release 2014.03.11

All the code uses now the urlopen method of YoutubeDL

test/test_InfoExtractor.py

@@ -0,0 +1,44 @@
+#!/usr/bin/env python
+
+from __future__ import unicode_literals
+
+# Allow direct execution
+import os
+import sys
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from test.helper import FakeYDL
+from youtube_dl.extractor.common import InfoExtractor
+from youtube_dl.extractor import YoutubeIE, get_info_extractor
+
+
+class TestIE(InfoExtractor):
+    pass
+
+
+class TestInfoExtractor(unittest.TestCase):
+    def setUp(self):
+        self.ie = TestIE(FakeYDL())
+
+    def test_ie_key(self):
+        self.assertEqual(get_info_extractor(YoutubeIE.ie_key()), YoutubeIE)
+
+    def test_html_search_regex(self):
+        html = '<p id="foo">Watch this <a href="http://www.youtube.com/watch?v=BaW_jenozKc">video</a></p>'
+        search = lambda re, *args: self.ie._html_search_regex(re, html, *args)
+        self.assertEqual(search(r'<p id="foo">(.+?)</p>', 'foo'), 'Watch this video')
+
+    def test_opengraph(self):
+        ie = self.ie
+        html = '''
+            <meta name="og:title" content='Foo'/>
+            <meta content="Some video's description " name="og:description"/>
+            <meta property='og:image' content='http://domain.com/pic.jpg?key1=val1&amp;key2=val2'/>
+            '''
+        self.assertEqual(ie._og_search_title(html), 'Foo')
+        self.assertEqual(ie._og_search_description(html), 'Some video\'s description ')
+        self.assertEqual(ie._og_search_thumbnail(html), 'http://domain.com/pic.jpg?key1=val1&key2=val2')
+
+if __name__ == '__main__':
+    unittest.main()

test/test_playlists.py

@@ -99,7 +99,7 @@ class TestPlaylists(unittest.TestCase):
         result = ie.extract('http://www.ustream.tv/channel/young-americans-for-liberty')
         self.assertIsPlaylist(result)
         self.assertEqual(result['id'], '5124905')
-        self.assertTrue(len(result['entries']) >= 11)
+        self.assertTrue(len(result['entries']) >= 6)
 
     def test_soundcloud_set(self):
         dl = FakeYDL()
@@ -254,9 +254,9 @@ class TestPlaylists(unittest.TestCase):
     def test_generic_rss_feed(self):
         dl = FakeYDL()
         ie = GenericIE(dl)
-        result = ie.extract('http://www.escapistmagazine.com/rss/videos/list/1.xml')
+        result = ie.extract('http://phihag.de/2014/youtube-dl/rss.xml')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], 'http://www.escapistmagazine.com/rss/videos/list/1.xml')
+        self.assertEqual(result['id'], 'http://phihag.de/2014/youtube-dl/rss.xml')
         self.assertEqual(result['title'], 'Zero Punctuation')
         self.assertTrue(len(result['entries']) > 10)
 

youtube_dl/YoutubeDL.py

@@ -370,12 +370,15 @@ class YoutubeDL(object):
         Print the message to stderr, it will be prefixed with 'WARNING:'
         If stderr is a tty file the 'WARNING:' will be colored
         '''
-        if self._err_file.isatty() and os.name != 'nt':
-            _msg_header = '\033[0;33mWARNING:\033[0m'
+        if self.params.get('logger') is not None:
+            self.params['logger'].warning(message)
         else:
-            _msg_header = 'WARNING:'
-        warning_message = '%s %s' % (_msg_header, message)
-        self.to_stderr(warning_message)
+            if self._err_file.isatty() and os.name != 'nt':
+                _msg_header = '\033[0;33mWARNING:\033[0m'
+            else:
+                _msg_header = 'WARNING:'
+            warning_message = '%s %s' % (_msg_header, message)
+            self.to_stderr(warning_message)
 
     def report_error(self, message, tb=None):
         '''
@@ -413,9 +416,9 @@ class YoutubeDL(object):
                 if template_dict.get('width') and template_dict.get('height'):
                     template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
                 elif template_dict.get('height'):
-                    res = '%sp' % template_dict['height']
+                    template_dict['resolution'] = '%sp' % template_dict['height']
                 elif template_dict.get('width'):
-                    res = '?x%d' % template_dict['width']
+                    template_dict['resolution'] = '?x%d' % template_dict['width']
 
             sanitize = lambda k, v: sanitize_filename(
                 compat_str(v),
@@ -698,8 +701,11 @@ class YoutubeDL(object):
         else:
             formats = info_dict['formats']
 
+        if not formats:
+            raise ExtractorError('No video formats found!')
+
         # We check that all the formats have the format and format_id fields
-        for (i, format) in enumerate(formats):
+        for i, format in enumerate(formats):
             if format.get('format_id') is None:
                 format['format_id'] = compat_str(i)
             if format.get('format') is None:
@@ -918,7 +924,7 @@ class YoutubeDL(object):
                     self.to_screen('[%s] %s: Downloading thumbnail ...' %
                                    (info_dict['extractor'], info_dict['id']))
                     try:
-                        uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
+                        uf = self.urlopen(info_dict['thumbnail'])
                         with open(thumb_filename, 'wb') as thumbf:
                             shutil.copyfileobj(uf, thumbf)
                         self.to_screen('[%s] %s: Writing thumbnail to: %s' %
@@ -1164,7 +1170,7 @@ class YoutubeDL(object):
 
     def urlopen(self, req):
         """ Start an HTTP download """
-        return self._opener.open(req)
+        return self._opener.open(req, timeout=self._socket_timeout)
 
     def print_debug_header(self):
         if not self.params.get('verbose'):
@@ -1195,7 +1201,7 @@ class YoutubeDL(object):
 
     def _setup_opener(self):
         timeout_val = self.params.get('socket_timeout')
-        timeout = 600 if timeout_val is None else float(timeout_val)
+        self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
 
         opts_cookiefile = self.params.get('cookiefile')
         opts_proxy = self.params.get('proxy')
@@ -1233,7 +1239,3 @@ class YoutubeDL(object):
         # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
         opener.addheaders = []
         self._opener = opener
-
-        # TODO remove this global modification
-        compat_urllib_request.install_opener(opener)
-        socket.setdefaulttimeout(timeout)

youtube_dl/__init__.py

@@ -50,6 +50,7 @@ __authors__  = (
     'Anthony Weems',
     'David Wagner',
     'Juan C. Olivares',
+    'Mattias Harrysson',
 )
 
 __license__ = 'Public Domain'

youtube_dl/downloader/http.py

@@ -49,7 +49,7 @@ class HttpFD(FileDownloader):
         while count <= retries:
             # Establish connection
             try:
-                data = compat_urllib_request.urlopen(request)
+                data = self.ydl.urlopen(request)
                 break
             except (compat_urllib_error.HTTPError, ) as err:
                 if (err.code < 500 or err.code >= 600) and err.code != 416:
@@ -59,7 +59,7 @@ class HttpFD(FileDownloader):
                     # Unable to resume (requested range not satisfiable)
                     try:
                         # Open the connection again without the range header
-                        data = compat_urllib_request.urlopen(basic_request)
+                        data = self.ydl.urlopen(basic_request)
                         content_length = data.info()['Content-Length']
                     except (compat_urllib_error.HTTPError, ) as err:
                         if err.code < 500 or err.code >= 600:

youtube_dl/extractor/__init__.py

@@ -1,5 +1,6 @@
 from .academicearth import AcademicEarthCourseIE
 from .addanime import AddAnimeIE
+from .aftonbladet import AftonbladetIE
 from .anitube import AnitubeIE
 from .aparat import AparatIE
 from .appletrailers import AppleTrailersIE
@@ -52,7 +53,6 @@ from .dailymotion import (
     DailymotionUserIE,
 )
 from .daum import DaumIE
-from .depositfiles import DepositFilesIE
 from .dotsub import DotsubIE
 from .dreisat import DreiSatIE
 from .defense import DefenseGouvFrIE
@@ -175,6 +175,7 @@ from .ooyala import OoyalaIE
 from .orf import ORFIE
 from .pbs import PBSIE
 from .photobucket import PhotobucketIE
+from .playvid import PlayvidIE
 from .podomatic import PodomaticIE
 from .pornhd import PornHdIE
 from .pornhub import PornHubIE
@@ -246,8 +247,8 @@ from .ustream import UstreamIE, UstreamChannelIE
 from .vbox7 import Vbox7IE
 from .veehd import VeeHDIE
 from .veoh import VeohIE
-from .vesti import VestiIE
 from .vevo import VevoIE
+from .vgtrk import VGTRKIE
 from .vice import ViceIE
 from .viddler import ViddlerIE
 from .videobam import VideoBamIE

youtube_dl/extractor/aftonbladet.py

@@ -0,0 +1,69 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import datetime
+import re
+
+from .common import InfoExtractor
+
+
+class AftonbladetIE(InfoExtractor):
+    _VALID_URL = r'^http://tv\.aftonbladet\.se/webbtv.+?(?P<video_id>article[0-9]+)\.ab(?:$|[?#])'
+    _TEST = {
+        'url': 'http://tv.aftonbladet.se/webbtv/nyheter/vetenskap/rymden/article36015.ab',
+        'info_dict': {
+            'id': 'article36015',
+            'ext': 'mp4',
+            'title': 'Vulkanutbrott i rymden - nu släpper NASA bilderna',
+            'description': 'Jupiters måne mest aktiv av alla himlakroppar',
+            'upload_date': '20140306',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.search(self._VALID_URL, url)
+
+        video_id = mobj.group('video_id')
+        webpage = self._download_webpage(url, video_id)
+
+        # find internal video meta data
+        META_URL = 'http://aftonbladet-play.drlib.aptoma.no/video/%s.json'
+        internal_meta_id = self._html_search_regex(
+            r'data-aptomaId="([\w\d]+)"', webpage, 'internal_meta_id')
+        internal_meta_url = META_URL % internal_meta_id
+        internal_meta_json = self._download_json(
+            internal_meta_url, video_id, 'Downloading video meta data')
+
+        # find internal video formats
+        FORMATS_URL = 'http://aftonbladet-play.videodata.drvideo.aptoma.no/actions/video/?id=%s'
+        internal_video_id = internal_meta_json['videoId']
+        internal_formats_url = FORMATS_URL % internal_video_id
+        internal_formats_json = self._download_json(
+            internal_formats_url, video_id, 'Downloading video formats')
+
+        formats = []
+        for fmt in internal_formats_json['formats']['http']['pseudostreaming']['mp4']:
+            p = fmt['paths'][0]
+            formats.append({
+                'url': 'http://%s:%d/%s/%s' % (p['address'], p['port'], p['path'], p['filename']),
+                'ext': 'mp4',
+                'width': fmt['width'],
+                'height': fmt['height'],
+                'tbr': fmt['bitrate'],
+                'protocol': 'http',
+            })
+        self._sort_formats(formats)
+
+        timestamp = datetime.datetime.fromtimestamp(internal_meta_json['timePublished'])
+        upload_date = timestamp.strftime('%Y%m%d')
+
+        return {
+            'id': video_id,
+            'title': internal_meta_json['title'],
+            'formats': formats,
+            'thumbnail': internal_meta_json['imageUrl'],
+            'description': internal_meta_json['shortPreamble'],
+            'upload_date': upload_date,
+            'duration': internal_meta_json['duration'],
+            'view_count': internal_meta_json['views'],
+        }

youtube_dl/extractor/arte.py

@@ -72,18 +72,22 @@ class ArteTvIE(InfoExtractor):
             return self._extract_liveweb(url, name, lang)
 
         if re.search(self._LIVE_URL, url) is not None:
-            raise ExtractorError(u'Arte live streams are not yet supported, sorry')
+            raise ExtractorError('Arte live streams are not yet supported, sorry')
             # self.extractLiveStream(url)
             # return
 
+        raise ExtractorError('No video found')
+
     def _extract_video(self, url, video_id, lang):
         """Extract from videos.arte.tv"""
         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
-        ref_xml_doc = self._download_xml(ref_xml_url, video_id, note=u'Downloading metadata')
+        ref_xml_doc = self._download_xml(
+            ref_xml_url, video_id, note='Downloading metadata')
         config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
         config_xml_url = config_node.attrib['ref']
-        config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')
+        config_xml = self._download_webpage(
+            config_xml_url, video_id, note='Downloading configuration')
 
         video_urls = list(re.finditer(r'<url quality="(?P<quality>.*?)">(?P<url>.*?)</url>', config_xml))
         def _key(m):

youtube_dl/extractor/collegehumor.py

@@ -35,15 +35,15 @@ class CollegeHumorIE(InfoExtractor):
     },
     # embedded youtube video
     {
-        'url': 'http://www.collegehumor.com/embed/6950457',
+        'url': 'http://www.collegehumor.com/embed/6950306',
         'info_dict': {
-            'id': 'W5gMp3ZjYg4',
+            'id': 'Z-bao9fg6Yc',
             'ext': 'mp4',
-            'title': 'Funny Dogs Protecting Babies Compilation 2014 [NEW HD]',
-            'uploader': 'FunnyPlox TV',
-            'uploader_id': 'funnyploxtv',
-            'description': 'md5:7ded37421526d54afdf005e25bc2b7a3',
-            'upload_date': '20140128',
+            'title': 'Young Americans Think President John F. Kennedy Died THIS MORNING IN A CAR ACCIDENT!!!',
+            'uploader': 'Mark Dice',
+            'uploader_id': 'MarkDice',
+            'description': 'md5:62c3dab9351fac7bb44b53b69511d87f',
+            'upload_date': '20140127',
         },
         'params': {
             'skip_download': True,

youtube_dl/extractor/common.py

@@ -118,9 +118,6 @@ class InfoExtractor(object):
     _real_extract() methods and define a _VALID_URL regexp.
     Probably, they should also be added to the list of extractors.
 
