Merge pull request #1380 from avaris/py34_warnings

Fix for Python 3.4 deprecation warnings while running tests

pelican/readers.py

@@ -9,6 +9,7 @@ import docutils
 import docutils.core
 import docutils.io
 from docutils.writers.html4css1 import HTMLTranslator
+import six
 
 # import the directives to have pygments support
 from pelican import rstdirectives  # NOQA
@@ -20,10 +21,7 @@ try:
     from html import escape
 except ImportError:
     from cgi import escape
-try:
+from six.moves.html_parser import HTMLParser
-    from html.parser import HTMLParser
-except ImportError:
-    from HTMLParser import HTMLParser
 
 from pelican import signals
 from pelican.contents import Page, Category, Tag, Author
@@ -42,7 +40,6 @@ METADATA_PROCESSORS = {
 
 logger = logging.getLogger(__name__)
 
-
 class BaseReader(object):
     """Base class to read files.
 
@@ -121,6 +118,19 @@ class RstReader(BaseReader):
     enabled = bool(docutils)
     file_extensions = ['rst']
 
+    class FileInput(docutils.io.FileInput):
+        """Patch docutils.io.FileInput to remove "U" mode in py3.
+
+        Universal newlines is enabled by default and "U" mode is deprecated
+        in py3.
+
+        """
+
+        def __init__(self, *args, **kwargs):
+            if six.PY3:
+                kwargs['mode'] = kwargs.get('mode', 'r').replace('U', '')
+            docutils.io.FileInput.__init__(self, *args, **kwargs)
+
     def __init__(self, *args, **kwargs):
         super(RstReader, self).__init__(*args, **kwargs)
 
@@ -152,12 +162,14 @@ class RstReader(BaseReader):
         extra_params = {'initial_header_level': '2',
                         'syntax_highlight': 'short',
                         'input_encoding': 'utf-8',
-                        'exit_status_level': 2}
+                        'exit_status_level': 2,
+                        'embed_stylesheet': False}
         user_params = self.settings.get('DOCUTILS_SETTINGS')
         if user_params:
             extra_params.update(user_params)
 
         pub = docutils.core.Publisher(
+            source_class=self.FileInput,
             destination_class=docutils.io.StringOutput)
         pub.set_components('standalone', 'restructuredtext', 'html')
         pub.writer.translator_class = PelicanHTMLTranslator
@@ -230,7 +242,11 @@ class HTMLReader(BaseReader):
 
     class _HTMLParser(HTMLParser):
         def __init__(self, settings, filename):
-            HTMLParser.__init__(self)
+            try:
+                # Python 3.4+
+                HTMLParser.__init__(self, convert_charrefs=False)
+            except TypeError:
+                HTMLParser.__init__(self)
             self.body = ''
             self.metadata = {}
             self.settings = settings

pelican/tools/pelican_import.py

@@ -4,11 +4,10 @@
 from __future__ import unicode_literals, print_function
 import argparse
 try:
-    # py3k import
+    from html import unescape  # py3.4+
-    from html.parser import HTMLParser
 except ImportError:
-    # py2 import
+    from six.moves.html_parser import HTMLParser
-    from HTMLParser import HTMLParser  # NOQA
+    unescape = HTMLParser().unescape
 import os
 import re
 import subprocess
@@ -129,7 +128,7 @@ def wp2fields(xml, wp_custpost=False):
 
             try:
                 # Use HTMLParser due to issues with BeautifulSoup 3
-                title = HTMLParser().unescape(item.title.contents[0])
+                title = unescape(item.title.contents[0])
             except IndexError:
                 title = 'No title [%s]' % item.find('post_name').string
                 logger.warning('Post "%s" is lacking a proper title' % title)