release 2014.08.24.5

This was broken in the mean time, so add a test case to make sure it doesn't break silently again.

youtube_dl/__init__.py

@@ -71,6 +71,7 @@ __authors__  = (
     'Sebastian Haas',
     'Alexander Kirk',
     'Erik Johnson',
+    'Keith Beckman',
 )
 
 __license__ = 'Public Domain'

youtube_dl/downloader/http.py

@@ -27,8 +27,16 @@ class HttpFD(FileDownloader):
             headers['Youtubedl-user-agent'] = info_dict['user_agent']
         if 'http_referer' in info_dict:
             headers['Referer'] = info_dict['http_referer']
-        basic_request = compat_urllib_request.Request(url, None, headers)
-        request = compat_urllib_request.Request(url, None, headers)
+        add_headers = info_dict.get('http_headers')
+        if add_headers:
+            headers.update(add_headers)
+        data = info_dict.get('http_post_data')
+        http_method = info_dict.get('http_method')
+        basic_request = compat_urllib_request.Request(url, data, headers)
+        request = compat_urllib_request.Request(url, data, headers)
+        if http_method is not None:
+            basic_request.get_method = lambda: http_method
+            request.get_method = lambda: http_method
 
         is_test = self.params.get('test', False)
 

youtube_dl/extractor/__init__.py

@@ -88,6 +88,7 @@ from .engadget import EngadgetIE
 from .escapist import EscapistIE
 from .everyonesmixtape import EveryonesMixtapeIE
 from .exfm import ExfmIE
+from .expotv import ExpoTVIE
 from .extremetube import ExtremeTubeIE
 from .facebook import FacebookIE
 from .faz import FazIE
@@ -126,6 +127,7 @@ from .googleplus import GooglePlusIE
 from .googlesearch import GoogleSearchIE
 from .gorillavid import GorillaVidIE
 from .goshgay import GoshgayIE
+from .grooveshark import GroovesharkIE
 from .hark import HarkIE
 from .helsinki import HelsinkiIE
 from .hentaistigma import HentaiStigmaIE
@@ -183,6 +185,7 @@ from .malemotion import MalemotionIE
 from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
+from .ministrygrid import MinistryGridIE
 from .mit import TechTVMITIE, MITIE, OCWMITIE
 from .mitele import MiTeleIE
 from .mixcloud import MixcloudIE
@@ -389,6 +392,7 @@ from .vuclip import VuClipIE
 from .vulture import VultureIE
 from .washingtonpost import WashingtonPostIE
 from .wat import WatIE
+from .wayofthemaster import WayOfTheMasterIE
 from .wdr import (
     WDRIE,
     WDRMobileIE,

youtube_dl/extractor/arte.py

@@ -177,16 +177,26 @@ class ArteTVPlus7IE(InfoExtractor):
 # It also uses the arte_vp_url url from the webpage to extract the information
 class ArteTVCreativeIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:creative'
-    _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/magazine?/(?P<id>.+)'
+    _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/(?:magazine?/)?(?P<id>[^?#]+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
         'info_dict': {
-            'id': '050489-002',
+            'id': '72176',
             'ext': 'mp4',
-            'title': 'Agentur Amateur / Agence Amateur #2 : Corporate Design',
+            'title': 'Folge 2 - Corporate Design',
+            'upload_date': '20131004',
         },
-    }
+    }, {
+        'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion',
+        'info_dict': {
+            'id': '160676',
+            'ext': 'mp4',
+            'title': 'Monty Python live (mostly)',
+            'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n',
+            'upload_date': '20140805',
+        }
+    }]
 
 
 class ArteTVFutureIE(ArteTVPlus7IE):

youtube_dl/extractor/bliptv.py

@@ -15,7 +15,7 @@ from ..utils import (
 
 
 class BlipTVIE(SubtitlesInfoExtractor):
-    _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+]+)))'
+    _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+_TESTS]+)))'
 
     _TESTS = [
         {
@@ -49,6 +49,21 @@ class BlipTVIE(SubtitlesInfoExtractor):
                 'uploader_id': '792887',
                 'duration': 279,
             }
+        },
+        {
+            # https://bugzilla.redhat.com/show_bug.cgi?id=967465
+            'url': 'http://a.blip.tv/api.swf#h6Uag5KbVwI',
+            'md5': '314e87b1ebe7a48fcbfdd51b791ce5a6',
+            'info_dict': {
+                'id': '6573122',
+                'ext': 'mov',
+                'upload_date': '20130520',
+                'description': 'Two hapless space marines argue over what to do when they realize they have an astronomically huge problem on their hands.',
+                'title': 'Red vs. Blue Season 11 Trailer',
+                'timestamp': 1369029609,
+                'uploader': 'redvsblue',
+                'uploader_id': '792887',
+            }
         }
     ]
 
