Create preprocess.py

preprocess.py

@@ -0,0 +1,18 @@
+import re  , string
+from textacy.preprocessing.resources import (RE_EMAIL , RE_URL , RE_NUMBER ,
+                                              RE_NUMBER , RE_EMOJI , RE_SHORT_URL , RE_PHONE_NUMBER
+                                            )
+
+NON_ARABIC_RE = re.compile(r"[%s]"%string.ascii_letters)
+
+def clean_text(text:str)->str:
+    '''remove unwanted data'''
+    patterns = [RE_EMAIL , RE_EMOJI , RE_NUMBER , RE_PHONE_NUMBER , RE_SHORT_URL , RE_URL , NON_ARABIC_RE]
+    
+    for pattern in patterns:
+        
+        text = pattern.sub("" , text)
+        
+    return text
+
+