release 2014.02.10

The ‘og:description’  property doesn’t contain the full description

test/test_utils.py

@@ -127,6 +127,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unified_strdate('8/7/2009'), '20090708')
         self.assertEqual(unified_strdate('Dec 14, 2012'), '20121214')
         self.assertEqual(unified_strdate('2012/10/11 01:56:38 +0000'), '20121011')
+        self.assertEqual(unified_strdate('1968-12-10'), '19681210')
 
     def test_find_xpath_attr(self):
         testxml = u'''<root>

youtube_dl/extractor/__init__.py

@@ -115,6 +115,7 @@ from .keezmovies import KeezMoviesIE
 from .khanacademy import KhanAcademyIE
 from .kickstarter import KickStarterIE
 from .keek import KeekIE
+from .kontrtube import KontrTubeIE
 from .la7 import LA7IE
 from .lifenews import LifeNewsIE
 from .liveleak import LiveLeakIE

youtube_dl/extractor/bbccouk.py

@@ -2,29 +2,160 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
 from ..utils import ExtractorError
 
 
-class BBCCoUkIE(InfoExtractor):
+class BBCCoUkIE(SubtitlesInfoExtractor):
     IE_NAME = 'bbc.co.uk'
-    IE_DESC = 'BBC - iPlayer Radio'
+    IE_DESC = 'BBC iPlayer'
     _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:programmes|iplayer/episode)/(?P<id>[\da-z]{8})'
 
-    _TEST = {
-        'url': 'http://www.bbc.co.uk/programmes/p01q7wz1',
-        'info_dict': {
-            'id': 'p01q7wz4',
-            'ext': 'flv',
-            'title': 'Friction: Blu Mar Ten guest mix: Blu Mar Ten - Guest Mix',
-            'description': 'Blu Mar Ten deliver a Guest Mix for Friction.',
-            'duration': 1936,
+    _TESTS = [
+        {
+            'url': 'http://www.bbc.co.uk/programmes/p01q7wz1',
+            'info_dict': {
+                'id': 'p01q7wz4',
+                'ext': 'flv',
+                'title': 'Friction: Blu Mar Ten guest mix: Blu Mar Ten - Guest Mix',
+                'description': 'Blu Mar Ten deliver a Guest Mix for Friction.',
+                'duration': 1936,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
+            'info_dict': {
+                'id': 'b00yng1d',
+                'ext': 'flv',
+                'title': 'The Man in Black: Series 3: The Printed Name',
+                'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
+                'duration': 1800,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
+            'info_dict': {
+                'id': 'b00yng1d',
+                'ext': 'flv',
+                'title': 'The Voice UK: Series 3: Blind Auditions 5',
+                'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
+                'duration': 5100,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
         }
-    }
+    ]
+
+    def _extract_asx_playlist(self, connection, programme_id):
+        asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
+        return [ref.get('href') for ref in asx.findall('./Entry/ref')]
+
+    def _extract_connection(self, connection, programme_id):
+        formats = []
+        protocol = connection.get('protocol')
+        supplier = connection.get('supplier')
+        if protocol == 'http':
+            href = connection.get('href')
+            # ASX playlist
+            if supplier == 'asx':
+                for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
+                    formats.append({
+                        'url': ref,
+                        'format_id': 'ref%s_%s' % (i, supplier),
+                    })
+            # Direct link
+            else:
+                formats.append({
+                    'url': href,
+                    'format_id': supplier,
+                })
+        elif protocol == 'rtmp':
+            application = connection.get('application', 'ondemand')
+            auth_string = connection.get('authString')
+            identifier = connection.get('identifier')
+            server = connection.get('server')
+            formats.append({
+                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
+                'play_path': identifier,
+                'app': '%s?%s' % (application, auth_string),
+                'page_url': 'http://www.bbc.co.uk',
+                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
+                'rtmp_live': False,
+                'ext': 'flv',
+                'format_id': supplier,
+            })
+        return formats
+
+    def _extract_items(self, playlist):
+        return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
+
+    def _extract_medias(self, media_selection):
+        return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
+
+    def _extract_connections(self, media):
+        return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
+
+    def _extract_video(self, media, programme_id):
+        formats = []
+        vbr = int(media.get('bitrate'))
+        vcodec = media.get('encoding')
+        service = media.get('service')
+        width = int(media.get('width'))
+        height = int(media.get('height'))
+        file_size = int(media.get('media_file_size'))
+        for connection in self._extract_connections(media):
+            conn_formats = self._extract_connection(connection, programme_id)
+            for format in conn_formats:
+                format.update({
+                    'format_id': '%s_%s' % (service, format['format_id']),
+                    'width': width,
+                    'height': height,
+                    'vbr': vbr,
+                    'vcodec': vcodec,
+                    'filesize': file_size,
+                })
+            formats.extend(conn_formats)
+        return formats
+
+    def _extract_audio(self, media, programme_id):
+        formats = []
+        abr = int(media.get('bitrate'))
+        acodec = media.get('encoding')
+        service = media.get('service')
+        for connection in self._extract_connections(media):
+            conn_formats = self._extract_connection(connection, programme_id)
+            for format in conn_formats:
+                format.update({
+                    'format_id': '%s_%s' % (service, format['format_id']),
+                    'abr': abr,
+                    'acodec': acodec,
+                })
+            formats.extend(conn_formats)
+        return formats
+
+    def _extract_captions(self, media, programme_id):
+        subtitles = {}
+        for connection in self._extract_connections(media):
+            captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
+            lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
+            ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
+            srt = ''
+            for pos, p in enumerate(ps):
+                srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'),
+                                                          p.text.strip() if p.text is not None else '')
+            subtitles[lang] = srt
+        return subtitles
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -33,84 +164,54 @@ class BBCCoUkIE(InfoExtractor):
         playlist = self._download_xml('http://www.bbc.co.uk/iplayer/playlist/%s' % group_id, group_id,
             'Downloading playlist XML')
 
-        item = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}item')
-        if item is None:
-            no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
-            if no_items is not None:
-                reason = no_items.get('reason')
-                if reason == 'preAvailability':
-                    msg = 'Episode %s is not yet available' % group_id
-                elif reason == 'postAvailability':
-                    msg = 'Episode %s is no longer available' % group_id
-                else:
-                    msg = 'Episode %s is not available: %s' % (group_id, reason)
-                raise ExtractorError(msg, expected=True)
-            raise ExtractorError('Failed to extract media for episode %s' % group_id, expected=True)
-
-        title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
-        description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
-
-        radio_programme_id = item.get('identifier')
-        duration = int(item.get('duration'))
-
-        media_selection = self._download_xml(
-            'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s'  % radio_programme_id,
-            radio_programme_id, 'Downloading media selection XML')
+        no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
+        if no_items is not None:
+            reason = no_items.get('reason')
+            if reason == 'preAvailability':
+                msg = 'Episode %s is not yet available' % group_id
+            elif reason == 'postAvailability':
+                msg = 'Episode %s is no longer available' % group_id
+            else:
+                msg = 'Episode %s is not available: %s' % (group_id, reason)
+            raise ExtractorError(msg, expected=True)
 
         formats = []
-        for media in media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media'):
-            bitrate = int(media.get('bitrate'))
-            encoding = media.get('encoding')
-            service = media.get('service')
-            connection = media.find('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
-            protocol = connection.get('protocol')
-            priority = connection.get('priority')
-            supplier = connection.get('supplier')
-            if protocol == 'http':
-                href = connection.get('href')
-                # ASX playlist
-                if supplier == 'asx':
-                    asx = self._download_xml(href, radio_programme_id, 'Downloading %s ASX playlist' % service)
-                    for i, ref in enumerate(asx.findall('./Entry/ref')):
-                        formats.append({
-                            'url': ref.get('href'),
-                            'format_id': '%s_ref%s' % (service, i),
-                            'abr': bitrate,
-                            'acodec': encoding,
-                            'preference': priority,
-                        })
-                    continue
-                # Direct link
-                formats.append({
-                    'url': href,
-                    'format_id': service,
-                    'abr': bitrate,
-                    'acodec': encoding,
-                    'preference': priority,
-                })
-            elif protocol == 'rtmp':
-                application = connection.get('application', 'ondemand')
-                auth_string = connection.get('authString')
-                identifier = connection.get('identifier')
-                server = connection.get('server')
-                formats.append({
-                    'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
-                    'play_path': identifier,
-                    'app': '%s?%s' % (application, auth_string),
-                    'rtmp_live': False,
-                    'ext': 'flv',
-                    'format_id': service,
-                    'abr': bitrate,
-                    'acodec': encoding,
-                    'preference': priority,
-                })
+        subtitles = None
+
+        for item in self._extract_items(playlist):
+            kind = item.get('kind')
+            if kind != 'programme' and kind != 'radioProgramme':
+                continue
+            title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
+            description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
+
+            programme_id = item.get('identifier')
+            duration = int(item.get('duration'))
+
+            media_selection = self._download_xml(
+                'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s'  % programme_id,
+                programme_id, 'Downloading media selection XML')
+
+            for media in self._extract_medias(media_selection):
+                kind = media.get('kind')
+                if kind == 'audio':
+                    formats.extend(self._extract_audio(media, programme_id))
+                elif kind == 'video':
+                    formats.extend(self._extract_video(media, programme_id))
+                elif kind == 'captions':
+                    subtitles = self._extract_captions(media, programme_id)
+
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(programme_id, subtitles)
+            return
 