@@ -150,7 +165,7 @@ class BlipTVIE(SubtitlesInfoExtractor):
 
 
 class BlipTVUserIE(InfoExtractor):
-    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
+    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$'
     _PAGE_SIZE = 12
     IE_NAME = 'blip.tv:user'
 

youtube_dl/extractor/brightcove.py

@@ -154,12 +154,14 @@ class BrightcoveIE(InfoExtractor):
     def _extract_brightcove_urls(cls, webpage):
         """Return a list of all Brightcove URLs from the webpage """
 
-        url_m = re.search(r'<meta\s+property="og:video"\s+content="(http://c.brightcove.com/[^"]+)"', webpage)
+        url_m = re.search(
+            r'<meta\s+property="og:video"\s+content="(https?://(?:secure|c)\.brightcove.com/[^"]+)"',
+            webpage)
         if url_m:
             url = unescapeHTML(url_m.group(1))
             # Some sites don't add it, we can't download with this url, for example:
             # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/
-            if 'playerKey' in url:
+            if 'playerKey' in url or 'videoId' in url:
                 return [url]
 
         matches = re.findall(
@@ -188,9 +190,13 @@ class BrightcoveIE(InfoExtractor):
             referer = smuggled_data.get('Referer', url)
             return self._get_video_info(
                 videoPlayer[0], query_str, query, referer=referer)
-        else:
+        elif 'playerKey' in query:
             player_key = query['playerKey']
             return self._get_playlist_info(player_key[0])
+        else:
+            raise ExtractorError(
+                'Cannot find playerKey= variable. Did you forget quotes in a shell invocation?',
+                expected=True)
 
     def _get_video_info(self, video_id, query_str, query, referer=None):
         request_url = self._FEDERATED_URL_TEMPLATE % query_str
@@ -202,6 +208,13 @@ class BrightcoveIE(InfoExtractor):
             req.add_header('Referer', referer)
         webpage = self._download_webpage(req, video_id)
 
+        error_msg = self._html_search_regex(
+            r"<h1>We're sorry.</h1>\s*<p>(.*?)</p>", webpage,
+            'error message', default=None)
+        if error_msg is not None:
+            raise ExtractorError(
+                'brightcove said: %s' % error_msg, expected=True)
+
         self.report_extraction(video_id)
         info = self._search_regex(r'var experienceJSON = ({.*});', webpage, 'json')
         info = json.loads(info)['data']

youtube_dl/extractor/common.py

@@ -84,6 +84,12 @@ class InfoExtractor(object):
                                  format, irrespective of the file format.
                                  -1 for default (order by other properties),
                                  -2 or smaller for less than default.
+                    * http_referer  HTTP Referer header value to set.
+                    * http_method  HTTP method to use for the download.
+                    * http_headers  A dictionary of additional HTTP headers
+                                 to add to the request.
+                    * http_post_data  Additional data to send with a POST
+                                 request.
     url:            Final video URL.
     ext:            Video filename extension.
     format:         The video format, defaults to ext (used for --get-format)

youtube_dl/extractor/ebaumsworld.py

@@ -1,19 +1,21 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
-from ..utils import determine_ext
 
 
 class EbaumsWorldIE(InfoExtractor):
     _VALID_URL = r'https?://www\.ebaumsworld\.com/video/watch/(?P<id>\d+)'
 
     _TEST = {
-        u'url': u'http://www.ebaumsworld.com/video/watch/83367677/',
-        u'file': u'83367677.mp4',
-        u'info_dict': {
-            u'title': u'A Giant Python Opens The Door',
-            u'description': u'This is how nightmares start...',
-            u'uploader': u'jihadpizza',
+        'url': 'http://www.ebaumsworld.com/video/watch/83367677/',
+        'info_dict': {
+            'id': '83367677',
+            'ext': 'mp4',
+            'title': 'A Giant Python Opens The Door',
+            'description': 'This is how nightmares start...',
+            'uploader': 'jihadpizza',
         },
     }
 
@@ -28,7 +30,6 @@ class EbaumsWorldIE(InfoExtractor):
             'id': video_id,
             'title': config.find('title').text,
             'url': video_url,
-            'ext': determine_ext(video_url),
             'description': config.find('description').text,
             'thumbnail': config.find('image').text,
             'uploader': config.find('username').text,

