修改抵押登记豁免函的字段处理

@@ -19,10 +19,18 @@ class HMHRetriever:
     def get_target_fields(self, pdf_text_list):
     def get_target_fields(self, pdf_text_list):
         result = dict()
         result = dict()
         is_find_name_id_company, is_find_application_no, is_find_name_date = False, False, False
         is_find_name_id_company, is_find_application_no, is_find_name_date = False, False, False
-        for bbox, text in pdf_text_list.pop(str(0), []):
-            # print(text)
+        # for bbox, text in pdf_text_list.pop(str(0), []):
+        pdf_text_items = pdf_text_list.pop(str(0), [])
+
+        for i in range(len(pdf_text_items)):
+            bbox, text = pdf_text_items[i]
+            combined_text = text
+            if i < len(pdf_text_items) - 1:
+                combined_text += pdf_text_items[i + 1][1]
+
             if not is_find_name_id_company:
             if not is_find_name_id_company:
-                name_id_company_list = re.findall(r'姓名(.*)证件号码(.*)与(.*公司)', text)
+                # name_id_company_list = re.findall(r'姓名(.*?)证件号码(.*?)与(.*?公司|.*)', combined_text)
+                name_id_company_list = re.findall(r'姓名(.*)证件号码(.*)与(.*公司)', combined_text)
                 for name_id_company_tuple in name_id_company_list:
                 for name_id_company_tuple in name_id_company_list:
                     if len(name_id_company_tuple) == 3: 
                     if len(name_id_company_tuple) == 3: 
                         result[self.search_fields_list[0][0]] = {
                         result[self.search_fields_list[0][0]] = {
@@ -40,7 +48,7 @@ class HMHRetriever:
                         is_find_name_id_company = True
                         is_find_name_id_company = True
                         break
                         break
             if not is_find_application_no:
             if not is_find_application_no:
-                application_no_list = re.findall(r'合同编号.*(CH-B\d*-\d*).*', text)
+                application_no_list = re.findall(r'合同编号.*(CH-B\d*-\d*).*', combined_text)
                 if len(application_no_list) == 1:
                 if len(application_no_list) == 1:
                     result[self.search_fields_list[3][0]] = {
                     result[self.search_fields_list[3][0]] = {
                         self.words_str: application_no_list[0],
                         self.words_str: application_no_list[0],
@@ -48,7 +56,7 @@ class HMHRetriever:
                     }
                     }
                     is_find_application_no = True
                     is_find_application_no = True
             if not is_find_name_date:
             if not is_find_name_date:
-                name_date_list = re.findall(r'(.*).*签署日期.*(\d{4}-\d{2}-\d{2})', text)
+                name_date_list = re.findall(r'(.*).*签署日期.*(\d{4}-\d{2}-\d{2})', combined_text)
                 for name_date_tuple in name_date_list:
                 for name_date_tuple in name_date_list:
                     if len(name_date_tuple) == 2: 
                     if len(name_date_tuple) == 2: 
                         result[self.search_fields_list[4][0]] = {
                         result[self.search_fields_list[4][0]] = {