Update 'chatbot/multipal_col.py'

chatbot/multipal_col.py

@@ -1,37 +1,51 @@
 import pandas as pd
 import yaml
+import re
 
 # Read the data from the CSV file
-df = pd.read_csv(r"D:\livecode\Book1.csv")
+df = pd.read_csv(r"D:\livecode\chatbot\AI Support Questionaire - Sheet1 (2).csv", encoding='latin')
 
-# df['Question'] = df['Question'].replace('\n', ' ', regex=True)
-# df['Question'] = df['Question'].str.strip()
 
-# df['Answers'] = df['Answers'].replace('\n', ' ', regex=True)
-# df['Answers'] = df['Answers'].str.strip()
+def remove_special_characters(df):
+    for column in df.columns:
+        df[column] = df[column].apply(lambda x: re.sub(r'[^a-zA-Z0-9->]', ' ', str(x)))
+    return df
 
-# # Extract the columns from the dataframe
-Question = df['Question'].tolist()
-Example1 = df['Example1'].tolist()
-Example2 = df['Example2'].tolist()
-Example3 = df['Example3'].tolist()
-uniqueid = df['uniqueid'].tolist()
-Answers = df['Answers'].tolist()
+df_cleaned = remove_special_characters(df)
+uniqueid = df['UniqueId'].tolist()
+# Answers = df['Answers'].tolist()
 
-question_cols = df.filter(regex='Example').columns
+question_cols = df.filter(regex='Question').columns
 answer_cols = df.filter(regex='Answers').columns
+
+
 #appending intents
 import ruamel.yaml as yaml
 
 # Load the domain YAML file
-with open(r'C:\Users\Bizgaze\Desktop\fileupdation\trail_update\domain.yml', 'r') as f:
+with open(r'D:\\livecode\\chatbot\\traildata\\domain.yml', 'r') as f:
     domain = yaml.safe_load(f)
     for i in uniqueid:
         domain['intents'].append(i)
-        with open(r'C:\Users\Bizgaze\Desktop\fileupdation\trail_update\domain.yml', 'w') as f:
+        with open(r'D:\\livecode\\chatbot\\traildata\\domain.yml', 'w') as f:
             yaml.dump(domain, f, default_flow_style=False, allow_unicode=True)
 
-#appending muitipal answers
+
+
+question_cols = df.filter(regex='Question').columns
+answer_cols = df.filter(regex='Answer').columns
+
+# for i in question_cols:
+#     df[i] = df[i].replace('\n', '$$', regex=True)
+#    # df[i] = df[i].str.strip()
+
+
+# for i in answer_cols:
+#     df[i] = df[i].replace('\n', '$$', regex=True)
+#    # df[i] = df[i].str.strip()
+
+
+
 from ruamel.yaml import YAML
 
 # Create a YAML object that preserves the formatting of the original YAML file
@@ -40,19 +54,19 @@ yaml.preserve_quotes = True
 yaml.indent(mapping=2, sequence=4, offset=2)
 
 # Read in the existing domain file
-with open(r'C:\Users\Bizgaze\Desktop\fileupdation\trail_update\domain.yml', 'r') as file:
+with open(r'D:\\livecode\\chatbot\\traildata\\domain.yml', 'r', encoding='utf-8') as file:
     domain = yaml.load(file)
 
 # Generate the new YAML code for each row of data
 new_content = {}
 for index, row in df.iterrows():
-    intent_name = row['uniqueid']
+    intent_name = row['UniqueId']
     examples = [row[col] for col in answer_cols ]
     separator = "_"
     for counter, example in enumerate(examples, start=1):
         #example=example.replace(':','')
         response_name = f"utter_{intent_name}{separator}{counter}"
-        response_text = f"- text: {example}"
+        response_text =f"- text: {example}"
         if 'responses' not in new_content:
             new_content['responses'] = {}
         new_content['responses'][response_name] = yaml.load(response_text)
@@ -61,14 +75,16 @@ for index, row in df.iterrows():
 domain['responses'].update(new_content['responses'])
 
 # Write the updated domain file back to disk, preserving the formatting of the original file
-with open(r'C:\Users\Bizgaze\Desktop\fileupdation\trail_update\domain.yml', 'w') as file:
+with open(r'D:\\livecode\\chatbot\\traildata\\domain.yml', 'w',encoding='utf-8') as file:
     yaml.dump(domain, file)
 
+
+
 #appending multipal action for each rule
-with open(r"C:\Users\Bizgaze\Desktop\fileupdation\trail_update\data\rules.yml", 'a') as f:
+with open(r"D:\\livecode\\chatbot\\traildata\\rules.yml", 'a') as f:
     for index, row in df.iterrows():
-        intent_name = row['uniqueid']
-        examples = [row[col] for col in question_cols]
+        intent_name = row['UniqueId']
+        examples = [row[col] for col in answer_cols]
         separator = "_"
         steps = {"intent": intent_name}
         gg=[]
@@ -83,25 +99,28 @@ with open(r"C:\Users\Bizgaze\Desktop\fileupdation\trail_update\data\rules.yml",
   steps:
   - intent: {steps['intent']}
   {output_str}
+  - action: action_service
   
 """)
 
 
-# appending multipal question
 
-with open(r'C:\Users\Bizgaze\Desktop\fileupdation\trail_update\data\nlu.yml', "a") as file:
+
+
+with open(r'D:\\livecode\\chatbot\\traildata\\nlu.yml', "a",encoding='utf-8') as file:
     for index, row in df.iterrows():
-        intent_name = row['uniqueid']
+        intent_name = row['UniqueId']
         examples = [row[col] for col in question_cols]
         file.write(f"""
 - intent: {intent_name}
   examples: |
 """)
         for example in examples:
-            file.write(f"   - \"{example}\"\n")
+            file.write(f"    - \"{example}\"\n")
         file.write("\n")
 
 
+
 # filename = r"C:\Users\Bizgaze\Desktop\fileupdation\trail_update\data\nlu.yml"
 # string_to_remove = '  - "nan"'