rename methods

source/services/ner/awscomprehend_2_ner_format.py

@@ -14,7 +14,7 @@ class Comprehend2NERFormat:
         return json_letter
     
     @staticmethod
-    def get_tokens(jsondata):
+    def _get_tokens(jsondata):
         data_token = []
         for block in jsondata['Blocks']:
             if block["BlockType"] == 'WORD':
@@ -23,7 +23,7 @@ class Comprehend2NERFormat:
         return df_token
     
     @staticmethod
-    def get_line_child_ids(jsondata):
+    def _get_line_child_ids(jsondata):
         df_line = pd.DataFrame(columns = ['lineid', 'childid'])
         for block in jsondata['Blocks']:
             if block["BlockType"] == 'LINE':
@@ -33,7 +33,7 @@ class Comprehend2NERFormat:
         return df_line
 
     @staticmethod
-    def get_ner_tags(jsondata):
+    def _get_ner_tags(jsondata):
         data_nertags = []
         for block in jsondata['Entities']:
             ner_tag = block['Type']
@@ -50,7 +50,7 @@ class Comprehend2NERFormat:
         return df_nertags
 
     @staticmethod
-    def insert_newline_char(df_prev):
+    def _insert_newline_char(df_prev):
         df = pd.DataFrame()
         df_insert = pd.DataFrame({'blockid' : 'newline', 'token' : '\n', 'ner_tag': 'O','lineid': 'newline'},index=[0])
         for group in  df_prev.groupby('lineid'):
@@ -62,16 +62,16 @@ class Comprehend2NERFormat:
 
     def __call__(self):
         json_letter = self.load_data()
-        df_token = self.get_tokens(jsondata=json_letter)
-        df_line = self.get_line_child_ids(jsondata=json_letter)
-        df_nertags = self.get_ner_tags(jsondata=json_letter)
+        df_token = self._get_tokens(jsondata=json_letter)
+        df_line = self._get_line_child_ids(jsondata=json_letter)
+        df_nertags = self._get_ner_tags(jsondata=json_letter)
 
         df1 = pd.merge(df_token, df_nertags, on='blockid', how='left')
         df1['ner_tag'][df1['ner_tag'].isna()] = 'O'
         df2 = pd.merge(df1, df_line, left_on='blockid', right_on='childid', how='left').drop(columns=['childid'])
         df2['linewordrank'] = np.arange(df2.shape[0])
 
-        df3 = self.insert_newline_char(df_prev=df2)
+        df3 = self._insert_newline_char(df_prev=df2)
         return {"tokens": df3['token'].tolist(),
                 "ner_tags": df3['ner_tag'].tolist(),
                 "filename": self.letterfilepath.name