youtube_dl/extractor/expotv.py

@@ -0,0 +1,73 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    unified_strdate,
+)
+
+
+class ExpoTVIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.expotv\.com/videos/[^?#]*/(?P<id>[0-9]+)($|[?#])'
+    _TEST = {
+        'url': 'http://www.expotv.com/videos/reviews/1/24/LinneCardscom/17561',
+        'md5': '2985e6d7a392b2f7a05e0ca350fe41d0',
+        'info_dict': {
+            'id': '17561',
+            'ext': 'mp4',
+            'upload_date': '20060212',
+            'title': 'My Favorite Online Scrapbook Store',
+            'view_count': int,
+            'description': 'You\'ll find most everything you need at this virtual store front.',
+            'uploader': 'Anna T.',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        player_key = self._search_regex(
+            r'<param name="playerKey" value="([^"]+)"', webpage, 'player key')
+        config_url = 'http://client.expotv.com/video/config/%s/%s' % (
+            video_id, player_key)
+        config = self._download_json(
+            config_url, video_id,
+            note='Downloading video configuration')
+
+        formats = [{
+            'url': fcfg['file'],
+            'height': int_or_none(fcfg.get('height')),
+            'format_note': fcfg.get('label'),
+            'ext': self._search_regex(
+                r'filename=.*\.([a-z0-9_A-Z]+)&', fcfg['file'],
+                'file extension', default=None),
+        } for fcfg in config['sources']]
+        self._sort_formats(formats)
+
+        title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+        thumbnail = config.get('image')
+        view_count = int_or_none(self._search_regex(
+            r'<h5>Plays: ([0-9]+)</h5>', webpage, 'view counts'))
+        uploader = self._search_regex(
+            r'<div class="reviewer">\s*<img alt="([^"]+)"', webpage, 'uploader',
+            fatal=False)
+        upload_date = unified_strdate(self._search_regex(
+            r'<h5>Reviewed on ([0-9/.]+)</h5>', webpage, 'upload date',
+            fatal=False))
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': title,
+            'description': description,
+            'view_count': view_count,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'upload_date': upload_date,
+        }

youtube_dl/extractor/generic.py

@@ -15,12 +15,14 @@ from ..utils import (
     compat_xml_parse_error,
 
     ExtractorError,
+    float_or_none,
     HEADRequest,
     orderedSet,
     parse_xml,
     smuggle_url,
     unescapeHTML,
     unified_strdate,
+    unsmuggle_url,
     url_basename,
 )
 from .brightcove import BrightcoveIE
@@ -305,6 +307,42 @@ class GenericIE(InfoExtractor):
             'params': {
                 'skip_download': True,
             }
+        },
+        # Camtasia studio
+        {
+            'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
+            'playlist': [{
+                'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
+                'info_dict': {
+                    'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
+                    'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
+                    'ext': 'flv',
+                    'duration': 2235.90,
+                }
+            }, {
+                'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
+                'info_dict': {
+                    'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
+                    'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
+                    'ext': 'flv',
+                    'duration': 2235.93,
+                }
+            }],
+            'info_dict': {
+                'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
+            }
+        },
+        # Flowplayer
+        {
+            'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
+            'md5': '9d65602bf31c6e20014319c7d07fba27',
+            'info_dict': {
+                'id': '5123ea6d5e5a7',
+                'ext': 'mp4',
+                'age_limit': 18,
+                'uploader': 'www.handjobhub.com',
+                'title': 'Busty Blonde Siri Tit Fuck While Wank at Handjob Hub',
+            }
         }
     ]
 
@@ -318,58 +356,6 @@ class GenericIE(InfoExtractor):
         """Report information extraction."""
         self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
 
-    def _send_head(self, url):
-        """Check if it is a redirect, like url shorteners, in case return the new url."""
-
-        class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
-            """
-            Subclass the HTTPRedirectHandler to make it use our
-            HEADRequest also on the redirected URL
-            """
-            def redirect_request(self, req, fp, code, msg, headers, newurl):
-                if code in (301, 302, 303, 307):
-                    newurl = newurl.replace(' ', '%20')
-                    newheaders = dict((k,v) for k,v in req.headers.items()
-                                      if k.lower() not in ("content-length", "content-type"))
-                    try:
-                        # This function was deprecated in python 3.3 and removed in 3.4
-                        origin_req_host = req.get_origin_req_host()
-                    except AttributeError:
-                        origin_req_host = req.origin_req_host
-                    return HEADRequest(newurl,
-                                       headers=newheaders,
-                                       origin_req_host=origin_req_host,
-                                       unverifiable=True)
-                else:
-                    raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp)
-
-        class HTTPMethodFallback(compat_urllib_request.BaseHandler):
-            """
-            Fallback to GET if HEAD is not allowed (405 HTTP error)
-            """
-            def http_error_405(self, req, fp, code, msg, headers):
-                fp.read()
-                fp.close()
-
-                newheaders = dict((k,v) for k,v in req.headers.items()
-                                  if k.lower() not in ("content-length", "content-type"))
-                return self.parent.open(compat_urllib_request.Request(req.get_full_url(),
-                                                 headers=newheaders,
-                                                 origin_req_host=req.get_origin_req_host(),
-                                                 unverifiable=True))
-
-        # Build our opener
-        opener = compat_urllib_request.OpenerDirector()
-        for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler,
-                        HTTPMethodFallback, HEADRedirectHandler,
-                        compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
-            opener.add_handler(handler())
-
-        response = opener.open(HEADRequest(url))
-        if response is None:
-            raise ExtractorError('Invalid URL protocol')
-        return response
-
     def _extract_rss(self, url, video_id, doc):
         playlist_title = doc.find('./channel/title').text
         playlist_desc_el = doc.find('./channel/description')
@@ -389,6 +375,43 @@ class GenericIE(InfoExtractor):
             'entries': entries,
         }
 
