release 2014.08.27

Don't rely on the XML being broken (if they fix it, our code wouldn't work anymore).
Instead, use the transform function we already have :)

This partially reverts commit c7bee2a725.

test/helper.py

@@ -102,7 +102,10 @@ def expect_info_dict(self, expected_dict, got_dict):
             match_rex = re.compile(match_str)
 
             self.assertTrue(
-                isinstance(got, compat_str) and match_rex.match(got),
+                isinstance(got, compat_str),
+                'Expected a %r object, but got %r' % (compat_str, type(got)))
+            self.assertTrue(
+                match_rex.match(got),
                 u'field %s (value: %r) should match %r' % (info_field, got, match_str))
         elif isinstance(expected, type):
             got = got_dict.get(info_field)

test/test_download.py

@@ -7,6 +7,7 @@ import unittest
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import (
+    assertGreaterEqual,
     get_params,
     gettestcases,
     expect_info_dict,
@@ -136,12 +137,18 @@ def generator(test_case):
                 self.assertEqual(res_dict['_type'], 'playlist')
                 expect_info_dict(self, test_case.get('info_dict', {}), res_dict)
             if 'playlist_mincount' in test_case:
-                self.assertGreaterEqual(
+                assertGreaterEqual(
+                    self,
                     len(res_dict['entries']),
                     test_case['playlist_mincount'],
                     'Expected at least %d in playlist %s, but got only %d' % (
                         test_case['playlist_mincount'], test_case['url'],
                         len(res_dict['entries'])))
+            if 'playlist_count' in test_case:
+                self.assertEqual(
+                    len(res_dict['entries']),
+                    test_case['playlist_count'],
+                    'Expected at %d in playlist %s, but got %d.')
 
             for tc in test_cases:
                 tc_filename = get_tc_filename(tc)

test/test_playlists.py

@@ -310,24 +310,6 @@ class TestPlaylists(unittest.TestCase):
         self.assertEqual(result['title'], 'Always/Never: A Little-Seen Movie About Nuclear Command and Control : The New Yorker')
         self.assertEqual(len(result['entries']), 3)
 
-    def test_GoogleSearch(self):
-        dl = FakeYDL()
-        ie = GoogleSearchIE(dl)
-        result = ie.extract('gvsearch15:python language')
-        self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], 'python language')
-        self.assertEqual(result['title'], 'python language')
-        self.assertEqual(len(result['entries']), 15)
-
-    def test_generic_rss_feed(self):
-        dl = FakeYDL()
-        ie = GenericIE(dl)
-        result = ie.extract('http://phihag.de/2014/youtube-dl/rss.xml')
-        self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], 'http://phihag.de/2014/youtube-dl/rss.xml')
-        self.assertEqual(result['title'], 'Zero Punctuation')
-        self.assertTrue(len(result['entries']) > 10)
-
     def test_ted_playlist(self):
         dl = FakeYDL()
         ie = TEDIE(dl)

youtube_dl/extractor/__init__.py

@@ -209,6 +209,7 @@ from .mtv import (
     MTVIggyIE,
 )
 from .musicplayon import MusicPlayOnIE
+from .musicvault import MusicVaultIE
 from .muzu import MuzuTVIE
 from .myspace import MySpaceIE
 from .myspass import MySpassIE
@@ -315,6 +316,7 @@ from .spankwire import SpankwireIE
 from .spiegel import SpiegelIE
 from .spiegeltv import SpiegeltvIE
 from .spike import SpikeIE
+from .sportdeutschland import SportDeutschlandIE
 from .stanfordoc import StanfordOpenClassroomIE
 from .steam import SteamIE
 from .streamcloud import StreamcloudIE

youtube_dl/extractor/bliptv.py

@@ -15,7 +15,7 @@ from ..utils import (
 
 
 class BlipTVIE(SubtitlesInfoExtractor):
-    _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+_TESTS]+)))'
+    _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+_]+)))'
 
