[yt-dlp.git] / youtube_dlc / extractor / rcs.py

# coding: utf-8
from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..utils import (
    clean_html,
    ExtractorError,
    js_to_json,
    base_url,
    url_basename,
    urljoin,
)


class RCSIE(InfoExtractor):
    _ALL_REPLACE = {
        'media2vam.corriere.it.edgesuite.net':
            'media2vam-corriere-it.akamaized.net',
        'media.youreporter.it.edgesuite.net':
            'media-youreporter-it.akamaized.net',
        'corrierepmd.corriere.it.edgesuite.net':
            'corrierepmd-corriere-it.akamaized.net',
        'media2vam-corriere-it.akamaized.net/fcs.quotidiani/vr/videos/':
            'video.corriere.it/vr360/videos/',
        '.net//': '.net/',
    }
    _MP4_REPLACE = {
        'media2vam.corbologna.corriere.it.edgesuite.net':
            'media2vam-bologna-corriere-it.akamaized.net',
        'media2vam.corfiorentino.corriere.it.edgesuite.net':
            'media2vam-fiorentino-corriere-it.akamaized.net',
        'media2vam.cormezzogiorno.corriere.it.edgesuite.net':
            'media2vam-mezzogiorno-corriere-it.akamaized.net',
        'media2vam.corveneto.corriere.it.edgesuite.net':
            'media2vam-veneto-corriere-it.akamaized.net',
        'media2.oggi.it.edgesuite.net':
            'media2-oggi-it.akamaized.net',
        'media2.quimamme.it.edgesuite.net':
            'media2-quimamme-it.akamaized.net',
        'media2.amica.it.edgesuite.net':
            'media2-amica-it.akamaized.net',
        'media2.living.corriere.it.edgesuite.net':
            'media2-living-corriere-it.akamaized.net',
        'media2.style.corriere.it.edgesuite.net':
            'media2-style-corriere-it.akamaized.net',
        'media2.iodonna.it.edgesuite.net':
            'media2-iodonna-it.akamaized.net',
        'media2.leitv.it.edgesuite.net':
            'media2-leitv-it.akamaized.net',
    }
    _MIGRATION_MAP = {
        'videoamica-vh.akamaihd': 'amica',
        'media2-amica-it.akamaized': 'amica',
        'corrierevam-vh.akamaihd': 'corriere',
        'media2vam-corriere-it.akamaized': 'corriere',
        'cormezzogiorno-vh.akamaihd': 'corrieredelmezzogiorno',
        'media2vam-mezzogiorno-corriere-it.akamaized': 'corrieredelmezzogiorno',
        'corveneto-vh.akamaihd': 'corrieredelveneto',
        'media2vam-veneto-corriere-it.akamaized': 'corrieredelveneto',
        'corbologna-vh.akamaihd': 'corrieredibologna',
        'media2vam-bologna-corriere-it.akamaized': 'corrieredibologna',
        'corfiorentino-vh.akamaihd': 'corrierefiorentino',
        'media2vam-fiorentino-corriere-it.akamaized': 'corrierefiorentino',
        'corinnovazione-vh.akamaihd': 'corriereinnovazione',
        'media2-gazzanet-gazzetta-it.akamaized': 'gazzanet',
        'videogazzanet-vh.akamaihd': 'gazzanet',
        'videogazzaworld-vh.akamaihd': 'gazzaworld',
        'gazzettavam-vh.akamaihd': 'gazzetta',
        'media2vam-gazzetta-it.akamaized': 'gazzetta',
        'videoiodonna-vh.akamaihd': 'iodonna',
        'media2-leitv-it.akamaized': 'leitv',
        'videoleitv-vh.akamaihd': 'leitv',
        'videoliving-vh.akamaihd': 'living',
        'media2-living-corriere-it.akamaized': 'living',
        'media2-oggi-it.akamaized': 'oggi',
        'videooggi-vh.akamaihd': 'oggi',
        'media2-quimamme-it.akamaized': 'quimamme',
        'quimamme-vh.akamaihd': 'quimamme',
        'videorunning-vh.akamaihd': 'running',
        'media2-style-corriere-it.akamaized': 'style',
        'style-vh.akamaihd': 'style',
        'videostyle-vh.akamaihd': 'style',
        'media2-stylepiccoli-it.akamaized': 'stylepiccoli',
        'stylepiccoli-vh.akamaihd': 'stylepiccoli',
        'doveviaggi-vh.akamaihd': 'viaggi',
        'media2-doveviaggi-it.akamaized': 'viaggi',
        'media2-vivimilano-corriere-it.akamaized': 'vivimilano',
        'vivimilano-vh.akamaihd': 'vivimilano',
        'media2-youreporter-it.akamaized': 'youreporter'
    }
    _MIGRATION_MEDIA = {
        'advrcs-vh.akamaihd': '',
        'corriere-f.akamaihd': '',
        'corrierepmd-corriere-it.akamaized': '',
        'corrprotetto-vh.akamaihd': '',
        'gazzetta-f.akamaihd': '',
        'gazzettapmd-gazzetta-it.akamaized': '',
        'gazzprotetto-vh.akamaihd': '',
        'periodici-f.akamaihd': '',
        'periodicisecure-vh.akamaihd': '',
        'videocoracademy-vh.akamaihd': ''
    }