+    def _extract_camtasia(self, url, video_id, webpage):
+        """ Returns None if no camtasia video can be found. """
+
+        camtasia_cfg = self._search_regex(
+            r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
+            webpage, 'camtasia configuration file', default=None)
+        if camtasia_cfg is None:
+            return None
+
+        title = self._html_search_meta('DC.title', webpage, fatal=True)
+
+        camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
+        camtasia_cfg = self._download_xml(
+            camtasia_url, video_id,
+            note='Downloading camtasia configuration',
+            errnote='Failed to download camtasia configuration')
+        fileset_node = camtasia_cfg.find('./playlist/array/fileset')
+
+        entries = []
+        for n in fileset_node.getchildren():
+            url_n = n.find('./uri')
+            if url_n is None:
+                continue
+
+            entries.append({
+                'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
+                'title': '%s - %s' % (title, n.tag),
+                'url': compat_urlparse.urljoin(url, url_n.text),
+                'duration': float_or_none(n.find('./duration').text),
+            })
+
+        return {
+            '_type': 'playlist',
+            'entries': entries,
+            'title': title,
+        }
+
     def _real_extract(self, url):
         if url.startswith('//'):
             return {
@@ -425,17 +448,31 @@ class GenericIE(InfoExtractor):
             else:
                 assert ':' in default_search
                 return self.url_result(default_search + url)
-        video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
+
+        url, smuggled_data = unsmuggle_url(url)
+        force_videoid = None
+        if smuggled_data and 'force_videoid' in smuggled_data:
+            force_videoid = smuggled_data['force_videoid']
+            video_id = force_videoid
+        else:
+            video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
 
         self.to_screen('%s: Requesting header' % video_id)
 
-        try:
-            response = self._send_head(url)
+        head_req = HEADRequest(url)
+        response = self._request_webpage(
+            head_req, video_id,
+            note=False, errnote='Could not send HEAD request to %s' % url,
+            fatal=False)
 
+        if response is not False:
             # Check for redirect
             new_url = response.geturl()
             if url != new_url:
                 self.report_following_redirect(new_url)
+                if force_videoid:
+                    new_url = smuggle_url(
+                        new_url, {'force_videoid': force_videoid})
                 return self.url_result(new_url)
 
             # Check for direct link to a video
@@ -456,10 +493,6 @@ class GenericIE(InfoExtractor):
                     'upload_date': upload_date,
                 }
 
-        except compat_urllib_error.HTTPError:
-            # This may be a stupid server that doesn't like HEAD, our UA, or so
-            pass
-
         try:
             webpage = self._download_webpage(url, video_id)
         except ValueError:
@@ -477,6 +510,11 @@ class GenericIE(InfoExtractor):
         except compat_xml_parse_error:
             pass
 
+        # Is it a Camtasia project?
+        camtasia_res = self._extract_camtasia(url, video_id, webpage)
+        if camtasia_res is not None:
+            return camtasia_res
+
         # Sometimes embedded video player is hidden behind percent encoding
         # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
         # Unescaping the whole page allows to handle those cases in a generic way
@@ -492,6 +530,16 @@ class GenericIE(InfoExtractor):
             r'(?s)<title>(.*?)</title>', webpage, 'video title',
             default='video')
 
+        # Try to detect age limit automatically
+        age_limit = self._rta_search(webpage)
+        # And then there are the jokers who advertise that they use RTA,
+        # but actually don't.
+        AGE_LIMIT_MARKERS = [
+            r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
+        ]
+        if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
+            age_limit = 18
+
         # video uploader is domain name
         video_uploader = self._search_regex(
             r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
@@ -573,7 +621,7 @@ class GenericIE(InfoExtractor):
         mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
         if mobj:
             return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
-        mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9]+)', webpage)
+        mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
         if mobj:
             return self.url_result(mobj.group(1), 'BlipTV')
 
