release 2014.04.21.6

The URL http://www.bilibili.tv/video/av636603/index_2.html does not work yet.

test/helper.py

@@ -85,7 +85,7 @@ def gettestcases(include_onlymatching=False):
         else:
             tests = getattr(ie, '_TESTS', [])
         for t in tests:
-            if not include_onlymatching and getattr(t, 'only_matching', False):
+            if not include_onlymatching and t.get('only_matching', False):
                 continue
             t['name'] = type(ie).__name__[:-len('IE')]
             yield t

test/test_subtitles.py

@@ -181,7 +181,7 @@ class TestTedSubtitles(BaseTestSubtitles):
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(len(subtitles.keys()), 28)
+        self.assertTrue(len(subtitles.keys()) >= 28)
 
     def test_list_subtitles(self):
         self.DL.expect_warning(u'Automatic Captions not supported by this server')

youtube_dl/extractor/__init__.py

@@ -20,6 +20,7 @@ from .auengine import AUEngineIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
 from .bbccouk import BBCCoUkIE
+from .bilibili import BiliBiliIE
 from .blinkx import BlinkxIE
 from .bliptv import BlipTVIE, BlipTVUserIE
 from .bloomberg import BloombergIE
@@ -40,6 +41,7 @@ from .cinemassacre import CinemassacreIE
 from .clipfish import ClipfishIE
 from .cliphunter import CliphunterIE
 from .clipsyndicate import ClipsyndicateIE
+from .clubic import ClubicIE
 from .cmt import CMTIE
 from .cnet import CNETIE
 from .cnn import (
@@ -295,6 +297,7 @@ from .vine import VineIE
 from .viki import VikiIE
 from .vk import VKIE
 from .vube import VubeIE
+from .vuclip import VuClipIE
 from .washingtonpost import WashingtonPostIE
 from .wat import WatIE
 from .wdr import (

youtube_dl/extractor/arte.py

@@ -74,7 +74,8 @@ class ArteTVPlus7IE(InfoExtractor):
         return self._extract_from_webpage(webpage, video_id, lang)
 
     def _extract_from_webpage(self, webpage, video_id, lang):
-        json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
+        json_url = self._html_search_regex(
+            r'arte_vp_url="(.*?)"', webpage, 'json vp url')
         return self._extract_from_json_url(json_url, video_id, lang)
 
     def _extract_from_json_url(self, json_url, video_id, lang):

