10.29截取工程标新逻辑

flask_app/main/基础信息整合快速版.py

@@ -53,7 +53,8 @@ def aggregate_basic_info_engineering(baseinfo_list):
     dynamic_key_handling(key_groups, relevant_keys_detected)
 
     # 创建一个副本以存储未分类的项目
-    unclassified_items = {k: v for k, v in combined_data.items() if k not in [item for sublist in key_groups.values() for item in sublist]}
+    unclassified_items = {k: v for k, v in combined_data.items() if
+                          k not in [item for sublist in key_groups.values() for item in sublist]}
 
     # 按键组分类并嵌套
     for group_name, keys in key_groups.items():
@@ -73,6 +74,7 @@ def aggregate_basic_info_engineering(baseinfo_list):
 
     return combined_data
 
+
 def dynamic_key_handling(key_groups, detected_keys):
     # 检查和调整键组配置
     for key in detected_keys:
@@ -120,6 +122,7 @@ def update_baseinfo_lists(baseinfo_list1, baseinfo_list2):
 
     return updated_list
 
+
 def process_judge_questions(judge_file_path, chosen_numbers, tobidders_notice_table, baseinfo_list1):
     judge_questions = read_questions_from_judge(judge_file_path, chosen_numbers)
     judge_consortium = judge_consortium_bidding(baseinfo_list1)
@@ -138,6 +141,7 @@ def process_judge_questions(judge_file_path, chosen_numbers, tobidders_notice_ta
         for question, response in res2:
             baseinfo_list1.append(clean_json_string(response))
 
+
 def process_baseinfo_list(baseinfo_list, tobidders_notice):
     questions_list = []
     for item in baseinfo_list:
@@ -157,6 +161,7 @@ def process_baseinfo_list(baseinfo_list, tobidders_notice):
     else:
         return []
 
+
 def combine_basic_info(merged_baseinfo_path, tobidders_notice_table, tobidders_notice, clause_path):
     """
     综合和处理基础信息，生成最终的基础信息字典。
@@ -189,7 +194,8 @@ def combine_basic_info(merged_baseinfo_path,tobidders_notice_table, tobidders_no
 
     with concurrent.futures.ThreadPoolExecutor(max_workers=2) as executor:
         # 提交两个任务
-        future1 = executor.submit(process_judge_questions, judge_file_path, chosen_numbers, tobidders_notice_table, baseinfo_list1)
+        future1 = executor.submit(process_judge_questions, judge_file_path, chosen_numbers, tobidders_notice_table,
+                                  baseinfo_list1)
         future2 = executor.submit(process_baseinfo_list, baseinfo_list1_copy, tobidders_notice)
 
         # 等待两个任务完成并获取结果
@@ -205,6 +211,7 @@ def combine_basic_info(merged_baseinfo_path,tobidders_notice_table, tobidders_no
     aggregated_baseinfo = aggregate_basic_info_engineering(updated_list)
     return {"基础信息": aggregated_baseinfo}
 
+
 # TODO:先不带投标人须知正文，如果是未知，再直接问正文，
 if __name__ == "__main__":
     start_time = time.time()
@@ -217,7 +224,3 @@ if __name__ == "__main__":
     print(json.dumps(res, ensure_ascii=False, indent=4))
     end_time = time.time()
     print("elapsed_time:" + str(end_time - start_time))
-
-
-
-

flask_app/main/截取pdf.py

@@ -196,7 +196,7 @@ def extract_pages_tobidders_notice(pdf_path, output_folder,begin_pattern, begin_
                             r'^附表(?:一)?[：:]',  # Match "附表一" with optional "一" and colon
                             re.MULTILINE
                         )
-                        print(f"使用默认的 end_pattern: {end_pattern.pattern}")  # 打印默认的 end_pattern
+                        # print(f"使用默认的 end_pattern: {end_pattern.pattern}")  # 打印默认的 end_pattern
 
                         # 定义基础的 mid_pattern
                         base_mid_pattern = r'^\s*(?:[（(]\s*[一1]?\s*[)）]\s*[、．.]*|[一1][、．.]+|[、．.]+)\s*(说\s*明|总\s*则)'
@@ -204,9 +204,6 @@ def extract_pages_tobidders_notice(pdf_path, output_folder,begin_pattern, begin_
                             rf'{base_mid_pattern}',
                             re.MULTILINE
                         )
-                        print(
-                            f"使用默认的 combined_mid_pattern: {combined_mid_pattern.pattern}")  # 打印默认的 combined_mid_pattern
-
                     continue
 
