feat: persist history and work caches

src/account_session.py

@@ -232,8 +232,9 @@ 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):
+    async def history_notices(self, stop_ids=None):
         notices = {}
+        stop_ids = set(stop_ids or ())
         min_time = max_time = 0
         seen_cursors = set()
         for _ in range(1000):
@@ -256,13 +257,14 @@ class Session:
                 not isinstance(row, dict) for row in rows
             ):
                 raise SessionError("历史通知列表格式无效")
+            reached_cache = False
             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)
+                nid = validate_uid(row.get("nid_str") or row.get("nid"))
+                reached_cache = reached_cache or nid in stop_ids
                 notices.setdefault(nid, row)
-            if not payload.get("has_more"):
+            if reached_cache or 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):
@@ -273,15 +275,16 @@ class Session:
             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):
+    async def works(self, all_pages=False, stop_ids=None):
         profile = await self.identity()
+        stop_ids = set(stop_ids or ())
         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
+        pages = 1000 if all_pages or stop_ids else 1
         for _ in range(pages):
             response = await self.json(
                 works_request_script(sec_uid, cursor), main_world=True
@@ -302,12 +305,13 @@ class Session:
                 not isinstance(row, dict) for row in rows
             ):
                 raise SessionError("作品列表格式无效")
+            reached_cache = False
             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)
+                ident = validate_uid(row.get("aweme_id") or row.get("awemeId"))
+                reached_cache = reached_cache or ident in stop_ids
                 cover = ((row.get("video") or {}).get("cover") or {}).get(
                     "url_list"
                 ) or []
@@ -324,7 +328,11 @@ class Session:
                         "business": row,
                     },
                 )
-            if not all_pages or not payload.get("has_more"):
+            if (
+                reached_cache
+                or (not all_pages and not stop_ids)
+                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: