X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fooyala.py;h=20b984288ac7bba05bef7d4a34d5f01b5cb0f851;hb=2e7e561c1d9dedf1a8e5a206e1ef86cfa4599956;hp=592cdc5648f13e6d77c7b2f32fd819114c9c4912;hpb=90bddb6cdd59107d137c13970dc50a6193d204a7;p=youtube-dl diff --git a/youtube_dl/extractor/ooyala.py b/youtube_dl/extractor/ooyala.py index 592cdc564..20b984288 100644 --- a/youtube_dl/extractor/ooyala.py +++ b/youtube_dl/extractor/ooyala.py @@ -6,14 +6,19 @@ from .common import InfoExtractor from ..utils import ( int_or_none, float_or_none, + ExtractorError, + unsmuggle_url, ) +from ..compat import compat_urllib_parse class OoyalaBaseIE(InfoExtractor): + _PLAYER_BASE = 'http://player.ooyala.com/' + _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/' + _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v1/authorization/embed_code/%s/%s?' - def _extract(self, player_url, video_id): - print(player_url) - content_tree = self._download_json(player_url, video_id)['content_tree'] + def _extract(self, content_tree_url, video_id, domain='example.org'): + content_tree = self._download_json(content_tree_url, video_id)['content_tree'] metadata = content_tree[list(content_tree)[0]] embed_code = metadata['embed_code'] pcode = metadata.get('asset_pcode') or embed_code @@ -22,39 +27,55 @@ class OoyalaBaseIE(InfoExtractor): 'title': metadata['title'], 'description': metadata.get('description'), 'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'), - 'duration': int_or_none(metadata.get('duration')), + 'duration': float_or_none(metadata.get('duration'), 1000), } + urls = [] formats = [] for supported_format in ('mp4', 'm3u8', 'hds', 'rtmp'): auth_data = self._download_json( - 'http://player.ooyala.com/sas/player_api/v1/authorization/embed_code/%s/%s?domain=www.example.org&supportedFormats=%s' % (pcode, embed_code, supported_format), + self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code) + + compat_urllib_parse.urlencode({ + 'domain': domain, + 'supportedFormats': supported_format + }), video_id, 'Downloading %s JSON' % supported_format) cur_auth_data = auth_data['authorization_data'][embed_code] - for stream in cur_auth_data['streams']: - url = base64.b64decode(stream['url']['data'].encode('ascii')).decode('utf-8') - delivery_type = stream['delivery_type'] - if delivery_type == 'remote_asset': - video_info['url'] = url - return video_info - if delivery_type == 'hls': - formats.extend(self._extract_m3u8_formats(url, embed_code, 'mp4', 'm3u8_native', 0, m3u8_id='hls', fatal=False)) - elif delivery_type == 'hds': - formats.extend(self._extract_f4m_formats(url, embed_code, f4m_id='hds', fatal=False)) - else: - formats.append({ - 'url': url, - 'ext': stream.get('delivery_type'), - 'vcodec': stream.get('video_codec'), - 'format_id': stream.get('profile'), - 'width': int_or_none(stream.get('width')), - 'height': int_or_none(stream.get('height')), - 'abr': int_or_none(stream.get('audio_bitrate')), - 'vbr': int_or_none(stream.get('video_bitrate')), - 'fps': float_or_none(stream.get('framerate')), - }) + if cur_auth_data['authorized']: + for stream in cur_auth_data['streams']: + url = base64.b64decode( + stream['url']['data'].encode('ascii')).decode('utf-8') + if url in urls: + continue + urls.append(url) + delivery_type = stream['delivery_type'] + if delivery_type == 'hls' or '.m3u8' in url: + formats.extend(self._extract_m3u8_formats( + url, embed_code, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + elif delivery_type == 'hds' or '.f4m' in url: + formats.extend(self._extract_f4m_formats( + url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False)) + elif '.smil' in url: + formats.extend(self._extract_smil_formats( + url, embed_code, fatal=False)) + else: + formats.append({ + 'url': url, + 'ext': stream.get('delivery_type'), + 'vcodec': stream.get('video_codec'), + 'format_id': delivery_type, + 'width': int_or_none(stream.get('width')), + 'height': int_or_none(stream.get('height')), + 'abr': int_or_none(stream.get('audio_bitrate')), + 'vbr': int_or_none(stream.get('video_bitrate')), + 'fps': float_or_none(stream.get('framerate')), + }) + else: + raise ExtractorError('%s said: %s' % ( + self.IE_NAME, cur_auth_data['message']), expected=True) self._sort_formats(formats) video_info['formats'] = formats @@ -73,7 +94,7 @@ class OoyalaIE(OoyalaBaseIE): 'ext': 'mp4', '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.', - 'duration': 853386, + 'duration': 853.386, }, }, { # Only available for ipad @@ -82,7 +103,7 @@ class OoyalaIE(OoyalaBaseIE): 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0', 'ext': 'mp4', 'title': 'Simulation Overview - Levels of Simulation', - 'duration': 194948, + 'duration': 194.948, }, }, { @@ -94,7 +115,7 @@ class OoyalaIE(OoyalaBaseIE): 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx', 'ext': 'mp4', 'title': 'Divide Tool Path.mp4', - 'duration': 204405, + 'duration': 204.405, } } ] @@ -109,9 +130,11 @@ class OoyalaIE(OoyalaBaseIE): ie=cls.ie_key()) def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) embed_code = self._match_id(url) - content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/embed_code/%s/%s' % (embed_code, embed_code) - return self._extract(content_tree_url, embed_code) + domain = smuggled_data.get('domain') + content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code) + return self._extract(content_tree_url, embed_code, domain) class OoyalaExternalIE(OoyalaBaseIE): @@ -137,7 +160,7 @@ class OoyalaExternalIE(OoyalaBaseIE): 'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG', 'ext': 'mp4', 'title': 'dm_140128_30for30Shorts___JudgingJewellv2', - 'duration': 1302000, + 'duration': 1302.0, }, 'params': { # m3u8 download @@ -147,5 +170,5 @@ class OoyalaExternalIE(OoyalaBaseIE): def _real_extract(self, url): partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups() - content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/external_id/%s/%s:%s' % (pcode, partner_id, video_id) + content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id) return self._extract(content_tree_url, video_id)