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

# coding: utf-8
from __future__ import unicode_literals

import re
import calendar
import datetime

from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
    HEADRequest,
    unified_strdate,
    strip_jsonp,
    int_or_none,
    float_or_none,
    determine_ext,
    remove_end,
    unescapeHTML,
)


class ORFTVthekIE(InfoExtractor):
    IE_NAME = 'orf:tvthek'
    IE_DESC = 'ORF TVthek'
    _VALID_URL = r'https?://tvthek\.orf\.at/(?:[^/]+/)+(?P<id>\d+)'

    _TESTS = [{
        'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389',
        'playlist': [{
            'md5': '2942210346ed779588f428a92db88712',
            'info_dict': {
                'id': '8896777',
                'ext': 'mp4',
                'title': 'Aufgetischt: Mit der Steirischen Tafelrunde',
                'description': 'md5:c1272f0245537812d4e36419c207b67d',
                'duration': 2668,
                'upload_date': '20141208',
            },
        }],
        'skip': 'Blocked outside of Austria / Germany',
    }, {
        'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256',
        'info_dict': {
            'id': '7982259',
            'ext': 'mp4',
            'title': 'Best of Ingrid Thurnher',
            'upload_date': '20140527',
            'description': 'Viele Jahre war Ingrid Thurnher das "Gesicht" der ZIB 2. Vor ihrem Wechsel zur ZIB 2 im Jahr 1995 moderierte sie unter anderem "Land und Leute", "Österreich-Bild" und "Niederösterreich heute".',
        },
        'params': {
            'skip_download': True,  # rtsp downloads
        },
        '_skip': 'Blocked outside of Austria / Germany',
    }, {
        'url': 'http://tvthek.orf.at/topic/Fluechtlingskrise/10463081/Heimat-Fremde-Heimat/13879132/Senioren-betreuen-Migrantenkinder/13879141',
        'skip_download': True,
    }, {
        'url': 'http://tvthek.orf.at/profile/Universum/35429',
        'skip_download': True,
    }]

    def _real_extract(self, url):
        playlist_id = self._match_id(url)
        webpage = self._download_webpage(url, playlist_id)

        data_jsb = self._parse_json(
            self._search_regex(
                r'<div[^>]+class=(["\']).*?VideoPlaylist.*?\1[^>]+data-jsb=(["\'])(?P<json>.+?)\2',
                webpage, 'playlist', group='json'),
            playlist_id, transform_source=unescapeHTML)['playlist']['videos']

        def quality_to_int(s):
            m = re.search('([0-9]+)', s)
            if m is None:
                return -1
            return int(m.group(1))

        entries = []
        for sd in data_jsb:
            video_id, title = sd.get('id'), sd.get('title')
            if not video_id or not title:
                continue
            video_id = compat_str(video_id)
            formats = [{
                'preference': -10 if fd['delivery'] == 'hls' else None,
                'format_id': '%s-%s-%s' % (
                    fd['delivery'], fd['quality'], fd['quality_string']),
                'url': fd['src'],
                'protocol': fd['protocol'],
                'quality': quality_to_int(fd['quality']),
            } for fd in sd['sources']]

            # Check for geoblocking.
            # There is a property is_geoprotection, but that's always false
            geo_str = sd.get('geoprotection_string')
            if geo_str:
                try:
                    http_url = next(
                        f['url']
                        for f in formats
                        if re.match(r'^https?://.*\.mp4$', f['url']))
                except StopIteration:
                    pass
                else:
                    req = HEADRequest(http_url)
                    self._request_webpage(
                        req, video_id,
                        note='Testing for geoblocking',
                        errnote=((
                            'This video seems to be blocked outside of %s. '
                            'You may want to try the streaming-* formats.')
                            % geo_str),
                        fatal=False)

            self._check_formats(formats, video_id)
            self._sort_formats(formats)

            subtitles = {}
            for sub in sd.get('subtitles', []):
                sub_src = sub.get('src')
                if not sub_src:
                    continue
                subtitles.setdefault(sub.get('lang', 'de-AT'), []).append({
                    'url': sub_src,
                })

            upload_date = unified_strdate(sd.get('created_date'))
            entries.append({
                '_type': 'video',
                'id': video_id,
                'title': title,
                'formats': formats,
                'subtitles': subtitles,
                'description': sd.get('description'),
                'duration': int_or_none(sd.get('duration_in_seconds')),
                'upload_date': upload_date,
                'thumbnail': sd.get('image_full_url'),
            })

        return {
            '_type': 'playlist',
            'entries': entries,
            'id': playlist_id,
        }


class ORFOE1IE(InfoExtractor):
    IE_NAME = 'orf:oe1'
    IE_DESC = 'Radio Österreich 1'
    _VALID_URL = r'https?://oe1\.orf\.at/(?:programm/|konsole\?.*?\btrack_id=)(?P<id>[0-9]+)'

