chore: collect latest dingTalk feishu docs

skills/dingtalk-feishu-collector/scripts/paths.py

@@ -2,6 +2,7 @@
 
 import os
 import sys
+import json
 
 def find_project_root():
     """向上查找项目根目录（包含 data/ 目录的最顶层）"""
@@ -56,12 +57,59 @@ def find_dws():
 def find_lark_cli():
     """查找 lark-cli"""
     import shutil
+    root = find_project_root()
+    ext = ".exe" if os.name == "nt" else ""
+
+    local = os.path.join(root, "tools", "lark-cli", "lark-cli" + ext)
+    if os.path.isfile(local):
+        return local
+
     found = shutil.which("lark-cli")
     if found:
         return found
     print("ERROR: 找不到 lark-cli，请运行 npm install -g @larksuite/cli", file=sys.stderr)
     sys.exit(1)
 
+def load_groups():
+    """读取群配置，优先使用 data/group_config.json，失败时返回默认群。"""
+    default = {
+        "dc战略问题研究院": "cidoUneRB4Db8TAXaTrKxkQAw==",
+        "创新组": "cidMuM+itt5PeY7xNSWsv3M0g==",
+        "天龙计划X小游戏矩阵": "cid/Vb3hSjdv4sIIs8Fp7sC4g==",
+    }
+    path = os.path.join(find_project_root(), "data", "group_config.json")
+    if not os.path.isfile(path):
+        return default
+    try:
+        with open(path, "r", encoding="utf-8") as f:
+            raw = json.load(f).get("groups", {})
+        groups = {}
+        for name, value in raw.items():
+            if isinstance(value, dict):
+                cid = value.get("cid") or value.get("id")
+            else:
+                cid = value
+            if cid:
+                groups[name] = cid
+        return groups or default
+    except Exception as e:
+        print(f"WARN: 读取群配置失败，使用默认配置: {e}", file=sys.stderr)
+        return default
+
+def normalize_messages(messages):
+    """兼容 list 和按群 dict 两种历史消息文件格式。"""
+    if isinstance(messages, dict):
+        return messages
+    grouped = {}
+    if not isinstance(messages, list):
+        return grouped
+    for msg in messages:
+        if not isinstance(msg, dict):
+            continue
+        group = msg.get("group") or msg.get("groupName") or msg.get("openConversationId") or "未分组"
+        grouped.setdefault(group, []).append(msg)
+    return grouped
+
 # 常用路径
 PROJECT_ROOT = find_project_root()
 DATA_DIR = os.path.join(PROJECT_ROOT, "data")