release 2014.08.21.2

This moves playlist tests where they belong, i.e. to the extractors themselves.
Additionally, all our network interaction configuration for tests in test_download now applies to playlist tests as well.

Makefile

@@ -6,10 +6,10 @@ clean:
 cleanall: clean
 	rm -f youtube-dl youtube-dl.exe
 
-PREFIX=/usr/local
-BINDIR=$(PREFIX)/bin
-MANDIR=$(PREFIX)/man
-PYTHON=/usr/bin/env python
+PREFIX ?= /usr/local
+BINDIR ?= $(PREFIX)/bin
+MANDIR ?= $(PREFIX)/man
+PYTHON ?= /usr/bin/env python
 
 # set SYSCONFDIR to /etc if PREFIX=/usr or PREFIX=/usr/local
 ifeq ($(PREFIX),/usr)

README.md

@@ -17,6 +17,14 @@ If you do not have curl, you can alternatively use a recent wget:
 
 Windows users can [download a .exe file](https://yt-dl.org/latest/youtube-dl.exe) and place it in their home directory or any other location on their [PATH](http://en.wikipedia.org/wiki/PATH_%28variable%29).
 
+OS X users can install **youtube-dl** with [Homebrew](http://brew.sh/).
+
+    brew install youtube-dl
+
+You can also use pip:
+
+    sudo pip install youtube-dl
+
 Alternatively, refer to the developer instructions below for how to check out and work with the git repository. For further options, including PGP signatures, see https://rg3.github.io/youtube-dl/download.html .
 
 # DESCRIPTION

test/helper.py

@@ -117,6 +117,7 @@ def expect_info_dict(self, expected_dict, got_dict):
                 u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got))
 
     # Check for the presence of mandatory fields
+    if got_dict.get('_type') != 'playlist':
         for key in ('id', 'url', 'title', 'ext'):
             self.assertTrue(got_dict.get(key), 'Missing mandatory field %s' % key)
     # Check for mandatory fields that are automatically set by YoutubeDL

test/test_all_urls.py

@@ -99,6 +99,7 @@ class TestAllURLsMatching(unittest.TestCase):
 
     def test_facebook_matching(self):
         self.assertTrue(FacebookIE.suitable('https://www.facebook.com/Shiniknoh#!/photo.php?v=10153317450565268'))
+        self.assertTrue(FacebookIE.suitable('https://www.facebook.com/cindyweather?fref=ts#!/photo.php?v=10152183998945793'))
 
     def test_no_duplicates(self):
         ies = gen_extractors()

test/test_download.py

@@ -63,15 +63,21 @@ def generator(test_case):
     def test_template(self):
         ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
         other_ies = [get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', [])]
+        is_playlist = any(k.startswith('playlist') for k in test_case)
+        test_cases = test_case.get(
+            'playlist', [] if is_playlist else [test_case])
+
         def print_skipping(reason):
             print('Skipping %s: %s' % (test_case['name'], reason))
         if not ie.working():
             print_skipping('IE marked as not _WORKING')
             return
-        if 'playlist' not in test_case:
-            info_dict = test_case.get('info_dict', {})
-            if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
+
+        for tc in test_cases:
+            info_dict = tc.get('info_dict', {})
+            if not tc.get('file') and not (info_dict.get('id') and info_dict.get('ext')):
                 raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?')
+
         if 'skip' in test_case:
             print_skipping(test_case['skip'])
             return
@@ -81,6 +87,9 @@ def generator(test_case):
                 return
 
         params = get_params(test_case.get('params', {}))
+        if is_playlist and 'playlist' not in test_case:
+            params.setdefault('extract_flat', True)
+            params.setdefault('skip_download', True)
 
         ydl = YoutubeDL(params)
         ydl.add_default_info_extractors()
@@ -93,7 +102,6 @@ def generator(test_case):
         def get_tc_filename(tc):
             return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
 
-        test_cases = test_case.get('playlist', [test_case])
         def try_rm_tcs_files():
             for tc in test_cases:
                 tc_filename = get_tc_filename(tc)
@@ -105,7 +113,10 @@ def generator(test_case):
             try_num = 1
             while True:
                 try:
-                    ydl.download([test_case['url']])
+                    # We're not using .download here sine that is just a shim
+                    # for outside error handling, and returns the exit code
+                    # instead of the result dict.
+                    res_dict = ydl.extract_info(test_case['url'])
                 except (DownloadError, ExtractorError) as err:
                     # Check if the exception is not a network related one
                     if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
@@ -121,6 +132,17 @@ def generator(test_case):
                 else:
                     break
 
+            if is_playlist:
+                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(
+                    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'])))
+
             for tc in test_cases:
                 tc_filename = get_tc_filename(tc)
                 if not test_case.get('params', {}).get('skip_download', False):

test/test_playlists.py

@@ -1,6 +1,17 @@
 #!/usr/bin/env python
 # encoding: utf-8
 
+## DEPRECATED FILE!
+# Add new tests to the extractors themselves, like this:
+# _TEST = {
+#    'url': 'http://example.com/playlist/42',
+#    'playlist_mincount': 99,
+#    'info_dict': {
+#        'id': '42',
+#        'title': 'Playlist number forty-two',
+#    }
+# }
+
 from __future__ import unicode_literals
 
 # Allow direct execution

test/test_utils.py

@@ -280,7 +280,7 @@ class TestUtil(unittest.TestCase):
         d = json.loads(stripped)
         self.assertEqual(d, [{"id": "532cb", "x": 3}])
 
-    def test_uppercase_escpae(self):
+    def test_uppercase_escape(self):
         self.assertEqual(uppercase_escape(u'aä'), u'aä')
         self.assertEqual(uppercase_escape(u'\\U0001d550'), u'𝕐')
 

youtube_dl/YoutubeDL.py

@@ -162,6 +162,7 @@ class YoutubeDL(object):
     default_search:    Prepend this string if an input url is not valid.
                        'auto' for elaborate guessing
     encoding:          Use this encoding instead of the system-specified.
+    extract_flat:      Do not resolve URLs, return the immediate result.
 
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
@@ -558,7 +559,12 @@ class YoutubeDL(object):
         Returns the resolved ie_result.
         """
 
-        result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system
+        result_type = ie_result.get('_type', 'video')
+
+        if self.params.get('extract_flat', False):
+            if result_type in ('url', 'url_transparent'):
+                return ie_result
+
         if result_type == 'video':
             self.add_extra_info(ie_result, extra_info)
             return self.process_video_result(ie_result, download=download)

youtube_dl/__init__.py

@@ -66,6 +66,10 @@ __authors__  = (
     'Naglis Jonaitis',
     'Charles Chen',
     'Hassaan Ali',
+    'Dobrosław Żybort',
+    'David Fabijan',
+    'Sebastian Haas',
+    'Alexander Kirk',
 )
 