    # Audios on ORF radio are only available for 7 days, so we can't add tests.
    _TESTS = [{
        'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
        'only_matching': True,
    }, {
        'url': 'http://oe1.orf.at/konsole?show=ondemand&track_id=443608&load_day=/programm/konsole/tag/20160726',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        show_id = self._match_id(url)
        data = self._download_json(
            'http://oe1.orf.at/programm/%s/konsole' % show_id,
            show_id
        )

        timestamp = datetime.datetime.strptime('%s %s' % (
            data['item']['day_label'],
            data['item']['time']
        ), '%d.%m.%Y %H:%M')
        unix_timestamp = calendar.timegm(timestamp.utctimetuple())

        return {
            'id': show_id,
            'title': data['item']['title'],
            'url': data['item']['url_stream'],
            'ext': 'mp3',
            'description': data['item'].get('info'),
            'timestamp': unix_timestamp
        }


class ORFFM4IE(InfoExtractor):
    IE_NAME = 'orf:fm4'
    IE_DESC = 'radio FM4'
    _VALID_URL = r'https?://fm4\.orf\.at/(?:7tage/?#|player/)(?P<date>[0-9]+)/(?P<show>\w+)'

    _TEST = {
        'url': 'http://fm4.orf.at/player/20160110/IS/',
        'md5': '01e736e8f1cef7e13246e880a59ad298',
        'info_dict': {
            'id': '2016-01-10_2100_tl_54_7DaysSun13_11244',
            'ext': 'mp3',
            'title': 'Im Sumpf',
            'description': 'md5:384c543f866c4e422a55f66a62d669cd',
            'duration': 7173,
            'timestamp': 1452456073,
            'upload_date': '20160110',
        },
        'skip': 'Live streams on FM4 got deleted soon',
    }

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        show_date = mobj.group('date')
        show_id = mobj.group('show')

        data = self._download_json(
            'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
            show_id
        )

        def extract_entry_dict(info, title, subtitle):
            return {
                'id': info['loopStreamId'].replace('.mp3', ''),
                'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
                'title': title,
                'description': subtitle,
                'duration': (info['end'] - info['start']) / 1000,
                'timestamp': info['start'] / 1000,
                'ext': 'mp3'
            }

        entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]

        return {
            '_type': 'playlist',
            'id': show_id,
            'title': data['title'],
            'description': data['subtitle'],
            'entries': entries
        }


class ORFIPTVIE(InfoExtractor):
    IE_NAME = 'orf:iptv'
    IE_DESC = 'iptv.ORF.at'
    _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'

    _TEST = {
        'url': 'http://iptv.orf.at/stories/2275236/',
        'md5': 'c8b22af4718a4b4af58342529453e3e5',
        'info_dict': {
            'id': '350612',
            'ext': 'flv',
            'title': 'Weitere Evakuierungen um Vulkan Calbuco',
            'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
            'duration': 68.197,
            'thumbnail': r're:^https?://.*\.jpg$',
            'upload_date': '20150425',
        },
    }

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

        webpage = self._download_webpage(
            'http://iptv.orf.at/stories/%s' % story_id, story_id)

        video_id = self._search_regex(
            r'data-video(?:id)?="(\d+)"', webpage, 'video id')

        data = self._download_json(
            'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
            video_id)[0]

        duration = float_or_none(data['duration'], 1000)

        video = data['sources']['default']
        load_balancer_url = video['loadBalancerUrl']
        abr = int_or_none(video.get('audioBitrate'))
        vbr = int_or_none(video.get('bitrate'))
        fps = int_or_none(video.get('videoFps'))
        width = int_or_none(video.get('videoWidth'))
        height = int_or_none(video.get('videoHeight'))
        thumbnail = video.get('preview')

        rendition = self._download_json(
            load_balancer_url, video_id, transform_source=strip_jsonp)

        f = {
            'abr': abr,
            'vbr': vbr,
            'fps': fps,
            'width': width,
            'height': height,
        }

        formats = []
        for format_id, format_url in rendition['redirect'].items():
            if format_id == 'rtmp':
                ff = f.copy()
                ff.update({
                    'url': format_url,
                    'format_id': format_id,
                })
                formats.append(ff)
            elif determine_ext(format_url) == 'f4m':
                formats.extend(self._extract_f4m_formats(
                    format_url, video_id, f4m_id=format_id))
            elif determine_ext(format_url) == 'm3u8':
                formats.extend(self._extract_m3u8_formats(
                    format_url, video_id, 'mp4', m3u8_id=format_id))
            else:
                continue
        self._sort_formats(formats)