-    _real_extract() must return a *list* of information dictionaries as
-    described above.
-
     Finally, the _WORKING attribute should be set to False for broken IEs
     in order to warn the users and skip the tests.
     """

youtube_dl/extractor/depositfiles.py

@@ -1,60 +0,0 @@
-import re
-import os
-import socket
-
-from .common import InfoExtractor
-from ..utils import (
-    compat_http_client,
-    compat_str,
-    compat_urllib_error,
-    compat_urllib_parse,
-    compat_urllib_request,
-
-    ExtractorError,
-)
-
-
-class DepositFilesIE(InfoExtractor):
-    """Information extractor for depositfiles.com"""
-
-    _VALID_URL = r'(?:http://)?(?:\w+\.)?depositfiles\.com/(?:../(?#locale))?files/(.+)'
-
-    def _real_extract(self, url):
-        file_id = url.split('/')[-1]
-        # Rebuild url in english locale
-        url = 'http://depositfiles.com/en/files/' + file_id
-
-        # Retrieve file webpage with 'Free download' button pressed
-        free_download_indication = {'gateway_result' : '1'}
-        request = compat_urllib_request.Request(url, compat_urllib_parse.urlencode(free_download_indication))
-        try:
-            self.report_download_webpage(file_id)
-            webpage = compat_urllib_request.urlopen(request).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            raise ExtractorError(u'Unable to retrieve file webpage: %s' % compat_str(err))
-
-        # Search for the real file URL
-        mobj = re.search(r'<form action="(http://fileshare.+?)"', webpage)
-        if (mobj is None) or (mobj.group(1) is None):
-            # Try to figure out reason of the error.
-            mobj = re.search(r'<strong>(Attention.*?)</strong>', webpage, re.DOTALL)
-            if (mobj is not None) and (mobj.group(1) is not None):
-                restriction_message = re.sub('\s+', ' ', mobj.group(1)).strip()
-                raise ExtractorError(u'%s' % restriction_message)
-            else:
-                raise ExtractorError(u'Unable to extract download URL from: %s' % url)
-
-        file_url = mobj.group(1)
-        file_extension = os.path.splitext(file_url)[1][1:]
-
-        # Search for file title
-        file_title = self._search_regex(r'<b title="(.*?)">', webpage, u'title')
-
-        return [{
-            'id':       file_id.decode('utf-8'),
-            'url':      file_url.decode('utf-8'),
-            'uploader': None,
-            'upload_date':  None,
-            'title':    file_title,
-            'ext':      file_extension.decode('utf-8'),
-        }]

youtube_dl/extractor/facebook.py

@@ -18,10 +18,8 @@ from ..utils import (
 
 
 class FacebookIE(InfoExtractor):
-    """Information Extractor for Facebook"""
-
     _VALID_URL = r'''(?x)
-        (?:https?://)?(?:\w+\.)?facebook\.com/
+        https?://(?:\w+\.)?facebook\.com/
         (?:[^#?]*\#!/)?
         (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
         (?:v|video_id)=(?P<id>[0-9]+)
@@ -37,14 +35,10 @@ class FacebookIE(InfoExtractor):
             'id': '120708114770723',
             'ext': 'mp4',
             'duration': 279,
-            'title': 'PEOPLE ARE AWESOME 2013'
+            'title': 'PEOPLE ARE AWESOME 2013',
         }
     }
 
-    def report_login(self):
-        """Report attempt to log in."""
-        self.to_screen('Logging in')
-
     def _login(self):
         (useremail, password) = self._get_login_info()
         if useremail is None:
@@ -101,8 +95,6 @@ class FacebookIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError('Invalid URL: %s' % url)
         video_id = mobj.group('id')
 
         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
@@ -128,18 +120,14 @@ class FacebookIE(InfoExtractor):
             video_url = video_data['sd_src']
         if not video_url:
             raise ExtractorError('Cannot find video URL')
-        video_duration = int(video_data['video_duration'])
-        thumbnail = video_data['thumbnail_src']
 
         video_title = self._html_search_regex(
             r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title')
 
-        info = {
+        return {
             'id': video_id,
             'title': video_title,
             'url': video_url,
-            'ext': 'mp4',
-            'duration': video_duration,
-            'thumbnail': thumbnail,
+            'duration': int(video_data['video_duration']),
+            'thumbnail': video_data['thumbnail_src'],
         }
-        return [info]

youtube_dl/extractor/funnyordie.py

@@ -1,12 +1,13 @@
 from __future__ import unicode_literals
 
+import json
 import re
 
 from .common import InfoExtractor
 
 
 class FunnyOrDieIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?funnyordie\.com/videos/(?P<id>[0-9a-f]+)/.*$'
+    _VALID_URL = r'https?://(?:www\.)?funnyordie\.com/(?P<type>embed|videos)/(?P<id>[0-9a-f]+)(?:$|[?#/])'
     _TEST = {
         'url': 'http://www.funnyordie.com/videos/0732f586d7/heart-shaped-box-literal-video-version',
         'file': '0732f586d7.mp4',
@@ -30,10 +31,20 @@ class FunnyOrDieIE(InfoExtractor):
             [r'type="video/mp4" src="(.*?)"', r'src="([^>]*?)" type=\'video/mp4\''],
             webpage, 'video URL', flags=re.DOTALL)
 
+        if mobj.group('type') == 'embed':
+            post_json = self._search_regex(
+                r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details')
+            post = json.loads(post_json)['attachment']
+            title = post['name']
+            description = post.get('description')
+        else:
+            title = self._og_search_title(webpage)
+            description = self._og_search_description(webpage)
+
         return {
             'id': video_id,
             'url': video_url,
             'ext': 'mp4',
-            'title': self._og_search_title(webpage),
-            'description': self._og_search_description(webpage),
+            'title': title,
+            'description': description,
         }

youtube_dl/extractor/gamekings.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -6,13 +8,14 @@ from .common import InfoExtractor
 class GamekingsIE(InfoExtractor):
     _VALID_URL = r'http://www\.gamekings\.tv/videos/(?P<name>[0-9a-z\-]+)'
     _TEST = {
-        u"url": u"http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/",
-        u'file': u'20130811.mp4',
+        'url': 'http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/',
         # MD5 is flaky, seems to change regularly
-        #u'md5': u'2f32b1f7b80fdc5cb616efb4f387f8a3',
+        # 'md5': '2f32b1f7b80fdc5cb616efb4f387f8a3',
         u'info_dict': {
-            u"title": u"Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review",
-            u"description": u"Melle en Steven hebben voor de review een week in de rechtbank doorbracht met Phoenix Wright: Ace Attorney - Dual Destinies.",
+            'id': '20130811',
+            'ext': 'mp4',
+            'title': 'Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review',
+            'description': 'md5:632e61a9f97d700e83f43d77ddafb6a4',
         }
     }
 

youtube_dl/extractor/generic.py

@@ -4,7 +4,6 @@ from __future__ import unicode_literals
 
 import os
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from .youtube import YoutubeIE
@@ -17,6 +16,7 @@ from ..utils import (
 
     ExtractorError,
     HEADRequest,
+    parse_xml,
     smuggle_url,
     unescapeHTML,
     unified_strdate,
@@ -134,6 +134,17 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        # funnyordie embed
+        {
+            'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
+            'md5': '7cf780be104d40fea7bae52eed4a470e',
+            'info_dict': {
+                'id': '18e820ec3f',
+                'ext': 'mp4',
+                'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
+                'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
+            }
+        },
     ]
 
     def report_download_webpage(self, video_id):
@@ -274,7 +285,7 @@ class GenericIE(InfoExtractor):
 
         # Is it an RSS feed?
         try:
-            doc = xml.etree.ElementTree.fromstring(webpage.encode('utf-8'))
+            doc = parse_xml(webpage)
             if doc.tag == 'rss':
                 return self._extract_rss(url, video_id, doc)
         except compat_xml_parse_error:
@@ -432,6 +443,14 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
 
