feat: 钉钉群聊飞书文档收集与知识图谱系统初始提交

- 通过悟空(dws CLI)拉取dc战略问题研究院+创新组两个群的消息(377条)
- 提取190个飞书链接、18个文件附件
- 下载HTML/MD/XLSX等报告文件到output/downloaded-files/
- 构建知识图谱(JSON+HTML可视化)
- 生成Obsidian知识库(28个页面,7大主题)
- 生成花园世界全量汇总报告
- 所有脚本路径改为相对路径,便于迁移

scripts/fetch_all.py

@@ -0,0 +1,75 @@
+import sys, io, json, subprocess, os
+sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding="utf-8")
+
+SCRIPT_DIR = os.path.dirname(os.path.abspath(__file__))
+PROJECT_ROOT = os.path.dirname(SCRIPT_DIR)
+
+DWS = r"C:\Users\admin\.real\.bin\dws\bin\dws.exe"
+
+def fetch(group_id, time_str, forward="true"):
+    cmd = [DWS, "chat", "message", "list", "--group", group_id, "--time", time_str, "--forward", forward, "--format", "json", "--limit", "200"]
+    result = subprocess.run(cmd, capture_output=True, timeout=60)
+    output = result.stdout.decode("utf-8", errors="replace")
+    json_start = output.find("{")
+    if json_start < 0: return [], False
+    data = json.loads(output[json_start:])
+    msgs = data.get("result", {}).get("messages", [])
+    has_more = data.get("result", {}).get("hasMore", False)
+    return msgs, has_more
+
+dc_id = "cidoUneRB4Db8TAXaTrKxkQAw=="
+
+# Fetch dc 05-24+
+msgs, hm = fetch(dc_id, "2026-05-24 00:00:00", "true")
+print("dc 05-24+: %d msgs, hasMore=%s" % (len(msgs), hm), flush=True)
+if msgs:
+    t0 = msgs[-1]["createTime"]
+    t1 = msgs[0]["createTime"]
+    print("  Range: %s ~ %s" % (t0, t1), flush=True)
+
+# Fetch backwards from 06-03
+msgs2, hm2 = fetch(dc_id, "2026-06-03 00:00:00", "false")
+print("dc before 06-03: %d msgs, hasMore=%s" % (len(msgs2), hm2), flush=True)
+if msgs2:
+    t0 = msgs2[-1]["createTime"]
+    t1 = msgs2[0]["createTime"]
+    print("  Range: %s ~ %s" % (t0, t1), flush=True)
+
+# Now combine ALL dc messages with dedup
+all_dc = {}
+
+# Segment 1: 05-01 forward
+msgs1, _ = fetch(dc_id, "2026-05-01 00:00:00", "true")
+for m in msgs1: all_dc[m["openMessageId"]] = m
+
+# Segment 2: 05-19 forward
+msgs2, _ = fetch(dc_id, "2026-05-19 00:00:00", "true")
+for m in msgs2: all_dc[m["openMessageId"]] = m
+
+# Segment 3: 05-24 forward
+msgs3, _ = fetch(dc_id, "2026-05-24 00:00:00", "true")
+for m in msgs3: all_dc[m["openMessageId"]] = m
+
+# Segment 4: backwards from 06-03
+msgs4, _ = fetch(dc_id, "2026-06-03 00:00:00", "false")
+for m in msgs4: all_dc[m["openMessageId"]] = m
+
+print("\nTotal unique dc messages: %d" % len(all_dc), flush=True)
+all_msgs = list(all_dc.values())
+if all_msgs:
+    times = sorted([m["createTime"] for m in all_msgs])
+    print("Full range: %s ~ %s" % (times[0], times[-1]), flush=True)
+
+# Also get cx
+cx_id = "cidMuM+itt5PeY7xNSWsv3M0g=="
+all_cx = {}
+msgs, _ = fetch(cx_id, "2026-05-01 00:00:00", "true")
+for m in msgs: all_cx[m["openMessageId"]] = m
+print("Total unique cx messages: %d" % len(all_cx), flush=True)
+
+# Save
+output = {"dc": list(all_dc.values()), "cx": list(all_cx.values())}
+with open(os.path.join(PROJECT_ROOT, "data", "raw-messages", "all_messages_combined.json"), "w", encoding="utf-8") as f:
+    json.dump(output, f, ensure_ascii=False, indent=2)
+print("Saved!")
+