Fix unicode errors when rendering to IPython notebook

redbaron.py

@@ -354,13 +354,13 @@ def __repr_html(self):
             for num, item in enumerate(self):
                 yield u"<tr>"
                 yield u"<td>"
-                yield str(num)
+                yield u"%d" % num
                 yield u"</td>"
                 yield u"<td>"
                 yield item._repr_html_() if hasattr(item, "_repr_html_") else str(item)
                 yield u"</td>"
                 yield u"</tr>"
-            yield "</table>"
+            yield u"</table>"
         return u''.join(__repr_html(self))
 
 
@@ -919,7 +919,7 @@ def __str__(self):
 
     def _repr_html_(self):
         return highlight(self.dumps(), PythonLexer(encode="Utf-8"),
-                         HtmlFormatter(noclasses=True, encoding="UTf-8"))
+                         HtmlFormatter(noclasses=True))
 
     def copy(self):
         # XXX not very optimised but at least very simple
@@ -1342,13 +1342,13 @@ def __repr_html(self):
             for num, item in enumerate(self):
                 yield u"<tr>"
                 yield u"<td>"
-                yield str(num)
+                yield u"%d" % num
                 yield u"</td>"
                 yield u"<td>"
                 yield item._repr_html_()
                 yield u"</td>"
                 yield u"</tr>"
-            yield "</table>"
+            yield u"</table>"
         return u''.join(__repr_html(self))
 
     def __str__(self):
@@ -2031,7 +2031,7 @@ def __repr__(self):
 
     def _repr_html_(self):
         return highlight(self.__repr__(), PythonLexer(encode="Utf-8"),
-                         HtmlFormatter(noclasses=True, encoding="UTf-8"))
+                         HtmlFormatter(noclasses=True))
 
 
 class ExceptNode(CodeBlockNode):

tests/test_redbaron.py

@@ -3,9 +3,20 @@
 
 """ Main redbaron test module """
 
+import re
+import sys
 from redbaron import RedBaron, truncate
 
 
+python_version = sys.version_info[0]
+if python_version == 3:
+    unicode_type = str
+    unicode_chr = chr
+else:
+    unicode_type = unicode
+    unicode_chr = unichr
+
+
 def test_other_name_assignment():
     red = RedBaron("a = b")
     assert red.assign is red[0]
@@ -36,3 +47,33 @@ def test_truncate():
     assert "1...6" == truncate("123456", 5)
     assert "123456...0" == truncate("12345678901234567890", 10)
 
+def test_html_repr():
+    def strip_html_tags(s):
+        assert isinstance(s, unicode_type)
+        s = re.sub(r'<[^>]+>', '', s)
+        s = re.sub(r'\n+', '\n', s)
+        s = re.sub(r'&#([0-9]+);', lambda m: unicode_chr(int(m.group(1))), s)
+        return s
+    source = (
+        b"first = line  # commentaire en fran\xC3\x87ais\n"
+        b'wait()\n'
+        b"if second_line:\n"
+        b"    # l'unicode est support\xC3\xA9\n"
+        b"    third(line)\n")
+    if python_version == 3:
+        source = source.decode('utf-8')
+    red = RedBaron(source)
+    assert strip_html_tags(red._repr_html_()) == (
+        u'Index' u'node'
+        u'0' u'first = line\n'
+        u'1' u'  # commentaire en fran\xC7ais\n'
+        u'2' u'wait()\n'
+        u'3' u'if second_line:\n'
+        u"    # l'unicode est support\xE9\n"
+        u'    third(line)\n')
+    assert strip_html_tags(red.node_list[1:4]._repr_html_()) == (
+        u'Index' u'node'
+        u'0' u'  # commentaire en fran\xC7ais\n'
+        u'1' u"'\\n'\n"
+        u'2' u'wait()\n')
+