 __license__ = 'Public Domain'

youtube_dl/downloader/common.py

@@ -292,7 +292,7 @@ class FileDownloader(object):
 
     def real_download(self, filename, info_dict):
         """Real download process. Redefine in subclasses."""
-        raise NotImplementedError(u'This method must be implemented by sublcasses')
+        raise NotImplementedError(u'This method must be implemented by subclasses')
 
     def _hook_progress(self, status):
         for ph in self._progress_hooks:

youtube_dl/extractor/__init__.py

@@ -77,6 +77,10 @@ from .ebaumsworld import EbaumsWorldIE
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .eitb import EitbIE
+from .ellentv import (
+    EllenTVIE,
+    EllenTVClipsIE,
+)
 from .elpais import ElPaisIE
 from .empflix import EmpflixIE
 from .engadget import EngadgetIE
@@ -126,6 +130,7 @@ from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
 from .hotnewhiphop import HotNewHipHopIE
 from .howcast import HowcastIE
+from .howstuffworks import HowStuffWorksIE
 from .huffpost import HuffPostIE
 from .hypem import HypemIE
 from .iconosquare import IconosquareIE
@@ -146,6 +151,7 @@ from .ivi import (
 from .izlesene import IzleseneIE
 from .jadorecettepub import JadoreCettePubIE
 from .jeuxvideo import JeuxVideoIE
+from .jove import JoveIE
 from .jukebox import JukeboxIE
 from .justintv import JustinTVIE
 from .jpopsukitv import JpopsukiIE
@@ -177,10 +183,12 @@ from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mit import TechTVMITIE, MITIE, OCWMITIE
+from .mitele import MiTeleIE
 from .mixcloud import MixcloudIE
 from .mlb import MLBIE
 from .mpora import MporaIE
 from .mofosex import MofosexIE
+from .mojvideo import MojvideoIE
 from .mooshare import MooshareIE
 from .morningstar import MorningstarIE
 from .motherless import MotherlessIE
@@ -224,9 +232,12 @@ from .nrk import (
 from .ntv import NTVIE
 from .nytimes import NYTimesIE
 from .nuvid import NuvidIE
-from .oe1 import OE1IE
 from .ooyala import OoyalaIE
-from .orf import ORFIE
+from .orf import (
+    ORFTVthekIE,
+    ORFOE1IE,
+    ORFFM4IE,
+)
 from .parliamentliveuk import ParliamentLiveUKIE
 from .pbs import PBSIE
 from .photobucket import PhotobucketIE
@@ -247,6 +258,7 @@ from .ro220 import Ro220IE
 from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
 from .rtbf import RTBFIE
+from .rtlnl import RtlXlIE
 from .rtlnow import RTLnowIE
 from .rts import RTSIE
 from .rtve import RTVEALaCartaIE
@@ -327,7 +339,7 @@ from .tutv import TutvIE
 from .tvigle import TvigleIE
 from .tvp import TvpIE
 from .tvplay import TVPlayIE
-from.ubu import UbuIE
+from .ubu import UbuIE
 from .udemy import (
     UdemyIE,
     UdemyCourseIE
@@ -383,6 +395,7 @@ from .wistia import WistiaIE
 from .worldstarhiphop import WorldStarHipHopIE
 from .wrzuta import WrzutaIE
 from .xbef import XBefIE
+from .xboxclips import XboxClipsIE
 from .xhamster import XHamsterIE
 from .xnxx import XNXXIE
 from .xvideos import XVideosIE

youtube_dl/extractor/appletrailers.py

@@ -6,6 +6,7 @@ import json
 from .common import InfoExtractor
 from ..utils import (
     compat_urlparse,
+    int_or_none,
 )
 
 
@@ -110,8 +111,8 @@ class AppleTrailersIE(InfoExtractor):
                 formats.append({
                     'url': format_url,
                     'format': format['type'],
-                    'width': format['width'],
-                    'height': int(format['height']),
+                    'width': int_or_none(format['width']),
+                    'height': int_or_none(format['height']),
                 })
 
             self._sort_formats(formats)

youtube_dl/extractor/ard.py

@@ -51,6 +51,9 @@ class ARDIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
+        if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage:
+            raise ExtractorError('Video %s is no longer available' % video_id, expected=True)
+
         title = self._html_search_regex(
             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
              r'<meta name="dcterms.title" content="(.*?)"/>',

youtube_dl/extractor/arte.py

@@ -109,15 +109,19 @@ class ArteTVPlus7IE(InfoExtractor):
             regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
             return any(re.match(r, f['versionCode']) for r in regexes)
         # Some formats may not be in the same language as the url
+        # TODO: Might want not to drop videos that does not match requested language
+        # but to process those formats with lower precedence
         formats = filter(_match_lang, all_formats)
         formats = list(formats)  # in python3 filter returns an iterator
         if not formats:
             # Some videos are only available in the 'Originalversion'
             # they aren't tagged as being in French or German
-            if all(f['versionCode'] == 'VO' or f['versionCode'] == 'VA' for f in all_formats):
+            # Sometimes there are neither videos of requested lang code
+            # nor original version videos available
+            # For such cases we just take all_formats as is
             formats = all_formats
-            else:
-                raise ExtractorError(u'The formats list is empty')
+            if not formats:
+                raise ExtractorError('The formats list is empty')
 
         if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
             def sort_key(f):

youtube_dl/extractor/common.py

@@ -463,8 +463,9 @@ class InfoExtractor(object):
         return self._og_search_property('title', html, **kargs)
 
     def _og_search_video_url(self, html, name='video url', secure=True, **kargs):
-        regexes = self._og_regexes('video')
-        if secure: regexes = self._og_regexes('video:secure_url') + regexes
+        regexes = self._og_regexes('video') + self._og_regexes('video:url')
+        if secure:
+            regexes = self._og_regexes('video:secure_url') + regexes
         return self._html_search_regex(regexes, html, name, **kargs)
 
     def _og_search_url(self, html, **kargs):

youtube_dl/extractor/dfb.py

@@ -30,7 +30,7 @@ class DFBIE(InfoExtractor):
             video_id)
         video_info = player_info.find('video')
 
-        f4m_info = self._download_xml(video_info.find('url').text, video_id)
+        f4m_info = self._download_xml(self._proto_relative_url(video_info.find('url').text.strip()), video_id)
         token_el = f4m_info.find('token')
         manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth'] + '&hdcore=3.2.0'
 

