Add bsdetector command-line entry point.

- Fix typo

bsdetector/__main__.py

@@ -0,0 +1,21 @@
+import argparse
+from bsdetector.bias import get_text_from_article_file, print_feature_data
+
+
+DESC = "Biased Statement Detection explains why english prose text is biased."
+
+
+def main():
+    parser = argparse.ArgumentParser(description=DESC)
+    parser.add_argument("-i", dest="filename", required=True,
+                        help="Input file", metavar="FILE")
+    parser.add_argument("-o", dest="output", required=False,
+                        help="Output type", metavar="OUTPUT",
+                        default='json', choices=['tsv', 'json', 'html'])
+    args = parser.parse_args()
+    sentence_list = get_text_from_article_file(args.filename).split('\n')
+    print_feature_data(sentence_list, output_type=args.output)
+
+
+if __name__ == '__main__':
+    main()

bsdetector/bias.py

@@ -585,7 +585,7 @@ def make_dict_output(list_of_sentences):
         if len(sent) >= 1:
             feature_data = extract_bias_features(sent)
             feature_data['text'] = sent
-            data.insert(0, feature_data)
+            data.append(feature_data)
     return data
 
 
@@ -636,25 +636,3 @@ def enumerate_sentences(fpath='input_text'):
             yield(biasq, statement)
         else:
             print('-- Statement is too short: {}'.format(statement))
-
-
-
-if __name__ == '__main__':
-    # Demo article file
-    # print(compute_avg_statement_bias_mp(get_text_from_article_file("news_articles/brexit_01.txt"), 4))
-    '''FPATH = 'input_text'
-    for bias, stmt in enumerate_sentences(FPATH):
-        msg = 'Bias: {}\t {}'.format(bias, stmt)
-        print(msg)
-
-    NEWSPATH = "news_articles/brexit_01.txt"
-    print('loading news article: {}'.format(NEWSPATH), file=sys.stderr)
-    STATEMENT = get_text_from_article_file(NEWSPATH)
-    print(compute_avg_statement_bias(STATEMENT))'''
-
-    demo_output_types = True
-    if demo_output_types:
-        sentence_list = get_text_from_article_file('input_text').split('\n')#[:2]
-        print_feature_data(sentence_list, output_type='json')
-        #cstr_summary_terms = get_caster(" ".join(sentence_list))
-        #print(cstr_summary_terms)

setup.py

@@ -50,6 +50,11 @@ def run(self):
     description="Detects biased statements in online media documents",
     url='url',
     cmdclass={
-        'install': PostInstallCommand,
+        'install': PostInstallCommand
+    },
+    entry_points={
+        'console_scripts': [
+            'bsdetector = bsdetector.__main__:main'
+        ]
     }
 )