[yt-dlp.git] / youtube_dl / extractor / limelight.py

# coding: utf-8
from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..utils import (
    determine_ext,
    float_or_none,
    int_or_none,
)


class LimelightBaseIE(InfoExtractor):
    _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
    _API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json'

    def _call_playlist_service(self, item_id, method, fatal=True):
        return self._download_json(
            self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
            item_id, 'Downloading PlaylistService %s JSON' % method, fatal=fatal)

    def _call_api(self, organization_id, item_id, method):
        return self._download_json(
            self._API_URL % (organization_id, self._API_PATH, item_id, method),
            item_id, 'Downloading API %s JSON' % method)

    def _extract(self, item_id, pc_method, mobile_method, meta_method):
        pc = self._call_playlist_service(item_id, pc_method)
        metadata = self._call_api(pc['orgId'], item_id, meta_method)
        mobile = self._call_playlist_service(item_id, mobile_method, fatal=False)
        return pc, mobile, metadata

    def _extract_info(self, streams, mobile_urls, properties):
        video_id = properties['media_id']
        formats = []

        for stream in streams:
            stream_url = stream.get('url')
            if not stream_url:
                continue
            if '.f4m' in stream_url:
                formats.extend(self._extract_f4m_formats(
                    stream_url, video_id, fatal=False))
            else:
                fmt = {
                    'url': stream_url,
                    'abr': float_or_none(stream.get('audioBitRate')),
                    'vbr': float_or_none(stream.get('videoBitRate')),
                    'fps': float_or_none(stream.get('videoFrameRate')),
                    'width': int_or_none(stream.get('videoWidthInPixels')),
                    'height': int_or_none(stream.get('videoHeightInPixels')),
                    'ext': determine_ext(stream_url)
                }
                rtmp = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url)
                if rtmp:
                    format_id = 'rtmp'
                    if stream.get('videoBitRate'):
                        format_id += '-%d' % int_or_none(stream['videoBitRate'])
                    fmt.update({
                        'url': rtmp.group('url'),
                        'play_path': rtmp.group('playpath'),
                        'app': rtmp.group('app'),
                        'ext': 'flv',
                        'format_id': format_id,
                    })
                formats.append(fmt)

        for mobile_url in mobile_urls:
            media_url = mobile_url.get('mobileUrl')
            if not media_url:
                continue
            format_id = mobile_url.get('targetMediaPlatform')
            if determine_ext(media_url) == 'm3u8':
                formats.extend(self._extract_m3u8_formats(
                    media_url, video_id, 'mp4', 'm3u8_native',
                    m3u8_id=format_id, fatal=False))
            else:
                formats.append({
                    'url': media_url,
                    'format_id': format_id,
                    'preference': -1,
                })

        self._sort_formats(formats)

        title = properties['title']
        description = properties.get('description')
        timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date'))
        duration = float_or_none(properties.get('duration_in_milliseconds'), 1000)
        filesize = int_or_none(properties.get('total_storage_in_bytes'))
        categories = [properties.get('category')]
        tags = properties.get('tags', [])
        thumbnails = [{
            'url': thumbnail['url'],
            'width': int_or_none(thumbnail.get('width')),
            'height': int_or_none(thumbnail.get('height')),
        } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]

        subtitles = {}
        for caption in properties.get('captions', {}):
            lang = caption.get('language_code')
            subtitles_url = caption.get('url')
            if lang and subtitles_url:
                subtitles[lang] = [{
                    'url': subtitles_url,
                }]

        return {
            'id': video_id,
            'title': title,
            'description': description,
            'formats': formats,
            'timestamp': timestamp,
            'duration': duration,
            'filesize': filesize,
            'categories': categories,
            'tags': tags,
            'thumbnails': thumbnails,
            'subtitles': subtitles,
        }


class LimelightMediaIE(LimelightBaseIE):
    IE_NAME = 'limelight'
    _VALID_URL = r'(?:limelight:media:|https?://link\.videoplatform\.limelight\.com/media/\??\bmediaId=)(?P<id>[a-z0-9]{32})'
    _TESTS = [{
        'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
        'info_dict': {
            'id': '3ffd040b522b4485b6d84effc750cd86',
            'ext': 'flv',
            'title': 'HaP and the HB Prince Trailer',
            'description': 'md5:8005b944181778e313d95c1237ddb640',
            'thumbnail': 're:^https?://.*\.jpeg$',
            'duration': 144.23,
            'timestamp': 1244136834,
            'upload_date': '20090604',
        },
        'params': {
            # rtmp download
            'skip_download': True,
        },
    }, {
        # video with subtitles
        'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
        'info_dict': {
            'id': 'a3e00274d4564ec4a9b29b9466432335',
            'ext': 'flv',
            'title': '3Play Media Overview Video',
            'description': '',
            'thumbnail': 're:^https?://.*\.jpeg$',
            'duration': 78.101,
            'timestamp': 1338929955,
            'upload_date': '20120605',
            'subtitles': 'mincount:9',
        },
        'params': {
            # rtmp download
            'skip_download': True,
        },
    }]
    _PLAYLIST_SERVICE_PATH = 'media'
    _API_PATH = 'media'

    def _real_extract(self, url):
        video_id = self._match_id(url)

        pc, mobile, metadata = self._extract(
            video_id, 'getPlaylistByMediaId', 'getMobilePlaylistByMediaId', 'properties')

        return self._extract_info(
            pc['playlistItems'][0].get('streams', []),
            mobile['mediaList'][0].get('mobileUrls', []) if mobile else [],
            metadata)


class LimelightChannelIE(LimelightBaseIE):
    IE_NAME = 'limelight:channel'
    _VALID_URL = r'(?:limelight:channel:|https?://link\.videoplatform\.limelight\.com/media/\??\bchannelId=)(?P<id>[a-z0-9]{32})'
    _TEST = {
        'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
        'info_dict': {
            'id': 'ab6a524c379342f9b23642917020c082',
            'title': 'Javascript Sample Code',
        },
        'playlist_mincount': 3,
    }
    _PLAYLIST_SERVICE_PATH = 'channel'
    _API_PATH = 'channels'

    def _real_extract(self, url):
        channel_id = self._match_id(url)

        pc, mobile, medias = self._extract(
            channel_id, 'getPlaylistByChannelId',
            'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1', 'media')

        entries = [
            self._extract_info(
                pc['playlistItems'][i].get('streams', []),
                mobile['mediaList'][i].get('mobileUrls', []) if mobile else [],
                medias['media_list'][i])
            for i in range(len(medias['media_list']))]

        return self.playlist_result(entries, channel_id, pc['title'])


class LimelightChannelListIE(LimelightBaseIE):
    IE_NAME = 'limelight:channel_list'
    _VALID_URL = r'(?:limelight:channel_list:|https?://link\.videoplatform\.limelight\.com/media/\?.*?\bchannelListId=)(?P<id>[a-z0-9]{32})'
    _TEST = {
        'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
        'info_dict': {
            'id': '301b117890c4465c8179ede21fd92e2b',
            'title': 'Website - Hero Player',
        },
        'playlist_mincount': 2,
    }
    _PLAYLIST_SERVICE_PATH = 'channel_list'

    def _real_extract(self, url):
        channel_list_id = self._match_id(url)

        channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')

        entries = [
            self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
            for channel in channel_list['channelList']]

        return self.playlist_result(entries, channel_list_id, channel_list['title'])
Commit	Line	Data
ef5acfe3	1	# coding: utf-8
	2	from __future__ import unicode_literals
	3
	4	import re
	5
	6	from .common import InfoExtractor
	7	from ..utils import (
ef5acfe3	8	determine_ext,
d7fc5631 S	9	float_or_none,
d7fc5631 S	10	int_or_none,
ef5acfe3	11	)
	12
	13
d7fc5631 S	14	class LimelightBaseIE(InfoExtractor):
	15	_PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
	16	_API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json'
ef5acfe3	17
d7fc5631 S	18	def _call_playlist_service(self, item_id, method, fatal=True):
	19	return self._download_json(
	20	self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
	21	item_id, 'Downloading PlaylistService %s JSON' % method, fatal=fatal)
ef5acfe3	22
d7fc5631 S	23	def _call_api(self, organization_id, item_id, method):
	24	return self._download_json(
	25	self._API_URL % (organization_id, self._API_PATH, item_id, method),
	26	item_id, 'Downloading API %s JSON' % method)
ef5acfe3	27
d7fc5631 S	28	def _extract(self, item_id, pc_method, mobile_method, meta_method):
	29	pc = self._call_playlist_service(item_id, pc_method)
	30	metadata = self._call_api(pc['orgId'], item_id, meta_method)
	31	mobile = self._call_playlist_service(item_id, mobile_method, fatal=False)
	32	return pc, mobile, metadata
	33
	34	def _extract_info(self, streams, mobile_urls, properties):
ef5acfe3	35	video_id = properties['media_id']
	36	formats = []
	37
ef5acfe3	38	for stream in streams:
d7fc5631 S	39	stream_url = stream.get('url')
	40	if not stream_url:
	41	continue
	42	if '.f4m' in stream_url:
8f1fddc8	43	formats.extend(self._extract_f4m_formats(
8f1fddc8	44	stream_url, video_id, fatal=False))
ef5acfe3	45	else:
ef5acfe3	46	fmt = {
d7fc5631 S	47	'url': stream_url,
	48	'abr': float_or_none(stream.get('audioBitRate')),
	49	'vbr': float_or_none(stream.get('videoBitRate')),
	50	'fps': float_or_none(stream.get('videoFrameRate')),
	51	'width': int_or_none(stream.get('videoWidthInPixels')),
	52	'height': int_or_none(stream.get('videoHeightInPixels')),
	53	'ext': determine_ext(stream_url)
ef5acfe3	54	}
d7fc5631	55	rtmp = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url)
ef5acfe3	56	if rtmp:
d7fc5631 S	57	format_id = 'rtmp'
	58	if stream.get('videoBitRate'):
	59	format_id += '-%d' % int_or_none(stream['videoBitRate'])
ef5acfe3	60	fmt.update({
	61	'url': rtmp.group('url'),
	62	'play_path': rtmp.group('playpath'),
	63	'app': rtmp.group('app'),
d7fc5631 S	64	'ext': 'flv',
d7fc5631 S	65	'format_id': format_id,
ef5acfe3	66	})
	67	formats.append(fmt)
	68
d7fc5631 S	69	for mobile_url in mobile_urls:
	70	media_url = mobile_url.get('mobileUrl')
	71	if not media_url:
	72	continue
	73	format_id = mobile_url.get('targetMediaPlatform')
	74	if determine_ext(media_url) == 'm3u8':
	75	formats.extend(self._extract_m3u8_formats(
8f1fddc8	76	media_url, video_id, 'mp4', 'm3u8_native',
8f1fddc8	77	m3u8_id=format_id, fatal=False))
d7fc5631 S	78	else:
	79	formats.append({
	80	'url': media_url,
	81	'format_id': format_id,
	82	'preference': -1,
	83	})
	84
ef5acfe3	85	self._sort_formats(formats)
	86
	87	title = properties['title']
	88	description = properties.get('description')
d7fc5631 S	89	timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date'))
	90	duration = float_or_none(properties.get('duration_in_milliseconds'), 1000)
	91	filesize = int_or_none(properties.get('total_storage_in_bytes'))
ef5acfe3	92	categories = [properties.get('category')]
d7fc5631	93	tags = properties.get('tags', [])
ef5acfe3	94	thumbnails = [{
d7fc5631	95	'url': thumbnail['url'],
ef5acfe3	96	'width': int_or_none(thumbnail.get('width')),
ef5acfe3	97	'height': int_or_none(thumbnail.get('height')),
d7fc5631 S	98	} for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]
	99
	100	subtitles = {}
	101	for caption in properties.get('captions', {}):
	102	lang = caption.get('language_code')
	103	subtitles_url = caption.get('url')
	104	if lang and subtitles_url:
	105	subtitles[lang] = [{
	106	'url': subtitles_url,
	107	}]
ef5acfe3	108
	109	return {
	110	'id': video_id,
	111	'title': title,
	112	'description': description,
	113	'formats': formats,
	114	'timestamp': timestamp,