youtube_dl/extractor/ellentv.py

@@ -0,0 +1,79 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    parse_iso8601,
+)
+
+
+class EllenTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?ellentv\.com/videos/(?P<id>[a-z0-9_-]+)'
+    _TEST = {
+        'url': 'http://www.ellentv.com/videos/0-7jqrsr18/',
+        'md5': 'e4af06f3bf0d5f471921a18db5764642',
+        'info_dict': {
+            'id': '0-7jqrsr18',
+            'ext': 'mp4',
+            'title': 'What\'s Wrong with These Photos? A Whole Lot',
+            'timestamp': 1406876400,
+            'upload_date': '20140801',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        timestamp = parse_iso8601(self._search_regex(
+            r'<span class="publish-date"><time datetime="([^"]+)">',
+            webpage, 'timestamp'))
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'url': self._html_search_meta('VideoURL', webpage, 'url'),
+            'timestamp': timestamp,
+        }
+
+
+class EllenTVClipsIE(InfoExtractor):
+    IE_NAME = 'EllenTV:clips'
+    _VALID_URL = r'https?://(?:www\.)?ellentv\.com/episodes/(?P<id>[a-z0-9_-]+)'
+    _TEST = {
+        'url': 'http://www.ellentv.com/episodes/meryl-streep-vanessa-hudgens/',
+        'info_dict': {
+            'id': 'meryl-streep-vanessa-hudgens',
+            'title': 'Meryl Streep, Vanessa Hudgens',
+        },
+        'playlist_mincount': 9,
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, playlist_id)
+        playlist = self._extract_playlist(webpage)
+
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'title': self._og_search_title(webpage),
+            'entries': self._extract_entries(playlist)
+        }
+
+    def _extract_playlist(self, webpage):
+        json_string = self._search_regex(r'playerView.addClips\(\[\{(.*?)\}\]\);', webpage, 'json')
+        try:
+            return json.loads("[{" + json_string + "}]")
+        except ValueError as ve:
+            raise ExtractorError('Failed to download JSON', cause=ve)
+
+    def _extract_entries(self, playlist):
+        return [self.url_result(item['url'], 'EllenTV') for item in playlist]

youtube_dl/extractor/escapist.py

@@ -36,7 +36,7 @@ class EscapistIE(InfoExtractor):
             r'<meta name="description" content="([^"]*)"',
             webpage, 'description', fatal=False)
 
-        playerUrl = self._og_search_video_url(webpage, name=u'player URL')
+        playerUrl = self._og_search_video_url(webpage, name='player URL')
 
         title = self._html_search_regex(
             r'<meta name="title" content="([^"]*)"',

youtube_dl/extractor/facebook.py

@@ -20,7 +20,7 @@ from ..utils import (
 class FacebookIE(InfoExtractor):
     _VALID_URL = r'''(?x)
         https?://(?:\w+\.)?facebook\.com/
-        (?:[^#?]*\#!/)?
+        (?:[^#]*?\#!/)?
         (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
         (?:v|video_id)=(?P<id>[0-9]+)
         (?:.*)'''

youtube_dl/extractor/firedrive.py

@@ -42,7 +42,6 @@ class FiredriveIE(InfoExtractor):
         fields = dict(re.findall(r'''(?x)<input\s+
             type="hidden"\s+
             name="([^"]+)"\s+
-            (?:id="[^"]+"\s+)?
             value="([^"]*)"
             ''', webpage))
 
@@ -66,7 +65,7 @@ class FiredriveIE(InfoExtractor):
         ext = self._search_regex(r'type:\s?\'([^\']+)\',',
                                  webpage, 'extension', fatal=False)
         video_url = self._search_regex(
-            r'file:\s?\'(http[^\']+)\',', webpage, 'file url')
+            r'file:\s?loadURL\(\'(http[^\']+)\'\),', webpage, 'file url')
 
         formats = [{
             'format_id': 'sd',

youtube_dl/extractor/generic.py

@@ -706,6 +706,13 @@ class GenericIE(InfoExtractor):
             url = unescapeHTML(mobj.group('url'))
             return self.url_result(url, ie='MTVServicesEmbedded')
 
+        # Look for embedded yahoo player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
+            webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Yahoo')
+
         # Start with something easy: JW Player in SWFObject
         found = re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
         if not found:

youtube_dl/extractor/howstuffworks.py

@@ -0,0 +1,134 @@
+from __future__ import unicode_literals
+
+import re
+import json
+import random
+import string
+
+from .common import InfoExtractor
+from ..utils import find_xpath_attr
+
+
+class HowStuffWorksIE(InfoExtractor):
+    _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P<id>.+?)-video\.htm'
+    _TESTS = [
+        {
+            'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm',
+            'info_dict': {
+                'id': '450221',
+                'display_id': 'cool-jobs-iditarod-musher',
+                'ext': 'flv',
+                'title': 'Cool Jobs - Iditarod Musher',
+                'description': 'md5:82bb58438a88027b8186a1fccb365f90',
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                # md5 is not consistent
+                'skip_download': True
+            }
+        },
+        {
+            'url': 'http://adventure.howstuffworks.com/39516-deadliest-catch-jakes-farewell-pots-video.htm',
+            'info_dict': {
+                'id': '553470',
+                'display_id': 'deadliest-catch-jakes-farewell-pots',
+                'ext': 'mp4',
+                'title': 'Deadliest Catch: Jake\'s Farewell Pots',
+                'description': 'md5:9632c346d5e43ee238028c9cefd8dbbc',
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                # md5 is not consistent
+                'skip_download': True
+            }
+        },
+        {
+            'url': 'http://entertainment.howstuffworks.com/arts/2706-sword-swallowing-1-by-dan-meyer-video.htm',
+            'info_dict': {
+                'id': '440011',
+                'display_id': 'sword-swallowing-1-by-dan-meyer',
+                'ext': 'flv',
+                'title': 'Sword Swallowing #1 by Dan Meyer',
+                'description': 'md5:b2409e88172913e2e7d3d1159b0ef735',
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                # md5 is not consistent
+                'skip_download': True
+            }
+        },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('id')
+        webpage = self._download_webpage(url, display_id)
+
+        content_id = self._search_regex(r'var siteSectionId="(\d+)";', webpage, 'content id')
+
+        mp4 = self._search_regex(
+            r'''(?xs)var\s+clip\s*=\s*{\s*
+                .+?\s*
+                content_id\s*:\s*%s\s*,\s*
+                .+?\s*
+                mp4\s*:\s*\[(.*?),?\]\s*
+                };\s*
+                videoData\.push\(clip\);''' % content_id,
+            webpage, 'mp4', fatal=False, default=None)
+
+        smil = self._download_xml(
+            'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % content_id,
+            content_id, 'Downloading video SMIL')
+
+        http_base = find_xpath_attr(
+            smil,
+            './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'),
+            'name',
+            'httpBase').get('content')
+
+        def random_string(str_len=0):
+            return ''.join([random.choice(string.ascii_uppercase) for _ in range(str_len)])
+
+        URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=%s&g=%s' % (random_string(5), random_string(12))
+
+        formats = []
+
+        if mp4:
+            for video in json.loads('[%s]' % mp4):
+                bitrate = video['bitrate']
+                fmt = {
+                    'url': video['src'].replace('http://pmd.video.howstuffworks.com', http_base) + URL_SUFFIX,
+                    'format_id': bitrate,
+                }
+                m = re.search(r'(?P<vbr>\d+)[Kk]', bitrate)
+                if m:
+                    fmt['vbr'] = int(m.group('vbr'))
+                formats.append(fmt)
+        else:
+            for video in smil.findall(
+                    './/{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')):
+                vbr = int(video.attrib['system-bitrate']) / 1000
+                formats.append({
+                    'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
+                    'format_id': '%dk' % vbr,
+                    'vbr': vbr,
+                })
+
+        self._sort_formats(formats)
+
+        title = self._og_search_title(webpage)
+        TITLE_SUFFIX = ' : HowStuffWorks'
+        if title.endswith(TITLE_SUFFIX):
+            title = title[:-len(TITLE_SUFFIX)]
+
+        description = self._og_search_description(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        return {
+            'id': content_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'formats': formats,
+        }