     _TESTS = [
         {

youtube_dl/extractor/common.py

@@ -114,7 +114,7 @@ class InfoExtractor(object):
     upload_date:    Video upload date (YYYYMMDD).
                     If not explicitly set, calculated from timestamp.
     uploader_id:    Nickname or id of the video uploader.
-    location:       Physical location of the video.
+    location:       Physical location where the video was filmed.
     subtitles:      The subtitle file contents as a dictionary in the format
                     {language: subtitles}.
     duration:       Length of the video in seconds, as an integer.
@@ -636,6 +636,55 @@ class InfoExtractor(object):
 
         return formats
 
+    def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None):
+        formats = [{
+            'format_id': 'm3u8-meta',
+            'url': m3u8_url,
+            'ext': ext,
+            'protocol': 'm3u8',
+            'preference': -1,
+            'resolution': 'multiple',
+            'format_note': 'Quality selection URL',
+        }]
+
+        m3u8_doc = self._download_webpage(m3u8_url, video_id)
+        last_info = None
+        kv_rex = re.compile(
+            r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
+        for line in m3u8_doc.splitlines():
+            if line.startswith('#EXT-X-STREAM-INF:'):
+                last_info = {}
+                for m in kv_rex.finditer(line):
+                    v = m.group('val')
+                    if v.startswith('"'):
+                        v = v[1:-1]
+                    last_info[m.group('key')] = v
+            elif line.startswith('#') or not line.strip():
+                continue
+            else:
+                tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000)
+
+                f = {
+                    'format_id': 'm3u8-%d' % (tbr if tbr else len(formats)),
+                    'url': line.strip(),
+                    'tbr': tbr,
+                    'ext': ext,
+                }
+                codecs = last_info.get('CODECS')
+                if codecs:
+                    video, audio = codecs.split(',')
+                    f['vcodec'] = video.partition('.')[0]
+                    f['acodec'] = audio.partition('.')[0]
+                resolution = last_info.get('RESOLUTION')
+                if resolution:
+                    width_str, height_str = resolution.split('x')
+                    f['width'] = int(width_str)
+                    f['height'] = int(height_str)
+                formats.append(f)
+                last_info = {}
+        self._sort_formats(formats)
+        return formats
+
 
 class SearchInfoExtractor(InfoExtractor):
     """

youtube_dl/extractor/eighttracks.py

@@ -1,10 +1,13 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
 import json
 import random
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
+    compat_str,
 )
 