+        # Look for funnyordie embed
+        matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
+        if matches:
+            urlrs = [self.url_result(unescapeHTML(eurl), 'FunnyOrDie')
+                     for eurl in matches]
+            return self.playlist_result(
+                urlrs, playlist_id=video_id, playlist_title=video_title)
+
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if mobj is None:

youtube_dl/extractor/jukebox.py

@@ -1,56 +1,61 @@
-# coding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    RegexNotFoundError,
     unescapeHTML,
 )
 
+
 class JukeboxIE(InfoExtractor):
     _VALID_URL = r'^http://www\.jukebox?\..+?\/.+[,](?P<video_id>[a-z0-9\-]+)\.html'
-    _IFRAME = r'<iframe .*src="(?P<iframe>[^"]*)".*>'
-    _VIDEO_URL = r'"config":{"file":"(?P<video_url>http:[^"]+[.](?P<video_ext>[^.?]+)[?]mdtk=[0-9]+)"'
-    _TITLE = r'<h1 class="inline">(?P<title>[^<]+)</h1>.*<span id="infos_article_artist">(?P<artist>[^<]+)</span>'
-    _IS_YOUTUBE = r'config":{"file":"(?P<youtube_url>http:[\\][/][\\][/]www[.]youtube[.]com[\\][/]watch[?]v=[^"]+)"'
+    _TEST = {
+        'url': 'http://www.jukebox.es/kosheen/videoclip,pride,r303r.html',
+        'md5': '5dc6477e74b1e37042ac5acedd8413e5',
+        'info_dict': {
+            'id': 'r303r',
+            'ext': 'flv',
+            'title': 'Kosheen-En Vivo Pride',
+            'uploader': 'Kosheen',
+        },
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('video_id')
 
         html = self._download_webpage(url, video_id)
-
-        mobj = re.search(self._IFRAME, html)
-        if mobj is None:
-            raise ExtractorError(u'Cannot extract iframe url')
-        iframe_url = unescapeHTML(mobj.group('iframe'))
+        iframe_url = unescapeHTML(self._search_regex(r'<iframe .*src="([^"]*)"', html, 'iframe url'))
 
         iframe_html = self._download_webpage(iframe_url, video_id, 'Downloading iframe')
-        mobj = re.search(r'class="jkb_waiting"', iframe_html)
-        if mobj is not None:
-            raise ExtractorError(u'Video is not available(in your country?)!')
+        if re.search(r'class="jkb_waiting"', iframe_html) is not None:
+            raise ExtractorError('Video is not available(in your country?)!')
 
         self.report_extraction(video_id)
 
-        mobj = re.search(self._VIDEO_URL, iframe_html)
-        if mobj is None:
-            mobj = re.search(self._IS_YOUTUBE, iframe_html)
-            if mobj is None:
-                raise ExtractorError(u'Cannot extract video url')
-            youtube_url = unescapeHTML(mobj.group('youtube_url')).replace('\/','/')
-            self.to_screen(u'Youtube video detected')
-            return self.url_result(youtube_url,ie='Youtube')
-        video_url = unescapeHTML(mobj.group('video_url')).replace('\/','/')
-        video_ext = unescapeHTML(mobj.group('video_ext'))
+        try:
+            video_url = self._search_regex(r'"config":{"file":"(?P<video_url>http:[^"]+\?mdtk=[0-9]+)"',
+                iframe_html, 'video url')
+            video_url = unescapeHTML(video_url).replace('\/', '/')
+        except RegexNotFoundError:
+            youtube_url = self._search_regex(
+                r'config":{"file":"(http:\\/\\/www\.youtube\.com\\/watch\?v=[^"]+)"',
+                iframe_html, 'youtube url')
+            youtube_url = unescapeHTML(youtube_url).replace('\/', '/')
+            self.to_screen('Youtube video detected')
+            return self.url_result(youtube_url, ie='Youtube')
 
-        mobj = re.search(self._TITLE, html)
-        if mobj is None:
-            raise ExtractorError(u'Cannot extract title')
-        title = unescapeHTML(mobj.group('title'))
-        artist = unescapeHTML(mobj.group('artist'))
+        title = self._html_search_regex(r'<h1 class="inline">([^<]+)</h1>',
+            html, 'title')
+        artist = self._html_search_regex(r'<span id="infos_article_artist">([^<]+)</span>',
+            html, 'artist')
 
-        return [{'id': video_id,
-                 'url': video_url,
-                 'title': artist + '-' + title,
-                 'ext': video_ext
-                 }]
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': artist + '-' + title,
+            'uploader': artist,
+        }

youtube_dl/extractor/lynda.py

@@ -10,6 +10,7 @@ from ..utils import (
     compat_urllib_request,
     ExtractorError,
     int_or_none,
+    compat_str,
 )
 
 
@@ -27,9 +28,10 @@ class LyndaIE(SubtitlesInfoExtractor):
 
     _TEST = {
         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
-        'file': '114408.mp4',
         'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
         'info_dict': {
+            'id': '114408',
+            'ext': 'mp4',
             'title': 'Using the exercise files',
             'duration': 68
         }
@@ -42,17 +44,18 @@ class LyndaIE(SubtitlesInfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1)
 
-        page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
-                                      video_id, 'Downloading video JSON')
+        page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id, video_id,
+            'Downloading video JSON')
         video_json = json.loads(page)
 
         if 'Status' in video_json:
             raise ExtractorError('lynda returned error: %s' % video_json['Message'], expected=True)
 
         if video_json['HasAccess'] is False:
-            raise ExtractorError('Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True)
+            raise ExtractorError(
+                'Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True)
 
-        video_id = video_json['ID']
+        video_id = compat_str(video_json['ID'])
         duration = video_json['DurationInSeconds']
         title = video_json['Title']
 
@@ -108,7 +111,7 @@ class LyndaIE(SubtitlesInfoExtractor):
             'stayPut': 'false'
         }        
         request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
-        login_page = self._download_webpage(request, None, note='Logging in as %s' % username)
+        login_page = self._download_webpage(request, None, 'Logging in as %s' % username)
 
         # Not (yet) logged in
         m = re.search(r'loginResultJson = \'(?P<json>[^\']+)\';', login_page)
@@ -133,7 +136,7 @@ class LyndaIE(SubtitlesInfoExtractor):
                     'stayPut': 'false',
                 }
                 request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form))
-                login_page = self._download_webpage(request, None, note='Confirming log in and log out from another device')
+                login_page = self._download_webpage(request, None, 'Confirming log in and log out from another device')
 
         if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
             raise ExtractorError('Unable to log in')
@@ -167,7 +170,7 @@ class LyndaIE(SubtitlesInfoExtractor):
 
     def _get_available_subtitles(self, video_id, webpage):
         url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
-        sub = self._download_webpage(url, None, note=False)
+        sub = self._download_webpage(url, None, False)
         sub_json = json.loads(sub)
         return {'en': url} if len(sub_json) > 0 else {}
 

youtube_dl/extractor/mtv.py

@@ -5,9 +5,12 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
+    compat_urllib_request,
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
+    HEADRequest,
+    unescapeHTML,
     url_basename,
     RegexNotFoundError,
 )
@@ -18,6 +21,7 @@ def _media_xml_tag(tag):
 
 
 class MTVServicesInfoExtractor(InfoExtractor):
+    _MOBILE_TEMPLATE = None
     @staticmethod
     def _id_from_uri(uri):
         return uri.split(':')[-1]
@@ -39,9 +43,29 @@ class MTVServicesInfoExtractor(InfoExtractor):
         else:
             return thumb_node.attrib['url']
 
