From cf9299e50381664a16c6de59e8ea6a2e995e2deb Mon Sep 17 00:00:00 2001
From: Rory Healy <rory.healy@student.unimelb.edu.au>
Date: Sun, 11 Apr 2021 19:43:49 +1000
Subject: [PATCH] partb2 complete

---
 partb2.py | 23 ++++++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)

diff --git a/partb2.py b/partb2.py
index 5f655e9..140080c 100644
--- a/partb2.py
+++ b/partb2.py
@@ -1,5 +1,22 @@
-# Part B Task 2
 import re
-import os
-import sys
+import argparse
 
+# parse input arguments
+parser = argparse.ArgumentParser()
+parser.add_argument('path_to_file', help = 'path to the csv file')
+args = parser.parse_args()
+
+# open file, add all lines to a single string
+file_given = open(args.path_to_file)
+f = ""
+for line in file_given:
+    f += line + " "
+file_given.close()
+
+# remove non-alphabetic characters, replace all whitespace characters with a 
+# single whitespace, and change all uppercase characters to lowercase
+f = re.sub(r'[^a-zA-Z\s]', r'', f)
+f = re.sub(r'\s+', r' ', f)
+f = f.lower()
+
+print(f)