aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl
diff options
context:
space:
mode:
authorSergey M <dstftw@gmail.com>2015-10-31 18:15:21 +0000
committerSergey M <dstftw@gmail.com>2015-10-31 18:15:21 +0000
commit30eecc6a044d4070d0e8ea4a6e0637867f0d3a28 (patch)
treec120cc9df90a468bb0d58bb7a7601cb8418eb32f /youtube_dl
parentdbd82a1d4fff1655920e111cc25a7fd526d7bf9a (diff)
parentae37338e681319a28d98dc551253d9fa1830969a (diff)
Merge pull request #7296 from jaimeMF/xml_attrib_unicode
Use a wrapper around xml.etree.ElementTree.fromstring in python 2.x (…
Diffstat (limited to 'youtube_dl')
-rw-r--r--youtube_dl/compat.py39
-rw-r--r--youtube_dl/downloader/f4m.py4
-rw-r--r--youtube_dl/extractor/ard.py4
-rw-r--r--youtube_dl/extractor/bbc.py8
-rw-r--r--youtube_dl/extractor/bilibili.py6
-rw-r--r--youtube_dl/extractor/brightcove.py4
-rw-r--r--youtube_dl/extractor/common.py4
-rw-r--r--youtube_dl/extractor/crunchyroll.py4
-rw-r--r--youtube_dl/extractor/generic.py4
-rw-r--r--youtube_dl/extractor/vevo.py6
-rw-r--r--youtube_dl/utils.py26
11 files changed, 65 insertions, 44 deletions
diff --git a/youtube_dl/compat.py b/youtube_dl/compat.py
index d103ab9ad..a3e85264a 100644
--- a/youtube_dl/compat.py
+++ b/youtube_dl/compat.py
@@ -14,6 +14,7 @@ import socket
import subprocess
import sys
import itertools
+import xml.etree.ElementTree
try:
@@ -212,6 +213,43 @@ try:
except ImportError: # Python 2.6
from xml.parsers.expat import ExpatError as compat_xml_parse_error
+if sys.version_info[0] >= 3:
+ compat_etree_fromstring = xml.etree.ElementTree.fromstring
+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
+ def _etree_iter(root):
+ for el in root.findall('*'):
+ yield el
+ for sub in _etree_iter(el):
+ yield sub
+
+ # on 2.6 XML doesn't have a parser argument, function copied from CPython
+ # 2.7 source
+ def _XML(text, parser=None):
+ if not parser:
+ parser = etree.XMLParser(target=etree.TreeBuilder())
+ parser.feed(text)
+ return parser.close()
+
+ def _element_factory(*args, **kwargs):
+ el = etree.Element(*args, **kwargs)
+ for k, v in el.items():
+ if isinstance(v, bytes):
+ el.set(k, v.decode('utf-8'))
+ return el
+
+ def compat_etree_fromstring(text):
+ doc = _XML(text, parser=etree.XMLParser(target=etree.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')
+ return doc
try:
from urllib.parse import parse_qs as compat_parse_qs
@@ -507,6 +545,7 @@ __all__ = [
'compat_chr',
'compat_cookiejar',
'compat_cookies',
+ 'compat_etree_fromstring',
'compat_expanduser',
'compat_get_terminal_size',
'compat_getenv',
diff --git a/youtube_dl/downloader/f4m.py b/youtube_dl/downloader/f4m.py
index 7f6143954..6170cc155 100644
--- a/youtube_dl/downloader/f4m.py
+++ b/youtube_dl/downloader/f4m.py
@@ -5,10 +5,10 @@ import io
import itertools
import os
import time
-import xml.etree.ElementTree as etree
from .fragment import FragmentFD
from ..compat import (
+ compat_etree_fromstring,
compat_urlparse,
compat_urllib_error,
compat_urllib_parse_urlparse,
@@ -290,7 +290,7 @@ class F4mFD(FragmentFD):
man_url = urlh.geturl()
manifest = urlh.read()
- doc = etree.fromstring(manifest)
+ doc = compat_etree_fromstring(manifest)
formats = [(int(f.attrib.get('bitrate', -1)), f)
for f in self._get_unencrypted_media(doc)]
if requested_bitrate is None:
diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py
index 6f465789b..73be6d204 100644
--- a/youtube_dl/extractor/ard.py
+++ b/youtube_dl/extractor/ard.py
@@ -14,8 +14,8 @@ from ..utils import (
parse_duration,
unified_strdate,
xpath_text,
- parse_xml,
)
+from ..compat import compat_etree_fromstring
class ARDMediathekIE(InfoExtractor):
@@ -161,7 +161,7 @@ class ARDMediathekIE(InfoExtractor):
raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True)
if re.search(r'[\?&]rss($|[=&])', url):
- doc = parse_xml(webpage)
+ doc = compat_etree_fromstring(webpage.encode('utf-8'))
if doc.tag == 'rss':
return GenericIE()._extract_rss(url, video_id, doc)
diff --git a/youtube_dl/extractor/bbc.py b/youtube_dl/extractor/bbc.py
index 2cdce1eb9..a55a6dbc9 100644
--- a/youtube_dl/extractor/bbc.py
+++ b/youtube_dl/extractor/bbc.py
@@ -2,7 +2,6 @@
from __future__ import unicode_literals
import re
