test positions

src/add_chunk_cli_pdf_img.py

@@ -10,7 +10,7 @@ from elasticsearch import Elasticsearch
 from minio import Minio
 from minio.error import S3Error
 
-from get_pos_pdf import smart_fuzzy_find_text_batch, find_text_positions_batch
+# from get_pos_pdf import smart_fuzzy_find_text_batch, find_text_positions_batch
 
 
 from dotenv import load_dotenv  # 新增
@@ -59,6 +59,7 @@ def update_positon_img_id_in_elasticsearch(tenant_id, doc_id, chunk_id, position
     :return: 更新结果
     """
     try:
+        
         # 构建索引名称
         index_name = f"ragflow_{tenant_id}"
 
@@ -93,7 +94,23 @@ def update_positon_img_id_in_elasticsearch(tenant_id, doc_id, chunk_id, position
             
         # 只有当 position 存在时才更新 positions
         if position is not None:
-            update_body["doc"]["positions"] = position
+            # 如果传入的是嵌套字典格式的 position
+            if isinstance(position, list) and all(isinstance(p, dict) for p in position):
+                # 将字典格式转换为整数列表格式
+                formatted_positions = []
+                for pos in position:
+                    pos_list = [
+                        pos.get('page', 0),  # 页码
+                        int(round(float(pos.get('x0', 0)))),  # x0
+                        int(round(float(pos.get('x1', 0)))),  # x1
+                        int(round(float(pos.get('y0', 0)))),  # y0
+                        int(round(float(pos.get('y1', 0))))   # y1
+                    ]
+                    formatted_positions.append(pos_list)
+                update_body["doc"]["positions"] = formatted_positions
+            # 如果已经是整数列表格式
+            elif isinstance(position, list):
+                update_body["doc"]["positions"] = position
 
         # 如果没有需要更新的字段，直接返回成功
         if not update_body["doc"]: