Merge branch 'remitamine-brightcove_in_page_embed'

youtube_dl/extractor/__init__.py

@@ -60,7 +60,10 @@
 from .bpb import BpbIE
 from .br import BRIE
 from .breakcom import BreakIE
-from .brightcove import BrightcoveIE
+from .brightcove import (
+    BrightcoveLegacyIE,
+    BrightcoveNewIE,
+)
 from .buzzfeed import BuzzFeedIE
 from .byutv import BYUtvIE
 from .c56 import C56IE

youtube_dl/extractor/aljazeera.py

@@ -15,7 +15,7 @@ class AlJazeeraIE(InfoExtractor):
             'description': 'As a birth attendant advocating for family planning, Remy is on the frontline of Tondo\'s battle with overcrowding.',
             'uploader': 'Al Jazeera English',
         },
-        'add_ie': ['Brightcove'],
+        'add_ie': ['BrightcoveLegacy'],
         'skip': 'Not accessible from Travis CI server',
     }
 
@@ -32,5 +32,5 @@ def _real_extract(self, url):
                 'playerKey=AQ~~%2CAAAAmtVJIFk~%2CTVGOQ5ZTwJbeMWnq5d_H4MOM57xfzApc'
                 '&%40videoPlayer={0}'.format(brightcove_id)
             ),
-            'ie_key': 'Brightcove',
+            'ie_key': 'BrightcoveLegacy',
         }

youtube_dl/extractor/brightcove.py

@@ -20,12 +20,17 @@
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
+    float_or_none,
+    js_to_json,
+    int_or_none,
+    parse_iso8601,
     unescapeHTML,
     unsmuggle_url,
 )
 
 
-class BrightcoveIE(InfoExtractor):
+class BrightcoveLegacyIE(InfoExtractor):
+    IE_NAME = 'brightcove:legacy'
     _VALID_URL = r'(?:https?://.*brightcove\.com/(services|viewer).*?\?|brightcove:)(?P<query>.*)'
     _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
 
@@ -346,3 +351,152 @@ def _extract_video_info(self, video_info):
         if 'url' not in info and not info.get('formats'):
             raise ExtractorError('Unable to extract video url for %s' % info['id'])
         return info
+
+
+class BrightcoveNewIE(InfoExtractor):
+    IE_NAME = 'brightcove:new'
+    _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*videoId=(?P<video_id>\d+)'
+    _TEST = {
+        'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
+        'md5': 'c8100925723840d4b0d243f7025703be',
+        'info_dict': {
+            'id': '4463358922001',
+            'ext': 'mp4',
+            'title': 'Meet the man behind Popcorn Time',
+            'description': 'md5:eac376a4fe366edc70279bfb681aea16',
+            'timestamp': 1441391203,
+            'upload_date': '20150904',
+            'duration': 165.768,
+            'uploader_id': '929656772001',
+        }
+    }
+
+    @staticmethod
+    def _extract_urls(webpage):
+        # Reference:
+        # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe
+        # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript)
+        # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html
+
+        entries = []
+
+        # Look for iframe embeds [1]
+        for _, url in re.findall(
+                r'<iframe[^>]+src=(["\'])((?:https?:)//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage):
+            entries.append(url)
+        # Look for embed_in_page embeds [2]
+        # According to examples from [3] it's unclear whether video id may be optional
+        # and what to do when it is
+        for video_id, account_id, player_id, embed in re.findall(
+                r'''(?sx)
+                    <video[^>]+
+                        data-video-id=["\'](\d+)["\'][^>]*>.*?
+                    </video>.*?
+                    <script[^>]+
+                        src=["\'](?:https?:)?//players\.brightcove\.net/
+                        (\d+)/([\da-f-]+)_([^/]+)/index\.min\.js
+                ''', webpage):
+            entries.append(
+                'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
+                % (account_id, player_id, embed, video_id))
+        return entries
+
+    def _real_extract(self, url):
+        account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups()
+
+        webpage = self._download_webpage(
+            'http://players.brightcove.net/%s/%s_%s/index.min.js'
+            % (account_id, player_id, embed), video_id)
+
+        policy_key = None
+
+        catalog = self._search_regex(
+            r'catalog\(({.+?})\);', webpage, 'catalog', default=None)
+        if catalog:
+            catalog = self._parse_json(
+                js_to_json(catalog), video_id, fatal=False)
+            if catalog:
+                policy_key = catalog.get('policyKey')
+
+        if not policy_key:
+            policy_key = self._search_regex(
+                r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1',
+                webpage, 'policy key', group='pk')
+
+        req = compat_urllib_request.Request(
+            'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s'
+            % (account_id, video_id),
+            headers={'Accept': 'application/json;pk=%s' % policy_key})
+        json_data = self._download_json(req, video_id)
+
+        title = json_data['name']
+
+        formats = []
+        for source in json_data.get('sources', []):
+            source_type = source.get('type')
+            src = source.get('src')
+            if source_type == 'application/x-mpegURL':
+                if not src:
+                    continue
+                m3u8_formats = self._extract_m3u8_formats(
+                    src, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False)
+                if m3u8_formats:
+                    formats.extend(m3u8_formats)
+            else:
+                streaming_src = source.get('streaming_src')
+                stream_name, app_name = source.get('stream_name'), source.get('app_name')
+                if not src and not streaming_src and (not stream_name or not app_name):
+                    continue
+                tbr = float_or_none(source.get('avg_bitrate'), 1000)
+                height = int_or_none(source.get('height'))
+                f = {
+                    'tbr': tbr,
+                    'width': int_or_none(source.get('width')),
+                    'height': height,
+                    'filesize': int_or_none(source.get('size')),
+                    'container': source.get('container'),
+                    'vcodec': source.get('codec'),
+                    'ext': source.get('container').lower(),
+                }
+
+                def build_format_id(kind):
+                    format_id = kind
+                    if tbr:
+                        format_id += '-%dk' % int(tbr)
+                    if height:
+                        format_id += '-%dp' % height
+                    return format_id
+
+                if src or streaming_src:
+                    f.update({
+                        'url': src or streaming_src,
+                        'format_id': build_format_id('http' if src else 'http-streaming'),
+                        'preference': 2 if src else 1,
+                    })
+                else:
+                    f.update({
+                        'url': app_name,
+                        'play_path': stream_name,
+                        'format_id': build_format_id('rtmp'),
+                    })
+                formats.append(f)
+        self._sort_formats(formats)
+
+        description = json_data.get('description')
+        thumbnail = json_data.get('thumbnail')
+        timestamp = parse_iso8601(json_data.get('published_at'))
+        duration = float_or_none(json_data.get('duration'), 1000)
+        tags = json_data.get('tags', [])
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'timestamp': timestamp,
+            'uploader_id': account_id,
+            'formats': formats,
+            'tags': tags,
+        }

youtube_dl/extractor/generic.py

@@ -30,7 +30,10 @@
     url_basename,
     xpath_text,
 )
-from .brightcove import BrightcoveIE
+from .brightcove import (
+    BrightcoveLegacyIE,
+    BrightcoveNewIE,
+)
 from .nbc import NBCSportsVPlayerIE
 from .ooyala import OoyalaIE
 from .rutv import RUTVIE
@@ -275,7 +278,7 @@ class GenericIE(InfoExtractor):
         # it also tests brightcove videos that need to set the 'Referer' in the
         # http requests
         {
-            'add_ie': ['Brightcove'],
+            'add_ie': ['BrightcoveLegacy'],
             'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
             'info_dict': {
                 'id': '2765128793001',
@@ -299,7 +302,7 @@ class GenericIE(InfoExtractor):
                 'uploader': 'thestar.com',
                 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.',
             },
-            'add_ie': ['Brightcove'],
+            'add_ie': ['BrightcoveLegacy'],
         },
         {
             'url': 'http://www.championat.com/video/football/v/87/87499.html',
@@ -314,7 +317,7 @@ class GenericIE(InfoExtractor):
         },
         {
             # https://github.com/rg3/youtube-dl/issues/3541
-            'add_ie': ['Brightcove'],
+            'add_ie': ['BrightcoveLegacy'],
             'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
             'info_dict': {
                 'id': '3866516442001',
@@ -1031,6 +1034,17 @@ class GenericIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'cinemasnob',
             },
+        },
+        # BrightcoveInPageEmbed embed
+        {
+            'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/',
+            'info_dict': {
+                'id': '4238694884001',
+                'ext': 'flv',
+                'title': 'Tabletop: Dread, Last Thoughts',
+                'description': 'Tabletop: Dread, Last Thoughts',
+                'duration': 51690,
+            },
         }
     ]
 
@@ -1290,14 +1304,14 @@ def _playlist_from_matches(matches, getter=None, ie=None):
             return self.playlist_result(
                 urlrs, playlist_id=video_id, playlist_title=video_title)
 
-        # Look for BrightCove:
+        # Look for Brightcove Legacy Studio embeds
-        bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
+        bc_urls = BrightcoveLegacyIE._extract_brightcove_urls(webpage)
         if bc_urls:
             self.to_screen('Brightcove video detected.')
             entries = [{
                 '_type': 'url',
                 'url': smuggle_url(bc_url, {'Referer': url}),
-                'ie_key': 'Brightcove'
+                'ie_key': 'BrightcoveLegacy'
             } for bc_url in bc_urls]
 
             return {
@@ -1307,6 +1321,11 @@ def _playlist_from_matches(matches, getter=None, ie=None):
                 'entries': entries,
             }
 
+        # Look for Brightcove New Studio embeds
+        bc_urls = BrightcoveNewIE._extract_urls(webpage)
+        if bc_urls:
+            return _playlist_from_matches(bc_urls, ie='BrightcoveNew')
+
         # Look for embedded rtl.nl player
         matches = re.findall(
             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',

youtube_dl/extractor/nowness.py

@@ -1,7 +1,7 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-from .brightcove import BrightcoveIE
+from .brightcove import BrightcoveLegacyIE
 from .common import InfoExtractor
 from ..utils import ExtractorError
 from ..compat import (
@@ -22,10 +22,10 @@ def _extract_url_result(self, post):
                             'http://www.nowness.com/iframe?id=%s' % video_id, video_id,
                             note='Downloading player JavaScript',
                             errnote='Unable to download player JavaScript')
-                        bc_url = BrightcoveIE._extract_brightcove_url(player_code)
+                        bc_url = BrightcoveLegacyIE._extract_brightcove_url(player_code)
                         if bc_url is None:
                             raise ExtractorError('Could not find player definition')
-                        return self.url_result(bc_url, 'Brightcove')
+                        return self.url_result(bc_url, 'BrightcoveLegacy')
                     elif source == 'vimeo':
                         return self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
                     elif source == 'youtube':

youtube_dl/extractor/safari.py

@@ -4,7 +4,7 @@
 import re
 
 from .common import InfoExtractor
-from .brightcove import BrightcoveIE
+from .brightcove import BrightcoveLegacyIE
 
 from ..compat import (
     compat_urllib_parse,
@@ -112,11 +112,11 @@ def _real_extract(self, url):
             '%s/%s/chapter-content/%s.html' % (self._API_BASE, course_id, part),
             part)
 
-        bc_url = BrightcoveIE._extract_brightcove_url(webpage)
+        bc_url = BrightcoveLegacyIE._extract_brightcove_url(webpage)
         if not bc_url:
             raise ExtractorError('Could not extract Brightcove URL from %s' % url, expected=True)
 
-        return self.url_result(smuggle_url(bc_url, {'Referer': url}), 'Brightcove')
+        return self.url_result(smuggle_url(bc_url, {'Referer': url}), 'BrightcoveLegacy')
 
 
 class SafariCourseIE(SafariBaseIE):

youtube_dl/extractor/space.py

@@ -3,14 +3,14 @@
 import re
 
 from .common import InfoExtractor
-from .brightcove import BrightcoveIE
+from .brightcove import BrightcoveLegacyIE
 from ..utils import RegexNotFoundError, ExtractorError
 
 
 class SpaceIE(InfoExtractor):
     _VALID_URL = r'https?://(?:(?:www|m)\.)?space\.com/\d+-(?P<title>[^/\.\?]*?)-video\.html'
     _TEST = {
-        'add_ie': ['Brightcove'],
+        'add_ie': ['BrightcoveLegacy'],
         'url': 'http://www.space.com/23373-huge-martian-landforms-detail-revealed-by-european-probe-video.html',
         'info_dict': {
             'id': '2780937028001',
@@ -31,8 +31,8 @@ def _real_extract(self, url):
             brightcove_url = self._og_search_video_url(webpage)
         except RegexNotFoundError:
             # Other videos works fine with the info from the object
-            brightcove_url = BrightcoveIE._extract_brightcove_url(webpage)
+            brightcove_url = BrightcoveLegacyIE._extract_brightcove_url(webpage)
         if brightcove_url is None:
             raise ExtractorError(
                 'The webpage does not contain a video', expected=True)
-        return self.url_result(brightcove_url, BrightcoveIE.ie_key())
+        return self.url_result(brightcove_url, BrightcoveLegacyIE.ie_key())

youtube_dl/extractor/tlc.py

@@ -3,7 +3,7 @@
 import re
 
 from .common import InfoExtractor
-from .brightcove import BrightcoveIE
+from .brightcove import BrightcoveLegacyIE
 from .discovery import DiscoveryIE
 from ..compat import compat_urlparse
 
@@ -66,6 +66,6 @@ def _real_extract(self, url):
 
         return {
             '_type': 'url',
-            'url': BrightcoveIE._extract_brightcove_url(iframe),
+            'url': BrightcoveLegacyIE._extract_brightcove_url(iframe),
-            'ie': BrightcoveIE.ie_key(),
+            'ie': BrightcoveLegacyIE.ie_key(),
         }