        title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
        description = self._og_search_description(webpage)
        upload_date = unified_strdate(self._html_search_meta(
            'dc.date', webpage, 'upload date'))

        return {
            'id': video_id,
            'title': title,
            'description': description,
            'duration': duration,
            'thumbnail': thumbnail,
            'upload_date': upload_date,
            'formats': formats,
        }
Commit	Line	Data
89284910	1	# coding: utf-8
5d73273f	2	from __future__ import unicode_literals
89284910	3
5d73273f	4	import re
eb368012 S	5	import calendar
eb368012 S	6	import datetime
54543467 JMF	7
54543467 JMF	8	from .common import InfoExtractor
73c801d6	9	from ..compat import compat_str
54543467	10	from ..utils import (
5d73273f PH	11	HEADRequest,
5d73273f PH	12	unified_strdate,
14137b57 S	13	strip_jsonp,
	14	int_or_none,
	15	float_or_none,
	16	determine_ext,
	17	remove_end,
73c801d6	18	unescapeHTML,
54543467 JMF	19	)
54543467 JMF	20
5d73273f	21
eb368012 S	22	class ORFTVthekIE(InfoExtractor):
	23	IE_NAME = 'orf:tvthek'
	24	IE_DESC = 'ORF TVthek'
73c801d6	25	_VALID_URL = r'https?://tvthek\.orf\.at/(?:[^/]+/)+(?P<id>\d+)'
5d73273f	26
13d27a42	27	_TESTS = [{
a6620ac2 PH	28	'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389',
	29	'playlist': [{
	30	'md5': '2942210346ed779588f428a92db88712',
	31	'info_dict': {
	32	'id': '8896777',
	33	'ext': 'mp4',
	34	'title': 'Aufgetischt: Mit der Steirischen Tafelrunde',
	35	'description': 'md5:c1272f0245537812d4e36419c207b67d',
	36	'duration': 2668,
	37	'upload_date': '20141208',
	38	},
	39	}],
13d27a42 PH	40	'skip': 'Blocked outside of Austria / Germany',
	41	}, {
	42	'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256',
7b0d333a NP	43	'info_dict': {
	44	'id': '7982259',
	45	'ext': 'mp4',
	46	'title': 'Best of Ingrid Thurnher',
	47	'upload_date': '20140527',
	48	'description': 'Viele Jahre war Ingrid Thurnher das "Gesicht" der ZIB 2. Vor ihrem Wechsel zur ZIB 2 im Jahr 1995 moderierte sie unter anderem "Land und Leute", "Österreich-Bild" und "Niederösterreich heute".',
	49	},
	50	'params': {
	51	'skip_download': True, # rtsp downloads
	52	},
13d27a42	53	'_skip': 'Blocked outside of Austria / Germany',
73c801d6 S	54	}, {
	55	'url': 'http://tvthek.orf.at/topic/Fluechtlingskrise/10463081/Heimat-Fremde-Heimat/13879132/Senioren-betreuen-Migrantenkinder/13879141',
	56	'skip_download': True,
	57	}, {
	58	'url': 'http://tvthek.orf.at/profile/Universum/35429',
	59	'skip_download': True,
13d27a42	60	}]
54543467	61
54543467	62	def _real_extract(self, url):
a6620ac2	63	playlist_id = self._match_id(url)
54543467 JMF	64	webpage = self._download_webpage(url, playlist_id)
54543467 JMF	65
73c801d6 S	66	data_jsb = self._parse_json(
	67	self._search_regex(
	68	r'<div[^>]+class=(["\']).?VideoPlaylist.?\1[^>]+data-jsb=(["\'])(?P<json>.+?)\2',
	69	webpage, 'playlist', group='json'),
	70	playlist_id, transform_source=unescapeHTML)['playlist']['videos']
5d73273f PH	71
	72	def quality_to_int(s):
	73	m = re.search('([0-9]+)', s)
	74	if m is None:
	75	return -1
	76	return int(m.group(1))
	77
	78	entries = []
73c801d6 S	79	for sd in data_jsb:
	80	video_id, title = sd.get('id'), sd.get('title')
	81	if not video_id or not title:
	82	continue
	83	video_id = compat_str(video_id)
5d73273f PH	84	formats = [{
	85	'preference': -10 if fd['delivery'] == 'hls' else None,
	86	'format_id': '%s-%s-%s' % (
	87	fd['delivery'], fd['quality'], fd['quality_string']),
	88	'url': fd['src'],
	89	'protocol': fd['protocol'],
	90	'quality': quality_to_int(fd['quality']),
