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

# coding: utf-8
from __future__ import unicode_literals

import functools
import re

from .common import InfoExtractor
from ..utils import (
    int_or_none,
    unified_strdate,
    OnDemandPagedList,
    xpath_text,
    determine_ext,
    qualities,
    float_or_none,
    ExtractorError,
)


class ZDFIE(InfoExtractor):
    _VALID_URL = r'(?:zdf:|zdf:video:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/(.*beitrag/(?:video/)?))(?P<id>[0-9]+)(?:/[^/?]+)?(?:\?.*)?'

    _TESTS = [{
        'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt',
        'info_dict': {
            'id': '2037704',
            'ext': 'webm',
            'title': 'ZDFspezial - Ende des Machtpokers',
            'description': 'Union und SPD haben sich auf einen Koalitionsvertrag geeinigt. Aber was bedeutet das für die Bürger? Sehen Sie hierzu das ZDFspezial "Ende des Machtpokers - Große Koalition für Deutschland".',
            'duration': 1022,
            'uploader': 'spezial',
            'uploader_id': '225948',
            'upload_date': '20131127',
        },
        'skip': 'Videos on ZDF.de are depublicised in short order',
    }]

    def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
        param_groups = {}
        for param_group in smil.findall(self._xpath_ns('./head/paramGroup', namespace)):
            group_id = param_group.attrib.get(self._xpath_ns('id', 'http://www.w3.org/XML/1998/namespace'))
            params = {}
            for param in param_group:
                params[param.get('name')] = param.get('value')
            param_groups[group_id] = params

        formats = []
        for video in smil.findall(self._xpath_ns('.//video', namespace)):
            src = video.get('src')
            if not src:
                continue
            bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
            group_id = video.get('paramGroup')
            param_group = param_groups[group_id]
            for proto in param_group['protocols'].split(','):
                formats.append({
                    'url': '%s://%s' % (proto, param_group['host']),
                    'app': param_group['app'],
                    'play_path': src,
                    'ext': 'flv',
                    'format_id': '%s-%d' % (proto, bitrate),
                    'tbr': bitrate,
                })
        self._sort_formats(formats)
        return formats

    def extract_from_xml_url(self, video_id, xml_url):
        doc = self._download_xml(
            xml_url, video_id,
            note='Downloading video info',
            errnote='Failed to download video info')

        status_code = doc.find('./status/statuscode')
        if status_code is not None and status_code.text != 'ok':
            code = status_code.text
            if code == 'notVisibleAnymore':
                message = 'Video %s is not available' % video_id
            else:
                message = '%s returned error: %s' % (self.IE_NAME, code)
            raise ExtractorError(message, expected=True)

        title = doc.find('.//information/title').text
        description = xpath_text(doc, './/information/detail', 'description')
        duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration'))
        uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader')
        uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id')
        upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date'))

        def xml_to_thumbnails(fnode):
            thumbnails = []
            for node in fnode:
                thumbnail_url = node.text
                if not thumbnail_url:
                    continue
                thumbnail = {
                    'url': thumbnail_url,
                }
                if 'key' in node.attrib:
                    m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key'])
                    if m:
                        thumbnail['width'] = int(m.group(1))
                        thumbnail['height'] = int(m.group(2))
                thumbnails.append(thumbnail)
            return thumbnails

        thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage'))

        format_nodes = doc.findall('.//formitaeten/formitaet')
        quality = qualities(['veryhigh', 'high', 'med', 'low'])

        def get_quality(elem):
            return quality(xpath_text(elem, 'quality'))
        format_nodes.sort(key=get_quality)
        format_ids = []
        formats = []
        for fnode in format_nodes:
            video_url = fnode.find('url').text
            is_available = 'http://www.metafilegenerator' not in video_url
            if not is_available:
                continue
            format_id = fnode.attrib['basetype']
            quality = xpath_text(fnode, './quality', 'quality')
            format_m = re.match(r'''(?x)
                (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
                (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
            ''', format_id)

            ext = determine_ext(video_url, None) or format_m.group('container')
            if ext not in ('smil', 'f4m', 'm3u8'):
                format_id = format_id + '-' + quality
            if format_id in format_ids:
                continue

            if ext == 'meta':
                continue
            elif ext == 'smil':
                formats.extend(self._extract_smil_formats(
                    video_url, video_id, fatal=False))
            elif ext == 'm3u8':
                # the certificates are misconfigured (see
                # https://github.com/rg3/youtube-dl/issues/8665)
                if video_url.startswith('https://'):
                    continue
                formats.extend(self._extract_m3u8_formats(
                    video_url, video_id, 'mp4', m3u8_id=format_id, fatal=False))
            elif ext == 'f4m':
                formats.extend(self._extract_f4m_formats(
                    video_url, video_id, f4m_id=format_id, fatal=False))
            else:
                proto = format_m.group('proto').lower()

                abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000)
                vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000)

                width = int_or_none(xpath_text(fnode, './width', 'width'))
                height = int_or_none(xpath_text(fnode, './height', 'height'))

                filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize'))

                format_note = ''
                if not format_note:
                    format_note = None

                formats.append({
                    'format_id': format_id,
                    'url': video_url,
                    'ext': ext,
                    'acodec': format_m.group('acodec'),
                    'vcodec': format_m.group('vcodec'),
                    'abr': abr,
                    'vbr': vbr,
                    'width': width,
                    'height': height,
                    'filesize': filesize,
                    'format_note': format_note,
                    'protocol': proto,
                    '_available': is_available,
                })
            format_ids.append(format_id)

        self._sort_formats(formats)

        return {
            'id': video_id,
            'title': title,
            'description': description,
            'duration': duration,
            'thumbnails': thumbnails,
            'uploader': uploader,
            'uploader_id': uploader_id,
            'upload_date': upload_date,
            'formats': formats,
        }

    def _real_extract(self, url):
        video_id = self._match_id(url)
        xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
        return self.extract_from_xml_url(video_id, xml_url)


class ZDFChannelIE(InfoExtractor):
    _VALID_URL = r'(?:zdf:topic:|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/(?:[^/]+/)?)(?P<id>[0-9]+)'
    _TESTS = [{
        'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic',
        'info_dict': {
            'id': '1586442',
        },
        'playlist_count': 3,
    }, {
        'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/aktuellste/332',
        'only_matching': True,
    }, {
        'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/meist-gesehen/332',
        'only_matching': True,
    }, {
        'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/_/1798716?bc=nrt;nrm?flash=off',
        'only_matching': True,
    }]
    _PAGE_SIZE = 50

    def _fetch_page(self, channel_id, page):
        offset = page * self._PAGE_SIZE
        xml_url = (
            'http://www.zdf.de/ZDFmediathek/xmlservice/web/aktuellste?ak=web&offset=%d&maxLength=%d&id=%s'
            % (offset, self._PAGE_SIZE, channel_id))
        doc = self._download_xml(
            xml_url, channel_id,
            note='Downloading channel info',
            errnote='Failed to download channel info')

        title = doc.find('.//information/title').text
        description = doc.find('.//information/detail').text
        for asset in doc.findall('.//teasers/teaser'):
            a_type = asset.find('./type').text
            a_id = asset.find('./details/assetId').text
            if a_type not in ('video', 'topic'):
                continue
            yield {
                '_type': 'url',
                'playlist_title': title,
                'playlist_description': description,
                'url': 'zdf:%s:%s' % (a_type, a_id),
            }

    def _real_extract(self, url):
        channel_id = self._match_id(url)
        entries = OnDemandPagedList(
            functools.partial(self._fetch_page, channel_id), self._PAGE_SIZE)