    def _get_video_src(self, video):
        mediaFiles = video['mediaProfile']['mediaFile']
        src = {}
        # audio
        if video['mediaType'] == 'AUDIO':
            for aud in mediaFiles:
                # todo: check
                src['mp3'] = aud['value']
        # video
        else:
            for vid in mediaFiles:
                if vid['mimeType'] == 'application/vnd.apple.mpegurl':
                    src['m3u8'] = vid['value']
                if vid['mimeType'] == 'video/mp4':
                    src['mp4'] = vid['value']

        # replace host
        for t in src:
            for s, r in self._ALL_REPLACE.items():
                src[t] = src[t].replace(s, r)
            for s, r in self._MP4_REPLACE.items():
                src[t] = src[t].replace(s, r)

        # switch cdn
        if 'mp4' in src and 'm3u8' in src:
            if '-lh.akamaihd' not in src['m3u8'] and 'akamai' in src['mp4']:
                if 'm3u8' in src:
                    matches = re.search(r'(?:https*:)?\/\/(?P<host>.*)\.net\/i(?P<path>.*)$', src['m3u8'])
                    src['m3u8'] = 'https://vod.rcsobjects.it/hls/%s%s' % (
                        self._MIGRATION_MAP[matches.group('host')],
                        matches.group('path').replace(
                            '///', '/').replace(
                            '//', '/').replace(
                            '.csmil', '.urlset'
                        )
                    )
                if 'mp4' in src:
                    matches = re.search(r'(?:https*:)?\/\/(?P<host>.*)\.net\/i(?P<path>.*)$', src['mp4'])
                    if matches:
                        if matches.group('host') in self._MIGRATION_MEDIA:
                            vh_stream = 'https://media2.corriereobjects.it'
                            if src['mp4'].find('fcs.quotidiani_!'):
                                vh_stream = 'https://media2-it.corriereobjects.it'
                            src['mp4'] = '%s%s' % (
                                vh_stream,
                                matches.group('path').replace(
                                    '///', '/').replace(
                                    '//', '/').replace(
                                    '/fcs.quotidiani/mediacenter', '').replace(
                                    '/fcs.quotidiani_!/mediacenter', '').replace(
                                    'corriere/content/mediacenter/', '').replace(
                                    'gazzetta/content/mediacenter/', '')
                            )
                        else:
                            src['mp4'] = 'https://vod.rcsobjects.it/%s%s' % (
                                self._MIGRATION_MAP[matches.group('host')],
                                matches.group('path').replace('///', '/').replace('//', '/')
                            )

        if 'mp3' in src:
            src['mp3'] = src['mp3'].replace(
                'media2vam-corriere-it.akamaized.net',
                'vod.rcsobjects.it/corriere')
        if 'mp4' in src:
            if src['mp4'].find('fcs.quotidiani_!'):
                src['mp4'] = src['mp4'].replace('vod.rcsobjects', 'vod-it.rcsobjects')
        if 'm3u8' in src:
            if src['m3u8'].find('fcs.quotidiani_!'):
                src['m3u8'] = src['m3u8'].replace('vod.rcsobjects', 'vod-it.rcsobjects')

        if 'geoblocking' in video['mediaProfile']:
            if 'm3u8' in src:
                src['m3u8'] = src['m3u8'].replace('vod.rcsobjects', 'vod-it.rcsobjects')
            if 'mp4' in src:
                src['mp4'] = src['mp4'].replace('vod.rcsobjects', 'vod-it.rcsobjects')
        if 'm3u8' in src:
            if src['m3u8'].find('csmil') and src['m3u8'].find('vod'):
                src['m3u8'] = src['m3u8'].replace('.csmil', '.urlset')

        return src

    def _create_formats(self, urls, video_id):
        formats = []
        formats = self._extract_m3u8_formats(
            urls['m3u8'], video_id, 'mp4', entry_protocol='m3u8_native',
            m3u8_id='hls', fatal=False)

        if not formats:
            formats.append({
                'format_id': 'http-mp4',
                'url': urls['mp4']
            })
        self._sort_formats(formats)
        return formats

    def _real_extract(self, url):
        video_id = self._match_id(url)
        mobj = re.search(self._VALID_URL, url).groupdict()

        if not mobj['cdn']:
            raise ExtractorError('CDN not found in url: %s' % url)