@@ -755,6 +803,15 @@ class GenericIE(InfoExtractor):
         if not found:
             # Broaden the findall a little bit: JWPlayer JS loader
             found = re.findall(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)
+        if not found:
+            # Flow player
+            found = re.findall(r'''(?xs)
+                flowplayer\("[^"]+",\s*
+                    \{[^}]+?\}\s*,
+                    \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
+                        ["']?url["']?\s*:\s*["']([^"']+)["']
+            ''', webpage)
+            assert found
         if not found:
             # Try to find twitter cards info
             found = re.findall(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
@@ -764,7 +821,12 @@ class GenericIE(InfoExtractor):
             m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
             # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
             if m_video_type is not None:
-                found = re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)
+                def check_video(vurl):
+                    vpath = compat_urlparse.urlparse(vurl).path
+                    return '.' in vpath and not vpath.endswith('.swf')
+                found = list(filter(
+                    check_video,
+                    re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)))
         if not found:
             # HTML5 video
             found = re.findall(r'(?s)<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage)
@@ -801,6 +863,7 @@ class GenericIE(InfoExtractor):
                 'url': video_url,
                 'uploader': video_uploader,
                 'title': video_title,
+                'age_limit': age_limit,
             })
 
         if len(entries) == 1:

youtube_dl/extractor/grooveshark.py

@@ -0,0 +1,190 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import time
+import math
+import os.path
+import re
+
+
+from .common import InfoExtractor
+from ..utils import ExtractorError, compat_urllib_request, compat_html_parser
+
+from ..utils import (
+    compat_urllib_parse,
+    compat_urlparse,
+)
+
+
+class GroovesharkHtmlParser(compat_html_parser.HTMLParser):
+    def __init__(self):
+        self._current_object = None
+        self.objects = []
+        compat_html_parser.HTMLParser.__init__(self)
+
+    def handle_starttag(self, tag, attrs):
+        attrs = dict((k, v) for k, v in attrs)
+        if tag == 'object':
+            self._current_object = {'attrs': attrs, 'params': []}
+        elif tag == 'param':
+            self._current_object['params'].append(attrs)
+
+    def handle_endtag(self, tag):
+        if tag == 'object':
+            self.objects.append(self._current_object)
+            self._current_object = None
+
+    @classmethod
+    def extract_object_tags(cls, html):
+        p = cls()
+        p.feed(html)
+        p.close()
+        return p.objects
+
+
+class GroovesharkIE(InfoExtractor):
+    _VALID_URL = r'https?://(www\.)?grooveshark\.com/#!/s/([^/]+)/([^/]+)'
+    _TEST = {
+        'url': 'http://grooveshark.com/#!/s/Jolene+Tenth+Key+Remix+Ft+Will+Sessions/6SS1DW?src=5',
+        'md5': '7ecf8aefa59d6b2098517e1baa530023',
+        'info_dict': {
+            'id': '6SS1DW',
+            'title': 'Jolene (Tenth Key Remix ft. Will Sessions)',
+            'ext': 'mp3',
+            'duration': 227,
+        }
+    }
+
+    do_playerpage_request = True
+    do_bootstrap_request = True
+
+    def _parse_target(self, target):
+        uri = compat_urlparse.urlparse(target)
+        hash = uri.fragment[1:].split('?')[0]
+        token = os.path.basename(hash.rstrip('/'))
+        return (uri, hash, token)
+
+    def _build_bootstrap_url(self, target):
+        (uri, hash, token) = self._parse_target(target)
+        query = 'getCommunicationToken=1&hash=%s&%d' % (compat_urllib_parse.quote(hash, safe=''), self.ts)
+        return (compat_urlparse.urlunparse((uri.scheme, uri.netloc, '/preload.php', None, query, None)), token)
+
+    def _build_meta_url(self, target):
+        (uri, hash, token) = self._parse_target(target)
+        query = 'hash=%s&%d' % (compat_urllib_parse.quote(hash, safe=''), self.ts)
+        return (compat_urlparse.urlunparse((uri.scheme, uri.netloc, '/preload.php', None, query, None)), token)
+
+    def _build_stream_url(self, meta):
+        return compat_urlparse.urlunparse(('http', meta['streamKey']['ip'], '/stream.php', None, None, None))
+
+    def _build_swf_referer(self, target, obj):
+        (uri, _, _) = self._parse_target(target)
+        return compat_urlparse.urlunparse((uri.scheme, uri.netloc, obj['attrs']['data'], None, None, None))
+
+    def _transform_bootstrap(self, js):
+        return re.split('(?m)^\s*try\s*{', js)[0] \
+                 .split(' = ', 1)[1].strip().rstrip(';')
+
+    def _transform_meta(self, js):
+        return js.split('\n')[0].split('=')[1].rstrip(';')
+
+    def _get_meta(self, target):
+        (meta_url, token) = self._build_meta_url(target)
+        self.to_screen('Metadata URL: %s' % meta_url)
+
+        headers = {'Referer': compat_urlparse.urldefrag(target)[0]}
+        req = compat_urllib_request.Request(meta_url, headers=headers)
+        res = self._download_json(req, token,
+                                  transform_source=self._transform_meta)
+
+        if 'getStreamKeyWithSong' not in res:
+            raise ExtractorError(
+                'Metadata not found. URL may be malformed, or Grooveshark API may have changed.')
+
+        if res['getStreamKeyWithSong'] is None:
+            raise ExtractorError(
+                'Metadata download failed, probably due to Grooveshark anti-abuse throttling. Wait at least an hour before retrying from this IP.',
+                expected=True)
+
+        return res['getStreamKeyWithSong']
+
+    def _get_bootstrap(self, target):
+        (bootstrap_url, token) = self._build_bootstrap_url(target)
+
+        headers = {'Referer': compat_urlparse.urldefrag(target)[0]}
+        req = compat_urllib_request.Request(bootstrap_url, headers=headers)
+        res = self._download_json(req, token, fatal=False,
+                                  note='Downloading player bootstrap data',
+                                  errnote='Unable to download player bootstrap data',
+                                  transform_source=self._transform_bootstrap)
+        return res
+
+    def _get_playerpage(self, target):
+        (_, _, token) = self._parse_target(target)
+
+        webpage = self._download_webpage(
+            target, token,
+            note='Downloading player page',
+            errnote='Unable to download player page',
+            fatal=False)
+
+        if webpage is not None:
+            # Search (for example German) error message
+            error_msg = self._html_search_regex(
+                r'<div id="content">\s*<h2>(.*?)</h2>', webpage,
+                'error message', default=None)
+            if error_msg is not None:
+                error_msg = error_msg.replace('\n', ' ')
+                raise ExtractorError('Grooveshark said: %s' % error_msg)
+
+        if webpage is not None:
+            o = GroovesharkHtmlParser.extract_object_tags(webpage)
+            return (webpage, [x for x in o if x['attrs']['id'] == 'jsPlayerEmbed'])
+
+        return (webpage, None)
+
+    def _real_initialize(self):
+        self.ts = int(time.time() * 1000)  # timestamp in millis
+
+    def _real_extract(self, url):
+        (target_uri, _, token) = self._parse_target(url)
+
+        # 1. Fill cookiejar by making a request to the player page
+        swf_referer = None
+        if self.do_playerpage_request:
+            (_, player_objs) = self._get_playerpage(url)
+            if player_objs is not None:
+                swf_referer = self._build_swf_referer(url, player_objs[0])
+                self.to_screen('SWF Referer: %s' % swf_referer)
+
+        # 2. Ask preload.php for swf bootstrap data to better mimic webapp
+        if self.do_bootstrap_request:
+            bootstrap = self._get_bootstrap(url)
+            self.to_screen('CommunicationToken: %s' % bootstrap['getCommunicationToken'])
+
+        # 3. Ask preload.php for track metadata.
+        meta = self._get_meta(url)
+
+        # 4. Construct stream request for track.
+        stream_url = self._build_stream_url(meta)
+        duration = int(math.ceil(float(meta['streamKey']['uSecs']) / 1000000))
+        post_dict = {'streamKey': meta['streamKey']['streamKey']}
+        post_data = compat_urllib_parse.urlencode(post_dict).encode('utf-8')
+        headers = {
+            'Content-Length': len(post_data),
+            'Content-Type': 'application/x-www-form-urlencoded'
+        }
+        if swf_referer is not None:
+            headers['Referer'] = swf_referer
+
+        return {
+            'id': token,
+            'title': meta['song']['Name'],
+            'http_method': 'POST',
+            'url': stream_url,
+            'ext': 'mp3',
+            'format': 'mp3 audio',
+            'duration': duration,
+            'http_post_data': post_data,
+            'http_headers': headers,
+        }

