阶段性更新

etl_billiards/tests/unit/task_test_utils.py

@@ -135,16 +135,41 @@ class FakeDBOperations:
 
     def __init__(self):
         self.upserts: List[Dict] = []
+        self.executes: List[Dict] = []
         self.commits = 0
         self.rollbacks = 0
         self.conn = FakeConnection()
 
     def batch_upsert_with_returning(self, sql: str, rows: List[Dict], page_size: int = 1000):
-        self.upserts.append({"sql": sql.strip(), "count": len(rows), "page_size": page_size})
+        self.upserts.append(
+            {
+                "sql": sql.strip(),
+                "count": len(rows),
+                "page_size": page_size,
+                "rows": [dict(row) for row in rows],
+            }
+        )
         return len(rows), 0
 
     def batch_execute(self, sql: str, rows: List[Dict], page_size: int = 1000):
-        self.upserts.append({"sql": sql.strip(), "count": len(rows), "page_size": page_size})
+        self.executes.append(
+            {
+                "sql": sql.strip(),
+                "count": len(rows),
+                "page_size": page_size,
+                "rows": [dict(row) for row in rows],
+            }
+        )
+
+    def execute(self, sql: str, params=None):
+        self.executes.append({"sql": sql.strip(), "params": params})
+
+    def query(self, sql: str, params=None):
+        self.executes.append({"sql": sql.strip(), "params": params, "type": "query"})
+        return []
+
+    def cursor(self):
+        return self.conn.cursor()
 
     def commit(self):
         self.commits += 1
@@ -161,22 +186,53 @@ class FakeAPIClient:
         self.calls: List[Dict] = []
 
     # pylint: disable=unused-argument
-    def get_paginated(self, endpoint: str, params=None, **kwargs):
+    def iter_paginated(
+        self,
+        endpoint: str,
+        params=None,
+        page_size: int = 200,
+        page_field: str = "pageIndex",
+        size_field: str = "pageSize",
+        data_path: Tuple[str, ...] = (),
+        list_key: str | None = None,
+    ):
         self.calls.append({"endpoint": endpoint, "params": params})
         if endpoint not in self.data_map:
             raise AssertionError(f"Missing fixture for endpoint {endpoint}")
-        return list(self.data_map[endpoint]), [{"page": 1, "size": len(self.data_map[endpoint])}]
+
+        records = list(self.data_map[endpoint])
+        yield 1, records, dict(params or {}), {"data": records}
+
+    def get_paginated(self, endpoint: str, params=None, **kwargs):
+        records = []
+        pages = []
+        for page_no, page_records, req, resp in self.iter_paginated(endpoint, params, **kwargs):
+            records.extend(page_records)
+            pages.append({"page": page_no, "request": req, "response": resp})
+        return records, pages
+
+    def get_source_hint(self, endpoint: str) -> str | None:
+        return None
 
 
 class OfflineAPIClient:
-    """离线模式专用 API Client，根据 endpoint 读取归档 JSON、套用 data_path 并回放列表数据。"""
+    """离线模式专用 API Client，根据 endpoint 读取归档 JSON、套入 data_path 并回放列表数据。"""
 
     def __init__(self, file_map: Dict[str, Path]):
         self.file_map = {k: Path(v) for k, v in file_map.items()}
         self.calls: List[Dict] = []
 
     # pylint: disable=unused-argument
-    def get_paginated(self, endpoint: str, params=None, page_size: int = 200, data_path: Tuple[str, ...] = (), **kwargs):
+    def iter_paginated(
+        self,
+        endpoint: str,
+        params=None,
+        page_size: int = 200,
+        page_field: str = "pageIndex",
+        size_field: str = "pageSize",
+        data_path: Tuple[str, ...] = (),
+        list_key: str | None = None,
+    ):
         self.calls.append({"endpoint": endpoint, "params": params})
         if endpoint not in self.file_map:
             raise AssertionError(f"Missing archive for endpoint {endpoint}")
@@ -188,17 +244,42 @@ class OfflineAPIClient:
         for key in data_path:
             if isinstance(data, dict):
                 data = data.get(key, [])
-            else:
-                data = []
-                break
+
+        if list_key and isinstance(data, dict):
+            data = data.get(list_key, [])
 
         if not isinstance(data, list):
             data = []
 
-        return data, [{"page": 1, "mode": "offline"}]
+        total = len(data)
+        start = 0
+        page = 1
+        while start < total or (start == 0 and total == 0):
+            chunk = data[start : start + page_size]
+            if not chunk and total != 0:
+                break
+            yield page, list(chunk), dict(params or {}), payload
+            if len(chunk) < page_size:
+                break
+            start += page_size
+            page += 1
+
+    def get_paginated(self, endpoint: str, params=None, **kwargs):
+        records = []
+        pages = []
+        for page_no, page_records, req, resp in self.iter_paginated(endpoint, params, **kwargs):
+            records.extend(page_records)
+            pages.append({"page": page_no, "request": req, "response": resp})
+        return records, pages
+
+    def get_source_hint(self, endpoint: str) -> str | None:
+        if endpoint not in self.file_map:
+            return None
+        return str(self.file_map[endpoint])
 
 
 class RealDBOperationsAdapter:
+
     """连接真实 PostgreSQL 的适配器，为任务提供 batch_upsert + 事务能力。"""
 
     def __init__(self, dsn: str):