ (ts, is_me, sender, text) ]}，按时间升序。"""
-    since = int((datetime.now(TZ) - timedelta(days=days)).timestamp())
-    out = defaultdict(list)
-    for conn in conns:
-        c = conn.cursor()
-        try:
-            c.execute('SELECT rowid, user_name FROM Name2Id')
-            id2name = {r[0]: r[1] for r in c.fetchall()}
-        except Exception:
-            continue
-        try:
-            c.execute('SELECT user_name FROM Name2Id WHERE is_session = 1')
-            sessions = [r[0] for r in c.fetchall() if r[0]]
-        except Exception:
-            sessions = []
-        for talker in sessions:
-            tbl = 'Msg_%s' % __import__('hashlib').md5(talker.encode()).hexdigest()
-            try:
-                c.execute('SELECT COUNT(*) FROM sqlite_master WHERE name=?', (tbl,))
-                if c.fetchone()[0] == 0:
-                    continue
-                c.execute('SELECT create_time, real_sender_id, local_type, message_content FROM "%s" '
-                          'WHERE create_time >= ? ORDER BY create_time' % tbl, (since,))
-                for ts, sender, ltype, content in c.fetchall():
-                    if (int(ltype or 0) & 0xFFFFFFFF) != 1:
-                        continue
-                    text = _clean(nd._decode_content(content))
-                    if not text:
-                        continue
-                    who = id2name.get(sender, '')
-                    out[talker].append((int(ts), same_account(who, my_wxid), who, text))
-            except Exception:
-                continue
-    return out
-
-
-def classify(follow):
-    """看承诺之后发生了什么：兑现 / 被催 / 只有确认 / 无后续。"""
-    if any(f[1] and DELIVER_RE.search(f[3]) for f in follow):
-        return 'delivered', '我后来提到"已发/已弄好"'
-    chases = [f for f in follow if not f[1] and CHASE_RE.search(f[3])]
-    if chases:
-        return 'chased', '对方后来催过：' + chases[0][3][:40]
-    if any(not f[1] and ACK_RE.search(f[3]) for f in follow):
-        return 'acked', '对方回复了确认'
-    return 'unknown', '没有后续证据'
-
-
-def main():
-    ap = argparse.ArgumentParser(description='我说过要给谁做什么，后来做了没有')
-    ap.add_argument('--days', type=int, default=60)
-    ap.add_argument('--talker', default='', help='只看某个会话（wxid 或昵称子串）')
-    ap.add_argument('--limit', type=int, default=30, help='最多列几条（默认 30）')
-    ap.add_argument('--json', action='store_true')
-    args = ap.parse_args()
-
-    cfg = load_config()
-    db = cfg.get('ntDbPath') or ''
-    my_wxid = cfg.get('wxid') or ''
-    if not db:
-        raise SystemExit('配置里没有 ntDbPath')
-    key = decrypt_lock(cfg.get('ntKey', '') or '')
-    salt = decrypt_lock(cfg.get('ntSalt', '') or '')
-    pw = decrypt_lock(cfg.get('favPassphrase', '') or cfg.get('decryptKey', '') or '')
-    conns = nd.connect_message_shards(db, key, salt, pw)
-    if not conns:
-        raise SystemExit('打不开消息分片')
-
-    try:
-        by_talker = load_messages(conns, args.days, my_wxid)
-        # 昵称在 **contact.db**，消息库里没有 `contact` 表 —— 从消息库找一定落空，
-        # 输出里就只剩 wxid（和 on_this_day.py 同一个坑）。
-        name_map = {}
-        try:
-            from network import load_graph
-            people, _rooms, _members = load_graph(cfg)
-            name_map = {p['username']: (p['remark'] or p['nick'] or p['username'])
-                        for p in people.values()}
-        except Exception as exc:
-            print(f'[WARN] 读联系人失败（{type(exc).__name__}: {exc}），下面显示 id',
-                  file=sys.stderr)
-    finally:
-        for conn in conns:
-            try:
-                conn.close()
-            except Exception:
-                pass
-
-    items = []
-    for talker, msgs in by_talker.items():
-        shown = name_map.get(talker, talker)
-        if args.talker and args.talker not in (talker, shown):
-            continue
-        for i, (ts, is_me, who, text) in enumerate(msgs):
-            if not is_me or not PROMISE_RE.search(text) or _NOT_PROMISE.search(text):
-                continue
-            follow = msgs[i + 1:i + 20]
