如何通过 API 获取 XChat DM 历史

curl --request POST \
  --url https://api.twexapi.io/v3/twitter/dm-history \
  --header "Authorization: Bearer $TWEXAPI_BEARER_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
    "recipient": "@example_user",
    "cookie": "auth_token=...; ct0=...; twid=...",
    "pin": "1234",
    "count": 50,
    "all": false
  }'

1{
2  "code": 200,
3  "msg": "success",
4  "data": {
5    "conversation_id": "1928096185664843776:1989842918455291904",
6    "has_more": true,
7    "messages": [
8      {
9        "id": "1234567890123456789",
10        "text": "谢谢，我确认导出现在正常了。",
11        "time": "2024-01-01T12:00:00Z",
12        "sequence_id": "2076661179334979585",
13        "sender_id": "1928096185664843776",
14        "recipient_id": "1989842918455291904",
15        "attachment": null,
16        "attachments": null
17      }
18    ]
19  }
20}

1import os
2from datetime import datetime, timezone
3from typing import Any
4
5import requests
6
7API_URL = "https://api.twexapi.io/v3/twitter/dm-history"
8BEARER_TOKEN = os.environ["TWEXAPI_BEARER_TOKEN"]
9X_COOKIE = os.environ["TWEXAPI_X_COOKIE"]
10XCHAT_PIN = os.environ.get("TWEXAPI_XCHAT_PIN", "1234")
11PROXY = os.environ.get("TWEXAPI_PROXY")
12
13def fetch_dm_page(
14    recipient: str,
15    *,
16    before: str | None = None,
17    count: int = 50,
18) -> dict[str, Any]:
19    if count < 10 or count > 200:
20        raise ValueError("count must be between 10 and 200")
21
22    payload: dict[str, Any] = {
23        "recipient": recipient,
24        "cookie": X_COOKIE,
25        "pin": XCHAT_PIN,
26        "count": count,
27        "all": False,
28    }
29    if before:
30        payload["before"] = before
31    if PROXY:
32        payload["proxy"] = PROXY
33
34    response = requests.post(
35        API_URL,
36        headers={
37            "Authorization": f"Bearer {BEARER_TOKEN}",
38            "Content-Type": "application/json",
39            "Accept": "application/json",
40        },
41        json=payload,
42        timeout=60,
43    )
44    response.raise_for_status()
45    data = response.json()
46
47    if data.get("code") != 200:
48        raise RuntimeError(data.get("msg", "TwexAPI returned an error"))
49
50    return data["data"]
51
52def normalize_message(conversation_id: str, message: dict[str, Any]) -> dict[str, Any]:
53    return {
54        "conversation_id": conversation_id,
55        "message_id": message["id"],
56        "text": message.get("text"),
57        "time": message.get("time"),
58        "sequence_id": message.get("sequence_id"),
59        "sender_id": message.get("sender_id"),
60        "recipient_id": message.get("recipient_id"),
61        "attachment": message.get("attachment"),
62        "attachments": message.get("attachments"),
63        "fetched_at": datetime.now(timezone.utc).isoformat(),
64        "raw": message,
65    }
66
67def export_history(recipient: str, *, max_pages: int = 10) -> list[dict[str, Any]]:
68    before = None
69    rows: list[dict[str, Any]] = []
70
71    for _ in range(max_pages):
72        page = fetch_dm_page(recipient, before=before)
73        conversation_id = page["conversation_id"]
74        messages = page.get("messages") or []
75
76        rows.extend(normalize_message(conversation_id, msg) for msg in messages)
77
78        if not page.get("has_more") or not messages:
79            break
80
81        oldest = min(
82            messages,
83            key=lambda msg: int(msg.get("sequence_id") or 0),
84        )
85        before = oldest.get("sequence_id")
86        if not before:
87            break
88
89    return rows
90
91if __name__ == "__main__":
92    exported = export_history("@example_user", max_pages=5)
93    print(f"Exported {len(exported)} messages")