Resolve issue first records only with share class name but without fund name (in previous page text).

core/data_extraction.py

@@ -130,12 +130,25 @@ class DataExtraction:
         data_list = []
         pdf_page_count = len(self.page_text_dict.keys())
         handled_page_num_list = []
+        
+        previous_page_num = -1
+        previous_page_datapoints = []
+        previous_page_fund_name = None
         for page_num, page_text in self.page_text_dict.items():
             if page_num in handled_page_num_list:
                 continue
             page_datapoints = self.get_datapoints_by_page_num(page_num)
             if len(page_datapoints) == 0:
                 continue
+            if previous_page_num == page_num - 1 and \
+                previous_page_datapoints == page_datapoints and \
+                    previous_page_fund_name is not None:
+                # Transfer previous page fund name to be the pre-fix of page text
+                # The purpose is to get fund name if the first records without fund name
+                # example document: 431073795, page index 1727 to 1728
+                logger.info(f"Transfer previous page fund name: {previous_page_fund_name} to be the pre-fix of page text")
+                page_text = f"\n{previous_page_fund_name}\n{page_text}"
+                
             extract_data = self.extract_data_by_page(
                 page_num,
                 page_text,
@@ -145,7 +158,12 @@ class DataExtraction:
             )
             data_list.append(extract_data)
             
+            
             page_data_list = extract_data.get("extract_data", {}).get("data", [])
+            if len(page_data_list) > 0:
+                previous_page_num = page_num
+                previous_page_fund_name = page_data_list[-1].get("fund_name", "")
+                previous_page_datapoints = page_datapoints
 
             current_page_data_count = len(page_data_list)
             if current_page_data_count > 0:

main.py

@@ -786,33 +786,33 @@ if __name__ == "__main__":
     # ]
     # Documents in EMEA Case 1.docx
     check_db_mapping_doc_id_list = [
-        "435128656",
-        "425480144",
-        "466528487",
-        "434902020",
-        "440029306",
-        "431073795",
-        "430240853",
-        "427637151",
-        "434924914",
-        "467595142",
-        "466859621",
-        "429564034",
         "424976833",
-        "466860852",
+        "425480144",
-        "466371135",
+        "427637151",
-        "470515549",
+        "429564034",
-        "434851173",
-        "434710819",
         "429950833",
-        "467788879"
+        "430240853",
+        "431073795",
+        "434710819",
+        "434851173",
+        "434902020",
+        "434924914",
+        "435128656",
+        "440029306",
+        "466371135",
+        "466528487",
+        "466859621",
+        "466860852",
+        "467595142",
+        "467788879",
+        "470515549"
     ]
     # special_doc_id_list = check_mapping_doc_id_list
     special_doc_id_list = check_db_mapping_doc_id_list
-    special_doc_id_list = ["514213638"]
+    special_doc_id_list = ["431073795"]
     output_mapping_child_folder = r"/data/emea_ar/output/mapping_data/docs/"
     output_mapping_total_folder = r"/data/emea_ar/output/mapping_data/total/"
-    re_run_extract_data = False
+    re_run_extract_data = True
     re_run_mapping_data = True
     force_save_total_data = False
     calculate_metrics = False