-    def _extract_video_formats(self, mdoc):
-        if re.match(r'.*/error_country_block\.swf$', mdoc.find('.//src').text) is not None:
-            raise ExtractorError('This video is not available from your country.', expected=True)
+    def _extract_mobile_video_formats(self, mtvn_id):
+        webpage_url = self._MOBILE_TEMPLATE % mtvn_id
+        req = compat_urllib_request.Request(webpage_url)
+        # Otherwise we get a webpage that would execute some javascript
+        req.add_header('Youtubedl-user-agent', 'curl/7')
+        webpage = self._download_webpage(req, mtvn_id,
+            'Downloading mobile page')
+        metrics_url = unescapeHTML(self._search_regex(r'<a href="(http://metrics.+?)"', webpage, 'url'))
+        req = HEADRequest(metrics_url)
+        response = self._request_webpage(req, mtvn_id, 'Resolving url')
+        url = response.geturl()
+        # Transform the url to get the best quality:
+        url = re.sub(r'.+pxE=mp4', 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=0+_pxK=18639+_pxE=mp4', url, 1)
+        return [{'url': url,'ext': 'mp4'}]
+
+    def _extract_video_formats(self, mdoc, mtvn_id):
+        if re.match(r'.*/(error_country_block\.swf|geoblock\.mp4)$', mdoc.find('.//src').text) is not None:
+            if mtvn_id is not None and self._MOBILE_TEMPLATE is not None:
+                self.to_screen('The normal version is not available from your '
+                    'country, trying with the mobile version')
+                return self._extract_mobile_video_formats(mtvn_id)
+            raise ExtractorError('This video is not available from your country.',
+                expected=True)
 
         formats = []
         for rendition in mdoc.findall('.//rendition'):
@@ -94,9 +118,16 @@ class MTVServicesInfoExtractor(InfoExtractor):
             raise ExtractorError('Could not find video title')
         title = title.strip()
 
+        # This a short id that's used in the webpage urls
+        mtvn_id = None
+        mtvn_id_node = find_xpath_attr(itemdoc, './/{http://search.yahoo.com/mrss/}category',
+                'scheme', 'urn:mtvn:id')
+        if mtvn_id_node is not None:
+            mtvn_id = mtvn_id_node.text
+
         return {
             'title': title,
-            'formats': self._extract_video_formats(mediagen_doc),
+            'formats': self._extract_video_formats(mediagen_doc, mtvn_id),
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
             'description': description,

youtube_dl/extractor/myvideo.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import binascii
 import base64
 import hashlib
@@ -14,18 +16,16 @@ from ..utils import (
 )
 
 
-
 class MyVideoIE(InfoExtractor):
-    """Information Extractor for myvideo.de."""
-
-    _VALID_URL = r'(?:http://)?(?:www\.)?myvideo\.de/(?:[^/]+/)?watch/([0-9]+)/([^?/]+).*'
-    IE_NAME = u'myvideo'
+    _VALID_URL = r'http://(?:www\.)?myvideo\.de/(?:[^/]+/)?watch/(?P<id>[0-9]+)/[^?/]+.*'
+    IE_NAME = 'myvideo'
     _TEST = {
-        u'url': u'http://www.myvideo.de/watch/8229274/bowling_fail_or_win',
-        u'file': u'8229274.flv',
-        u'md5': u'2d2753e8130479ba2cb7e0a37002053e',
-        u'info_dict': {
-            u"title": u"bowling-fail-or-win"
+        'url': 'http://www.myvideo.de/watch/8229274/bowling_fail_or_win',
+        'md5': '2d2753e8130479ba2cb7e0a37002053e',
+        'info_dict': {
+            'id': '8229274',
+            'ext': 'flv',
+            'title': 'bowling-fail-or-win',
         }
     }
 
@@ -53,10 +53,7 @@ class MyVideoIE(InfoExtractor):
 
     def _real_extract(self,url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'invalid URL: %s' % url)
-
-        video_id = mobj.group(1)
+        video_id = mobj.group('id')
 
         GK = (
           b'WXpnME1EZGhNRGhpTTJNM01XVmhOREU0WldNNVpHTTJOakpt'
@@ -74,37 +71,33 @@ class MyVideoIE(InfoExtractor):
             video_url = mobj.group(1) + '.flv'
 
             video_title = self._html_search_regex('<title>([^<]+)</title>',
-                webpage, u'title')
+                webpage, 'title')
 
-            video_ext = self._search_regex('[.](.+?)$', video_url, u'extension')
-
-            return [{
-                'id':       video_id,
-                'url':      video_url,
-                'uploader': None,
-                'upload_date':  None,
-                'title':    video_title,
-                'ext':      video_ext,
-            }]
+            return {
+                'id': video_id,
+                'url': video_url,
+                'title': video_title,
+            }
 
         mobj = re.search(r'data-video-service="/service/data/video/%s/config' % video_id, webpage)
         if mobj is not None:
             request = compat_urllib_request.Request('http://www.myvideo.de/service/data/video/%s/config' % video_id, '')
             response = self._download_webpage(request, video_id,
-                                              u'Downloading video info')
+                                              'Downloading video info')
             info = json.loads(base64.b64decode(response).decode('utf-8'))
-            return {'id': video_id,
-                    'title': info['title'],
-                    'url': info['streaming_url'].replace('rtmpe', 'rtmpt'),
-                    'play_path': info['filename'],
-                    'ext': 'flv',
-                    'thumbnail': info['thumbnail'][0]['url'],
-                    }
+            return {
+                'id': video_id,
+                'title': info['title'],
+                'url': info['streaming_url'].replace('rtmpe', 'rtmpt'),
+                'play_path': info['filename'],
+                'ext': 'flv',
+                'thumbnail': info['thumbnail'][0]['url'],
+            }
 
         # try encxml
         mobj = re.search('var flashvars={(.+?)}', webpage)
         if mobj is None:
-            raise ExtractorError(u'Unable to extract video')
+            raise ExtractorError('Unable to extract video')
 
         params = {}
         encxml = ''
@@ -118,7 +111,7 @@ class MyVideoIE(InfoExtractor):
             params['domain'] = 'www.myvideo.de'
         xmldata_url = '%s?%s' % (encxml, compat_urllib_parse.urlencode(params))
         if 'flash_playertype=MTV' in xmldata_url:
-            self._downloader.report_warning(u'avoiding MTV player')
+            self._downloader.report_warning('avoiding MTV player')
             xmldata_url = (
                 'http://www.myvideo.de/dynamic/get_player_video_xml.php'
                 '?flash_playertype=D&ID=%s&_countlimit=4&autorun=yes'
@@ -144,7 +137,7 @@ class MyVideoIE(InfoExtractor):
             video_url = compat_urllib_parse.unquote(mobj.group(1))
             if 'myvideo2flash' in video_url:
                 self.report_warning(
-                    u'Rewriting URL to use unencrypted rtmp:// ...',
+                    'Rewriting URL to use unencrypted rtmp:// ...',
                     video_id)
                 video_url = video_url.replace('rtmpe://', 'rtmp://')
 
@@ -152,39 +145,31 @@ class MyVideoIE(InfoExtractor):
             # extract non rtmp videos
             mobj = re.search('path=\'(http.*?)\' source=\'(.*?)\'', dec_data)
             if mobj is None:
-                raise ExtractorError(u'unable to extract url')
+                raise ExtractorError('unable to extract url')
             video_url = compat_urllib_parse.unquote(mobj.group(1)) + compat_urllib_parse.unquote(mobj.group(2))
 