youtube_dl/extractor/ministrygrid.py

@@ -0,0 +1,57 @@
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    smuggle_url,
+)
+
+
+class MinistryGridIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.ministrygrid.com/([^/?#]*/)*(?P<id>[^/#?]+)/?(?:$|[?#])'
+
+    _TEST = {
+        'url': 'http://www.ministrygrid.com/training-viewer/-/training/t4g-2014-conference/the-gospel-by-numbers-4/the-gospel-by-numbers',
+        'md5': '844be0d2a1340422759c2a9101bab017',
+        'info_dict': {
+            'id': '3453494717001',
+            'ext': 'mp4',
+            'title': 'The Gospel by Numbers',
+            'description': 'Coming soon from T4G 2014!',
+            'uploader': 'LifeWay Christian Resources (MG)',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        portlets_json = self._search_regex(
+            r'Liferay\.Portlet\.list=(\[.+?\])', webpage, 'portlet list')
+        portlets = json.loads(portlets_json)
+        pl_id = self._search_regex(
+            r'<!--\s*p_l_id - ([0-9]+)<br>', webpage, 'p_l_id')
+
+        for i, portlet in enumerate(portlets):
+            portlet_url = 'http://www.ministrygrid.com/c/portal/render_portlet?p_l_id=%s&p_p_id=%s' % (pl_id, portlet)
+            portlet_code = self._download_webpage(
+                portlet_url, video_id,
+                note='Looking in portlet %s (%d/%d)' % (portlet, i + 1, len(portlets)),
+                fatal=False)
+            video_iframe_url = self._search_regex(
+                r'<iframe.*?src="([^"]+)"', portlet_code, 'video iframe',
+                default=None)
+            if video_iframe_url:
+                surl = smuggle_url(
+                    video_iframe_url, {'force_videoid': video_id})
+                return {
+                    '_type': 'url',
+                    'id': video_id,
+                    'url': surl,
+                }
+
+        raise ExtractorError('Could not find video iframe in any portlets')