youtube_dl/extractor/bilibili.py

@@ -0,0 +1,106 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_parse_qs,
+    ExtractorError,
+    int_or_none,
+    unified_strdate,
+)
+
+
+class BiliBiliIE(InfoExtractor):
+    _VALID_URL = r'http://www\.bilibili\.tv/video/av(?P<id>[0-9]+)/'
+
+    _TEST = {
+        'url': 'http://www.bilibili.tv/video/av1074402/',
+        'md5': '2c301e4dab317596e837c3e7633e7d86',
+        'info_dict': {
+            'id': '1074402',
+            'ext': 'flv',
+            'title': '【金坷垃】金泡沫',
+            'duration': 308,
+            'upload_date': '20140420',
+            'thumbnail': 're:^https?://.+\.jpg',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        video_code = self._search_regex(
+            r'(?s)<div itemprop="video".*?>(.*?)</div>', webpage, 'video code')
+
+        title = self._html_search_meta(
+            'media:title', video_code, 'title', fatal=True)
+        duration_str = self._html_search_meta(
+            'duration', video_code, 'duration')
+        if duration_str is None:
+            duration = None
+        else:
+            duration_mobj = re.match(
+                r'^T(?:(?P<hours>[0-9]+)H)?(?P<minutes>[0-9]+)M(?P<seconds>[0-9]+)S$',
+                duration_str)
+            duration = (
+                int_or_none(duration_mobj.group('hours'), default=0) * 3600 +
+                int(duration_mobj.group('minutes')) * 60 +
+                int(duration_mobj.group('seconds')))
+        upload_date = unified_strdate(self._html_search_meta(
+            'uploadDate', video_code, fatal=False))
+        thumbnail = self._html_search_meta(
+            'thumbnailUrl', video_code, 'thumbnail', fatal=False)
+
+        player_params = compat_parse_qs(self._html_search_regex(
+            r'<iframe .*?class="player" src="https://secure.bilibili.tv/secure,([^"]+)"',
+            webpage, 'player params'))
+
+        if 'cid' in player_params:
+            cid = player_params['cid'][0]
+
+            lq_doc = self._download_xml(
+                'http://interface.bilibili.cn/v_cdn_play?cid=%s' % cid,
+                video_id,
+                note='Downloading LQ video info'
+            )
+            lq_durl = lq_doc.find('.//durl')
+            formats = [{
+                'format_id': 'lq',
+                'quality': 1,
+                'url': lq_durl.find('./url').text,
+                'filesize': int_or_none(
+                    lq_durl.find('./size'), get_attr='text'),
+            }]
+
+            hq_doc = self._download_xml(
+                'http://interface.bilibili.cn/playurl?cid=%s' % cid,
+                video_id,
+                note='Downloading HQ video info',
+                fatal=False,
+            )
+            if hq_doc is not False:
+                hq_durl = hq_doc.find('.//durl')
+                formats.append({
+                    'format_id': 'hq',
+                    'quality': 2,
+                    'ext': 'flv',
+                    'url': hq_durl.find('./url').text,
+                    'filesize': int_or_none(
+                        hq_durl.find('./size'), get_attr='text'),
+                })
+        else:
+            raise ExtractorError('Unsupported player parameters: %r' % (player_params,))
+
+        self._sort_formats(formats)
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'duration': duration,
+            'upload_date': upload_date,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/clubic.py

@@ -0,0 +1,58 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    qualities,
+)
+
+
+class ClubicIE(InfoExtractor):
+    _VALID_URL = r'http://(?:www\.)?clubic\.com/video/[^/]+/video.*-(?P<id>[0-9]+)\.html'
+
+    _TEST = {
+        'url': 'http://www.clubic.com/video/clubic-week/video-clubic-week-2-0-le-fbi-se-lance-dans-la-photo-d-identite-448474.html',
+        'md5': '1592b694ba586036efac1776b0b43cd3',
+        'info_dict': {
+            'id': '448474',
+            'ext': 'mp4',
+            'title': 'Clubic Week 2.0 : le FBI se lance dans la photo d\u0092identité',
+            'description': 're:Gueule de bois chez Nokia. Le constructeur a indiqué cette.*',
+            'thumbnail': 're:^http://img\.clubic\.com/.*\.jpg$',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        player_url = 'http://player.m6web.fr/v1/player/clubic/%s.html' % video_id
+        player_page = self._download_webpage(player_url, video_id)
+
+        config_json = self._search_regex(
+            r'(?m)M6\.Player\.config\s*=\s*(\{.+?\});$', player_page,
+            'configuration')
+        config = json.loads(config_json)
+
+        video_info = config['videoInfo']
+        sources = config['sources']
+        quality_order = qualities(['sd', 'hq'])
+
+        formats = [{
+            'format_id': src['streamQuality'],
+            'url': src['src'],
+            'quality': quality_order(src['streamQuality']),
+        } for src in sources]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': video_info['title'],
+            'formats': formats,
+            'description': clean_html(video_info.get('description')),
+            'thumbnail': config.get('poster'),
+        }

youtube_dl/extractor/cnet.py

@@ -33,7 +33,7 @@ class CNETIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
         data_json = self._html_search_regex(
-            r"<div class=\"cnetVideoPlayer\" data-cnet-video-options='([^']+)'",
+            r"<div class=\"cnetVideoPlayer\"\s+.*?data-cnet-video-options='([^']+)'",
             webpage, 'data json')
         data = json.loads(data_json)
         vdata = data['video']

