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

# coding: utf-8
from __future__ import unicode_literals

import json
import re
import calendar
import datetime

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


class ORFTVthekIE(InfoExtractor):
    IE_NAME = 'orf:tvthek'
    IE_DESC = 'ORF TVthek'
    _VALID_URL = r'https?://tvthek\.orf\.at/(?:programs/.+?/episodes|topics?/.+?|program/[^/]+)/(?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',
        'playlist': [{
            'md5': '68f543909aea49d621dfc7703a11cfaf',
            '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".',
            }
        }],
        '_skip': 'Blocked outside of Austria / Germany',
    }]

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

        data_json = self._search_regex(
            r'initializeAdworx\((.+?)\);\n', webpage, 'video info')
        all_data = json.loads(data_json)

        def get_segments(all_data):
            for data in all_data:
                if data['name'] in (
                        'Tracker::EPISODE_DETAIL_PAGE_OVER_PROGRAM',
                        'Tracker::EPISODE_DETAIL_PAGE_OVER_TOPIC'):
                    return data['values']['segments']

        sdata = get_segments(all_data)
        if not sdata:
            raise ExtractorError('Unable to extract segments')

        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 sdata:
            video_id = sd['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['playlist_item_array']['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)

            upload_date = unified_strdate(sd['created_date'])
            entries.append({
                '_type': 'video',
                'id': video_id,
                'title': sd['header'],
                'formats': formats,
                'description': sd.get('description'),
                'duration': int(sd['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'http://oe1\.orf\.at/(?:programm/|konsole.*?#\?track_id=)(?P<id>[0-9]+)'

    # Audios on ORF radio are only available for 7 days, so we can't add tests.
    _TEST = {
        'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
        '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'http://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',
        },
    }

    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'http://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': '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
54543467	4	import json
5d73273f	5	import re
eb368012 S	6	import calendar
eb368012 S	7	import datetime
54543467 JMF	8
	9	from .common import InfoExtractor
	10	from ..utils import (
5d73273f PH	11	HEADRequest,
5d73273f PH	12	unified_strdate,
2acc1f8f	13	ExtractorError,
14137b57 S	14	strip_jsonp,
	15	int_or_none,
	16	float_or_none,
	17	determine_ext,
	18	remove_end,
54543467 JMF	19	)
54543467 JMF	20
5d73273f	21
eb368012 S	22	class ORFTVthekIE(InfoExtractor):
	23	IE_NAME = 'orf:tvthek'
	24	IE_DESC = 'ORF TVthek'
13d27a42	25	_VALID_URL = r'https?://tvthek\.orf\.at/(?:programs/.+?/episodes\|topics?/.+?\|program/[^/]+)/(?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',
	43	'playlist': [{
	44	'md5': '68f543909aea49d621dfc7703a11cfaf',
	45	'info_dict': {
	46	'id': '7982259',
	47	'ext': 'mp4',
	48	'title': 'Best of Ingrid Thurnher',
	49	'upload_date': '20140527',
	50	'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".',
	51	}
	52	}],
	53	'_skip': 'Blocked outside of Austria / Germany',
	54	}]
54543467	55
54543467	56	def _real_extract(self, url):
a6620ac2	57	playlist_id = self._match_id(url)
54543467 JMF	58	webpage = self._download_webpage(url, playlist_id)
54543467 JMF	59
5d73273f PH	60	data_json = self._search_regex(
	61	r'initializeAdworx\((.+?)\);\n', webpage, 'video info')
	62	all_data = json.loads(data_json)
2acc1f8f S	63
	64	def get_segments(all_data):
	65	for data in all_data:
13d27a42 PH	66	if data['name'] in (
	67	'Tracker::EPISODE_DETAIL_PAGE_OVER_PROGRAM',
	68	'Tracker::EPISODE_DETAIL_PAGE_OVER_TOPIC'):
2acc1f8f S	69	return data['values']['segments']
	70
	71	sdata = get_segments(all_data)
	72	if not sdata:
	73	raise ExtractorError('Unable to extract segments')
5d73273f PH	74
	75	def quality_to_int(s):
	76	m = re.search('([0-9]+)', s)
	77	if m is None:
	78	return -1
	79	return int(m.group(1))
	80
	81	entries = []
	82	for sd in sdata:
	83	video_id = sd['id']
	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']),
	91	} for fd in sd['playlist_item_array']['sources']]
	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)
	117
	118	upload_date = unified_strdate(sd['created_date'])
	119	entries.append({
54543467	120	'_type': 'video',
5d73273f PH	121	'id': video_id,
	122	'title': sd['header'],
	123	'formats': formats,
	124	'description': sd.get('description'),
	125	'duration': int(sd['duration_in_seconds']),
	126	'upload_date': upload_date,
	127	'thumbnail': sd.get('image_full_url'),
	128	})
	129
	130	return {
	131	'_type': 'playlist',
	132	'entries': entries,
	133	'id': playlist_id,
	134	}
eb368012 S	135
eb368012 S	136
eb368012 S	137	class ORFOE1IE(InfoExtractor):
	138	IE_NAME = 'orf:oe1'
	139	IE_DESC = 'Radio Österreich 1'
27ca82eb PH	140	_VALID_URL = r'http://oe1\.orf\.at/(?:programm/\|konsole.*?#\?track_id=)(?P<id>[0-9]+)'
	141
	142	# Audios on ORF radio are only available for 7 days, so we can't add tests.
	143	_TEST = {
	144	'url': 'http://oe1.orf.at/konsole?show=on_demand#?track_id=394211',
	145	'only_matching': True,
	146	}
eb368012 S	147
eb368012 S	148	def _real_extract(self, url):
a6620ac2	149	show_id = self._match_id(url)
eb368012 S	150	data = self._download_json(
	151	'http://oe1.orf.at/programm/%s/konsole' % show_id,
	152	show_id
	153	)
	154
	155	timestamp = datetime.datetime.strptime('%s %s' % (
	156	data['item']['day_label'],
	157	data['item']['time']
	158	), '%d.%m.%Y %H:%M')
	159	unix_timestamp = calendar.timegm(timestamp.utctimetuple())
	160
	161	return {
	162	'id': show_id,
	163	'title': data['item']['title'],
	164	'url': data['item']['url_stream'],
	165	'ext': 'mp3',
	166	'description': data['item'].get('info'),
	167	'timestamp': unix_timestamp
	168	}
	169
	170
	171	class ORFFM4IE(InfoExtractor):
b9f030cc	172	IE_NAME = 'orf:fm4'
eb368012	173	IE_DESC = 'radio FM4'
4654c1d0	174	_VALID_URL = r'http://fm4\.orf\.at/(?:7tage/?#\|player/)(?P<date>[0-9]+)/(?P<show>\w+)'
eb368012	175
98698606 S	176	_TEST = {