        return {
            '_type': 'playlist',
            'id': channel_id,
            'entries': entries,
        }
Commit	Line	Data
a2e6db36	1	# coding: utf-8
919052d0	2	from __future__ import unicode_literals
a2e6db36	3
9abd500a	4	import functools
d5822b96 PH	5	import re
	6
	7	from .common import InfoExtractor
	8	from ..utils import (
61224dbc	9	int_or_none,
02dbf93f	10	unified_strdate,
9abd500a	11	OnDemandPagedList,
7b091c37	12	xpath_text,
a5c1d955	13	determine_ext,
	14	qualities,
	15	float_or_none,
6b461026	16	ExtractorError,
d5822b96 PH	17	)
d5822b96 PH	18
0b7c2485	19
d5822b96	20	class ZDFIE(InfoExtractor):
9abd500a	21	_VALID_URL = r'(?:zdf:\|zdf:video:\|https?://www\.zdf\.de/ZDFmediathek(?:#)?/(.beitrag/(?:video/)?))(?P<id>[0-9]+)(?:/[^/?]+)?(?:\?.)?'
a2e6db36	22
a5c1d955	23	_TESTS = [{
919052d0 S	24	'url': 'http://www.zdf.de/ZDFmediathek/beitrag/video/2037704/ZDFspezial---Ende-des-Machtpokers--?bc=sts;stt',
	25	'info_dict': {
	26	'id': '2037704',
	27	'ext': 'webm',
	28	'title': 'ZDFspezial - Ende des Machtpokers',
	29	'description': 'Union und SPD haben sich auf einen Koalitionsvertrag geeinigt. Aber was bedeutet das für die Bürger? Sehen Sie hierzu das ZDFspezial "Ende des Machtpokers - Große Koalition für Deutschland".',
	30	'duration': 1022,
	31	'uploader': 'spezial',
	32	'uploader_id': '225948',
	33	'upload_date': '20131127',
a2e6db36	34	},
919052d0	35	'skip': 'Videos on ZDF.de are depublicised in short order',
a5c1d955	36	}]
	37
	38	def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
	39	param_groups = {}
	40	for param_group in smil.findall(self._xpath_ns('./head/paramGroup', namespace)):
	41	group_id = param_group.attrib.get(self._xpath_ns('id', 'http://www.w3.org/XML/1998/namespace'))
	42	params = {}
	43	for param in param_group:
	44	params[param.get('name')] = param.get('value')
	45	param_groups[group_id] = params
	46
	47	formats = []
	48	for video in smil.findall(self._xpath_ns('.//video', namespace)):
	49	src = video.get('src')
	50	if not src:
	51	continue
	52	bitrate = float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
	53	group_id = video.get('paramGroup')
	54	param_group = param_groups[group_id]
	55	for proto in param_group['protocols'].split(','):
	56	formats.append({
	57	'url': '%s://%s' % (proto, param_group['host']),
	58	'app': param_group['app'],
	59	'play_path': src,
	60	'ext': 'flv',
	61	'format_id': '%s-%d' % (proto, bitrate),
	62	'tbr': bitrate,
a5c1d955	63	})
	64	self._sort_formats(formats)
	65	return formats
	66
	67	def extract_from_xml_url(self, video_id, xml_url):
	68	doc = self._download_xml(
	69	xml_url, video_id,
	70	note='Downloading video info',
	71	errnote='Failed to download video info')
	72
6b461026	73	status_code = doc.find('./status/statuscode')
	74	if status_code is not None and status_code.text != 'ok':
	75	code = status_code.text
	76	if code == 'notVisibleAnymore':
	77	message = 'Video %s is not available' % video_id
	78	else:
	79	message = '%s returned error: %s' % (self.IE_NAME, code)
	80	raise ExtractorError(message, expected=True)
	81
a5c1d955	82	title = doc.find('.//information/title').text
	83	description = xpath_text(doc, './/information/detail', 'description')
	84	duration = int_or_none(xpath_text(doc, './/details/lengthSec', 'duration'))
	85	uploader = xpath_text(doc, './/details/originChannelTitle', 'uploader')
	86	uploader_id = xpath_text(doc, './/details/originChannelId', 'uploader id')
	87	upload_date = unified_strdate(xpath_text(doc, './/details/airtime', 'upload date'))
	88
	89	def xml_to_thumbnails(fnode):
	90	thumbnails = []
	91	for node in fnode:
	92	thumbnail_url = node.text
	93	if not thumbnail_url:
	94	continue
	95	thumbnail = {
	96	'url': thumbnail_url,
	97	}
	98	if 'key' in node.attrib:
	99	m = re.match('^([0-9]+)x([0-9]+)$', node.attrib['key'])
	100	if m:
	101	thumbnail['width'] = int(m.group(1))
	102	thumbnail['height'] = int(m.group(2))
	103	thumbnails.append(thumbnail)
	104	return thumbnails
	105
	106	thumbnails = xml_to_thumbnails(doc.findall('.//teaserimages/teaserimage'))
	107
	108	format_nodes = doc.findall('.//formitaeten/formitaet')
	109	quality = qualities(['veryhigh', 'high', 'med', 'low'])
	110
	111	def get_quality(elem):
	112	return quality(xpath_text(elem, 'quality'))
	113	format_nodes.sort(key=get_quality)
	114	format_ids = []
	115	formats = []
	116	for fnode in format_nodes:
	117	video_url = fnode.find('url').text
	118	is_available = 'http://www.metafilegenerator' not in video_url
	119	if not is_available:
	120	continue
	121	format_id = fnode.attrib['basetype']
	122	quality = xpath_text(fnode, './quality', 'quality')
