Merge pull request #438 from NaanProphet/stopwords-unicode

unicode stopwords files now supported in python 2.7

test/test_wordcloud_cli.py

@@ -121,6 +121,15 @@ def test_unicode_text_file():
     assert len(text) == 16
 
 
+def test_unicode_with_stopwords():
+    unicode_file = os.path.join(os.path.dirname(__file__), "unicode_text.txt")
+    stopwords_file = os.path.join(os.path.dirname(__file__), "unicode_stopwords.txt")
+    args, text, image_file = cli.parse_args(['--text', unicode_file, '--stopwords', stopwords_file])
+
+    # expect the unicode character from stopwords file was correctly read in
+    assert u'\u304D' in args['stopwords']
+
+
 def test_cli_writes_image(tmpdir, tmp_text_file):
     # ensure writing works with all python versions
     tmp_image_file = tmpdir.join("word_cloud.png")

test/unicode_stopwords.txt

@@ -0,0 +1 @@
+き

wordcloud/wordcloud_cli.py

@@ -175,7 +175,7 @@ def parse_args(arguments):
 
     if args['stopwords']:
         with args.pop('stopwords') as f:
-            args['stopwords'] = set(map(str.strip, f.readlines()))
+            args['stopwords'] = set(map(lambda l: l.strip(), f.readlines()))
 
     if args['mask']:
         mask = args.pop('mask')