-import xml.etree.ElementTree
from .common import InfoExtractor
from ..utils import (
@@ -14,7 +13,10 @@ from ..utils import (
remove_end,
unescapeHTML,
)
-from ..compat import compat_HTTPError
+from ..compat import (
+ compat_etree_fromstring,
+ compat_HTTPError,
+)
class BBCCoUkIE(InfoExtractor):
@@ -344,7 +346,7 @@ class BBCCoUkIE(InfoExtractor):
url, programme_id, 'Downloading media selection XML')
except ExtractorError as ee:
if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
- media_selection = xml.etree.ElementTree.fromstring(ee.cause.read().decode('utf-8'))
+ media_selection = compat_etree_fromstring(ee.cause.read().decode('utf-8'))
else:
raise
return self._process_media_selector(media_selection, programme_id)
diff --git a/youtube_dl/extractor/bilibili.py b/youtube_dl/extractor/bilibili.py
index ecc17ebeb..6c66a1236 100644
--- a/youtube_dl/extractor/bilibili.py
+++ b/youtube_dl/extractor/bilibili.py
@@ -4,9 +4,11 @@ from __future__ import unicode_literals
import re
import itertools
import json
-import xml.etree.ElementTree as ET
from .common import InfoExtractor
+from ..compat import (
+ compat_etree_fromstring,
+)
from ..utils import (
int_or_none,
unified_strdate,
@@ -88,7 +90,7 @@ class BiliBiliIE(InfoExtractor):
except ValueError:
pass
- lq_doc = ET.fromstring(lq_page)
+ lq_doc = compat_etree_fromstring(lq_page)
lq_durls = lq_doc.findall('./durl')
hq_doc = self._download_xml(
diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py
index 4721c2293..1686cdde1 100644
--- a/youtube_dl/extractor/brightcove.py
+++ b/youtube_dl/extractor/brightcove.py
@@ -3,10 +3,10 @@ from __future__ import unicode_literals
import re
import json
-import xml.etree.ElementTree
from .common import InfoExtractor
from ..compat import (
+ compat_etree_fromstring,
compat_parse_qs,
compat_str,
compat_urllib_parse,
@@ -119,7 +119,7 @@ class BrightcoveIE(InfoExtractor):
object_str = fix_xml_ampersands(object_str)
try:
- object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8'))
+ object_doc = compat_etree_fromstring(object_str.encode('utf-8'))
except compat_xml_parse_error:
return
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 1f09fbb47..5e263f8b5 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -10,7 +10,6 @@ import re
import socket
import sys
import time
-import xml.etree.ElementTree
from ..compat import (
compat_cookiejar,
@@ -23,6 +22,7 @@ from ..compat import (
compat_urllib_request,
compat_urlparse,
compat_str,
+ compat_etree_fromstring,
)
from ..utils import (
NO_DEFAULT,
@@ -461,7 +461,7 @@ class InfoExtractor(object):
return xml_string
if transform_source:
xml_string = transform_source(xml_string)
- return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))
+ return compat_etree_fromstring(xml_string.encode('utf-8'))
def _download_json(self, url_or_request, video_id,
note='Downloading JSON metadata',
diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py
index f8ce10111..0c9b8ca02 100644
--- a/youtube_dl/extractor/crunchyroll.py
+++ b/youtube_dl/extractor/crunchyroll.py
@@ -5,12 +5,12 @@ import re
import json
import base64
import zlib
-import xml.etree.ElementTree
from hashlib import sha1
from math import pow, sqrt, floor
from .common import InfoExtractor
from ..compat import (
+ compat_etree_fromstring,
compat_urllib_parse,
compat_urllib_parse_unquote,
compat_urllib_request,
@@ -234,7 +234,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
return output
def _extract_subtitles(self, subtitle):
- sub_root = xml.etree.ElementTree.fromstring(subtitle)
+ sub_root = compat_etree_fromstring(subtitle)
return [{
'ext': 'srt',
'data': self._convert_subtitles_to_srt(sub_root),
diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index a84135032..ee5419f51 100644
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -9,6 +9,7 @@ import sys
from .common import InfoExtractor
from .youtube import YoutubeIE
from ..compat import (
+ compat_etree_fromstring,
compat_urllib_parse_unquote,
compat_urllib_request,
compat_urlparse,
@@ -21,7 +22,6 @@ from ..utils import (
HEADRequest,
is_html,
orderedSet,
- parse_xml,
smuggle_url,
unescapeHTML,
unified_strdate,
@@ -1238,7 +1238,7 @@ class GenericIE(InfoExtractor):