         self._sort_formats(formats)
 
         return {
-            'id': radio_programme_id,
+            'id': programme_id,
             'title': title,
             'description': description,
             'duration': duration,
             'formats': formats,
+            'subtitles': subtitles,
         }

youtube_dl/extractor/bloomberg.py

@@ -24,5 +24,7 @@ class BloombergIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         name = mobj.group('name')
         webpage = self._download_webpage(url, name)
-        ooyala_url = self._twitter_search_player(webpage)
-        return self.url_result(ooyala_url, OoyalaIE.ie_key())
+        embed_code = self._search_regex(
+            r'<source src="https?://[^/]+/[^/]+/[^/]+/([^/]+)', webpage,
+            'embed code')
+        return OoyalaIE._build_url_result(embed_code)

youtube_dl/extractor/common.py

@@ -271,8 +271,11 @@ class InfoExtractor(object):
 
     def _download_json(self, url_or_request, video_id,
                        note=u'Downloading JSON metadata',
-                       errnote=u'Unable to download JSON metadata'):
+                       errnote=u'Unable to download JSON metadata',
+                       transform_source=None):
         json_string = self._download_webpage(url_or_request, video_id, note, errnote)
+        if transform_source:
+            json_string = transform_source(json_string)
         try:
             return json.loads(json_string)
         except ValueError as ve:

youtube_dl/extractor/kontrtube.py

@@ -0,0 +1,66 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class KontrTubeIE(InfoExtractor):
+    IE_NAME = 'kontrtube'
+    IE_DESC = 'KontrTube.ru - Труба зовёт'
+    _VALID_URL = r'http://(?:www\.)?kontrtube\.ru/videos/(?P<id>\d+)/.+'
+
+    _TEST = {
+        'url': 'http://www.kontrtube.ru/videos/2678/nad-olimpiyskoy-derevney-v-sochi-podnyat-rossiyskiy-flag/',
+        'md5': '975a991a4926c9a85f383a736a2e6b80',
+        'info_dict': {
+            'id': '2678',
+            'ext': 'mp4',
+            'title': 'Над олимпийской деревней в Сочи поднят российский флаг',
+            'description': 'md5:80edc4c613d5887ae8ccf1d59432be41',
+            'thumbnail': 'http://www.kontrtube.ru/contents/videos_screenshots/2000/2678/preview.mp4.jpg',
+            'duration': 270,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id, 'Downloading page')
+
+        video_url = self._html_search_regex(r"video_url: '(.+?)/?',", webpage, 'video URL')
+        thumbnail = self._html_search_regex(r"preview_url: '(.+?)/?',", webpage, 'video thumbnail', fatal=False)
+        title = self._html_search_regex(r'<title>(.+?) - Труба зовёт - Интересный видеохостинг</title>', webpage,
+            'video title')
+        description = self._html_search_meta('description', webpage, 'video description')
+
+        mobj = re.search(r'<div class="col_2">Длительность: <span>(?P<minutes>\d+)м:(?P<seconds>\d+)с</span></div>',
+            webpage)
+        duration = int(mobj.group('minutes')) * 60 + int(mobj.group('seconds')) if mobj else None
+
+        view_count = self._html_search_regex(r'<div class="col_2">Просмотров: <span>(\d+)</span></div>', webpage,
+            'view count', fatal=False)
+        view_count = int(view_count) if view_count is not None else None
+
+        comment_count = None
+        comment_str = self._html_search_regex(r'Комментарии: <span>([^<]+)</span>', webpage, 'comment count',
+            fatal=False)
+        if comment_str.startswith('комментариев нет'):
+            comment_count = 0
+        else:
+            mobj = re.search(r'\d+ из (?P<total>\d+) комментариев', comment_str)
+            if mobj:
+                comment_count = int(mobj.group('total'))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'thumbnail': thumbnail,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'view_count': view_count,
+            'comment_count': comment_count,
+        }

youtube_dl/extractor/slideshare.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -12,11 +14,12 @@ class SlideshareIE(InfoExtractor):
     _VALID_URL = r'https?://www\.slideshare\.net/[^/]+?/(?P<title>.+?)($|\?)'
 