youtube_dl/extractor/common.py

@@ -279,9 +279,12 @@ class InfoExtractor(object):
 
     def _download_xml(self, url_or_request, video_id,
                       note=u'Downloading XML', errnote=u'Unable to download XML',
-                      transform_source=None):
+                      transform_source=None, fatal=True):
         """Return the xml as an xml.etree.ElementTree.Element"""
-        xml_string = self._download_webpage(url_or_request, video_id, note, errnote)
+        xml_string = self._download_webpage(
+            url_or_request, video_id, note, errnote, fatal=fatal)
+        if xml_string is False:
+            return xml_string
         if transform_source:
             xml_string = transform_source(xml_string)
         return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))

youtube_dl/extractor/facebook.py

@@ -76,9 +76,8 @@ class FacebookIE(InfoExtractor):
 
             check_form = {
                 'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
-                'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, 'nh'),
+                'h': self._search_regex(r'name="h" value="(\w*?)"', login_results, 'h'),
                 'name_action_selected': 'dont_save',
-                'submit[Continue]': self._search_regex(r'<button[^>]+value="(.*?)"[^>]+name="submit\[Continue\]"', login_results, 'continue'),
             }
             check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
             check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')

youtube_dl/extractor/generic.py

@@ -531,7 +531,7 @@ class GenericIE(InfoExtractor):
             return OoyalaIE._build_url_result(mobj.group('ec'))
 
         # Look for Aparat videos
-        mobj = re.search(r'<iframe src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
+        mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
         if mobj is not None:
             return self.url_result(mobj.group(1), 'Aparat')
 
@@ -619,7 +619,13 @@ class GenericIE(InfoExtractor):
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:
             # Look for gorilla-vid style embedding
-            mobj = re.search(r'(?s)(?:jw_plugins|JWPlayerOptions).*?file\s*:\s*["\'](.*?)["\']', webpage)
+            mobj = re.search(r'''(?sx)
+                (?:
+                    jw_plugins|
+                    JWPlayerOptions|
+                    jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
+                )
+                .*?file\s*:\s*["\'](.*?)["\']''', webpage)
         if mobj is None:
             # Broaden the search a little bit
             mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)

youtube_dl/extractor/jukebox.py

@@ -14,7 +14,7 @@ class JukeboxIE(InfoExtractor):
     _VALID_URL = r'^http://www\.jukebox?\..+?\/.+[,](?P<video_id>[a-z0-9\-]+)\.html'
     _TEST = {
         'url': 'http://www.jukebox.es/kosheen/videoclip,pride,r303r.html',
-        'md5': '5dc6477e74b1e37042ac5acedd8413e5',
+        'md5': '1574e9b4d6438446d5b7dbcdf2786276',
         'info_dict': {
             'id': 'r303r',
             'ext': 'flv',

youtube_dl/extractor/mdr.py

@@ -3,9 +3,6 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-)
 
 
 class MDRIE(InfoExtractor):

youtube_dl/extractor/rutube.py

@@ -47,7 +47,7 @@ class RutubeIE(InfoExtractor):
         author = video.get('author') or {}
 
         options = self._download_json(
-            'http://rutube.ru/api/play/options/%s/?format=json' %video_id,
+            'http://rutube.ru/api/play/options/%s/?format=json' % video_id,
             video_id, 'Downloading options JSON')
 
         m3u8_url = options['video_balancer'].get('m3u8')

youtube_dl/extractor/ted.py

@@ -97,7 +97,7 @@ class TEDIE(SubtitlesInfoExtractor):
         playlist_info = info['playlist']
 
         playlist_entries = [
-            self.url_result(u'http://www.ted.com/talks/' + talk['slug'], self.ie_key())
+            self.url_result('http://www.ted.com/talks/' + talk['slug'], self.ie_key())
             for talk in info['talks']
         ]
         return self.playlist_result(
@@ -163,7 +163,7 @@ class TEDIE(SubtitlesInfoExtractor):
                 sub_lang_list[l] = url
             return sub_lang_list
         else:
-            self._downloader.report_warning(u'video doesn\'t have subtitles')
+            self._downloader.report_warning('video doesn\'t have subtitles')
             return {}
 