youtube_dl/extractor/pornotube.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -9,15 +11,16 @@ from ..utils import (
 
 
 class PornotubeIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:\w+\.)?pornotube\.com(/c/(?P<channel>[0-9]+))?(/m/(?P<videoid>[0-9]+))(/(?P<title>.+))$'
+    _VALID_URL = r'https?://(?:\w+\.)?pornotube\.com(/c/(?P<channel>[0-9]+))?(/m/(?P<videoid>[0-9]+))(/(?P<title>.+))$'
     _TEST = {
-        u'url': u'http://pornotube.com/c/173/m/1689755/Marilyn-Monroe-Bathing',
-        u'file': u'1689755.flv',
-        u'md5': u'374dd6dcedd24234453b295209aa69b6',
-        u'info_dict': {
-            u"upload_date": u"20090708", 
-            u"title": u"Marilyn-Monroe-Bathing",
-            u"age_limit": 18
+        'url': 'http://pornotube.com/c/173/m/1689755/Marilyn-Monroe-Bathing',
+        'md5': '374dd6dcedd24234453b295209aa69b6',
+        'info_dict': {
+            'id': '1689755',
+            'ext': 'flv',
+            'upload_date': '20090708',
+            'title': 'Marilyn-Monroe-Bathing',
+            'age_limit': 18
         }
     }
 
@@ -32,22 +35,22 @@ class PornotubeIE(InfoExtractor):
 
         # Get the video URL
         VIDEO_URL_RE = r'url: "(?P<url>http://video[0-9].pornotube.com/.+\.flv)",'
-        video_url = self._search_regex(VIDEO_URL_RE, webpage, u'video url')
+        video_url = self._search_regex(VIDEO_URL_RE, webpage, 'video url')
         video_url = compat_urllib_parse.unquote(video_url)
 
         #Get the uploaded date
         VIDEO_UPLOADED_RE = r'<div class="video_added_by">Added (?P<date>[0-9\/]+) by'
-        upload_date = self._html_search_regex(VIDEO_UPLOADED_RE, webpage, u'upload date', fatal=False)
-        if upload_date: upload_date = unified_strdate(upload_date)
+        upload_date = self._html_search_regex(VIDEO_UPLOADED_RE, webpage, 'upload date', fatal=False)
+        if upload_date:
+            upload_date = unified_strdate(upload_date)
         age_limit = self._rta_search(webpage)
 
-        info = {'id': video_id,
-                'url': video_url,
-                'uploader': None,
-                'upload_date': upload_date,
-                'title': video_title,
-                'ext': 'flv',
-                'format': 'flv',
-                'age_limit': age_limit}
-
-        return [info]
+        return {
+            'id': video_id,
+            'url': video_url,
+            'upload_date': upload_date,
+            'title': video_title,
+            'ext': 'flv',
+            'format': 'flv',
+            'age_limit': age_limit,
+        }

youtube_dl/extractor/wat.py

@@ -2,27 +2,30 @@
 from __future__ import unicode_literals
 
 import re
+import time
+import hashlib
 
 from .common import InfoExtractor
 from ..utils import (
+    ExtractorError,
     unified_strdate,
 )
 
 
 class WatIE(InfoExtractor):
-    _VALID_URL = r'http://www\.wat\.tv/.*-(?P<shortID>.*?)_.*?\.html'
+    _VALID_URL = r'http://www\.wat\.tv/video/(?P<display_id>.*)-(?P<short_id>.*?)_.*?\.html'
     IE_NAME = 'wat.tv'
     _TEST = {
-        'url': 'http://www.wat.tv/video/world-war-philadelphia-vost-6bv55_2fjr7_.html',
+        'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html',
+        'md5': 'ce70e9223945ed26a8056d413ca55dc9',
         'info_dict': {
-            'id': '10631273',
+            'id': '11713067',
+            'display_id': 'soupe-figues-l-orange-aux-epices',
             'ext': 'mp4',
-            'title': 'World War Z - Philadelphia VOST',
-            'description': 'La menace est partout. Que se passe-t-il à Philadelphia ?\r\nWORLD WAR Z, avec Brad Pitt, au cinéma le 3 juillet.\r\nhttp://www.worldwarz.fr',
-        },
-        'params': {
-            # Sometimes wat serves the whole file with the --test option
-            'skip_download': True,
+            'title': 'Soupe de figues à l\'orange et aux épices',
+            'description': 'Retrouvez l\'émission "Petits plats en équilibre", diffusée le 18 août 2014.',
+            'upload_date': '20140819',
+            'duration': 120,
         },
     }
 
