Added method to clean input csv of NBSP unicode chars

The CSV file gets copied and pasted, into and out of Slack. While we would not recoomend this, it happened.
This addition strips thos characters from the input csv and writes a clean file.

src/main.py

@@ -5,6 +5,13 @@
 import csv
 from geocode import geocode, find_timezone, find_country_code
 
+# Strip non-breaking-space invisible characters
+def replace_non_breaking_spaces(unsanitised, clean):
+    with open(unsanitised, "r") as input:
+        with open(clean, "w") as output:
+            for line in input:
+                line = line.replace("\xa0", " ")
+                output.write(line)
 
 # Convert CSV file to JSON object.
 def convert_csv_to_json(file_path):
@@ -43,7 +50,11 @@ def add_geocoding_to_json(data):
 
 if __name__ == '__main__':
 
-    csv_file_path = os.getcwd() + '/../data_src/sites_with_clients.csv'
+    unsanitised_csv_file_path = os.getcwd() + '/../data_src/sites_with_clients.csv'
+    csv_file_path = os.getcwd() + '/../data_src/sites_with_clients.clean.csv'
+
+    # Strip non-breaking-space invisible characters
+    replace_non_breaking_spaces(unsanitised_csv_file_path, csv_file_path)
 
     # Convert CSV to valid JSON
     json_data_without_geocoding = convert_csv_to_json(csv_file_path)