        # for leitv/youreporter/viaggi don't use the embed page
        if (mobj['cdn'] not in ['leitv.it', 'youreporter.it']) and (mobj['vid'] == 'video'):
            url = 'https://video.%s/video-embed/%s' % (mobj['cdn'], video_id)

        page = self._download_webpage(url, video_id)

        video_data = None
        # look for json video data url
        json = self._search_regex(
            r'''var url\s*=\s*["']((?:https?:)?//video\.rcs\.it/fragment-includes/video-includes/.+?\.json)["'];''',
            page, video_id, default=None)
        if json:
            if json.startswith('//'):
                json = 'https:%s' % json
            video_data = self._download_json(json, video_id)

        # if url not found, look for json video data directly in the page
        else:
            json = self._search_regex(
                r'[\s;]video\s*=\s*({[\s\S]+?})(?:;|,playlist=)',
                page, video_id, default=None)
            if json:
                video_data = self._parse_json(
                    json, video_id, transform_source=js_to_json)
            else:
                # if no video data found try search for iframes
                emb = RCSEmbedsIE._extract_url(page)
                if emb:
                    return self._real_extract(emb)

        if not video_data:
            raise ExtractorError('Video data not found in the page')

        formats = self._create_formats(
            self._get_video_src(video_data), video_id)

        return {
            'id': video_id,
            'title': video_data['title'],
            'description': video_data['description'] or clean_html(video_data['htmlDescription']),
            'uploader': video_data['provider'] if video_data['provider'] else mobj['cdn'],
            'formats': formats
        }


class RCSEmbedsIE(RCSIE):
    IE_NAME = 'rcs:rcs'
    _VALID_URL = r'''(?x)
                    https?://(?P<vid>video)\.
                    (?P<cdn>
                    (?:
                        rcs|
                        (?:corriere\w+\.)?corriere|
                        (?:gazzanet\.)?gazzetta
                    )\.it)
                    /video-embed/(?P<id>[^/=&\?]+?)(?:$|\?)'''
    _TESTS = [{
        'url': 'https://video.rcs.it/video-embed/iodonna-0001585037',
        'md5': '623ecc8ffe7299b2d0c1046d8331a9df',
        'info_dict': {
            'id': 'iodonna-0001585037',
            'ext': 'mp4',
            'title': 'Sky Arte racconta Madonna nella serie "Artist to icon"',
            'description': 'md5:65b09633df9ffee57f48b39e34c9e067',
            'uploader': 'rcs.it',
        }
    }, {
        'url': 'https://video.corriere.it/video-embed/b727632a-f9d0-11ea-91b0-38d50a849abb?player',
        'match_only': True
    }, {
        'url': 'https://video.gazzetta.it/video-embed/49612410-00ca-11eb-bcd8-30d4253e0140',
        'match_only': True
    }]

    @staticmethod
    def _sanitize_urls(urls):
        # add protocol if missing
        for i, e in enumerate(urls):
            if e.startswith('//'):
                urls[i] = 'https:%s' % e
        # clean iframes urls
        for i, e in enumerate(urls):
            urls[i] = urljoin(base_url(e), url_basename(e))
        return urls

    @staticmethod
    def _extract_urls(webpage):
        entries = [
            mobj.group('url')
            for mobj in re.finditer(r'''(?x)
            (?:
                data-frame-src=|
                <iframe[^\n]+src=
            )
            (["'])
                (?P<url>(?:https?:)?//video\.
                    (?:
                        rcs|
                        (?:corriere\w+\.)?corriere|
                        (?:gazzanet\.)?gazzetta
                    )
                \.it/video-embed/.+?)
            \1''', webpage)]
        return RCSEmbedsIE._sanitize_urls(entries)