youtube_dl/extractor/jove.py

@@ -0,0 +1,80 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    unified_strdate
+)
+
+
+class JoveIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?jove\.com/video/(?P<id>[0-9]+)'
+    _CHAPTERS_URL = 'http://www.jove.com/video-chapters?videoid={video_id:}'
+    _TESTS = [
+        {
+            'url': 'http://www.jove.com/video/2744/electrode-positioning-montage-transcranial-direct-current',
+            'md5': '93723888d82dbd6ba8b3d7d0cd65dd2b',
+            'info_dict': {
+                'id': '2744',
+                'ext': 'mp4',
+                'title': 'Electrode Positioning and Montage in Transcranial Direct Current Stimulation',
+                'description': 'md5:015dd4509649c0908bc27f049e0262c6',
+                'thumbnail': 're:^https?://.*\.png$',
+                'upload_date': '20110523',
+            }
+        },
+        {
+            'url': 'http://www.jove.com/video/51796/culturing-caenorhabditis-elegans-axenic-liquid-media-creation',
+            'md5': '914aeb356f416811d911996434811beb',
+            'info_dict': {
+                'id': '51796',
+                'ext': 'mp4',
+                'title': 'Culturing Caenorhabditis elegans in Axenic Liquid Media and Creation of Transgenic Worms by Microparticle Bombardment',
+                'description': 'md5:35ff029261900583970c4023b70f1dc9',
+                'thumbnail': 're:^https?://.*\.png$',
+                'upload_date': '20140802',
+            }
+        },
+
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        chapters_id = self._html_search_regex(
+            r'/video-chapters\?videoid=([0-9]+)', webpage, 'chapters id')
+
+        chapters_xml = self._download_xml(
+            self._CHAPTERS_URL.format(video_id=chapters_id),
+            video_id, note='Downloading chapters XML',
+            errnote='Failed to download chapters XML')
+
+        video_url = chapters_xml.attrib.get('video')
+        if not video_url:
+            raise ExtractorError('Failed to get the video URL')
+
+        title = self._html_search_meta('citation_title', webpage, 'title')
+        thumbnail = self._og_search_thumbnail(webpage)
+        description = self._html_search_regex(
+            r'<div id="section_body_summary"><p class="jove_content">(.+?)</p>',
+            webpage, 'description', fatal=False)
+        publish_date = unified_strdate(self._html_search_meta(
+            'citation_publication_date', webpage, 'publish date', fatal=False))
+        comment_count = self._html_search_regex(
+            r'<meta name="num_comments" content="(\d+) Comments?"',
+            webpage, 'comment count', fatal=False)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+            'thumbnail': thumbnail,
+            'description': description,
+            'upload_date': publish_date,
+            'comment_count': comment_count,
+        }

youtube_dl/extractor/livestream.py

@@ -111,17 +111,28 @@ class LivestreamIE(InfoExtractor):
         event_name = mobj.group('event_name')
         webpage = self._download_webpage(url, video_id or event_name)
 
-        if video_id is None:
-            # This is an event page:
+        og_video = self._og_search_video_url(webpage, 'player url', fatal=False, default=None)
+        if og_video is None:
             config_json = self._search_regex(
                 r'window.config = ({.*?});', webpage, 'window config')
             info = json.loads(config_json)['event']
+
+            def is_relevant(vdata, vid):
+                result = vdata['type'] == 'video'
+                if video_id is not None:
+                    result = result and compat_str(vdata['data']['id']) == vid
+                return result
+
             videos = [self._extract_video_info(video_data['data'])
                       for video_data in info['feed']['data']
-                if video_data['type'] == 'video']
+                      if is_relevant(video_data, video_id)]
+            if video_id is None:
+                # This is an event page:
                 return self.playlist_result(videos, info['id'], info['full_name'])
             else:
-            og_video = self._og_search_video_url(webpage, 'player url')
+                if videos:
+                    return videos[0]
+        else:
             query_str = compat_urllib_parse_urlparse(og_video).query
             query = compat_urlparse.parse_qs(query_str)
             api_url = query['play_url'][0].replace('.smil', '')

youtube_dl/extractor/mitele.py

@@ -0,0 +1,60 @@
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    get_element_by_attribute,
+    parse_duration,
+    strip_jsonp,
+)
+
+
+class MiTeleIE(InfoExtractor):
+    IE_NAME = 'mitele.es'
+    _VALID_URL = r'http://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<episode>[^/]+)/'
+
+    _TEST = {
+        'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
+        'md5': '6a75fe9d0d3275bead0cb683c616fddb',
+        'info_dict': {
+            'id': '0fce117d',
+            'ext': 'mp4',
+            'title': 'Programa 144 - Tor, la web invisible',
+            'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
+            'display_id': 'programa-144',
+            'duration': 2913,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        episode = mobj.group('episode')
+        webpage = self._download_webpage(url, episode)
+        embed_data_json = self._search_regex(
+            r'MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data',
+            flags=re.DOTALL
+        ).replace('\'', '"')
+        embed_data = json.loads(embed_data_json)
+
+        info_url = embed_data['flashvars']['host']
+        info_el = self._download_xml(info_url, episode).find('./video/info')
+
+        video_link = info_el.find('videoUrl/link').text
+        token_query = compat_urllib_parse.urlencode({'id': video_link})
+        token_info = self._download_json(
+            'http://token.mitele.es/?' + token_query, episode,
+            transform_source=strip_jsonp
+        )
+
+        return {
+            'id': embed_data['videoId'],
+            'display_id': episode,
+            'title': info_el.find('title').text,
+            'url': token_info['tokenizedUrl'],
+            'description': get_element_by_attribute('class', 'text', webpage),
+            'thumbnail': info_el.find('thumb').text,
+            'duration': parse_duration(info_el.find('duration').text),
+        }

youtube_dl/extractor/mojvideo.py

@@ -0,0 +1,58 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    parse_duration,
+)
+
+
+class MojvideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?mojvideo\.com/video-(?P<display_id>[^/]+)/(?P<id>[a-f0-9]+)'
+    _TEST = {
+        'url': 'http://www.mojvideo.com/video-v-avtu-pred-mano-rdecelaska-alfi-nipic/3d1ed4497707730b2906',
+        'md5': 'f7fd662cc8ce2be107b0d4f2c0483ae7',
+        'info_dict': {
+            'id': '3d1ed4497707730b2906',
+            'display_id': 'v-avtu-pred-mano-rdecelaska-alfi-nipic',
+            'ext': 'mp4',
+            'title': 'V avtu pred mano rdečelaska - Alfi Nipič',
+            'thumbnail': 're:^http://.*\.jpg$',
+            'duration': 242,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        # XML is malformed
+        playerapi = self._download_webpage(
+            'http://www.mojvideo.com/playerapi.php?v=%s&t=1' % video_id, display_id)
+
+        if '<error>true</error>' in playerapi:
+            error_desc = self._html_search_regex(
+                r'<errordesc>([^<]*)</errordesc>', playerapi, 'error description', fatal=False)
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, error_desc), expected=True)
+
+        title = self._html_search_regex(
+            r'<title>([^<]+)</title>', playerapi, 'title')
+        video_url = self._html_search_regex(
+            r'<file>([^<]+)</file>', playerapi, 'video URL')
+        thumbnail = self._html_search_regex(
+            r'<preview>([^<]+)</preview>', playerapi, 'thumbnail', fatal=False)
+        duration = parse_duration(self._html_search_regex(
+            r'<duration>([^<]+)</duration>', playerapi, 'duration', fatal=False))
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'url': video_url,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+        }

youtube_dl/extractor/nowness.py

@@ -1,3 +1,4 @@
+# encoding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -8,19 +9,34 @@ from ..utils import ExtractorError
 
 
 class NownessIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?nowness\.com/[^?#]*?/(?P<id>[0-9]+)/(?P<slug>[^/]+?)(?:$|[?#])'
+    _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/[^?#]*?/(?P<id>[0-9]+)/(?P<slug>[^/]+?)(?:$|[?#])'
 