73c801d6	91	} for fd in sd['sources']]
5d73273f PH	92
	93	# Check for geoblocking.
	94	# There is a property is_geoprotection, but that's always false
	95	geo_str = sd.get('geoprotection_string')
	96	if geo_str:
	97	try:
	98	http_url = next(
	99	f['url']
	100	for f in formats
	101	if re.match(r'^https?://.*\.mp4$', f['url']))
	102	except StopIteration:
	103	pass
	104	else:
	105	req = HEADRequest(http_url)
4f81667d	106	self._request_webpage(
5d73273f PH	107	req, video_id,
	108	note='Testing for geoblocking',
	109	errnote=((
	110	'This video seems to be blocked outside of %s. '
	111	'You may want to try the streaming-* formats.')
	112	% geo_str),
	113	fatal=False)
	114
e277f2a6	115	self._check_formats(formats, video_id)
5d73273f PH	116	self._sort_formats(formats)
5d73273f PH	117
791d29db RA	118	subtitles = {}
	119	for sub in sd.get('subtitles', []):
	120	sub_src = sub.get('src')
	121	if not sub_src:
	122	continue
	123	subtitles.setdefault(sub.get('lang', 'de-AT'), []).append({
	124	'url': sub_src,
	125	})
	126
73c801d6	127	upload_date = unified_strdate(sd.get('created_date'))
5d73273f	128	entries.append({
54543467	129	'_type': 'video',
5d73273f	130	'id': video_id,
73c801d6	131	'title': title,
5d73273f	132	'formats': formats,
791d29db	133	'subtitles': subtitles,
5d73273f	134	'description': sd.get('description'),
73c801d6	135	'duration': int_or_none(sd.get('duration_in_seconds')),
5d73273f PH	136	'upload_date': upload_date,
	137	'thumbnail': sd.get('image_full_url'),
	138	})
	139
	140	return {
	141	'_type': 'playlist',
	142	'entries': entries,
	143	'id': playlist_id,
	144	}
eb368012 S	145
eb368012 S	146
eb368012 S	147	class ORFOE1IE(InfoExtractor):
	148	IE_NAME = 'orf:oe1'
	149	IE_DESC = 'Radio Österreich 1'
3e050d51	150	_VALID_URL = r'https?://oe1\.orf\.at/(?:programm/\|konsole\?.*?\btrack_id=)(?P<id>[0-9]+)'
27ca82eb PH	151
27ca82eb PH	152	# Audios on ORF radio are only available for 7 days, so we can't add tests.
3e050d51	153	_TESTS = [{
27ca82eb PH	154	'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
27ca82eb PH	155	'only_matching': True,
3e050d51 S	156	}, {
	157	'url': 'http://oe1.orf.at/konsole?show=ondemand&track_id=443608&load_day=/programm/konsole/tag/20160726',
	158	'only_matching': True,
	159	}]
eb368012 S	160
eb368012 S	161	def _real_extract(self, url):
a6620ac2	162	show_id = self._match_id(url)
eb368012 S	163	data = self._download_json(
	164	'http://oe1.orf.at/programm/%s/konsole' % show_id,
	165	show_id
	166	)
	167
	168	timestamp = datetime.datetime.strptime('%s %s' % (
	169	data['item']['day_label'],
	170	data['item']['time']
	171	), '%d.%m.%Y %H:%M')
	172	unix_timestamp = calendar.timegm(timestamp.utctimetuple())
	173
	174	return {
	175	'id': show_id,
	176	'title': data['item']['title'],
	177	'url': data['item']['url_stream'],
	178	'ext': 'mp3',
	179	'description': data['item'].get('info'),
	180	'timestamp': unix_timestamp
	181	}
	182
	183
	184	class ORFFM4IE(InfoExtractor):
b9f030cc	185	IE_NAME = 'orf:fm4'
eb368012	186	IE_DESC = 'radio FM4'
5886b38d	187	_VALID_URL = r'https?://fm4\.orf\.at/(?:7tage/?#\|player/)(?P<date>[0-9]+)/(?P<show>\w+)'
eb368012	188
98698606 S	189	_TEST = {
	190	'url': 'http://fm4.orf.at/player/20160110/IS/',
	191	'md5': '01e736e8f1cef7e13246e880a59ad298',
	192	'info_dict': {
	193	'id': '2016-01-10_2100_tl_54_7DaysSun13_11244',
	194	'ext': 'mp3',
	195	'title': 'Im Sumpf',
	196	'description': 'md5:384c543f866c4e422a55f66a62d669cd',
	197	'duration': 7173,
	198	'timestamp': 1452456073,
	199	'upload_date': '20160110',
	200	},
f5535ed0	201	'skip': 'Live streams on FM4 got deleted soon',
98698606 S	202	}
98698606 S	203
