aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl
diff options
context:
space:
mode:
authorYen Chi Hsuan <yan12125@gmail.com>2016-05-23 01:34:08 +0800
committerYen Chi Hsuan <yan12125@gmail.com>2016-05-23 01:40:11 +0800
commiteb7941e3e6e92bac26f5d21525fc8ac89c934abe (patch)
tree8eae588751968cb8293de55b1957d4cecbdcc8dc /youtube_dl
parentdb3b8b2103099a8859402f2167d7ad1a8fa66829 (diff)
[compat] Fix for XML with <!DOCTYPE> in Python 2.7 and 3.2
Such XML documents cause DeprecationWarning if python is run with `-W error`
Diffstat (limited to 'youtube_dl')
-rw-r--r--youtube_dl/compat.py17
1 files changed, 12 insertions, 5 deletions
diff --git a/youtube_dl/compat.py b/youtube_dl/compat.py
index 1392361a1..06e5f3ff6 100644
--- a/youtube_dl/compat.py
+++ b/youtube_dl/compat.py
@@ -245,13 +245,20 @@ try:
except ImportError: # Python 2.6
from xml.parsers.expat import ExpatError as compat_xml_parse_error
+
+etree = xml.etree.ElementTree
+
+
+class _TreeBuilder(etree.TreeBuilder):
+ def doctype(self, name, pubid, system):
+ pass
+
if sys.version_info[0] >= 3:
- compat_etree_fromstring = xml.etree.ElementTree.fromstring
+ def compat_etree_fromstring(text):
+ return etree.XML(text, parser=etree.XMLParser(target=_TreeBuilder()))
else:
# python 2.x tries to encode unicode strings with ascii (see the
# XMLParser._fixtext method)
- etree = xml.etree.ElementTree
-
try:
_etree_iter = etree.Element.iter
except AttributeError: # Python <=2.6
@@ -265,7 +272,7 @@ else:
# 2.7 source
def _XML(text, parser=None):
if not parser:
- parser = etree.XMLParser(target=etree.TreeBuilder())
+ parser = etree.XMLParser(target=_TreeBuilder())
parser.feed(text)
return parser.close()
@@ -277,7 +284,7 @@ else:
return el
def compat_etree_fromstring(text):
- doc = _XML(text, parser=etree.XMLParser(target=etree.TreeBuilder(element_factory=_element_factory)))
+ doc = _XML(text, parser=etree.XMLParser(target=_TreeBuilder(element_factory=_element_factory)))
for el in _etree_iter(doc):
if el.text is not None and isinstance(el.text, bytes):
el.text = el.text.decode('utf-8')