     _TEST = {
-        u'url': u'http://www.slideshare.net/Dataversity/keynote-presentation-managing-scale-and-complexity',
-        u'file': u'25665706.mp4',
-        u'info_dict': {
-            u'title': u'Managing Scale and Complexity',
-            u'description': u'This was a keynote presentation at the NoSQL Now! 2013 Conference & Expo (http://www.nosqlnow.com). This presentation was given by Adrian Cockcroft from Netflix',
+        'url': 'http://www.slideshare.net/Dataversity/keynote-presentation-managing-scale-and-complexity',
+        'info_dict': {
+            'id': '25665706',
+            'ext': 'mp4',
+            'title': 'Managing Scale and Complexity',
+            'description': 'This was a keynote presentation at the NoSQL Now! 2013 Conference & Expo (http://www.nosqlnow.com). This presentation was given by Adrian Cockcroft from Netflix.',
         },
     }
 
@@ -26,15 +29,17 @@ class SlideshareIE(InfoExtractor):
         webpage = self._download_webpage(url, page_title)
         slideshare_obj = self._search_regex(
             r'var slideshare_object =  ({.*?}); var user_info =',
-            webpage, u'slideshare object')
+            webpage, 'slideshare object')
         info = json.loads(slideshare_obj)
-        if info['slideshow']['type'] != u'video':
-            raise ExtractorError(u'Webpage type is "%s": only video extraction is supported for Slideshare' % info['slideshow']['type'], expected=True)
+        if info['slideshow']['type'] != 'video':
+            raise ExtractorError('Webpage type is "%s": only video extraction is supported for Slideshare' % info['slideshow']['type'], expected=True)
 
         doc = info['doc']
         bucket = info['jsplayer']['video_bucket']
         ext = info['jsplayer']['video_extension']
         video_url = compat_urlparse.urljoin(bucket, doc + '-SD.' + ext)
+        description = self._html_search_regex(
+            r'<p class="description.*?"[^>]*>(.*?)</p>', webpage, 'description')
 
         return {
             '_type': 'video',
@@ -43,5 +48,5 @@ class SlideshareIE(InfoExtractor):
             'ext': ext,
             'url': video_url,
             'thumbnail': info['slideshow']['pin_image_url'],
-            'description': self._og_search_description(webpage),
+            'description': description,
         }

youtube_dl/extractor/youtube.py

@@ -34,6 +34,7 @@ from ..utils import (
     unified_strdate,
     orderedSet,
     write_json_file,
+    uppercase_escape,
 )
 
 class YoutubeBaseInfoExtractor(InfoExtractor):
@@ -136,7 +137,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                          (?:https?://|//)?                                    # http(s):// or protocol-independent URL (optional)
                          (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
                             (?:www\.)?deturl\.com/www\.youtube\.com/|
-                            (?:www\.)?pwnyoutube\.com|
+                            (?:www\.)?pwnyoutube\.com/|
                             tube\.majestyc\.net/|
                             youtube\.googleapis\.com/)                        # the various hostnames, with wildcard subdomains
                          (?:.*?\#/)?                                          # handle anchor (#/) redirect urls
@@ -1590,11 +1591,10 @@ class YoutubeChannelIE(InfoExtractor):
             # Download all channel pages using the json-based channel_ajax query
             for pagenum in itertools.count(1):
                 url = self._MORE_PAGES_URL % (pagenum, channel_id)
-                page = self._download_webpage(url, channel_id,
-                                              u'Downloading page #%s' % pagenum)
-    
-                page = json.loads(page)
-    
+                page = self._download_json(
+                    url, channel_id, note=u'Downloading page #%s' % pagenum,
+                    transform_source=uppercase_escape)
+
                 ids_in_page = self.extract_videos_from_page(page['content_html'])
                 video_ids.extend(ids_in_page)
     

youtube_dl/utils.py

@@ -756,9 +756,9 @@ def unified_strdate(date_str):
     """Return a string with the date in the format YYYYMMDD"""
     upload_date = None
     #Replace commas
-    date_str = date_str.replace(',',' ')
+    date_str = date_str.replace(',', ' ')
     # %z (UTC offset) is only supported in python>=3.2
-    date_str = re.sub(r' ?(\+|-)[0-9:]*$', '', date_str)
+    date_str = re.sub(r' ?(\+|-)[0-9]{2}:?[0-9]{2}$', '', date_str)
     format_expressions = [
         '%d %B %Y',
         '%B %d %Y',
@@ -1214,3 +1214,9 @@ class PagedList(object):
             if end == nextfirstid:
                 break
         return res
+
+
+def uppercase_escape(s):
+    return re.sub(
+        r'\\U([0-9a-fA-F]{8})',
+        lambda m: compat_chr(int(m.group(1), base=16)), s)

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.02.08.2'
+__version__ = '2014.02.10'