-        video_file = self._search_regex('source=\'(.*?)\'', dec_data, u'video file')
+        video_file = self._search_regex('source=\'(.*?)\'', dec_data, 'video file')
         video_file = compat_urllib_parse.unquote(video_file)
 
         if not video_file.endswith('f4m'):
             ppath, prefix = video_file.split('.')
             video_playpath = '%s:%s' % (prefix, ppath)
-            video_hls_playlist = ''
         else:
             video_playpath = ''
-            video_hls_playlist = (
-                video_file
-            ).replace('.f4m', '.m3u8')
 
-        video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, u'swfobj')
+        video_swfobj = self._search_regex('swfobject.embedSWF\(\'(.+?)\'', webpage, 'swfobj')
         video_swfobj = compat_urllib_parse.unquote(video_swfobj)
 
         video_title = self._html_search_regex("<h1(?: class='globalHd')?>(.*?)</h1>",
-            webpage, u'title')
+            webpage, 'title')
 
-        return [{
-            'id':                 video_id,
-            'url':                video_url,
-            'tc_url':             video_url,
-            'uploader':           None,
-            'upload_date':        None,
-            'title':              video_title,
-            'ext':                u'flv',
-            'play_path':          video_playpath,
-            'video_file':         video_file,
-            'video_hls_playlist': video_hls_playlist,
-            'player_url':         video_swfobj,
-        }]
+        return {
+            'id': video_id,
+            'url': video_url,
+            'tc_url': video_url,
+            'title': video_title,
+            'ext': 'flv',
+            'play_path': video_playpath,
+            'player_url': video_swfobj,
+        }
 

youtube_dl/extractor/photobucket.py

@@ -1,76 +1,43 @@
+from __future__ import unicode_literals
+
 import datetime
 import json
 import re
 
 from .common import InfoExtractor
 
-from ..utils import (
-    ExtractorError,
-)
 
 class PhotobucketIE(InfoExtractor):
-    """Information extractor for photobucket.com."""
-
-    # TODO: the original _VALID_URL was:
-    # r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*[\?\&]current=(.*\.flv)'
-    # Check if it's necessary to keep the old extracion process
-    _VALID_URL = r'(?:http://)?(?:[a-z0-9]+\.)?photobucket\.com/.*(([\?\&]current=)|_)(?P<id>.*)\.(?P<ext>(flv)|(mp4))'
-    IE_NAME = u'photobucket'
+    _VALID_URL = r'http://(?:[a-z0-9]+\.)?photobucket\.com/.*(([\?\&]current=)|_)(?P<id>.*)\.(?P<ext>(flv)|(mp4))'
     _TEST = {
-        u'url': u'http://media.photobucket.com/user/rachaneronas/media/TiredofLinkBuildingTryBacklinkMyDomaincom_zpsc0c3b9fa.mp4.html?filters[term]=search&filters[primary]=videos&filters[secondary]=images&sort=1&o=0',
-        u'file': u'zpsc0c3b9fa.mp4',
-        u'md5': u'7dabfb92b0a31f6c16cebc0f8e60ff99',
-        u'info_dict': {
-            u"upload_date": u"20130504", 
-            u"uploader": u"rachaneronas", 
-            u"title": u"Tired of Link Building? Try BacklinkMyDomain.com!"
+        'url': 'http://media.photobucket.com/user/rachaneronas/media/TiredofLinkBuildingTryBacklinkMyDomaincom_zpsc0c3b9fa.mp4.html?filters[term]=search&filters[primary]=videos&filters[secondary]=images&sort=1&o=0',
+        'file': 'zpsc0c3b9fa.mp4',
+        'md5': '7dabfb92b0a31f6c16cebc0f8e60ff99',
+        'info_dict': {
+            'upload_date': '20130504',
+            'uploader': 'rachaneronas',
+            'title': 'Tired of Link Building? Try BacklinkMyDomain.com!',
         }
     }
 
     def _real_extract(self, url):
-        # Extract id from URL
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-
         video_id = mobj.group('id')
-
         video_extension = mobj.group('ext')
 
-        # Retrieve video webpage to extract further information
         webpage = self._download_webpage(url, video_id)
 
         # Extract URL, uploader, and title from webpage
         self.report_extraction(video_id)
-        # We try first by looking the javascript code:
-        mobj = re.search(r'Pb\.Data\.Shared\.put\(Pb\.Data\.Shared\.MEDIA, (?P<json>.*?)\);', webpage)
-        if mobj is not None:
-            info = json.loads(mobj.group('json'))
-            return [{
-                'id':       video_id,
-                'url':      info[u'downloadUrl'],
-                'uploader': info[u'username'],
-                'upload_date':  datetime.date.fromtimestamp(info[u'creationDate']).strftime('%Y%m%d'),
-                'title':    info[u'title'],
-                'ext':      video_extension,
-                'thumbnail': info[u'thumbUrl'],
-            }]
-
-        # We try looking in other parts of the webpage
-        video_url = self._search_regex(r'<link rel="video_src" href=".*\?file=([^"]+)" />',
-            webpage, u'video URL')
-
-        mobj = re.search(r'<title>(.*) video by (.*) - Photobucket</title>', webpage)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract title')
-        video_title = mobj.group(1).decode('utf-8')
-        video_uploader = mobj.group(2).decode('utf-8')
-
-        return [{
-            'id':       video_id.decode('utf-8'),
-            'url':      video_url.decode('utf-8'),
-            'uploader': video_uploader,
-            'upload_date':  None,
-            'title':    video_title,
-            'ext':      video_extension.decode('utf-8'),
-        }]
+        info_json = self._search_regex(r'Pb\.Data\.Shared\.put\(Pb\.Data\.Shared\.MEDIA, (.*?)\);',
+            webpage, 'info json')
+        info = json.loads(info_json)
+        return {
+            'id': video_id,
+            'url': info['downloadUrl'],
+            'uploader': info['username'],
+            'upload_date': datetime.date.fromtimestamp(info['creationDate']).strftime('%Y%m%d'),
+            'title': info['title'],
+            'ext': video_extension,
+            'thumbnail': info['thumbUrl'],
+        }

youtube_dl/extractor/playvid.py

@@ -0,0 +1,80 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+)
+
+
+class PlayvidIE(InfoExtractor):
+    _VALID_URL = r'^https?://www\.playvid\.com/watch(\?v=|/)(?P<id>.+?)(?:#|$)'
+    _TEST = {
+        'url': 'http://www.playvid.com/watch/agbDDi7WZTV',
+        'md5': '44930f8afa616efdf9482daf4fe53e1e',
+        'info_dict': {
+            'id': 'agbDDi7WZTV',
+            'ext': 'mp4',
+            'title': 'Michelle Lewin in Miami Beach',
+            'duration': 240,
+            'age_limit': 18,
+        }
+    }
+
+    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_title = None
+        duration = None
+        video_thumbnail = None
+        formats = []
+
+        # most of the information is stored in the flashvars
+        flashvars = self._html_search_regex(
+            r'flashvars="(.+?)"', webpage, 'flashvars')
+
+        infos = compat_urllib_parse.unquote(flashvars).split(r'&')
+        for info in infos:
+            videovars_match = re.match(r'^video_vars\[(.+?)\]=(.+?)$', info)
+            if videovars_match:
+                key = videovars_match.group(1)
+                val = videovars_match.group(2)
+
+                if key == 'title':
+                    video_title = compat_urllib_parse.unquote_plus(val)
+                if key == 'duration':
+                    try:
+                        duration = int(val)
+                    except ValueError:
+                        pass
+                if key == 'big_thumb':
+                    video_thumbnail = val
+
+                videourl_match = re.match(
+                    r'^video_urls\]\[(?P<resolution>[0-9]+)p', key)
+                if videourl_match:
+                    height = int(videourl_match.group('resolution'))
+                    formats.append({
+                        'height': height,
+                        'url': val,
+                    })
+        self._sort_formats(formats)
+
+        # Extract title - should be in the flashvars; if not, look elsewhere
+        if video_title is None:
+            video_title = self._html_search_regex(
+                r'<title>(.*?)</title', webpage, 'title')
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': video_title,
+            'thumbnail': video_thumbnail,
+            'duration': duration,
+            'description': None,
+            'age_limit': 18
+        }