	177	'url': 'http://fm4.orf.at/player/20160110/IS/',
	178	'md5': '01e736e8f1cef7e13246e880a59ad298',
	179	'info_dict': {
	180	'id': '2016-01-10_2100_tl_54_7DaysSun13_11244',
	181	'ext': 'mp3',
	182	'title': 'Im Sumpf',
	183	'description': 'md5:384c543f866c4e422a55f66a62d669cd',
	184	'duration': 7173,
	185	'timestamp': 1452456073,
	186	'upload_date': '20160110',
	187	},
	188	}
	189
eb368012 S	190	def _real_extract(self, url):
	191	mobj = re.match(self._VALID_URL, url)
	192	show_date = mobj.group('date')
	193	show_id = mobj.group('show')
	194
	195	data = self._download_json(
	196	'http://audioapi.orf.at/fm4/json/2.0/broadcasts/%s/4%s' % (show_date, show_id),
	197	show_id
	198	)
	199
	200	def extract_entry_dict(info, title, subtitle):
	201	return {
	202	'id': info['loopStreamId'].replace('.mp3', ''),
	203	'url': 'http://loopstream01.apa.at/?channel=fm4&id=%s' % info['loopStreamId'],
	204	'title': title,
	205	'description': subtitle,
	206	'duration': (info['end'] - info['start']) / 1000,
	207	'timestamp': info['start'] / 1000,
	208	'ext': 'mp3'
	209	}
	210
	211	entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
	212
	213	return {
	214	'_type': 'playlist',
	215	'id': show_id,
	216	'title': data['title'],
	217	'description': data['subtitle'],
	218	'entries': entries
5f6a1245	219	}
14137b57 S	220
	221
	222	class ORFIPTVIE(InfoExtractor):
	223	IE_NAME = 'orf:iptv'
	224	IE_DESC = 'iptv.ORF.at'
	225	_VALID_URL = r'http://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
	226
	227	_TEST = {
529d26c3 S	228	'url': 'http://iptv.orf.at/stories/2275236/',
529d26c3 S	229	'md5': 'c8b22af4718a4b4af58342529453e3e5',
14137b57	230	'info_dict': {
529d26c3	231	'id': '350612',
14137b57	232	'ext': 'flv',
529d26c3 S	233	'title': 'Weitere Evakuierungen um Vulkan Calbuco',
	234	'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
	235	'duration': 68.197,
14137b57	236	'thumbnail': 're:^https?://.*\.jpg$',
529d26c3	237	'upload_date': '20150425',
14137b57 S	238	},
	239	}
	240
	241	def _real_extract(self, url):
	242	story_id = self._match_id(url)
	243
	244	webpage = self._download_webpage(
	245	'http://iptv.orf.at/stories/%s' % story_id, story_id)
	246
	247	video_id = self._search_regex(
	248	r'data-video(?:id)?="(\d+)"', webpage, 'video id')
	249
	250	data = self._download_json(
	251	'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
	252	video_id)[0]
	253
	254	duration = float_or_none(data['duration'], 1000)
	255
	256	video = data['sources']['default']
	257	load_balancer_url = video['loadBalancerUrl']
	258	abr = int_or_none(video.get('audioBitrate'))
	259	vbr = int_or_none(video.get('bitrate'))
	260	fps = int_or_none(video.get('videoFps'))
	261	width = int_or_none(video.get('videoWidth'))
	262	height = int_or_none(video.get('videoHeight'))
	263	thumbnail = video.get('preview')
	264
	265	rendition = self._download_json(
	266	load_balancer_url, video_id, transform_source=strip_jsonp)
	267
	268	f = {
	269	'abr': abr,
	270	'vbr': vbr,
	271	'fps': fps,
	272	'width': width,
	273	'height': height,
	274	}
	275
	276	formats = []
	277	for format_id, format_url in rendition['redirect'].items():
	278	if format_id == 'rtmp':
	279	ff = f.copy()
	280	ff.update({
	281	'url': format_url,
	282	'format_id': format_id,
	283	})
	284	formats.append(ff)
	285	elif determine_ext(format_url) == 'f4m':
	286	formats.extend(self._extract_f4m_formats(
	287	format_url, video_id, f4m_id=format_id))
	288	elif determine_ext(format_url) == 'm3u8':
	289	formats.extend(self._extract_m3u8_formats(
	290	format_url, video_id, 'mp4', m3u8_id=format_id))
	291	else:
	292	continue
	293	self._sort_formats(formats)
	294
	295	title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
	296	description = self._og_search_description(webpage)
	297	upload_date = unified_strdate(self._html_search_meta(
	298	'dc.date', webpage, 'upload date'))
	299
	300	return {
	301	'id': video_id,
302	'title': title,
303	'description': description,
304	'duration': duration,
305	'thumbnail': thumbnail,
306	'upload_date': upload_date,
307	'formats': formats,
308	}