 
@@ -12,86 +15,98 @@ class EightTracksIE(InfoExtractor):
     IE_NAME = '8tracks'
     _VALID_URL = r'https?://8tracks\.com/(?P<user>[^/]+)/(?P<id>[^/#]+)(?:#.*)?$'
     _TEST = {
-        u"name": u"EightTracks",
-        u"url": u"http://8tracks.com/ytdl/youtube-dl-test-tracks-a",
-        u"playlist": [
+        "name": "EightTracks",
+        "url": "http://8tracks.com/ytdl/youtube-dl-test-tracks-a",
+        "info_dict": {
+            'id': '1336550',
+            'display_id': 'youtube-dl-test-tracks-a',
+            "description": "test chars:  \"'/\\ä↭",
+            "title": "youtube-dl test tracks \"'/\\ä↭<>",
+        },
+        "playlist": [
             {
-                u"file": u"11885610.m4a",
-                u"md5": u"96ce57f24389fc8734ce47f4c1abcc55",
-                u"info_dict": {
-                    u"title": u"youtue-dl project<>\"' - youtube-dl test track 1 \"'/\\\u00e4\u21ad",
-                    u"uploader_id": u"ytdl"
+                "md5": "96ce57f24389fc8734ce47f4c1abcc55",
+                "info_dict": {
+                    "id": "11885610",
+                    "ext": "m4a",
+                    "title": "youtue-dl project<>\"' - youtube-dl test track 1 \"'/\\\u00e4\u21ad",
+                    "uploader_id": "ytdl"
                 }
             },
             {
-                u"file": u"11885608.m4a",
-                u"md5": u"4ab26f05c1f7291ea460a3920be8021f",
-                u"info_dict": {
-                    u"title": u"youtube-dl project - youtube-dl test track 2 \"'/\\\u00e4\u21ad",
-                    u"uploader_id": u"ytdl"
+                "md5": "4ab26f05c1f7291ea460a3920be8021f",
+                "info_dict": {
+                    "id": "11885608",
+                    "ext": "m4a",
+                    "title": "youtube-dl project - youtube-dl test track 2 \"'/\\\u00e4\u21ad",
+                    "uploader_id": "ytdl"
                 }
             },
             {
-                u"file": u"11885679.m4a",
-                u"md5": u"d30b5b5f74217410f4689605c35d1fd7",
-                u"info_dict": {
-                    u"title": u"youtube-dl project as well - youtube-dl test track 3 \"'/\\\u00e4\u21ad",
-                    u"uploader_id": u"ytdl"
+                "md5": "d30b5b5f74217410f4689605c35d1fd7",
+                "info_dict": {
+                    "id": "11885679",
+                    "ext": "m4a",
+                    "title": "youtube-dl project as well - youtube-dl test track 3 \"'/\\\u00e4\u21ad",
+                    "uploader_id": "ytdl"
                 }
             },
             {
-                u"file": u"11885680.m4a",
-                u"md5": u"4eb0a669317cd725f6bbd336a29f923a",
-                u"info_dict": {
-                    u"title": u"youtube-dl project as well - youtube-dl test track 4 \"'/\\\u00e4\u21ad",
-                    u"uploader_id": u"ytdl"
+                "md5": "4eb0a669317cd725f6bbd336a29f923a",
+                "info_dict": {
+                    "id": "11885680",
+                    "ext": "m4a",
+                    "title": "youtube-dl project as well - youtube-dl test track 4 \"'/\\\u00e4\u21ad",
+                    "uploader_id": "ytdl"
                 }
             },
             {
-                u"file": u"11885682.m4a",
-                u"md5": u"1893e872e263a2705558d1d319ad19e8",
-                u"info_dict": {
-                    u"title": u"PH - youtube-dl test track 5 \"'/\\\u00e4\u21ad",
-                    u"uploader_id": u"ytdl"
+                "md5": "1893e872e263a2705558d1d319ad19e8",
+                "info_dict": {
+                    "id": "11885682",
+                    "ext": "m4a",
+                    "title": "PH - youtube-dl test track 5 \"'/\\\u00e4\u21ad",
+                    "uploader_id": "ytdl"
                 }
             },
             {
-                u"file": u"11885683.m4a",
-                u"md5": u"b673c46f47a216ab1741ae8836af5899",
-                u"info_dict": {
-                    u"title": u"PH - youtube-dl test track 6 \"'/\\\u00e4\u21ad",
-                    u"uploader_id": u"ytdl"
+                "md5": "b673c46f47a216ab1741ae8836af5899",
+                "info_dict": {
+                    "id": "11885683",
+                    "ext": "m4a",
+                    "title": "PH - youtube-dl test track 6 \"'/\\\u00e4\u21ad",
+                    "uploader_id": "ytdl"
                 }
             },
             {
-                u"file": u"11885684.m4a",
-                u"md5": u"1d74534e95df54986da7f5abf7d842b7",
-                u"info_dict": {
-                    u"title": u"phihag - youtube-dl test track 7 \"'/\\\u00e4\u21ad",
-                    u"uploader_id": u"ytdl"
+                "md5": "1d74534e95df54986da7f5abf7d842b7",
+                "info_dict": {
+                    "id": "11885684",
+                    "ext": "m4a",
+                    "title": "phihag - youtube-dl test track 7 \"'/\\\u00e4\u21ad",
+                    "uploader_id": "ytdl"
                 }
             },
             {
-                u"file": u"11885685.m4a",
-                u"md5": u"f081f47af8f6ae782ed131d38b9cd1c0",
-                u"info_dict": {
-                    u"title": u"phihag - youtube-dl test track 8 \"'/\\\u00e4\u21ad",
-                    u"uploader_id": u"ytdl"
+                "md5": "f081f47af8f6ae782ed131d38b9cd1c0",
+                "info_dict": {
+                    "id": "11885685",
+                    "ext": "m4a",
+                    "title": "phihag - youtube-dl test track 8 \"'/\\\u00e4\u21ad",
+                    "uploader_id": "ytdl"
                 }
             }
         ]
     }
 
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
         playlist_id = mobj.group('id')
 
         webpage = self._download_webpage(url, playlist_id)
 
-        json_like = self._search_regex(r"PAGE.mix = (.*?);\n", webpage, u'trax information', flags=re.DOTALL)
+        json_like = self._search_regex(
+            r"(?s)PAGE.mix = (.*?);\n", webpage, 'trax information')
         data = json.loads(json_like)
 
         session = str(random.randint(0, 1000000000))
@@ -99,21 +114,30 @@ class EightTracksIE(InfoExtractor):
         track_count = data['tracks_count']
         first_url = 'http://8tracks.com/sets/%s/play?player=sm&mix_id=%s&format=jsonh' % (session, mix_id)
         next_url = first_url
-        res = []
+        entries = []
         for i in range(track_count):
-            api_json = self._download_webpage(next_url, playlist_id,
-                note=u'Downloading song information %s/%s' % (str(i+1), track_count),
-                errnote=u'Failed to download song information')
+            api_json = self._download_webpage(
+                next_url, playlist_id,
+                note='Downloading song information %d/%d' % (i + 1, track_count),
+                errnote='Failed to download song information')
             api_data = json.loads(api_json)
-            track_data = api_data[u'set']['track']
+            track_data = api_data['set']['track']
             info = {
-                'id': track_data['id'],
+                'id': compat_str(track_data['id']),
                 'url': track_data['track_file_stream_url'],
                 'title': track_data['performer'] + u' - ' + track_data['name'],
                 'raw_title': track_data['name'],
                 'uploader_id': data['user']['login'],
                 'ext': 'm4a',
             }
-            res.append(info)
-            next_url = 'http://8tracks.com/sets/%s/next?player=sm&mix_id=%s&format=jsonh&track_id=%s' % (session, mix_id, track_data['id'])
-        return res
+            entries.append(info)
+            next_url = 'http://8tracks.com/sets/%s/next?player=sm&mix_id=%s&format=jsonh&track_id=%s' % (
+                session, mix_id, track_data['id'])
+        return {
+            '_type': 'playlist',
+            'entries': entries,
+            'id': compat_str(mix_id),
+            'display_id': playlist_id,
+            'title': data.get('name'),
+            'description': data.get('description'),
+        }

youtube_dl/extractor/empflix.py

@@ -3,6 +3,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import fix_xml_ampersands
 
 
 class EmpflixIE(InfoExtractor):
@@ -36,7 +37,8 @@ class EmpflixIE(InfoExtractor):
             webpage, 'flashvars.config')
 
         cfg_xml = self._download_xml(
-            cfg_url, video_id, note='Downloading metadata')
+            cfg_url, video_id, note='Downloading metadata',
+            transform_source=fix_xml_ampersands)
 
         formats = [
             {
@@ -44,11 +46,13 @@ class EmpflixIE(InfoExtractor):
                 'format_id': item.find('res').text,
             } for item in cfg_xml.findall('./quality/item')
         ]
+        thumbnail = cfg_xml.find('./startThumb').text
 
         return {
             'id': video_id,
             'title': video_title,
             'description': video_description,
+            'thumbnail': thumbnail,
             'formats': formats,
             'age_limit': age_limit,
         }

youtube_dl/extractor/generic.py

@@ -341,6 +341,16 @@ class GenericIE(InfoExtractor):
                 'uploader': 'www.handjobhub.com',
                 'title': 'Busty Blonde Siri Tit Fuck While Wank at Handjob Hub',
             }
+        },
+        # RSS feed
+        {
+            'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
+            'info_dict': {
+                'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
+                'title': 'Zero Punctuation',
+                'description': 're:'
+            },
+            'playlist_mincount': 11,
         }
     ]
 
@@ -809,7 +819,6 @@ class GenericIE(InfoExtractor):
                     \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
                         ["']?url["']?\s*:\s*["']([^"']+)["']
             ''', webpage)
-            assert found
         if not found:
             # Try to find twitter cards info
             found = re.findall(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)

youtube_dl/extractor/googlesearch.py

@@ -14,6 +14,14 @@ class GoogleSearchIE(SearchInfoExtractor):
     _MAX_RESULTS = 1000
     IE_NAME = 'video.google:search'
     _SEARCH_KEY = 'gvsearch'
+    _TEST = {
+        'url': 'gvsearch15:python language',
+        'info_dict': {
+            'id': 'python language',
+            'title': 'python language',
+        },
+        'playlist_count': 15,
+    }
 
     def _get_n_results(self, query, n):
         """Get a specified number of results for a query"""

youtube_dl/extractor/ign.py

@@ -18,6 +18,7 @@ class IGNIE(InfoExtractor):
     _DESCRIPTION_RE = [
         r'<span class="page-object-description">(.+?)</span>',
         r'id="my_show_video">.*?<p>(.*?)</p>',
+        r'<meta name="description" content="(.*?)"',
     ]
 
     _TESTS = [
@@ -55,6 +56,17 @@ class IGNIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        {
+            'url': 'http://www.ign.com/articles/2014/08/15/rewind-theater-wild-trailer-gamescom-2014?watch',
+            'md5': '4e9a0bda1e5eebd31ddcf86ec0b9b3c7',
+            'info_dict': {
+                'id': '078fdd005f6d3c02f63d795faa1b984f',
+                'ext': 'mp4',
+                'title': 'Rewind Theater - Wild Trailer Gamescom 2014',
+                'description': 'Giant skeletons, bloody hunts, and captivating'
+                    ' natural beauty take our breath away.',
+            },
+        },
     ]
 
     def _find_video_id(self, webpage):
@@ -62,6 +74,7 @@ class IGNIE(InfoExtractor):
             r'data-video-id="(.+?)"',
             r'<object id="vid_(.+?)"',
             r'<meta name="og:image" content=".*/(.+?)-(.+?)/.+.jpg"',
+            r'class="hero-poster[^"]*?"[^>]*id="(.+?)"',
         ]
         return self._search_regex(res_id, webpage, 'video id')
 
@@ -70,10 +83,7 @@ class IGNIE(InfoExtractor):
         name_or_id = mobj.group('name_or_id')
         page_type = mobj.group('type')
         webpage = self._download_webpage(url, name_or_id)
-        if page_type == 'articles':
-            video_url = self._search_regex(r'var videoUrl = "(.+?)"', webpage, 'video url')
-            return self.url_result(video_url, ie='IGN')
-        elif page_type != 'video':
+        if page_type != 'video':
             multiple_urls = re.findall(
                 '<param name="flashvars" value="[^"]*?url=(https?://www\.ign\.com/videos/.*?)["&]',
                 webpage)

youtube_dl/extractor/mixcloud.py

@@ -6,6 +6,7 @@ from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
     ExtractorError,
+    HEADRequest,
     int_or_none,
     parse_iso8601,
 )
@@ -38,7 +39,7 @@ class MixcloudIE(InfoExtractor):
             try:
                 # We only want to know if the request succeed
                 # don't download the whole file
-                self._request_webpage(url, None, False)
+                self._request_webpage(HEADRequest(url), None, False)
                 return url
             except ExtractorError:
                 url = None

youtube_dl/extractor/mofosex.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import os
 import re
 
@@ -8,15 +10,17 @@ from ..utils import (
     compat_urllib_parse,
 )
 
+
 class MofosexIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>mofosex\.com/videos/(?P<videoid>[0-9]+)/.*?\.html)'
+    _VALID_URL = r'^https?://(?:www\.)?(?P<url>mofosex\.com/videos/(?P<videoid>[0-9]+)/.*?\.html)'
     _TEST = {
-        u'url': u'http://www.mofosex.com/videos/5018/japanese-teen-music-video.html',
-        u'file': u'5018.mp4',
-        u'md5': u'1b2eb47ac33cc75d4a80e3026b613c5a',
-        u'info_dict': {
-            u"title": u"Japanese Teen Music Video",
-            u"age_limit": 18,
+        'url': 'http://www.mofosex.com/videos/5018/japanese-teen-music-video.html',
+        'md5': '1b2eb47ac33cc75d4a80e3026b613c5a',
+        'info_dict': {
+            'id': '5018',
+            'ext': 'mp4',
+            'title': 'Japanese Teen Music Video',
+            'age_limit': 18,
         }
     }
 
@@ -29,8 +33,8 @@ class MofosexIE(InfoExtractor):
         req.add_header('Cookie', 'age_verified=1')
         webpage = self._download_webpage(req, video_id)
 
-        video_title = self._html_search_regex(r'<h1>(.+?)<', webpage, u'title')
-        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'flashvars.video_url = \'([^\']+)', webpage, u'video_url'))
+        video_title = self._html_search_regex(r'<h1>(.+?)<', webpage, 'title')
+        video_url = compat_urllib_parse.unquote(self._html_search_regex(r'flashvars.video_url = \'([^\']+)', webpage, 'video_url'))
         path = compat_urllib_parse_urlparse(video_url).path
         extension = os.path.splitext(path)[1][1:]
         format = path.split('/')[5].split('_')[:2]

youtube_dl/extractor/musicvault.py

@@ -0,0 +1,78 @@
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    strip_jsonp,
+    parse_duration,
+    unified_strdate,
+)
+
+
+class MusicVaultIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.musicvault\.com/(?P<uploader_id>[^/?#]*)/video/(?P<display_id>[^/?#]*)_(?P<id>[0-9]+)\.html'
+    _TEST = {
+        'url': 'http://www.musicvault.com/the-allman-brothers-band/video/straight-from-the-heart_1010863.html',
+        'md5': '2cdbb3ae75f7fb3519821507d2fb3c15',
+        'info_dict': {
+            'id': '1010863',
+            'ext': 'mp4',
+            'uploader_id': 'the-allman-brothers-band',
+            'title': 'Straight from the Heart',
+            'duration': 244,
+            'uploader': 'The Allman Brothers Band',
+            'thumbnail': 're:^https?://.*/thumbnail/.*',
+            'upload_date': '19811216',
+            'location': 'Capitol Theatre (Passaic, NJ)',
+            'description': 'Listen to The Allman Brothers Band perform Straight from the Heart at Capitol Theatre (Passaic, NJ) on Dec 16, 1981',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('display_id')
+        webpage = self._download_webpage(url, display_id)
+
+        thumbnail = self._search_regex(
+            r'<meta itemprop="thumbnail" content="([^"]+)"',
+            webpage, 'thumbnail', fatal=False)
+
+        data_div = self._search_regex(
+            r'(?s)<div class="data">(.*?)</div>', webpage, 'data fields')
+        uploader = self._html_search_regex(
+            r'<h1.*?>(.*?)</h1>', data_div, 'uploader', fatal=False)
+        title = self._html_search_regex(
+            r'<h2.*?>(.*?)</h2>', data_div, 'title')
+        upload_date = unified_strdate(self._html_search_regex(
+            r'<h3.*?>(.*?)</h3>', data_div, 'uploader', fatal=False))
+        location = self._html_search_regex(
+            r'<h4.*?>(.*?)</h4>', data_div, 'location', fatal=False)
+
+        duration = parse_duration(self._html_search_meta('duration', webpage))
+
+        VIDEO_URL_TEMPLATE = 'http://cdnapi.kaltura.com/p/%(uid)s/sp/%(wid)s/playManifest/entryId/%(entry_id)s/format/url/protocol/http'
+        kaltura_id = self._search_regex(
+            r'<div id="video-detail-player" data-kaltura-id="([^"]+)"',
+            webpage, 'kaltura ID')
+        video_url = VIDEO_URL_TEMPLATE % {
+            'entry_id': kaltura_id,
+            'wid': self._search_regex(r'/wid/_([0-9]+)/', webpage, 'wid'),
+            'uid': self._search_regex(r'uiconf_id/([0-9]+)/', webpage, 'uid'),
+        }
+
+        return {
+            'id': mobj.group('id'),
+            'url': video_url,
+            'ext': 'mp4',
+            'display_id': display_id,
+            'uploader_id': mobj.group('uploader_id'),
+            'thumbnail': thumbnail,
+            'description': self._html_search_meta('description', webpage),
+            'upload_date': upload_date,
+            'location': location,
+            'title': title,
+            'uploader': uploader,
+            'duration': duration,
+        }