	123	format_m = re.match(r'''(?x)
	124	(?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_
	125	(?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+)
	126	''', format_id)
	127
	128	ext = determine_ext(video_url, None) or format_m.group('container')
	129	if ext not in ('smil', 'f4m', 'm3u8'):
	130	format_id = format_id + '-' + quality
	131	if format_id in format_ids:
	132	continue
	133
	134	if ext == 'meta':
	135	continue
	136	elif ext == 'smil':
7e5edcfd S	137	formats.extend(self._extract_smil_formats(
7e5edcfd S	138	video_url, video_id, fatal=False))
a5c1d955	139	elif ext == 'm3u8':
9173202b JMF	140	# the certificates are misconfigured (see
	141	# https://github.com/rg3/youtube-dl/issues/8665)
	142	if video_url.startswith('https://'):
	143	continue
7e5edcfd	144	formats.extend(self._extract_m3u8_formats(
6b461026	145	video_url, video_id, 'mp4', m3u8_id=format_id, fatal=False))
a5c1d955	146	elif ext == 'f4m':
7e5edcfd	147	formats.extend(self._extract_f4m_formats(
6b461026	148	video_url, video_id, f4m_id=format_id, fatal=False))
a5c1d955	149	else:
	150	proto = format_m.group('proto').lower()
	151
	152	abr = int_or_none(xpath_text(fnode, './audioBitrate', 'abr'), 1000)
	153	vbr = int_or_none(xpath_text(fnode, './videoBitrate', 'vbr'), 1000)
	154
	155	width = int_or_none(xpath_text(fnode, './width', 'width'))
	156	height = int_or_none(xpath_text(fnode, './height', 'height'))
	157
	158	filesize = int_or_none(xpath_text(fnode, './filesize', 'filesize'))
	159
	160	format_note = ''
	161	if not format_note:
	162	format_note = None
	163
	164	formats.append({
	165	'format_id': format_id,
	166	'url': video_url,
	167	'ext': ext,
	168	'acodec': format_m.group('acodec'),
	169	'vcodec': format_m.group('vcodec'),
	170	'abr': abr,
	171	'vbr': vbr,
	172	'width': width,
	173	'height': height,
	174	'filesize': filesize,
	175	'format_note': format_note,
	176	'protocol': proto,
	177	'_available': is_available,
	178	})
	179	format_ids.append(format_id)
	180
	181	self._sort_formats(formats)
	182
	183	return {
	184	'id': video_id,
	185	'title': title,
	186	'description': description,
	187	'duration': duration,
	188	'thumbnails': thumbnails,
	189	'uploader': uploader,
	190	'uploader_id': uploader_id,
	191	'upload_date': upload_date,
	192	'formats': formats,
	193	}
d5822b96	194
9abd500a PH	195	def _real_extract(self, url):
9abd500a PH	196	video_id = self._match_id(url)
919052d0	197	xml_url = 'http://www.zdf.de/ZDFmediathek/xmlservice/web/beitragsDetails?ak=web&id=%s' % video_id
a5c1d955	198	return self.extract_from_xml_url(video_id, xml_url)
8560c618	199
8560c618	200
9abd500a	201	class ZDFChannelIE(InfoExtractor):
67ba388e	202	_VALID_URL = r'(?:zdf:topic:\|https?://www\.zdf\.de/ZDFmediathek(?:#)?/.*kanaluebersicht/(?:[^/]+/)?)(?P<id>[0-9]+)'
c2404463	203	_TESTS = [{
8560c618 AK	204	'url': 'http://www.zdf.de/ZDFmediathek#/kanaluebersicht/1586442/sendung/Titanic',
	205	'info_dict': {
	206	'id': '1586442',
8560c618	207	},
d51a853d	208	'playlist_count': 3,
c2404463 S	209	}, {
	210	'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/aktuellste/332',
	211	'only_matching': True,
	212	}, {
	213	'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/meist-gesehen/332',
	214	'only_matching': True,
	215	}, {
	216	'url': 'http://www.zdf.de/ZDFmediathek/kanaluebersicht/_/1798716?bc=nrt;nrm?flash=off',
	217	'only_matching': True,
	218	}]
9abd500a PH	219	_PAGE_SIZE = 50
	220
	221	def _fetch_page(self, channel_id, page):
	222	offset = page * self._PAGE_SIZE
	223	xml_url = (
	224	'http://www.zdf.de/ZDFmediathek/xmlservice/web/aktuellste?ak=web&offset=%d&maxLength=%d&id=%s'
	225	% (offset, self._PAGE_SIZE, channel_id))
	226	doc = self._download_xml(
	227	xml_url, channel_id,
	228	note='Downloading channel info',
	229	errnote='Failed to download channel info')
	230
	231	title = doc.find('.//information/title').text
	232	description = doc.find('.//information/detail').text
	233	for asset in doc.findall('.//teasers/teaser'):
	234	a_type = asset.find('./type').text
	235	a_id = asset.find('./details/assetId').text
	236	if a_type not in ('video', 'topic'):
	237	continue
	238	yield {
	239	'_type': 'url',
	240	'playlist_title': title,
	241	'playlist_description': description,
	242	'url': 'zdf:%s:%s' % (a_type, a_id),
	243	}
	244
	245	def _real_extract(self, url):
	246	channel_id = self._match_id(url)
	247	entries = OnDemandPagedList(
	248	functools.partial(self._fetch_page, channel_id), self._PAGE_SIZE)
8560c618	249
8560c618 AK	250	return {
8560c618 AK	251	'_type': 'playlist',
9abd500a PH	252	'id': channel_id,
9abd500a PH	253	'entries': entries,
8560c618	254	}