Merge pull request #35 from rguptauw/string_fix

string fix

pybcoin/SentimentAnalyzer/sentiment_scorer.py

@@ -24,6 +24,7 @@ def __init__(self, config):
         self.logger = logging.getLogger('simpleExample')
         self.analyzer = SentimentIntensityAnalyzer()
         self.path = config['Sentiment']['text_csv_path']
+        self.write_format = '_sentiment.csv'
 
     def __sentiment_count__(self, grouped_Data):
         """
@@ -61,6 +62,6 @@ def sentiment_scorer(self, keyword='tweets'):
 
         text_df = text_df.groupby(['Date']).apply(
             self.__sentiment_count__).reset_index().drop(['level_1'], axis=1)
-        hist_sent = pd.read_csv(self.path + keyword + '_sentiment.csv')
+        hist_sent = pd.read_csv(self.path + keyword + self.write_format)
         hist_sent = hist_sent.append(text_df)
-        hist_sent.to_csv(self.path + keyword + '_sentiment.csv', index=False)
+        hist_sent.to_csv(self.path + keyword + self.write_format, index=False)

pybcoin/test/test_sentiment.py

@@ -21,6 +21,10 @@ def setUp(self):
         self.text = ['neo', 'bitcoin', 'bitcoin', 'fork']
         self.date = ['22/05/2016']
         self.path = './pybcoin/test/data/'
+        self.tweets_path = 'tweets_sentiment.csv'
+        self.image_path = './pybcoin/static/date_22-05-2016.png'
+        self.static_path = './pybcoin/static/'
+        self.image_name = 'date_22-05-2016.png'
 
     """
     Test function for text sentiments.
@@ -34,10 +38,10 @@ def test_sentiment_score(self):
         """
         self.collector.sentiment_scorer(keyword='tweets')
         test_sentiment = pd.read_csv(self.collector.path +
-                                     'tweets_sentiment.csv')
+                                     self.tweets_path)
         flag = test_sentiment.empty
         self.assertEqual(flag, False)
-        os.remove('./pybcoin/static/date_22-05-2016.png')
+        os.remove(self.image_path)
 
     def test_sentiment_column_names(self):
 
@@ -46,17 +50,17 @@ def test_sentiment_column_names(self):
         """
         self.collector.sentiment_scorer(keyword='tweets')
         test_col_names = pd.read_csv(self.collector.path +
-                                     'tweets_sentiment.csv').columns
+                                     self.tweets_path).columns
         self.assertEqual(sorted(test_col_names),
                          ['Date', 'Negative', 'Positive'])
-        os.remove('./pybcoin/static/date_22-05-2016.png')
+        os.remove(self.image_path)
 
     def test_sentiment_wordcloud(self):
 
         """
         Checks that wordcloud is generated.
         """
         create_word_cloud(self.text, self.date)
-        self.assertEqual(os.path.isfile('./pybcoin/static/' +
-                                        'date_22-05-2016.png'), True)
-        os.remove('./pybcoin/static/date_22-05-2016.png')
+        self.assertEqual(os.path.isfile(self.static_path +
+                                        self.image_name), True)
+        os.remove(self.image_path)

pybcoin/utils/pre_processing.py

@@ -35,10 +35,11 @@ def remove_urls(text):
         : param path
         : return text(str)
     """
+    url_pattern = r'(?:(?:http|https|https://www|https://|http://|):\/\/)' \
+        '?([-a-zA-Z0-9.]{2,256}\.[a-z]{2,4})\b(?:\/[-a-zA-Z0-9@:' \
+        '%_\+.~#?&//=]*)?'
     text = re.sub(r"http\S+", "", text)
-    text = re.sub(r'(?:(?:http|https|https://www|https://|http://|):\/\/)'
-                  '?([-a-zA-Z0-9.]{2,256}\.[a-z]{2,4})\b(?:\/[-a-zA-Z0-9@:'
-                  '%_\+.~#?&//=]*)?', "", text, flags=re.MULTILINE)
+    text = re.sub(url_pattern, "", text, flags=re.MULTILINE)
     text = '\n'.join([a for a in text.split("\n") if a.strip()])
     return text
 
@@ -100,18 +101,18 @@ def create_word_cloud(text, date_generated):
         : param text
         : return png(matplotlib image)
     """
+    word_cloud_path = './pybcoin/static/date_'
     text = ' '.join(text)
     wordcloud = WordCloud(stopwords=STOPWORDS,
-                          background_color='black',
+                          background_color='white',
                           width=2500,
                           height=2000
                           ).generate(text)
     plt.figure(1, figsize=(13, 13))
     plt.imshow(wordcloud)
     plt.axis('off')
-    plt.savefig('./pybcoin/static/date_' +
-                date_generated[0].replace('/', '-') +
-                '.png', facecolor='k', bbox_inches='tight')
+    plt.savefig(word_cloud_path + date_generated[0].replace('/', '-') +
+                '.png', bbox_inches='tight')
 
 
 def pre_process_data(path):