-    _TEST = {
+    _TESTS = [
+        {
             'url': 'http://www.nowness.com/day/2013/6/27/3131/candor--the-art-of-gesticulation',
             'md5': '068bc0202558c2e391924cb8cc470676',
             'info_dict': {
                 'id': '2520295746001',
                 'ext': 'mp4',
-            'description': 'Candor: The Art of Gesticulation',
-            'uploader': 'Nowness',
                 'title': 'Candor: The Art of Gesticulation',
+                'description': 'Candor: The Art of Gesticulation',
+                'thumbnail': 're:^https?://.*\.jpg',
+                'uploader': 'Nowness',
             }
+        },
+        {
+            'url': 'http://cn.nowness.com/day/2014/8/7/4069/kasper-bj-rke-ft-jaakko-eino-kalevi--tnr',
+            'md5': 'e79cf125e387216f86b2e0a5b5c63aa3',
+            'info_dict': {
+                'id': '3716354522001',
+                'ext': 'mp4',
+                'title': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR',
+                'description': 'Kasper Bjørke ft. Jaakko Eino Kalevi: TNR',
+                'thumbnail': 're:^https?://.*\.jpg',
+                'uploader': 'Nowness',
             }
+        },
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/oe1.py

@@ -1,40 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import calendar
-import datetime
-import re
-
-from .common import InfoExtractor
-
-# audios on oe1.orf.at are only available for 7 days, so we can't
-# add tests.
-
-
-class OE1IE(InfoExtractor):
-    IE_DESC = 'oe1.orf.at'
-    _VALID_URL = r'http://oe1\.orf\.at/programm/(?P<id>[0-9]+)'
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        show_id = mobj.group('id')
-
-        data = self._download_json(
-            'http://oe1.orf.at/programm/%s/konsole' % show_id,
-            show_id
-        )
-
-        timestamp = datetime.datetime.strptime('%s %s' % (
-            data['item']['day_label'],
-            data['item']['time']
-        ), '%d.%m.%Y %H:%M')
-        unix_timestamp = calendar.timegm(timestamp.utctimetuple())
-
-        return {
-            'id': show_id,
-            'title': data['item']['title'],
-            'url': data['item']['url_stream'],
-            'ext': 'mp3',
-            'description': data['item'].get('info'),
-            'timestamp': unix_timestamp
-        }

youtube_dl/extractor/ooyala.py

@@ -3,13 +3,17 @@ import re
 import json
 
 from .common import InfoExtractor
-from ..utils import unescapeHTML
+from ..utils import (
+    unescapeHTML,
+    ExtractorError,
+)
 
 
 class OoyalaIE(InfoExtractor):
     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
 
-    _TEST = {
+    _TESTS = [
+        {
             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
             'md5': '3f5cceb3a7bf461d6c29dc466cf8033c',
@@ -19,7 +23,18 @@ class OoyalaIE(InfoExtractor):
                 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
             },
-    }
+        }, {
+            # Only available for ipad
+            'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
+            'md5': '4b9754921fddb68106e48c142e2a01e6',
+            'info_dict': {
+                'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
+                'ext': 'mp4',
+                'title': 'Simulation Overview - Levels of Simulation',
+                'description': '',
+            },
+        },
+    ]
 
     @staticmethod
     def _url_for_embed_code(embed_code):
@@ -47,13 +62,30 @@ class OoyalaIE(InfoExtractor):
         player = self._download_webpage(player_url, embedCode)
         mobile_url = self._search_regex(r'mobile_player_url="(.+?)&device="',
                                         player, 'mobile player url')
-        mobile_player = self._download_webpage(mobile_url, embedCode)
+        # Looks like some videos are only available for particular devices
+        # (e.g. http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0
+        # is only available for ipad)
+        # Working around with fetching URLs for all the devices found starting with 'unknown'
+        # until we succeed or eventually fail for each device.
+        devices = re.findall(r'device\s*=\s*"([^"]+)";', player)
+        devices.remove('unknown')
+        devices.insert(0, 'unknown')
+        for device in devices:
+            mobile_player = self._download_webpage(
+                '%s&device=%s' % (mobile_url, device), embedCode,
+                'Downloading mobile player JS for %s device' % device)
             videos_info = self._search_regex(
                 r'var streams=window.oo_testEnv\?\[\]:eval\("\((\[{.*?}\])\)"\);',
-            mobile_player, 'info').replace('\\"','"')
-        videos_more_info = self._search_regex(r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, 'more info').replace('\\"','"')
+                mobile_player, 'info', fatal=False, default=None)
+            if videos_info:
+                break
+        if not videos_info:
+            raise ExtractorError('Unable to extract info')
+        videos_info = videos_info.replace('\\"', '"')
+        videos_more_info = self._search_regex(
+            r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, 'more info').replace('\\"', '"')
         videos_info = json.loads(videos_info)
-        videos_more_info =json.loads(videos_more_info)
+        videos_more_info = json.loads(videos_more_info)
 
         if videos_more_info.get('lineup'):
             videos = [self._extract_result(info, more_info) for (info, more_info) in zip(videos_info, videos_more_info['lineup'])]

youtube_dl/extractor/orf.py

@@ -3,6 +3,8 @@ from __future__ import unicode_literals
 
 import json
 import re
+import calendar
+import datetime
 
 from .common import InfoExtractor
 from ..utils import (
@@ -12,7 +14,9 @@ from ..utils import (
 )
 
 
-class ORFIE(InfoExtractor):
+class ORFTVthekIE(InfoExtractor):
+    IE_NAME = 'orf:tvthek'
+    IE_DESC = 'ORF TVthek'
     _VALID_URL = r'https?://tvthek\.orf\.at/(?:programs/.+?/episodes|topics/.+?|program/[^/]+)/(?P<id>\d+)'
 