	115	'duration': duration,
	116	'filesize': filesize,
	117	'categories': categories,
d7fc5631	118	'tags': tags,
ef5acfe3	119	'thumbnails': thumbnails,
	120	'subtitles': subtitles,
	121	}
	122
	123
d7fc5631	124	class LimelightMediaIE(LimelightBaseIE):
ef5acfe3	125	IE_NAME = 'limelight'
5886b38d	126	_VALID_URL = r'(?:limelight:media:\|https?://link\.videoplatform\.limelight\.com/media/\??\bmediaId=)(?P<id>[a-z0-9]{32})'
9c544e25	127	_TESTS = [{
ef5acfe3	128	'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
ef5acfe3	129	'info_dict': {
ef5acfe3	130	'id': '3ffd040b522b4485b6d84effc750cd86',
d7fc5631	131	'ext': 'flv',
ef5acfe3	132	'title': 'HaP and the HB Prince Trailer',
9c544e25	133	'description': 'md5:8005b944181778e313d95c1237ddb640',
ef5acfe3	134	'thumbnail': 're:^https?://.*\.jpeg$',
d7fc5631	135	'duration': 144.23,
ef5acfe3	136	'timestamp': 1244136834,
d7fc5631 S	137	'upload_date': '20090604',
	138	},
	139	'params': {
	140	# rtmp download
	141	'skip_download': True,
	142	},
9c544e25 S	143	}, {
	144	# video with subtitles
	145	'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
	146	'info_dict': {
	147	'id': 'a3e00274d4564ec4a9b29b9466432335',
	148	'ext': 'flv',
	149	'title': '3Play Media Overview Video',
	150	'description': '',
	151	'thumbnail': 're:^https?://.*\.jpeg$',
	152	'duration': 78.101,
	153	'timestamp': 1338929955,
	154	'upload_date': '20120605',
	155	'subtitles': 'mincount:9',
	156	},
	157	'params': {
	158	# rtmp download
	159	'skip_download': True,
	160	},
	161	}]
d7fc5631 S	162	_PLAYLIST_SERVICE_PATH = 'media'
d7fc5631 S	163	_API_PATH = 'media'
ef5acfe3	164
	165	def _real_extract(self, url):
	166	video_id = self._match_id(url)
	167
d7fc5631 S	168	pc, mobile, metadata = self._extract(
d7fc5631 S	169	video_id, 'getPlaylistByMediaId', 'getMobilePlaylistByMediaId', 'properties')
ef5acfe3	170
d7fc5631 S	171	return self._extract_info(
	172	pc['playlistItems'][0].get('streams', []),
	173	mobile['mediaList'][0].get('mobileUrls', []) if mobile else [],
	174	metadata)
ef5acfe3	175
ef5acfe3	176
d7fc5631	177	class LimelightChannelIE(LimelightBaseIE):
ef5acfe3	178	IE_NAME = 'limelight:channel'
5886b38d	179	_VALID_URL = r'(?:limelight:channel:\|https?://link\.videoplatform\.limelight\.com/media/\??\bchannelId=)(?P<id>[a-z0-9]{32})'
ef5acfe3	180	_TEST = {
	181	'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
	182	'info_dict': {
	183	'id': 'ab6a524c379342f9b23642917020c082',
	184	'title': 'Javascript Sample Code',
	185	},
	186	'playlist_mincount': 3,
	187	}
d7fc5631 S	188	_PLAYLIST_SERVICE_PATH = 'channel'
d7fc5631 S	189	_API_PATH = 'channels'
ef5acfe3	190
	191	def _real_extract(self, url):
	192	channel_id = self._match_id(url)
	193
d7fc5631 S	194	pc, mobile, medias = self._extract(
	195	channel_id, 'getPlaylistByChannelId',
	196	'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1', 'media')
ef5acfe3	197
d7fc5631 S	198	entries = [
	199	self._extract_info(
	200	pc['playlistItems'][i].get('streams', []),
	201	mobile['mediaList'][i].get('mobileUrls', []) if mobile else [],
	202	medias['media_list'][i])
	203	for i in range(len(medias['media_list']))]
ef5acfe3	204
d7fc5631	205	return self.playlist_result(entries, channel_id, pc['title'])
ef5acfe3	206
ef5acfe3	207
d7fc5631	208	class LimelightChannelListIE(LimelightBaseIE):
ef5acfe3	209	IE_NAME = 'limelight:channel_list'
5886b38d	210	_VALID_URL = r'(?:limelight:channel_list:\|https?://link\.videoplatform\.limelight\.com/media/\?.*?\bchannelListId=)(?P<id>[a-z0-9]{32})'
ef5acfe3	211	_TEST = {
	212	'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
	213	'info_dict': {
	214	'id': '301b117890c4465c8179ede21fd92e2b',
	215	'title': 'Website - Hero Player',
	216	},
	217	'playlist_mincount': 2,
	218	}
d7fc5631	219	_PLAYLIST_SERVICE_PATH = 'channel_list'
ef5acfe3	220
	221	def _real_extract(self, url):
	222	channel_list_id = self._match_id(url)
	223
d7fc5631	224	channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')
ef5acfe3	225
d7fc5631 S	226	entries = [
	227	self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
	228	for channel in channel_list['channelList']]
ef5acfe3	229
d7fc5631	230	return self.playlist_result(entries, channel_list_id, channel_list['title'])