youtube_dl/extractor/sockshare.py

@@ -61,7 +61,10 @@ class SockshareIE(InfoExtractor):
             r'<a href="([^"]*)".+class="download_file_link"',
             webpage, 'file url')
         video_url = "http://www.sockshare.com" + video_url
-        title = self._html_search_regex(r'<h1>(.+)<strong>', webpage, 'title')
+        title = self._html_search_regex((
+            r'<h1>(.+)<strong>',
+            r'var name = "([^"]+)";'),
+            webpage, 'title', default=None)
         thumbnail = self._html_search_regex(
             r'<img\s+src="([^"]*)".+?name="bg"',
             webpage, 'thumbnail')

youtube_dl/extractor/sportdeutschland.py

@@ -0,0 +1,77 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+    parse_iso8601,
+)
+
+
+class SportDeutschlandIE(InfoExtractor):
+    _VALID_URL = r'https?://sportdeutschland\.tv/(?P<sport>[^/?#]+)/(?P<id>[^?#/]+)(?:$|[?#])'
+    _TEST = {
+        'url': 'http://sportdeutschland.tv/badminton/live-li-ning-badminton-weltmeisterschaft-2014-kopenhagen',
+        'info_dict': {
+            'id': 'live-li-ning-badminton-weltmeisterschaft-2014-kopenhagen',
+            'ext': 'mp4',
+            'title': 'LIVE: Li-Ning Badminton Weltmeisterschaft 2014 Kopenhagen',
+            'categories': ['Badminton'],
+            'view_count': int,
+            'thumbnail': 're:^https?://.*\.jpg',
+            'description': 're:^Die Badminton-WM 2014 aus Kopenhagen LIVE',
+            'timestamp': 1409043600,
+            'upload_date': '20140826',
+        },
+        'params': {
+            'skip_download': 'Live stream',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        sport_id = mobj.group('sport')
+
+        api_url = 'http://splink.tv/api/permalinks/%s/%s' % (
+            sport_id, video_id)
+        req = compat_urllib_request.Request(api_url, headers={
+            'Accept': 'application/vnd.vidibus.v2.html+json',
+            'Referer': url,
+        })
+        data = self._download_json(req, video_id)
+
+        categories = list(data.get('section', {}).get('tags', {}).values())
+        asset = data['asset']
+
+        smil_url = asset['video']
+        m3u8_url = smil_url.replace('.smil', '.m3u8')
+        formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4')
+
+        smil_doc = self._download_xml(
+            smil_url, video_id, note='Downloading SMIL metadata')
+        base_url = smil_doc.find('./head/meta').attrib['base']
+        formats.extend([{
+            'format_id': 'rmtp',
+            'url': base_url,
+            'play_path': n.attrib['src'],
+            'ext': 'flv',
+            'preference': -100,
+            'format_note': 'Seems to fail at example stream',
+        } for n in smil_doc.findall('./body/video')])
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': asset['title'],
+            'thumbnail': asset.get('image'),
+            'description': asset.get('teaser'),
+            'categories': categories,
+            'view_count': asset.get('views'),
+            'rtmp_live': asset['live'],
+            'timestamp': parse_iso8601(asset.get('date')),
+        }
+

youtube_dl/utils.py

@@ -766,10 +766,9 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
         return ret
 
     def http_request(self, req):
-        for h,v in std_headers.items():
-            if h in req.headers:
-                del req.headers[h]
-            req.add_header(h, v)
+        for h, v in std_headers.items():
+            if h not in req.headers:
+                req.add_header(h, v)
         if 'Youtubedl-no-compression' in req.headers:
             if 'Accept-encoding' in req.headers:
                 del req.headers['Accept-encoding']
@@ -1458,6 +1457,12 @@ def urlencode_postdata(*args, **kargs):
     return compat_urllib_parse.urlencode(*args, **kargs).encode('ascii')
 
 
+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):
@@ -1465,7 +1470,14 @@ def parse_xml(s):
 
     parser = xml.etree.ElementTree.XMLParser(target=TreeBuilder())
     kwargs = {'parser': parser} if sys.version_info >= (2, 7) else {}
-    return xml.etree.ElementTree.XML(s.encode('utf-8'), **kwargs)
+    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
 
 
 if sys.version_info < (3, 0) and sys.platform == 'win32':

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.08.25.2'
+__version__ = '2014.08.27'