youtube_dl/extractor/pornhub.py

@@ -44,7 +44,7 @@ class PornHubIE(InfoExtractor):
 
         video_urls = list(map(compat_urllib_parse.unquote , re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage)))
         if webpage.find('"encrypted":true') != -1:
-            password = self._html_search_regex(r'"video_title":"([^"]+)', webpage, 'password').replace('+', ' ')
+            password = compat_urllib_parse.unquote_plus(self._html_search_regex(r'"video_title":"([^"]+)', webpage, 'password'))
             video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls))
 
         formats = []

youtube_dl/extractor/soundcloud.py

@@ -54,6 +54,7 @@ class SoundcloudIE(InfoExtractor):
                 'id': '47127627',
                 'ext': 'mp3',
                 'title': 'Goldrushed',
+                'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
                 'uploader': 'The Royal Concept',
                 'upload_date': '20120521',
             },

youtube_dl/extractor/spike.py

@@ -1,10 +1,15 @@
 from __future__ import unicode_literals
 
+import re
+
 from .mtv import MTVServicesInfoExtractor
 
 
 class SpikeIE(MTVServicesInfoExtractor):
-    _VALID_URL = r'https?://www\.spike\.com/(video-clips|episodes)/.+'
+    _VALID_URL = r'''(?x)https?://
+        (www\.spike\.com/(video-clips|episodes)/.+|
+         m\.spike\.com/videos/video.rbml\?id=(?P<mobile_id>[^&]+))
+        '''
     _TEST = {
         'url': 'http://www.spike.com/video-clips/lhtu8m/auction-hunters-can-allen-ride-a-hundred-year-old-motorcycle',
         'md5': '1a9265f32b0c375793d6c4ce45255256',
@@ -17,3 +22,11 @@ class SpikeIE(MTVServicesInfoExtractor):
     }
 
     _FEED_URL = 'http://www.spike.com/feeds/mrss/'
+    _MOBILE_TEMPLATE = 'http://m.spike.com/videos/video.rbml?id=%s'
+
+    def _real_extract(self, url):
+        mobj = re.search(self._VALID_URL, url)
+        mobile_id = mobj.group('mobile_id')
+        if mobile_id is not None:
+            url = 'http://www.spike.com/video-clips/%s' % mobile_id
+        return super(SpikeIE, self)._real_extract(url)

youtube_dl/extractor/vevo.py

@@ -57,7 +57,7 @@ class VevoIE(InfoExtractor):
             'age_limit': 18,
             'title': 'Tunnel Vision (Explicit)',
             'uploader': 'Justin Timberlake',
-            'upload_date': '20130704',
+            'upload_date': '20130703',
         },
         'params': {
             'skip_download': 'true',
@@ -169,7 +169,7 @@ class VevoIE(InfoExtractor):
 
         timestamp_ms = int(self._search_regex(
             r'/Date\((\d+)\)/', video_info['launchDate'], 'launch date'))
-        upload_date = datetime.datetime.fromtimestamp(timestamp_ms // 1000)
+        upload_date = datetime.datetime.utcfromtimestamp(timestamp_ms // 1000)
         return {
             'id': video_id,
             'title': video_info['title'],

youtube_dl/extractor/vgtrk.py

@@ -10,10 +10,9 @@ from ..utils import (
 )
 
 
-class VestiIE(InfoExtractor):
-    IE_NAME = 'vesti'
-    IE_DESC = 'Вести.Ru'
-    _VALID_URL = r'http://(?:.+?\.)?(?:vesti\.ru|russia\.tv)/(?P<id>.+)'
+class VGTRKIE(InfoExtractor):
+    IE_DESC = 'ВГТРК'
+    _VALID_URL = r'http://(?:.+?\.)?(?:vesti\.ru|russia2?\.tv|tvkultura\.ru|rutv\.ru)/(?P<id>.+)'
 
     _TESTS = [
         {
@@ -72,6 +71,35 @@ class VestiIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        {
+            'url': 'http://sochi2014.vesti.ru/video/index/video_id/766403',
+            'info_dict': {
+                'id': '766403',
+                'ext': 'mp4',
+                'title': 'XXII зимние Олимпийские игры. Российские хоккеисты стартовали на Олимпиаде с победы',
+                'description': 'md5:55805dfd35763a890ff50fa9e35e31b3',
+                'duration': 271,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+            'skip': 'Blocked outside Russia',
+        },
+        {
+            'url': 'http://sochi2014.vesti.ru/live/play/live_id/301',
+            'info_dict': {
+                'id': '51499',
+                'ext': 'flv',
+                'title': 'Сочи-2014. Биатлон. Индивидуальная гонка. Мужчины ',
+                'description': 'md5:9e0ed5c9d2fa1efbfdfed90c9a6d179c',
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Translation has finished'
+        },
         {
             'url': 'http://russia.tv/video/show/brand_id/5169/episode_id/970443/video_id/975648',
             'info_dict': {
@@ -101,34 +129,48 @@ class VestiIE(InfoExtractor):
             },
         },
         {
-            'url': 'http://sochi2014.vesti.ru/video/index/video_id/766403',
+            'url': 'http://2.russia.tv/video/show/brand_id/48863/episode_id/972920/video_id/978667/viewtype/picture',
             'info_dict': {
-                'id': '766403',
+                'id': '775081',
                 'ext': 'mp4',
-                'title': 'XXII зимние Олимпийские игры. Российские хоккеисты стартовали на Олимпиаде с победы',
-                'description': 'md5:55805dfd35763a890ff50fa9e35e31b3',
-                'duration': 271,
+                'title': 'XXII зимние Олимпийские игры. Россияне заняли весь пьедестал в лыжных гонках',
+                'description': 'md5:15d3741dd8d04b203fbc031c6a47fb0f',
+                'duration': 101,
             },
             'params': {
                 # m3u8 download
                 'skip_download': True,
             },
-            'skip': 'Blocked outside Russia'
+            'skip': 'Blocked outside Russia',
         },
         {
-            'url': 'http://sochi2014.vesti.ru/live/play/live_id/301',
+            'url': 'http://tvkultura.ru/video/show/brand_id/31724/episode_id/972347/video_id/978186',
             'info_dict': {
-                'id': '51499',
-                'ext': 'flv',
-                'title': 'Сочи-2014. Биатлон. Индивидуальная гонка. Мужчины ',
-                'description': 'md5:9e0ed5c9d2fa1efbfdfed90c9a6d179c',
+                'id': '774471',
+                'ext': 'mp4',
+                'title': 'Монологи на все времена',
+                'description': 'md5:18d8b5e6a41fb1faa53819471852d5d5',
+                'duration': 2906,
             },
             'params': {
-                # rtmp download
+                # m3u8 download
                 'skip_download': True,
             },
-            'skip': 'Translation has finished'
-        }
+        },
+        {
+            'url': 'http://rutv.ru/brand/show/id/6792/channel/75',
+            'info_dict': {
+                'id': '125521',
+                'ext': 'mp4',
+                'title': 'Грустная дама червей. Х/ф',
+                'description': '',
+                'duration': 4882,
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
+        },
     ]
 