@@ -36,13 +39,20 @@ class WatIE(InfoExtractor):
         def real_id_for_chapter(chapter):
             return chapter['tc_start'].split('-')[0]
         mobj = re.match(self._VALID_URL, url)
-        short_id = mobj.group('shortID')
-        webpage = self._download_webpage(url, short_id)
+        short_id = mobj.group('short_id')
+        display_id = mobj.group('display_id')
+        webpage = self._download_webpage(url, display_id or short_id)
         real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id')
 
         video_info = self.download_video_info(real_id)
+
+        if video_info.get('geolock'):
+            raise ExtractorError('This content is not available in your area', expected=True)
+
         chapters = video_info['chapters']
         first_chapter = chapters[0]
+        files = video_info['files']
+        first_file = files[0]
 
         if real_id_for_chapter(first_chapter) != real_id:
             self.to_screen('Multipart video detected')
@@ -61,12 +71,45 @@ class WatIE(InfoExtractor):
             upload_date = unified_strdate(first_chapter['date_diffusion'])
         # Otherwise we can continue and extract just one part, we have to use
         # the short id for getting the video url
+
+        formats = [{
+            'url': 'http://wat.tv/get/android5/%s.mp4' % real_id,
+            'format_id': 'Mobile',
+        }]
+
+        fmts = [('SD', 'web')]
+        if first_file.get('hasHD'):
+            fmts.append(('HD', 'webhd'))
+
+        def compute_token(param):
+            timestamp = '%08x' % int(time.time())
+            magic = '9b673b13fa4682ed14c3cfa5af5310274b514c4133e9b3a81e6e3aba009l2564'
+            return '%s/%s' % (hashlib.md5((magic + param + timestamp).encode('ascii')).hexdigest(), timestamp)
+
+        for fmt in fmts:
+            webid = '/%s/%s' % (fmt[1], real_id)
+            video_url = self._download_webpage(
+                'http://www.wat.tv/get%s?token=%s&getURL=1' % (webid, compute_token(webid)),
+                real_id,
+                'Downloding %s video URL' % fmt[0],
+                'Failed to download %s video URL' % fmt[0],
+                False)
+            if not video_url:
+                continue
+            formats.append({
+                'url': video_url,
+                'ext': 'mp4',
+                'format_id': fmt[0],
+            })
+
         return {
             'id': real_id,
-            'url': 'http://wat.tv/get/android5/%s.mp4' % real_id,
+            'display_id': display_id,
             'title': first_chapter['title'],
             'thumbnail': first_chapter['preview'],
             'description': first_chapter['description'],
             'view_count': video_info['views'],
             'upload_date': upload_date,
+            'duration': first_file['duration'],
+            'formats': formats,
         }

youtube_dl/extractor/wayofthemaster.py

@@ -0,0 +1,52 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class WayOfTheMasterIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.wayofthemaster\.com/([^/?#]*/)*(?P<id>[^/?#]+)\.s?html(?:$|[?#])'
+
+    _TEST = {
+        'url': 'http://www.wayofthemaster.com/hbks.shtml',
+        'md5': '5316b57487ada8480606a93cb3d18d24',
+        'info_dict': {
+            'id': 'hbks',
+            'ext': 'mp4',
+            'title': 'Intelligent Design vs. Evolution',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._search_regex(
+            r'<img src="images/title_[^"]+".*?alt="([^"]+)"',
+            webpage, 'title', default=None)
+        if title is None:
+            title = self._html_search_regex(
+                r'<title>(.*?)</title>', webpage, 'page title')
+
+        url_base = self._search_regex(
+            r'<param\s+name="?movie"?\s+value=".*?/wotm_videoplayer_highlow[0-9]*\.swf\?vid=([^"]+)"',
+            webpage, 'URL base')
+        formats = [{
+            'format_id': 'low',
+            'quality': 1,
+            'url': url_base + '_low.mp4',
+        }, {
+            'format_id': 'high',
+            'quality': 2,
+            'url': url_base + '_high.mp4',
+        }]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+        }

youtube_dl/utils.py

@@ -855,6 +855,7 @@ def unified_strdate(date_str):
         '%Y/%m/%d',
         '%d.%m.%Y',
         '%d/%m/%Y',
+        '%d/%m/%y',
         '%Y/%m/%d %H:%M:%S',
         '%Y-%m-%d %H:%M:%S',
         '%d.%m.%Y %H:%M',

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.08.23'
+__version__ = '2014.08.24.5'