     _TEST = {
@@ -105,3 +109,73 @@ class ORFIE(InfoExtractor):
             'entries': entries,
             'id': playlist_id,
         }
+
+
+# Audios on ORF radio are only available for 7 days, so we can't add tests.
+
+
+class ORFOE1IE(InfoExtractor):
+    IE_NAME = 'orf:oe1'
+    IE_DESC = 'Radio Österreich 1'
+    _VALID_URL = r'http://oe1\.orf\.at/programm/(?P<id>[0-9]+)'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        show_id = mobj.group('id')
+
+        data = self._download_json(
+            'http://oe1.orf.at/programm/%s/konsole' % show_id,
+            show_id
+        )
+
+        timestamp = datetime.datetime.strptime('%s %s' % (
+            data['item']['day_label'],
+            data['item']['time']
+        ), '%d.%m.%Y %H:%M')
+        unix_timestamp = calendar.timegm(timestamp.utctimetuple())
+
+        return {
+            'id': show_id,
+            'title': data['item']['title'],
+            'url': data['item']['url_stream'],
+            'ext': 'mp3',
+            'description': data['item'].get('info'),
+            'timestamp': unix_timestamp
+        }
+
+
+class ORFFM4IE(InfoExtractor):
+    IE_DESC = 'orf:fm4'
+    IE_DESC = 'radio FM4'
+    _VALID_URL = r'http://fm4\.orf\.at/7tage/?#(?P<date>[0-9]+)/(?P<show>\w+)'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        show_date = mobj.group('date')
+        show_id = mobj.group('show')
+
+        data = self._download_json(
+            'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
+            show_id
+        )
+
+        def extract_entry_dict(info, title, subtitle):
+            return {
+                'id': info['loopStreamId'].replace('.mp3', ''),
+                'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
+                'title': title,
+                'description': subtitle,
+                'duration': (info['end'] - info['start']) / 1000,
+                'timestamp': info['start'] / 1000,
+                'ext': 'mp3'
+            }
+
+        entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
+
+        return {
+            '_type': 'playlist',
+            'id': show_id,
+            'title': data['title'],
+            'description': data['subtitle'],
+            'entries': entries
+        }

youtube_dl/extractor/pbs.py

@@ -20,7 +20,8 @@ class PBSIE(InfoExtractor):
         )
     '''
 
-    _TEST = {
+    _TESTS = [
+        {
             'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
             'md5': 'ce1888486f0908d555a8093cac9a7362',
             'info_dict': {
@@ -30,7 +31,30 @@ class PBSIE(InfoExtractor):
                 'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
                 'duration': 3190,
             },
-    }
+        },
+        {
+            'url': 'http://www.pbs.org/wgbh/pages/frontline/losing-iraq/',
+            'md5': '143c98aa54a346738a3d78f54c925321',
+            'info_dict': {
+                'id': '2365297690',
+                'ext': 'mp4',
+                'title': 'Losing Iraq',
+                'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
+                'duration': 5050,
+            },
+        },
+        {
+            'url': 'http://www.pbs.org/newshour/bb/education-jan-june12-cyberschools_02-23/',
+            'md5': 'b19856d7f5351b17a5ab1dc6a64be633',
+            'info_dict': {
+                'id': '2201174722',
+                'ext': 'mp4',
+                'title': 'Cyber Schools Gain Popularity, but Quality Questions Persist',
+                'description': 'md5:5871c15cba347c1b3d28ac47a73c7c28',
+                'duration': 801,
+            },
+        },
+    ]
 
     def _extract_ids(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -40,10 +64,13 @@ class PBSIE(InfoExtractor):
         if presumptive_id:
             webpage = self._download_webpage(url, display_id)
 
-            # frontline video embed
+            MEDIA_ID_REGEXES = [
+                r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",  # frontline video embed
+                r'class="coveplayerid">([^<]+)<',                       # coveplayer
+            ]
+
             media_id = self._search_regex(
-                r"div\s*:\s*'videoembed'\s*,\s*mediaid\s*:\s*'(\d+)'",
-                webpage, 'frontline video ID', fatal=False, default=None)
+                MEDIA_ID_REGEXES, webpage, 'media ID', fatal=False, default=None)
             if media_id:
                 return media_id, presumptive_id
 

youtube_dl/extractor/reverbnation.py

@@ -1,23 +1,23 @@
 from __future__ import unicode_literals
 
 import re
-import time
 
 from .common import InfoExtractor
-from ..utils import strip_jsonp
+from ..utils import str_or_none
 
 
 class ReverbNationIE(InfoExtractor):
     _VALID_URL = r'^https?://(?:www\.)?reverbnation\.com/.*?/song/(?P<id>\d+).*?$'
     _TESTS = [{
         'url': 'http://www.reverbnation.com/alkilados/song/16965047-mona-lisa',
-        'file': '16965047.mp3',
         'md5': '3da12ebca28c67c111a7f8b262d3f7a7',
         'info_dict': {
+            "id": "16965047",
+            "ext": "mp3",
             "title": "MONA LISA",
             "uploader": "ALKILADOS",
-            "uploader_id": 216429,
-            "thumbnail": "//gp1.wac.edgecastcdn.net/802892/production_public/Photo/13761700/image/1366002176_AVATAR_MONA_LISA.jpg"
+            "uploader_id": "216429",
+            "thumbnail": "re:^https://gp1\.wac\.edgecastcdn\.net/.*?\.jpg$"
         },
     }]
 
@@ -26,10 +26,8 @@ class ReverbNationIE(InfoExtractor):
         song_id = mobj.group('id')
 
         api_res = self._download_json(
-            'https://api.reverbnation.com/song/%s?callback=api_response_5&_=%d'
-                % (song_id, int(time.time() * 1000)),
+            'https://api.reverbnation.com/song/%s' % song_id,
             song_id,
-            transform_source=strip_jsonp,
             note='Downloading information of song %s' % song_id
         )
 
@@ -38,8 +36,9 @@ class ReverbNationIE(InfoExtractor):
             'title': api_res.get('name'),
             'url': api_res.get('url'),
             'uploader': api_res.get('artist', {}).get('name'),
-            'uploader_id': api_res.get('artist', {}).get('id'),
-            'thumbnail': api_res.get('image', api_res.get('thumbnail')),
+            'uploader_id': str_or_none(api_res.get('artist', {}).get('id')),
+            'thumbnail': self._proto_relative_url(
+                api_res.get('image', api_res.get('thumbnail'))),
             'ext': 'mp3',
             'vcodec': 'none',
         }

youtube_dl/extractor/rtlnl.py

@@ -0,0 +1,52 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class RtlXlIE(InfoExtractor):
+    IE_NAME = 'rtlxl.nl'
+    _VALID_URL = r'https?://www\.rtlxl\.nl/#!/[^/]+/(?P<uuid>[^/?]+)'
+
+    _TEST = {
+        'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/6e4203a6-0a5e-3596-8424-c599a59e0677',
+        'info_dict': {
+            'id': '6e4203a6-0a5e-3596-8424-c599a59e0677',
+            'ext': 'flv',
+            'title': 'RTL Nieuws - Laat',
+            'description': 'Dagelijks het laatste nieuws uit binnen- en '
+                'buitenland. Voor nog meer nieuws kunt u ook gebruikmaken van '
+                'onze mobiele apps.',
+            'timestamp': 1408051800,
+            'upload_date': '20140814',
+        },
+        'params': {
+            # We download the first bytes of the first fragment, it can't be
+            # processed by the f4m downloader beacuse it isn't complete
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        uuid = mobj.group('uuid')
+
+        info = self._download_json(
+            'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=flash/' % uuid,
+            uuid)
+        meta = info['meta']
+        material = info['material'][0]
+        episode_info = info['episodes'][0]
+
+        f4m_url = 'http://manifest.us.rtl.nl' + material['videopath']
+        progname = info['abstracts'][0]['name']
+        subtitle = material['title'] or info['episodes'][0]['name']
+
+        return {
+            'id': uuid,
+            'title': '%s - %s' % (progname, subtitle), 
+            'formats': self._extract_f4m_formats(f4m_url, uuid),
+            'timestamp': material['original_date'],
+            'description': episode_info['synopsis'],
+        }