# Is it an RSS feed, a SMIL file or a XSPF playlist?
try:
- doc = parse_xml(webpage)
+ doc = compat_etree_fromstring(webpage.encode('utf-8'))
if doc.tag == 'rss':
return self._extract_rss(url, video_id, doc)
elif re.match(r'^(?:{[^}]+})?smil$', doc.tag):
diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py
index c17094f81..4c0de354f 100644
--- a/youtube_dl/extractor/vevo.py
+++ b/youtube_dl/extractor/vevo.py
@@ -1,10 +1,10 @@
from __future__ import unicode_literals
import re
-import xml.etree.ElementTree
from .common import InfoExtractor
from ..compat import (
+ compat_etree_fromstring,
compat_urllib_request,
)
from ..utils import (
@@ -97,7 +97,7 @@ class VevoIE(InfoExtractor):
if last_version['version'] == -1:
raise ExtractorError('Unable to extract last version of the video')
- renditions = xml.etree.ElementTree.fromstring(last_version['data'])
+ renditions = compat_etree_fromstring(last_version['data'])
formats = []
# Already sorted from worst to best quality
for rend in renditions.findall('rendition'):
@@ -114,7 +114,7 @@ class VevoIE(InfoExtractor):
def _formats_from_smil(self, smil_xml):
formats = []
- smil_doc = xml.etree.ElementTree.fromstring(smil_xml.encode('utf-8'))
+ smil_doc = compat_etree_fromstring(smil_xml.encode('utf-8'))
els = smil_doc.findall('.//{http://www.w3.org/2001/SMIL20/Language}video')
for el in els:
src = el.attrib['src']
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 764a89cca..efd5f4ae1 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -36,6 +36,7 @@ import zlib
from .compat import (
compat_basestring,
compat_chr,
+ compat_etree_fromstring,
compat_html_entities,
compat_http_client,
compat_kwargs,
@@ -1665,29 +1666,6 @@ def encode_dict(d, encoding='utf-8'):
return dict((k.encode(encoding), v.encode(encoding)) for k, v in d.items())
-try:
- etree_iter = xml.etree.ElementTree.Element.iter
-except AttributeError: # Python <=2.6
- etree_iter = lambda n: n.findall('.//*')
-
-
-def parse_xml(s):
- class TreeBuilder(xml.etree.ElementTree.TreeBuilder):
- def doctype(self, name, pubid, system):
- pass # Ignore doctypes
-
- parser = xml.etree.ElementTree.XMLParser(target=TreeBuilder())
- kwargs = {'parser': parser} if sys.version_info >= (2, 7) else {}
- tree = xml.etree.ElementTree.XML(s.encode('utf-8'), **kwargs)
- # Fix up XML parser in Python 2.x
- if sys.version_info < (3, 0):
- for n in etree_iter(tree):
- if n.text is not None:
- if not isinstance(n.text, compat_str):
- n.text = n.text.decode('utf-8')
- return tree
-
-
US_RATINGS = {
'G': 0,
'PG': 10,
@@ -1988,7 +1966,7 @@ def dfxp2srt(dfxp_data):
return out
- dfxp = xml.etree.ElementTree.fromstring(dfxp_data.encode('utf-8'))
+ dfxp = compat_etree_fromstring(dfxp_data.encode('utf-8'))
out = []
paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall(_x('.//ttaf1:p')) or dfxp.findall('.//p')