     def _watch_info(self, url, name):
@@ -178,7 +178,10 @@ class TEDIE(SubtitlesInfoExtractor):
         title = self._html_search_regex(
             r"(?s)<h1(?:\s+class='[^']+')?>(.+?)</h1>", webpage, 'title')
         description = self._html_search_regex(
-            r'(?s)<h4 class="[^"]+" id="h3--about-this-talk">.*?</h4>(.*?)</div>',
+            [
+                r'(?s)<h4 class="[^"]+" id="h3--about-this-talk">.*?</h4>(.*?)</div>',
+                r'(?s)<p><strong>About this talk:</strong>\s+(.*?)</p>',
+            ],
             webpage, 'description', fatal=False)
 
         return {

youtube_dl/extractor/vuclip.py

@@ -0,0 +1,66 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse_urlparse,
+    parse_duration,
+    qualities,
+)
+
+
+class VuClipIE(InfoExtractor):
+    _VALID_URL = r'http://(?:m\.)?vuclip\.com/w\?.*?cid=(?P<id>[0-9]+)'
+
+    _TEST = {
+        'url': 'http://m.vuclip.com/w?cid=843902317&fid=63532&z=1007&nvar&frm=index.html&bu=4757321434',
+        'md5': '92ac9d1ccefec4f0bb474661ab144fcf',
+        'info_dict': {
+            'id': '843902317',
+            'ext': '3gp',
+            'title': 'Movie Trailer: Noah',
+            'duration': 139,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        ad_m = re.search(
+            r'''value="No.*?" onClick="location.href='([^"']+)'"''', webpage)
+        if ad_m:
+            urlr = compat_urllib_parse_urlparse(url)
+            adfree_url = urlr.scheme + '://' + urlr.netloc + ad_m.group(1)
+            webpage = self._download_webpage(
+                adfree_url, video_id, note='Download post-ad page')
+
+        links_code = self._search_regex(
+            r'(?s)<div class="social align_c".*?>(.*?)<hr\s*/?>', webpage,
+            'links')
+        title = self._html_search_regex(
+            r'<title>(.*?)-\s*Vuclip</title>', webpage, 'title').strip()
+
+        quality_order = qualities(['Reg', 'Hi'])
+        formats = []
+        for url, q in re.findall(
+                r'<a href="(?P<url>[^"]+)".*?>(?P<q>[^<]+)</a>', links_code):
+            format_id = compat_urllib_parse_urlparse(url).scheme + '-' + q
+            formats.append({
+                'format_id': format_id,
+                'url': url,
+                'quality': quality_order(q),
+            })
+        self._sort_formats(formats)
+
+        duration = parse_duration(self._search_regex(
+            r'\(([0-9:]+)\)</span></h1>', webpage, 'duration', fatal=False))
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': title,
+            'duration': duration,
+        }

youtube_dl/utils.py

@@ -1245,7 +1245,10 @@ class HEADRequest(compat_urllib_request.Request):
         return "HEAD"
 
 
-def int_or_none(v, scale=1, default=None):
+def int_or_none(v, scale=1, default=None, get_attr=None):
+    if get_attr:
+        if v is not None:
+            v = getattr(v, get_attr, None)
     return default if v is None else (int(v) // scale)
 
 
@@ -1406,3 +1409,14 @@ US_RATINGS = {
 
 def strip_jsonp(code):
     return re.sub(r'(?s)^[a-zA-Z_]+\s*\(\s*(.*)\);\s*?\s*$', r'\1', code)
+
+
+def qualities(quality_ids):
+    """ Get a numeric quality value out of a list of possible values """
+    def q(qid):
+        try:
+            return quality_ids.index(qid)
+        except ValueError:
+            return -1
+    return q
+

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.04.21.1'
+__version__ = '2014.04.21.6'