youtube_dl/extractor/shared.py

@@ -17,11 +17,11 @@ class SharedIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://shared.sx/0060718775',
-        'md5': '53e1c58fc3e777ae1dfe9e57ba2f9c72',
+        'md5': '106fefed92a8a2adb8c98e6a0652f49b',
         'info_dict': {
             'id': '0060718775',
             'ext': 'mp4',
-            'title': 'Big Buck Bunny Trailer',
+            'title': 'Bmp4',
         },
     }
 

youtube_dl/extractor/teamcoco.py

@@ -37,7 +37,7 @@ class TeamcocoIE(InfoExtractor):
         video_id = mobj.group("video_id")
         if not video_id:
             video_id = self._html_search_regex(
-                r'<article class="video" data-id="(\d+?)"',
+                r'data-node-id="(\d+?)"',
                 webpage, 'video id')
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id

youtube_dl/extractor/vimeo.py

@@ -121,6 +121,21 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
                 'videopassword': 'youtube-dl',
             },
         },
+        {
+            'url': 'http://vimeo.com/channels/keypeele/75629013',
+            'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
+            'note': 'Video is freely available via original URL '
+                    'and protected with password when accessed via http://vimeo.com/75629013',
+            'info_dict': {
+                'id': '75629013',
+                'ext': 'mp4',
+                'title': 'Key & Peele: Terrorist Interrogation',
+                'description': 'md5:8678b246399b070816b12313e8b4eb5c',
+                'uploader_id': 'atencio',
+                'uploader': 'Peter Atencio',
+                'duration': 187,
+            },
+        },
         {
             'url': 'http://vimeo.com/76979871',
             'md5': '3363dd6ffebe3784d56f4132317fd446',
@@ -196,8 +211,6 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
         video_id = mobj.group('id')
         if mobj.group('pro') or mobj.group('player'):
             url = 'http://player.vimeo.com/video/' + video_id
-        else:
-            url = 'https://vimeo.com/' + video_id
 
         # Retrieve video webpage to extract further information
         request = compat_urllib_request.Request(url, None, headers)
@@ -263,7 +276,7 @@ class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
         if video_thumbnail is None:
             video_thumbs = config["video"].get("thumbs")
             if video_thumbs and isinstance(video_thumbs, dict):
-                _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in video_thumbs.items())[-1]
+                _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 
         # Extract video description
         video_description = None

youtube_dl/extractor/vube.py

@@ -1,10 +1,12 @@
 from __future__ import unicode_literals
 
-import json
 import re
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    compat_str,
+)
 
 
 class VubeIE(InfoExtractor):
@@ -29,6 +31,7 @@ class VubeIE(InfoExtractor):
                 'like_count': int,
                 'dislike_count': int,
                 'comment_count': int,
+                'categories': ['pop', 'music', 'cover', 'singing', 'jessie j', 'price tag', 'chiara grispo'],
             }
         },
         {
@@ -47,6 +50,7 @@ class VubeIE(InfoExtractor):
                 'like_count': int,
                 'dislike_count': int,
                 'comment_count': int,
+                'categories': ['seraina', 'jessica', 'krewella', 'alive'],
             }
         }, {
             'url': 'http://vube.com/vote/Siren+Gene/0nmsMY5vEq?n=2&t=s',
@@ -56,13 +60,15 @@ class VubeIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'Frozen - Let It Go Cover by Siren Gene',
                 'description': 'My rendition of "Let It Go" originally sung by Idina Menzel.',
-                'uploader': 'Siren Gene',
-                'uploader_id': 'Siren',
                 'thumbnail': 're:^http://frame\.thestaticvube\.com/snap/[0-9x]+/10283ab622a-86c9-4681-51f2-30d1f65774af\.jpg$',
+                'uploader': 'Siren',
+                'timestamp': 1395448018,
+                'upload_date': '20140322',
                 'duration': 221.788,
                 'like_count': int,
                 'dislike_count': int,
                 'comment_count': int,
+                'categories': ['let it go', 'cover', 'idina menzel', 'frozen', 'singing', 'disney', 'siren gene'],
             }
         }
     ]
@@ -71,46 +77,39 @@ class VubeIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
 
-        webpage = self._download_webpage(url, video_id)
-        data_json = self._search_regex(
-            r'(?s)window\["(?:tapiVideoData|vubeOriginalVideoData)"\]\s*=\s*(\{.*?\n});\n',
-            webpage, 'video data'
-        )
-        data = json.loads(data_json)
-        video = (
-            data.get('video') or
-            data)
-        assert isinstance(video, dict)
+        video = self._download_json(
+            'http://vube.com/t-api/v1/video/%s' % video_id, video_id, 'Downloading video JSON')
 
         public_id = video['public_id']
 
-        formats = [
-            {
-                'url': 'http://video.thestaticvube.com/video/%s/%s.mp4' % (fmt['media_resolution_id'], public_id),
-                'height': int(fmt['height']),
-                'abr': int(fmt['audio_bitrate']),
-                'vbr': int(fmt['video_bitrate']),
-                'format_id': fmt['media_resolution_id']
-            } for fmt in video['mtm'] if fmt['transcoding_status'] == 'processed'
-        ]
+        formats = []
+
+        for media in video['media'].get('video', []) + video['media'].get('audio', []):
+            if media['transcoding_status'] != 'processed':
+                continue
+            fmt = {
+                'url': 'http://video.thestaticvube.com/video/%s/%s.mp4' % (media['media_resolution_id'], public_id),
+                'abr': int(media['audio_bitrate']),
+                'format_id': compat_str(media['media_resolution_id']),
+            }
+            vbr = int(media['video_bitrate'])
+            if vbr:
+                fmt.update({
+                    'vbr': vbr,
+                    'height': int(media['height']),
+                })
+            formats.append(fmt)
 
         self._sort_formats(formats)
 
         title = video['title']
         description = video.get('description')
-        thumbnail = self._proto_relative_url(
-            video.get('thumbnail') or video.get('thumbnail_src'),
-            scheme='http:')
-        uploader = data.get('user', {}).get('channel', {}).get('name') or video.get('user_alias')
-        uploader_id = data.get('user', {}).get('name')
+        thumbnail = self._proto_relative_url(video.get('thumbnail_src'), scheme='http:')
+        uploader = video.get('user_alias') or video.get('channel')
         timestamp = int_or_none(video.get('upload_time'))
         duration = video['duration']
         view_count = video.get('raw_view_count')
-        like_count = video.get('rlikes')
-        if like_count is None:
         like_count = video.get('total_likes')
-        dislike_count = video.get('rhates')
-        if dislike_count is None:
         dislike_count = video.get('total_hates')
 
         comments = video.get('comments')
@@ -124,6 +123,8 @@ class VubeIE(InfoExtractor):
         else:
             comment_count = len(comments)
 