-            state, why = classify(follow)
-            items.append({
-                'time': datetime.fromtimestamp(ts, tz=TZ).strftime('%Y-%m-%d %H:%M'),
-                'talker': shown,
-                'said': text[:100],
-                'state': state,
-                'evidence': why,
-            })
-    items.sort(key=lambda x: x['time'], reverse=True)
-    items = items[:args.limit]
-
-    if args.json:
-        print(json.dumps({'success': True, 'days': args.days, 'items': items}, ensure_ascii=False, indent=2))
-        return
-
-    label = {'delivered': '已兑现', 'chased': '可能欠着', 'acked': '已确认', 'unknown': '看不清'}
-    order = {'chased': 0, 'unknown': 1, 'delivered': 2, 'acked': 3}
-    items.sort(key=lambda x: (order[x['state']], x['time']))
-    print(f'最近 {args.days} 天里，"我"承诺过 {len(items)} 件事：\n')
-    for it in items:
-        print(f"[{label[it['state']]}] {it['time']}  {it['talker']}")
-        print(f"    我说：{it['said']}")
-        print(f"    后续：{it['evidence']}")
-        print()
-
-
-if __name__ == '__main__':
-    main()
+"""欠账清单：我说过要给谁做什么，后来做了没有。
+
+这是微信永远回答不了的问题 —— 它知道"我明天发你"，但不知道"兑现"是什么。
+
+**默认不调模型**：先用规则把候选捞出来（我发的 + 承诺句式），再在同一个会话之后的对话里
+找"完成信号 / 对方催问 / 对方确认"三类证据。规则会误报，所以每条都带着原话和后续，
+让读者自己判 —— 而不是给一个"AI 认为你欠了 7 件事"的黑箱。
+
+用法：
+    python scripts/promises.py --days 60
+    python scripts/promises.py --days 90 --json
+    python scripts/promises.py --days 60 --talker 老表
+"""
+from __future__ import annotations
+
+import argparse
+import json
+import pathlib
+import re
+import sys
+from collections import defaultdict
+from datetime import datetime, timezone, timedelta
+
+SCRIPT_DIR = pathlib.Path(__file__).resolve().parent
+sys.path.insert(0, str(SCRIPT_DIR))
+
+from _utils import load_config, decrypt_lock  # noqa: E402
+import nt_decrypt as nd  # noqa: E402
+
+TZ = timezone(timedelta(hours=8))
+
+# 承诺句式：我 + 将来/稍后 + 一个交付动作。宁可多收，后面还有证据过滤。
+PROMISE_RE = re.compile(
+    r'(我|这边)[^。！？\n]{0,12}'
+    r'(明天|后天|下周|下周一|晚点|稍后|回头|等会|等会儿|下午|晚上|今天之内|这周)[^。！？\n]{0,12}'
+    r'(发|给|寄|做|弄|整理|传|送|安排|联系|确认|回复|处理)')
+DELIVER_RE = re.compile(r'(发了|发你了|发您了|已发|寄了|弄好了|做好了|整理好了|搞定了|已上传|传了|安排了|回复了|处理好了)')
+CHASE_RE = re.compile(r'(发了吗|寄了吗|好了吗|弄好了吗|什么时候|还没|催一下|怎么还没|在吗.{0,6}(发|给))')
+ACK_RE = re.compile(r'^(收到|好的|好滴|谢谢|多谢|ok|OK|嗯嗯|好嘞)')
+_NOT_PROMISE = re.compile(r'(我明天要|我下周要|我想|我觉得|我昨天|我上次|我已经)')
+
+
+def _clean(text):
+    text = ' '.join(str(text or '').split())
+    return re.sub(r'^[A-Za-z0-9_\-]+@?(chatroom|openim)?\s*:\s*', '', text)
+
+
+def same_account(name, wxid):
+    """配置里的 wxid 与 `Name2Id` 里的账号名**可能是同一个人的两种写法**。
+
+    实测本机：配置是 `wxid_账号A_后缀`，而 `Name2Id` 里只有 `wxid_账号A`（没有后缀）——
+    精确比较一个都匹配不上，于是"我发的消息"一条也捞不到（第一版就是这么得到 0 条承诺的）。
+    """
+    n, w = (name or '').strip(), (wxid or '').strip()
+    if not n or not w:
+        return False
+    return n == w or n.startswith(w + '_') or w.startswith(n + '_')
+
+
+def load_messages(conns, days, my_wxid):
+    """{talker: [ (ts, is_me, sender, text) ]}，按时间升序。"""
+    since = int((datetime.now(TZ) - timedelta(days=days)).timestamp())
+    out = defaultdict(list)
+    for conn in conns:
+        c = conn.cursor()
+        try:
+            c.execute('SELECT rowid, user_name FROM Name2Id')
+            id2name = {r[0]: r[1] for r in c.fetchall()}
+        except Exception:
+            continue
+        try:
+            c.execute('SELECT user_name FROM Name2Id WHERE is_session = 1')
+            sessions = [r[0] for r in c.fetchall() if r[0]]
+        except Exception:
+            sessions = []
+        for talker in sessions:
+            tbl = 'Msg_%s' % __import__('hashlib').md5(talker.encode()).hexdigest()
+            try:
+                c.execute('SELECT COUNT(*) FROM sqlite_master WHERE name=?', (tbl,))
+                if c.fetchone()[0] == 0:
+                    continue
+                c.execute('SELECT create_time, real_sender_id, local_type, message_content FROM "%s" '
+                          'WHERE create_time >= ? ORDER BY create_time' % tbl, (since,))
