Merge branch 'master' of http://62.234.201.16/nzy/papertools

niurenshaixuan/main.py

@@ -73,14 +73,11 @@ def load_fake_niuren():
     return fake_niuren_names, fake_niuren_papers
 
 
-def check_niuren(authors, title):
+def check_niuren(authors, title, niuren_pool_names, true_niuren_names, true_niuren_papers, fake_niuren_names, fake_niuren_papers):
     niuren = []
     order = []
     suspected_niuren = []
 
-    niuren_pool_names = load_niuren_pool()
-    true_niuren_names, true_niuren_papers = load_true_niuren()
-    fake_niuren_names, fake_niuren_papers = load_fake_niuren()
 
     for author_idx, author in enumerate(authors):
         if author == "":
@@ -112,13 +109,13 @@ def check_niuren(authors, title):
     return niuren_str, order_str, suspected_niuren_str
 
 
-def process_row(index, row):
+def process_row(index, row, niuren_pool_names, true_niuren_names, true_niuren_papers, fake_niuren_names, fake_niuren_papers):
     authors = row.get('引文作者', '')
     title = row.get('引文名称', '')
     if not isinstance(authors, str):
         return index, "", "", ""
     authors = [i.strip() for i in authors.split(";") if i!=""]
-    niuren_true, niuren_true_order, suspected_niuren = check_niuren(authors, title)
+    niuren_true, niuren_true_order, suspected_niuren = check_niuren(authors, title, niuren_pool_names, true_niuren_names, true_niuren_papers, fake_niuren_names, fake_niuren_papers)
     return index, niuren_true, niuren_true_order, suspected_niuren
 
 
@@ -126,6 +123,10 @@ if __name__ == "__main__":
     # 读取原始 Excel 文件的前两行
     original_header = pd.read_excel(input_file_path, nrows=2, header=None)
 
+    niuren_pool_names = load_niuren_pool()
+    true_niuren_names, true_niuren_papers = load_true_niuren()
+    fake_niuren_names, fake_niuren_papers = load_fake_niuren()
+
     # 读取数据并跳过前两行
     input_df = pd.read_excel(input_file_path, skiprows=2)
 
@@ -135,17 +136,12 @@ if __name__ == "__main__":
     input_df['疑似牛人'] = None
 
 
-    results = Parallel(n_jobs=-1)(delayed(process_row)(index, row) for index, row in tqdm(input_df.iterrows()))
+    results = Parallel(n_jobs=-1)(delayed(process_row)(index, row, niuren_pool_names, true_niuren_names, true_niuren_papers, fake_niuren_names, fake_niuren_papers) for index, row in tqdm(input_df.iterrows()))
     for index, niuren_true, niuren_true_order, suspected_niuren in results:
         input_df.at[index, '牛人'] = niuren_true
         input_df.at[index, '牛人署名顺序'] = niuren_true_order
         input_df.at[index, '疑似牛人'] = suspected_niuren
 
-    # for index, row in tqdm(input_df.iterrows()):
-    #     index, niuren_true, niuren_true_order, suspected_niuren = process_row(index, row)
-    #     input_df.at[index, '牛人'] = niuren_true
-    #     input_df.at[index, '牛人署名顺序'] = niuren_true_order
-    #     input_df.at[index, '疑似牛人'] = suspected_niuren
 
 
     max_cols = max(original_header.shape[1], input_df.shape[1])