+        categories = [tag['text'] for tag in video['tags']]
+
         return {
             'id': video_id,
             'formats': formats,
@@ -131,11 +132,11 @@ class VubeIE(InfoExtractor):
             'description': description,
             'thumbnail': thumbnail,
             'uploader': uploader,
-            'uploader_id': uploader_id,
             'timestamp': timestamp,
             'duration': duration,
             'view_count': view_count,
             'like_count': like_count,
             'dislike_count': dislike_count,
             'comment_count': comment_count,
+            'categories': categories,
         }

youtube_dl/extractor/xboxclips.py

@@ -0,0 +1,57 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_iso8601,
+    float_or_none,
+    int_or_none,
+)
+
+
+class XboxClipsIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?xboxclips\.com/video\.php\?.*vid=(?P<id>[\w-]{36})'
+    _TEST = {
+        'url': 'https://xboxclips.com/video.php?uid=2533274823424419&gamertag=Iabdulelah&vid=074a69a9-5faf-46aa-b93b-9909c1720325',
+        'md5': 'fbe1ec805e920aeb8eced3c3e657df5d',
+        'info_dict': {
+            'id': '074a69a9-5faf-46aa-b93b-9909c1720325',
+            'ext': 'mp4',
+            'title': 'Iabdulelah playing Upload Studio',
+            'filesize_approx': 28101836.8,
+            'timestamp': 1407388500,
+            'upload_date': '20140807',
+            'duration': 56,
+        }
+    }
+
+    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_url = self._html_search_regex(
+            r'>Link: <a href="([^"]+)">', webpage, 'video URL')
+        title = self._html_search_regex(
+            r'<title>XboxClips \| ([^<]+)</title>', webpage, 'title')
+        timestamp = parse_iso8601(self._html_search_regex(
+            r'>Recorded: ([^<]+)<', webpage, 'upload date', fatal=False))
+        filesize = float_or_none(self._html_search_regex(
+            r'>Size: ([\d\.]+)MB<', webpage, 'file size', fatal=False), invscale=1024 * 1024)
+        duration = int_or_none(self._html_search_regex(
+            r'>Duration: (\d+) Seconds<', webpage, 'duration', fatal=False))
+        view_count = int_or_none(self._html_search_regex(
+            r'>Views: (\d+)<', webpage, 'view count', fatal=False))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'timestamp': timestamp,
+            'filesize_approx': filesize,
+            'duration': duration,
+            'view_count': view_count,
+        }

youtube_dl/extractor/yahoo.py

@@ -15,7 +15,7 @@ from ..utils import (
 
 class YahooIE(InfoExtractor):
     IE_DESC = 'Yahoo screen and movies'
-    _VALID_URL = r'https?://(?:screen|movies)\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html'
+    _VALID_URL = r'(?P<url>https?://(?:screen|movies)\.yahoo\.com/.*?-(?P<id>[0-9]+)(?:-[a-z]+)?\.html)'
     _TESTS = [
         {
             'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
@@ -46,12 +46,23 @@ class YahooIE(InfoExtractor):
                 'title': 'The World Loves Spider-Man',
                 'description': '''People all over the world are celebrating the release of \"The Amazing Spider-Man 2.\" We're taking a look at the enthusiastic response Spider-Man has received from viewers all over the world.''',
             }
+        },
+        {
+            'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
+            'md5': '60e8ac193d8fb71997caa8fce54c6460',
+            'info_dict': {
+                'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
+                'ext': 'mp4',
+                'title': "Yahoo Saves 'Community'",
+                'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
             }
+        },
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
+        url = mobj.group('url')
         webpage = self._download_webpage(url, video_id)
 
         items_json = self._search_regex(

youtube_dl/extractor/youtube.py

@@ -374,6 +374,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                 return lambda s: u''.join(s[i] for i in cache_spec)
             except IOError:
                 pass  # No cache available
+            except ValueError:
+                try:
+                    file_size = os.path.getsize(cache_fn)
+                except (OSError, IOError) as oe:
+                    file_size = str(oe)
+                self._downloader.report_warning(
+                    u'Cache %s failed (%s)' % (cache_fn, file_size))
 
         if player_type == 'js':
             code = self._download_webpage(

youtube_dl/utils.py

@@ -24,6 +24,7 @@ import socket
 import struct
 import subprocess
 import sys
+import tempfile
 import traceback
 import xml.etree.ElementTree
 import zlib
@@ -228,18 +229,36 @@ else:
         assert type(s) == type(u'')
         print(s)
 
-# In Python 2.x, json.dump expects a bytestream.
-# In Python 3.x, it writes to a character stream
-if sys.version_info < (3,0):
-    def write_json_file(obj, fn):
-        with open(fn, 'wb') as f:
-            json.dump(obj, f)
-else:
-    def write_json_file(obj, fn):
-        with open(fn, 'w', encoding='utf-8') as f:
-            json.dump(obj, f)
 
-if sys.version_info >= (2,7):
+def write_json_file(obj, fn):
+    """ Encode obj as JSON and write it to fn, atomically """
+
+    # In Python 2.x, json.dump expects a bytestream.
+    # In Python 3.x, it writes to a character stream
+    if sys.version_info < (3, 0):
+        mode = 'wb'
+        encoding = None
+    else:
+        mode = 'w'
+        encoding = 'utf-8'
+    tf = tempfile.NamedTemporaryFile(
+        suffix='.tmp', prefix=os.path.basename(fn) + '.',
+        dir=os.path.dirname(fn),
+        delete=False)
+
+    try:
+        with tf:
+            json.dump(obj, tf)
+        os.rename(tf.name, fn)
+    except:
+        try:
+            os.remove(tf.name)
+        except OSError:
+            pass
+        raise
+
+
+if sys.version_info >= (2, 7):
     def find_xpath_attr(node, xpath, key, val):
         """ Find the xpath xpath[@key=val] """
         assert re.match(r'^[a-zA-Z-]+$', key)
@@ -827,6 +846,7 @@ def unified_strdate(date_str):
         '%b %dnd %Y %I:%M%p',
         '%b %dth %Y %I:%M%p',
         '%Y-%m-%d',
+        '%Y/%m/%d',
         '%d.%m.%Y',
         '%d/%m/%Y',
         '%Y/%m/%d %H:%M:%S',
@@ -1273,9 +1293,15 @@ def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1):
     if get_attr:
         if v is not None:
             v = getattr(v, get_attr, None)
+    if v == '':
+        v = None
     return default if v is None else (int(v) * invscale // scale)
 
 
+def str_or_none(v, default=None):
+    return default if v is None else compat_str(v)
+
+
 def str_to_int(int_str):
     if int_str is None:
         return None

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.08.02.1'
+__version__ = '2014.08.21.2'