Refactoring and modularization

setup.py

@@ -51,6 +51,6 @@ setup(
     test_suite='tests',
     tests_require=test_requirements,
     url='https://github.com/Serbaf/tweet_model',
-    version='0.3.3',
+    version='0.4.0',
     zip_safe=False,
 )

tweet_model/tweet_model.py

@@ -2,7 +2,6 @@
 """Main module."""
 
 import logging
-import sys
 
 from tweet_manager.lib import format_csv
 
@@ -368,6 +367,37 @@ class NotValidTweetError(Exception):
     pass
 
 
+def get_tweet_from_csv_raw_line(header, line):
+    """
+    Given a CSV header and a CSV line in raw format (strings with comma
+    separated values), extract the values for every field and then calls
+    get_tweet_from_csv_line to instance a Tweet.
+    Returns a Tweet object
+    """
+
+    header_fields = format_csv.split_csv_line(header)
+    line_fields = format_csv.split_csv_line(line)
+
+    return get_tweet_from_csv_line(header_fields, line_fields)
+
+
+def get_tweet_from_csv_line(header_fields, line_fields):
+    """
+    Given the fields of a CSV line and header, the function instances a Tweet
+    object with all the non-empty attributes initialized to the values
+    indicated in the CSV entry.
+    Returns a Tweet object
+    """
+
+    tweet_contents = {}
+    for i in range(len(line_fields)):
+        if line_fields[i] != '':
+            tweet_contents[header_fields[i].replace(".", "__")] =\
+                line_fields[i]
+
+    return Tweet(**tweet_contents)
+
+
 def get_tweets_from_csv(csv_file):
     """
     Take one argument: a path pointing to a valid CSV file.
@@ -384,18 +414,18 @@ def get_tweets_from_csv(csv_file):
         header = csv_object.readline()
         body = csv_object.readlines()
 
-    header = format_csv.split_csv_line(header)
+    header_fields = format_csv.split_csv_line(header)
 
     # Check that the header contains valid fields
     test_tweet = Tweet()
-    for field in header:
+    for field in header_fields:
         field_components = field.split(".")
         checking_dict = test_tweet.__dict__
         error_string = ""
         for component in field_components:
             error_string += component
             if (checking_dict is None) or (component not in checking_dict):
-                logger.error('The field in the header "{error_string}" ' +
+                logger.error(f'The field in the header "{error_string}" ' +
                              'is not a valid element of a Tweet')
                 raise NotValidTweetError("Header contains field which doesn't"
                                          + " belong to tweet specification: "
@@ -406,12 +436,7 @@ def get_tweets_from_csv(csv_file):
     # Go through every tweet in the file, instance it using the 'Tweet' class
     # and add it to the list 'tweets'
     for j in range(len(body)):
-        body[j] = format_csv.split_csv_line(body[j])
-        tweet_contents = {}
-        for i in range(len(body[j])):
-            if body[j][i] != '':
-                tweet_contents[header[i].replace(".", "__")] = body[j][i]
-
-        tweets.append(Tweet(**tweet_contents))
+        line_fields = format_csv.split_csv_line(body[j])
+        tweets.append(get_tweet_from_csv_line(header_fields, line_fields))
 
     return tweets