+                for ts, sender, ltype, content in c.fetchall():
+                    if (int(ltype or 0) & 0xFFFFFFFF) != 1:
+                        continue
+                    text = _clean(nd._decode_content(content))
+                    if not text:
+                        continue
+                    who = id2name.get(sender, '')
+                    out[talker].append((int(ts), same_account(who, my_wxid), who, text))
+            except Exception:
+                continue
+    return out
+
+
+def classify(follow):
+    """看承诺之后发生了什么：兑现 / 被催 / 只有确认 / 无后续。"""
+    if any(f[1] and DELIVER_RE.search(f[3]) for f in follow):
+        return 'delivered', '我后来提到"已发/已弄好"'
+    chases = [f for f in follow if not f[1] and CHASE_RE.search(f[3])]
+    if chases:
+        return 'chased', '对方后来催过：' + chases[0][3][:40]
+    if any(not f[1] and ACK_RE.search(f[3]) for f in follow):
+        return 'acked', '对方回复了确认'
+    return 'unknown', '没有后续证据'
+
+
+def main():
+    ap = argparse.ArgumentParser(description='我说过要给谁做什么，后来做了没有')
+    ap.add_argument('--days', type=int, default=60)
+    ap.add_argument('--talker', default=os.environ.get('WEFLOW_TALKER', ''), help='只看某个会话（wxid 或昵称子串）')
+    ap.add_argument('--limit', type=int, default=30, help='最多列几条（默认 30）')
+    ap.add_argument('--json', action='store_true')
+    args = ap.parse_args()
+
+    cfg = load_config()
+    db = cfg.get('ntDbPath') or ''
+    my_wxid = cfg.get('wxid') or ''
+    if not db:
+        raise SystemExit('配置里没有 ntDbPath')
+    key = decrypt_lock(cfg.get('ntKey', '') or '')
+    salt = decrypt_lock(cfg.get('ntSalt', '') or '')
+    pw = decrypt_lock(cfg.get('favPassphrase', '') or cfg.get('decryptKey', '') or '')
+    conns = nd.connect_message_shards(db, key, salt, pw)
+    if not conns:
+        raise SystemExit('打不开消息分片')
+
+    try:
+        by_talker = load_messages(conns, args.days, my_wxid)
+        # 昵称在 **contact.db**，消息库里没有 `contact` 表 —— 从消息库找一定落空，
+        # 输出里就只剩 wxid（和 on_this_day.py 同一个坑）。
+        name_map = {}
+        try:
+            from network import load_graph
+            people, _rooms, _members = load_graph(cfg)
+            name_map = {p['username']: (p['remark'] or p['nick'] or p['username'])
+                        for p in people.values()}
+        except Exception as exc:
+            print(f'[WARN] 读联系人失败（{type(exc).__name__}: {exc}），下面显示 id',
+                  file=sys.stderr)
+    finally:
+        for conn in conns:
+            try:
+                conn.close()
+            except Exception:
+                pass
+
+    items = []
+    for talker, msgs in by_talker.items():
+        shown = name_map.get(talker, talker)
+        if args.talker and args.talker not in (talker, shown):
+            continue
+        for i, (ts, is_me, who, text) in enumerate(msgs):
+            if not is_me or not PROMISE_RE.search(text) or _NOT_PROMISE.search(text):
+                continue
+            follow = msgs[i + 1:i + 20]
+            state, why = classify(follow)
+            items.append({
+                'time': datetime.fromtimestamp(ts, tz=TZ).strftime('%Y-%m-%d %H:%M'),
+                'talker': shown,
+                'said': text[:100],
+                'state': state,
+                'evidence': why,
+            })
+    items.sort(key=lambda x: x['time'], reverse=True)
+    items = items[:args.limit]
+
+    if args.json:
+        print(json.dumps({'success': True, 'days': args.days, 'items': items}, ensure_ascii=False, indent=2))
+        return
+
+    label = {'delivered': '已兑现', 'chased': '可能欠着', 'acked': '已确认', 'unknown': '看不清'}
+    order = {'chased': 0, 'unknown': 1, 'delivered': 2, 'acked': 3}
+    items.sort(key=lambda x: (order[x['state']], x['time']))
+    print(f'最近 {args.days} 天里，"我"承诺过 {len(items)} 件事：\n')
+    for it in items:
+        print(f"[{label[it['state']]}] {it['time']}  {it['talker']}")
+        print(f"    我说：{it['said']}")
+        print(f"    后续：{it['evidence']}")
+        print()
+
+
+if __name__ == '__main__':
+    main()