feature: support python3.5 python3.6, closes #5

.travis.yml

@@ -1,6 +1,8 @@
 language: python
 python:
   - '2.7'
+  - '3.5'
+  - '3.6'
 sudo: false
 install:
   python setup.py install

rich_text_diff/__init__.py

@@ -1,16 +1,21 @@
 # -*- coding: utf-8 -*-
 import copy
+import sys
 import logging
 import re
-import urllib
 
 from bidict import bidict
 import diff_match_patch as dmp_module
 from lxml.html import fromstring, tostring, fragment_fromstring
 from lxml import etree
 
+if sys.version_info < (3,):
+    chr = unichr
+    unicode_type = unicode
+else:
+    unicode_type = str
 
-UNICODE_KEY = [unichr(item) for item in range(0xE000, 0xFFFF + 1)]
+UNICODE_KEY = [chr(item) for item in range(0xE000, 0xFFFF + 1)]
 # unicode spec not in use
 DMP = dmp_module.diff_match_patch()
 
@@ -45,31 +50,31 @@ def _map_tag(self, content):
                     self.tag_map[self.code_key.pop()] = tag
 
     def _map_media_tag(self, element, raw_tag):
-        url = urlencode(element.attrib)
-        if url in self.media_url.values():
-            code = self.media_url.inv[url]
+        tag_key = gen_tag_key(element.attrib)
+        if tag_key in self.media_url.values():
+            code = self.media_url.inv[tag_key]
             self.tag_map[code].append(raw_tag)
             return
         code = self.code_key.pop()
         self.tag_map[code] = [raw_tag]
-        self.media_url[code] = url
+        self.media_url[code] = tag_key
 
     def _replace(self, new_content, old_content):
         self._map_tag(new_content)
-        for code, tag in self.tag_map.iteritems():
+        for code, tag in self.tag_map.items():
             if not isinstance(tag, list):
                 tag = [tag]
             for item in tag:
                 new_content = new_content.replace(item, code)
-        for code, tag in self.tag_map.iteritems():
+        for code, tag in self.tag_map.items():
             if not isinstance(tag, list):
                 tag = [tag]
             for item in tag:
                 old_content = old_content.replace(item, code)
         return to_unicode(new_content), to_unicode(old_content)
 
     def _recover(self, content):
-        for code, tag in self.tag_map.iteritems():
+        for code, tag in self.tag_map.items():
             if isinstance(tag, list):
                 tag = tag[0]
             content = content.replace(code, tag)
@@ -88,15 +93,16 @@ def _diff(self, old_content, new_content):
         for (op, data) in diffs:
             text = self._recover(data)
             if op == self.INSERT:
-                html.append("<ins style=\"background:#e6ffe6;\">{}</ins>".format(text))
+                html.append(u"<ins style=\"background:#e6ffe6;\">{}</ins>".format(text))
             elif op == self.DELETE:
-                html.append("<del style=\"background:#ffe6e6;\">{}</del>".format(text))
+                html.append(u"<del style=\"background:#ffe6e6;\">{}</del>".format(text))
             elif op == self.EQUAL:
                 html.append(text)
-        return "".join(html)
+        return utf8(u"".join(html))
 
 
-_TO_UNICODE_TYPES = (unicode, type(None))
+_TO_UNICODE_TYPES = (unicode_type, type(None))
+_UTF8_TYPES = (bytes, type(None))
 
 
 def to_unicode(value):
@@ -109,21 +115,23 @@ def to_unicode(value):
     return value.decode("utf-8")
 
 
+def utf8(value):
+    if isinstance(value, _UTF8_TYPES):
+        return value
+    if not isinstance(value, unicode_type):
+        raise TypeError("Expected bytes, unicode, or None; got %r" % type(value))
+    return value.encode("utf-8")
+
+
 def ensure_closed_tag(html):
     try:
         element = fromstring(html)
     except etree.ParserError as e:
-        logging.warn('fromstring error: {}, use fragment_fromstring'.format(e))
+        logging.warning('fromstring error: {}, use fragment_fromstring'.format(e))
         element = fragment_fromstring(html, create_parent='div')
-    return tostring(element, encoding='utf-8')
+    return to_unicode(tostring(element, encoding='utf-8'))
 
 
-def urlencode(query):
-    l = []
-    for k, v in query.items():
-        if isinstance(v, unicode):
-            v = v.encode('utf-8')
-        k = urllib.quote_plus(str(k))
-        v = urllib.quote_plus(str(v))
-        l.append(k + '=' + v)
+def gen_tag_key(query):
+    l = ["{}={}".format(to_unicode(k), to_unicode(v)) for k, v in query.items()]
     return '&'.join(l)

setup.py

@@ -3,13 +3,13 @@
 
 setup(
     name='rich_text_diff',
-    version='0.0.5',
+    version='0.0.6',
     author='liukai',
     author_email='liukai@zhihu.com',
     description='support rich text diff',
     packages=['rich_text_diff'],
     test_suite='nose.collector',
-    python_requires='>=2.7',
+    python_requires='>=2.7,!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*',
     long_description='readme',
     long_description_content_type='text/markdown',
     url='https://github.com/c1ay/rich_text_diff',