             if start_page is not None and mid_page is None and combined_mid_pattern:
@@ -306,6 +303,8 @@ def get_start_and_common_header(input_path):
     begin_pattern = re.compile(r'.*(?:招标公告|投标邀请书|投标邀请函)\s*$',re.MULTILINE)
     pdf_document = PdfReader(input_path)
     for i, page in enumerate(pdf_document.pages):
+        if i > 25:
+            return common_header, 0  # 如果页码大于25，直接返回last_begin_index=0
         text = page.extract_text()
         if text:
             cleaned_text = clean_page_content(text, common_header)
@@ -371,7 +370,7 @@ def truncate_pdf_main(input_path, output_folder, selection):
                     re.compile(r'^第[一二三四五六七八九十百千]+(?:章|部分)\s*[\u4e00-\u9fff]+', re.MULTILINE)
                 ),
                 (
-                    re.compile(r'.*(?:招标公告|投标邀请书|投标邀请函)\s*$', re.MULTILINE),
+                    re.compile(r'.*(?:招标公告|投标邀请书|投标邀请函|第一卷)\s*$', re.MULTILINE),
                     re.compile(r".*(?:投标人须知|投标人须知前附表)\s*$", re.MULTILINE)
                 )
             ]
@@ -381,7 +380,12 @@ def truncate_pdf_main(input_path, output_folder, selection):
             # Selection 5: 无效标
             pattern_pairs = [
                 (
-                    re.compile(r'第[一二三四五六七八九十]+章\s*(招标公告|.*邀请.*)|第一卷|招标编号:|招标编号：'),
+                    re.compile(
+                        r'^第[一二三四五六七八九十百千]+(?:章|部分).*?(?:公告|邀请书|邀请函).*|^第一卷|^投标邀请书'),
+                    re.compile(r'第[一二三四五六七八九十]+章\s*合同|[：:]清标报告|^第二卷', re.MULTILINE)
+                ),
+                (
+                    re.compile(r'.*(?:招标公告|投标邀请书|投标邀请函|第一卷)\s*$', re.MULTILINE),
                     re.compile(r'第[一二三四五六七八九十]+章\s*合同|[：:]清标报告|^第二卷', re.MULTILINE)
                 )
             ]
@@ -393,8 +397,14 @@ def truncate_pdf_main(input_path, output_folder, selection):
 
         for idx, (begin_pattern, end_pattern) in enumerate(pattern_pairs, start=1):
             is_secondary_match = (idx == 2)
+            begin_page = last_begin_index if last_begin_index != 0 else {
+                1: 3,   #前附表
+                2: 10,  #评标
+                3: 5,   #资格
+                4: 0,   #公告
+                5: 0    #无效标
+            }.get(selection, 0)
             if selection == 1:      #投标人须知
-                begin_page = 3
                 output_paths = list(
                     extract_pages_tobidders_notice(input_path,output_folder, begin_pattern, begin_page, common_header))
                 if output_paths and any(os.path.isfile(f) for f in output_paths):
@@ -402,10 +412,6 @@ def truncate_pdf_main(input_path, output_folder, selection):
                 else:
                     # print(f"Selection {selection}: 使用组合提取函数失败。尝试下一个模式对。")
                     continue
-            if selection == 2:    #评分细则
-                begin_page = 10
-            else:
-                begin_page = last_begin_index
             output_paths = extract_pages(
                 input_path,
                 output_folder,
@@ -669,7 +675,7 @@ def truncate_pdf_specific_engineering(pdf_path, output_folder, selections, uniqu
 #TODO:目前merged_baseinfo没有包含投标人须知正文。
 #投标人须知前附表改为货物标一样的
 if __name__ == "__main__":
-    input_path = "C:\\Users\\Administrator\\Desktop\\招标文件\\招标test文件夹\\zbtest8.pdf"  # 可以是单个PDF文件路径或文件夹路径  #zbtest20.pdf zbtest4_evaluation_method.pdf zbtest8.pdf
+    input_path = "C:\\Users\\Administrator\\Desktop\\招标文件\\招标test文件夹\\zbtest7.pdf"  # 可以是单个PDF文件路径或文件夹路径  #zbtest20.pdf zbtest4_evaluation_method.pdf zbtest8.pdf
     # input_path="C:\\Users\\Administrator\\Desktop\\fsdownload\\68549b0b-e892-41a9-897c-c3694535ee61\\ztbfile.pdf"
     # input_path = "C:\\Users\\Administrator\\Desktop\\货物标\\zbfiles\\2-招标文件.pdf"
     output_folder="C:\\Users\\Administrator\\Desktop\\招标文件\\new_test\\新建文件夹"
@@ -679,5 +685,5 @@ if __name__ == "__main__":
     # print(files)
     selection = 5 # 例如：1 - 投标人须知前附表+正文, 2 - 评标办法, 3 -资格审查条件 4-招标公告 5-无效标
     generated_files = truncate_pdf_main(input_path, output_folder, selection)
-    print(generated_files)
+
     # print("生成的文件:", generated_files)

flask_app/main/投标人须知正文条款提取成json文件.py

@@ -180,7 +180,7 @@ def convert_clause_to_json(input_path,output_folder,type=1):
             r'^附(?:录|件|表)(?:一)?[:：]'
         ]
     else:
-        start_word = r'第[一二三四五六七八九十]+章\s*招标公告|第一卷|招标编号:|招标编号：'
+        start_word = r'第[一二三四五六七八九十]+章\s*招标公告|第一卷|招标编号:|招标编号：|投标邀请书|投标邀请函'
         end_phrases=[r'第[一二三四五六七八九十]+章\s*投标人须知',r'投标人须知前附表']
     result = convert_to_json(input_path, start_word, end_phrases)   #过滤无关信息
     # 检查输出文件夹是否存在，如果不存在则创建