feat: add history replay and work filters

src/account_session.py

@@ -9,8 +9,14 @@ from urllib.parse import urlsplit
 from account_store import decode
 from douyin_im import EXPRESSION
 from follow_user import validate_uid
-from get_current_user import BROWSER_SCRIPT, compact_user, parse_user_response
-from subscribe_notifications import INSTALL, WAIT, detail_request_script
+from get_current_user import BROWSER_SCRIPT, business_user, parse_user_response
+from subscribe_notifications import (
+    INSTALL,
+    WAIT,
+    detail_request_script,
+    history_request_script,
+    works_request_script,
+)
 
 
 class SessionError(RuntimeError):
@@ -139,7 +145,7 @@ class Session:
 
     async def profile(self):
         try:
-            user = compact_user(
+            user = business_user(
                 parse_user_response(
                     await self.evaluate(BROWSER_SCRIPT, main_world=True)
                 )
@@ -226,6 +232,109 @@ class Session:
             raise SessionError("通知详情返回了未请求的 ID，已停止处理")
         return notices
 
+    async def history_notices(self):
+        notices = {}
+        min_time = max_time = 0
+        seen_cursors = set()
+        for _ in range(1000):
+            response = await self.json(
+                history_request_script(min_time, max_time), main_world=True
+            )
+            try:
+                status = int(response["status"])
+                payload = json.loads(response["body"])
+            except (KeyError, TypeError, ValueError, json.JSONDecodeError) as exc:
+                raise SessionError("历史通知响应格式无效") from exc
+            if status != 200 or payload.get("status_code") != 0:
+                raise SessionError(
+                    f"历史通知请求失败：HTTP {status}，业务码 {payload.get('status_code')}"
+                )
+            rows = payload.get("notice_list_v2")
+            if rows is None:
+                rows = []
+            if not isinstance(rows, list) or any(
+                not isinstance(row, dict) for row in rows
+            ):
+                raise SessionError("历史通知列表格式无效")
+            for row in rows:
+                if str(row.get("user_id")) != self.uid:
+                    raise SessionError("历史通知所属身份不符")
+                nid = row.get("nid_str") or row.get("nid")
+                validate_uid(nid)
+                notices.setdefault(nid, row)
+            if not payload.get("has_more"):
+                return list(notices.values())
+            cursor = (payload.get("min_time"), payload.get("max_time"))
+            if cursor in seen_cursors or cursor == (min_time, max_time):
+                raise SessionError("历史通知分页游标未推进")
+            if any(type(value) not in (int, float) for value in cursor):
+                raise SessionError("历史通知分页游标格式无效")
+            seen_cursors.add(cursor)
+            min_time, max_time = cursor
+        raise SessionError("历史通知超过 50000 条，已停止以避免无限分页")
+
+    async def works(self, all_pages=False):
+        profile = await self.identity()
+        sec_uid = profile.get("sec_uid") or profile.get("secUid")
+        if not isinstance(sec_uid, str) or not sec_uid:
+            raise SessionError("当前账号缺少作品列表身份参数")
+        works = {}
+        cursor = 0
+        seen_cursors = set()
+        pages = 1000 if all_pages else 1
+        for _ in range(pages):
+            response = await self.json(
+                works_request_script(sec_uid, cursor), main_world=True
+            )
+            try:
+                status = int(response["status"])
+                payload = json.loads(response["body"])
+            except (KeyError, TypeError, ValueError, json.JSONDecodeError) as exc:
+                raise SessionError("作品列表响应格式无效") from exc
+            if status != 200 or payload.get("status_code") != 0:
+                raise SessionError(
+                    f"作品列表请求失败：HTTP {status}，业务码 {payload.get('status_code')}"
+                )
+            rows = payload.get("aweme_list")
+            if rows is None:
+                rows = []
+            if not isinstance(rows, list) or any(
+                not isinstance(row, dict) for row in rows
+            ):
+                raise SessionError("作品列表格式无效")
+            for row in rows:
+                author = row.get("author") or {}
+                if str(author.get("uid")) != self.uid:
+                    raise SessionError("作品列表所属身份不符")
+                ident = row.get("aweme_id") or row.get("awemeId")
+                validate_uid(ident)
+                cover = ((row.get("video") or {}).get("cover") or {}).get(
+                    "url_list"
+                ) or []
+                if not cover and row.get("images"):
+                    cover = (row["images"][0] or {}).get("url_list") or []
+                works.setdefault(
+                    ident,
+                    {
+                        "aweme_id": ident,
+                        "desc": row.get("desc") or "",
+                        "create_time": row.get("create_time") or "",
+                        "statistics": row.get("statistics") or {},
+                        "cover": cover[0] if cover else "",
+                        "business": row,
+                    },
+                )
+            if not all_pages or not payload.get("has_more"):
+                return list(works.values())
+            next_cursor = payload.get("max_cursor")
+            if type(next_cursor) not in (int, float) or next_cursor < 0:
+                raise SessionError("作品列表分页游标格式无效")
+            if next_cursor in seen_cursors or next_cursor == cursor:
+                raise SessionError("作品列表分页游标未推进")
+            seen_cursors.add(next_cursor)
+            cursor = next_cursor
+        raise SessionError("作品列表超过 18000 条，已停止以避免无限分页")
+
     async def follow(self, target, check=False):
         self.require_bound()
         validate_uid(target)