     def _real_extract(self, url):

youtube_dl/extractor/videodetective.py

@@ -1,22 +1,23 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from .internetvideoarchive import InternetVideoArchiveIE
-from ..utils import (
-    compat_urlparse,
-)
+from ..utils import compat_urlparse
 
 
 class VideoDetectiveIE(InfoExtractor):
     _VALID_URL = r'https?://www\.videodetective\.com/[^/]+/[^/]+/(?P<id>\d+)'
 
     _TEST = {
-        u'url': u'http://www.videodetective.com/movies/kick-ass-2/194487',
-        u'file': u'194487.mp4',
-        u'info_dict': {
-            u'title': u'KICK-ASS 2',
-            u'description': u'md5:65ba37ad619165afac7d432eaded6013',
-            u'duration': 135,
+        'url': 'http://www.videodetective.com/movies/kick-ass-2/194487',
+        'info_dict': {
+            'id': '194487',
+            'ext': 'mp4',
+            'title': 'KICK-ASS 2',
+            'description': 'md5:65ba37ad619165afac7d432eaded6013',
+            'duration': 135,
         },
     }
 
@@ -26,5 +27,4 @@ class VideoDetectiveIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         og_video = self._og_search_video_url(webpage)
         query = compat_urlparse.urlparse(og_video).query
-        return self.url_result(InternetVideoArchiveIE._build_url(query),
-            ie=InternetVideoArchiveIE.ie_key())
+        return self.url_result(InternetVideoArchiveIE._build_url(query), ie=InternetVideoArchiveIE.ie_key())

youtube_dl/extractor/vimeo.py

@@ -8,6 +8,7 @@ import itertools
 from .common import InfoExtractor
 from .subtitles import SubtitlesInfoExtractor
 from ..utils import (
+    compat_HTTPError,
     compat_urllib_parse,
     compat_urllib_request,
     clean_html,
@@ -172,7 +173,18 @@ class VimeoIE(SubtitlesInfoExtractor):
 
         # Retrieve video webpage to extract further information
         request = compat_urllib_request.Request(url, None, headers)
-        webpage = self._download_webpage(request, video_id)
+        try:
+            webpage = self._download_webpage(request, video_id)
+        except ExtractorError as ee:
+            if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
+                errmsg = ee.cause.read()
+                if b'Because of its privacy settings, this video cannot be played here' in errmsg:
+                    raise ExtractorError(
+                        'Cannot download embed-only video without embedding '
+                        'URL. Please call youtube-dl with the URL of the page '
+                        'that embeds this video.',
+                        expected=True)
+            raise
 
         # Now we begin extracting as much information as we can from what we
         # retrieved. First we extract the information common to all extractors,

youtube_dl/extractor/vube.py

@@ -13,7 +13,7 @@ class VubeIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://vube.com/Chiara+Grispo+Video+Channel/YL2qNPkqon',
-        'md5': 'f81dcf6d0448e3291f54380181695821',
+        'md5': 'db7aba89d4603dadd627e9d1973946fe',
         'info_dict': {
             'id': 'YL2qNPkqon',
             'ext': 'mp4',
@@ -77,4 +77,4 @@ class VubeIE(InfoExtractor):
             'like_count': like_count,
             'dislike_count': dislike_count,
             'comment_count': comment_count,
-        }
+        }

youtube_dl/extractor/xnxx.py

@@ -1,55 +1,49 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
-
-    ExtractorError,
 )
 
 
 class XNXXIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:video|www)\.xnxx\.com/video([0-9]+)/(.*)'
-    VIDEO_URL_RE = r'flv_url=(.*?)&'
-    VIDEO_TITLE_RE = r'<title>(.*?)\s+-\s+XNXX.COM'
-    VIDEO_THUMB_RE = r'url_bigthumb=(.*?)&amp;'
+    _VALID_URL = r'^https?://(?:video|www)\.xnxx\.com/video(?P<id>[0-9]+)/(.*)'
     _TEST = {
-        u'url': u'http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_',
-        u'file': u'1135332.flv',
-        u'md5': u'0831677e2b4761795f68d417e0b7b445',
-        u'info_dict': {
-            u"title": u"lida \u00bb Naked Funny Actress  (5)",
-            u"age_limit": 18,
+        'url': 'http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_',
+        'md5': '0831677e2b4761795f68d417e0b7b445',
+        'info_dict': {
+            'id': '1135332',
+            'ext': 'flv',
+            'title': 'lida » Naked Funny Actress  (5)',
+            'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-        video_id = mobj.group(1)
+        video_id = mobj.group('id')
 
         # Get webpage content
         webpage = self._download_webpage(url, video_id)
 
-        video_url = self._search_regex(self.VIDEO_URL_RE,
-            webpage, u'video URL')
+        video_url = self._search_regex(r'flv_url=(.*?)&amp;',
+            webpage, 'video URL')
         video_url = compat_urllib_parse.unquote(video_url)
 
-        video_title = self._html_search_regex(self.VIDEO_TITLE_RE,
-            webpage, u'title')
+        video_title = self._html_search_regex(r'<title>(.*?)\s+-\s+XNXX.COM',
+            webpage, 'title')
 
-        video_thumbnail = self._search_regex(self.VIDEO_THUMB_RE,
-            webpage, u'thumbnail', fatal=False)
+        video_thumbnail = self._search_regex(r'url_bigthumb=(.*?)&amp;',
+            webpage, 'thumbnail', fatal=False)
 
-        return [{
+        return {
             'id': video_id,
             'url': video_url,
-            'uploader': None,
-            'upload_date': None,
             'title': video_title,
             'ext': 'flv',
             'thumbnail': video_thumbnail,
-            'description': None,
             'age_limit': 18,
-        }]
+        }

youtube_dl/extractor/youtube.py

@@ -1285,10 +1285,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
 
         # Decide which formats to download
         try:
-            mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
+            mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
             if not mobj:
                 raise ValueError('Could not find vevo ID')
-            ytplayer_config = json.loads(mobj.group(1))
+            json_code = uppercase_escape(mobj.group(1))
+            ytplayer_config = json.loads(json_code)
             args = ytplayer_config['args']
             # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
             # this signatures are encrypted

youtube_dl/utils.py

@@ -22,6 +22,7 @@ import struct
 import subprocess
 import sys
 import traceback
+import xml.etree.ElementTree
 import zlib
 
 try:
@@ -1267,3 +1268,13 @@ def read_batch_urls(batch_fd):
 
 def urlencode_postdata(*args, **kargs):
     return compat_urllib_parse.urlencode(*args, **kargs).encode('ascii')
+
+
+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 {}
+    return xml.etree.ElementTree.XML(s.encode('utf-8'), **kwargs)

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.03.07.1'
+__version__ = '2014.03.11'