eb368012 S	204	def _real_extract(self, url):
	205	mobj = re.match(self._VALID_URL, url)
	206	show_date = mobj.group('date')
	207	show_id = mobj.group('show')
	208
	209	data = self._download_json(
	210	'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
	211	show_id
	212	)
	213
	214	def extract_entry_dict(info, title, subtitle):
	215	return {
	216	'id': info['loopStreamId'].replace('.mp3', ''),
	217	'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
	218	'title': title,
	219	'description': subtitle,
	220	'duration': (info['end'] - info['start']) / 1000,
	221	'timestamp': info['start'] / 1000,
	222	'ext': 'mp3'
	223	}
	224
	225	entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
	226
	227	return {
	228	'_type': 'playlist',
	229	'id': show_id,
	230	'title': data['title'],
	231	'description': data['subtitle'],
	232	'entries': entries
5f6a1245	233	}
14137b57 S	234
	235
	236	class ORFIPTVIE(InfoExtractor):
	237	IE_NAME = 'orf:iptv'
	238	IE_DESC = 'iptv.ORF.at'
5886b38d	239	_VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
14137b57 S	240
14137b57 S	241	_TEST = {
529d26c3 S	242	'url': 'http://iptv.orf.at/stories/2275236/',
529d26c3 S	243	'md5': 'c8b22af4718a4b4af58342529453e3e5',
14137b57	244	'info_dict': {
529d26c3	245	'id': '350612',
14137b57	246	'ext': 'flv',
529d26c3 S	247	'title': 'Weitere Evakuierungen um Vulkan Calbuco',
	248	'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
	249	'duration': 68.197,
ec85ded8	250	'thumbnail': r're:^https?://.*\.jpg$',
529d26c3	251	'upload_date': '20150425',
14137b57 S	252	},
	253	}
	254
	255	def _real_extract(self, url):
	256	story_id = self._match_id(url)
	257
	258	webpage = self._download_webpage(
	259	'http://iptv.orf.at/stories/%s' % story_id, story_id)
	260
	261	video_id = self._search_regex(
	262	r'data-video(?:id)?="(\d+)"', webpage, 'video id')
	263
	264	data = self._download_json(
	265	'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
	266	video_id)[0]
	267
	268	duration = float_or_none(data['duration'], 1000)
	269
	270	video = data['sources']['default']
	271	load_balancer_url = video['loadBalancerUrl']
	272	abr = int_or_none(video.get('audioBitrate'))
	273	vbr = int_or_none(video.get('bitrate'))
	274	fps = int_or_none(video.get('videoFps'))
	275	width = int_or_none(video.get('videoWidth'))
	276	height = int_or_none(video.get('videoHeight'))
	277	thumbnail = video.get('preview')
	278
	279	rendition = self._download_json(
	280	load_balancer_url, video_id, transform_source=strip_jsonp)
	281
	282	f = {
	283	'abr': abr,
	284	'vbr': vbr,
	285	'fps': fps,
	286	'width': width,
	287	'height': height,
	288	}
	289
	290	formats = []
	291	for format_id, format_url in rendition['redirect'].items():
	292	if format_id == 'rtmp':
	293	ff = f.copy()
	294	ff.update({
	295	'url': format_url,
	296	'format_id': format_id,
	297	})
	298	formats.append(ff)
	299	elif determine_ext(format_url) == 'f4m':
	300	formats.extend(self._extract_f4m_formats(
	301	format_url, video_id, f4m_id=format_id))
	302	elif determine_ext(format_url) == 'm3u8':
	303	formats.extend(self._extract_m3u8_formats(
	304	format_url, video_id, 'mp4', m3u8_id=format_id))
	305	else:
	306	continue
	307	self._sort_formats(formats)
	308
	309	title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
	310	description = self._og_search_description(webpage)
	311	upload_date = unified_strdate(self._html_search_meta(
	312	'dc.date', webpage, 'upload date'))
	313
	314	return {
	315	'id': video_id,
316	'title': title,
317	'description': description,
318	'duration': duration,
319	'thumbnail': thumbnail,
320	'upload_date': upload_date,
321	'formats': formats,
322	}