    @staticmethod
    def _extract_url(webpage):
        urls = RCSEmbedsIE._extract_urls(webpage)
        return urls[0] if urls else None


class CorriereIE(RCSIE):
    IE_NAME = 'rcs:corriere'
    _VALID_URL = r'''(?x)https?://(?P<vid>video|viaggi)\.
                    (?P<cdn>
                    (?:
                        corrieredelmezzogiorno\.|
                        corrieredelveneto\.|
                        corrieredibologna\.|
                        corrierefiorentino\.
                    )?
                    corriere\.it)/.+?/(?P<id>[^/]+)(?=\?|/$|$)'''
    _TESTS = [{
        'url': 'https://video.corriere.it/sport/formula-1/vettel-guida-ferrari-sf90-mugello-suo-fianco-c-elecrerc-bendato-video-esilarante/b727632a-f9d0-11ea-91b0-38d50a849abb',
        'md5': '0f4ededc202b0f00b6e509d831e2dcda',
        'info_dict': {
            'id': 'b727632a-f9d0-11ea-91b0-38d50a849abb',
            'ext': 'mp4',
            'title': 'Vettel guida la Ferrari SF90 al Mugello e al suo fianco c\'è Leclerc (bendato): il video è esilarante',
            'description': 'md5:93b51c9161ac8a64fb2f997b054d0152',
            'uploader': 'Corriere Tv',
        }
    }, {
        'url': 'https://viaggi.corriere.it/video/norvegia-il-nuovo-ponte-spettacolare-sopra-la-cascata-di-voringsfossen/',
        'md5': 'da378e4918d2afbf7d61c35abb948d4c',
        'info_dict': {
            'id': '5b7cd134-e2c1-11ea-89b3-b56dd0df2aa2',
            'ext': 'mp4',
            'title': 'La nuova spettacolare attrazione in Norvegia: il ponte sopra Vøringsfossen',
            'description': 'md5:18b35a291f6746c0c8dacd16e5f5f4f8',
            'uploader': 'DOVE Viaggi',
        }
    }, {
        'url': 'https://video.corriere.it/video-embed/b727632a-f9d0-11ea-91b0-38d50a849abb?player',
        'match_only': True
    }, {
        'url': 'https://video.corriere.it/video-360/metro-copenaghen-tutta-italiana/a248a7f0-e2db-11e9-9830-af2de6b1f945',
        'match_only': True
    }]


class GazzettaIE(RCSIE):
    IE_NAME = 'rcs:gazzetta'
    _VALID_URL = r'https?://(?P<vid>video)\.(?P<cdn>(?:gazzanet\.)?gazzetta\.it)/.+?/(?P<id>[^/]+?)(?:$|\?)'
    _TESTS = [{
        'url': 'https://video.gazzetta.it/video-motogp-catalogna-cadute-dovizioso-vale-rossi/49612410-00ca-11eb-bcd8-30d4253e0140?vclk=Videobar',
        'md5': 'eedc1b5defd18e67383afef51ff7bdf9',
        'info_dict': {
            'id': '49612410-00ca-11eb-bcd8-30d4253e0140',
            'ext': 'mp4',
            'title': 'Dovizioso, il contatto con Zarco e la caduta. E anche Vale finisce a terra',
            'description': 'md5:8c6e905dc3b9413218beca11ebd69778',
            'uploader': 'AMorici',
        }
    }, {
        'url': 'https://video.gazzetta.it/video-embed/49612410-00ca-11eb-bcd8-30d4253e0140',
        'match_only': True
    }, {
        'url': 'https://video.gazzanet.gazzetta.it/video-embed/gazzanet-mo05-0000260789',
        'match_only': True
    }]


class RCSVariousIE(RCSIE):
    IE_NAME = 'rcs:various'
    _VALID_URL = r'''(?x)https?://www\.
                    (?P<cdn>
                        leitv\.it|
                        youreporter\.it
                    )/(?:video/)?(?P<id>[^/]+?)(?:$|\?|/)'''
    _TESTS = [{
        'url': 'https://www.leitv.it/video/marmellata-di-ciliegie-fatta-in-casa/',
        'md5': '618aaabac32152199c1af86784d4d554',
        'info_dict': {
            'id': 'marmellata-di-ciliegie-fatta-in-casa',
            'ext': 'mp4',
            'title': 'Marmellata di ciliegie fatta in casa',
            'description': 'md5:89133864d6aad456dbcf6e7a29f86263',
            'uploader': 'leitv.it',
        }
    }, {
        'url': 'https://www.youreporter.it/fiume-sesia-3-ottobre-2020/',
        'md5': '8dccd436b47a830bab5b4a88232f391a',
        'info_dict': {
            'id': 'fiume-sesia-3-ottobre-2020',
            'ext': 'mp4',
            'title': 'Fiume Sesia 3 ottobre 2020',
            'description': 'md5:0070eef1cc884d13c970a4125063de55',
            'uploader': 'youreporter.it',
        }
    }]
Commit	Line	Data
a85e131b	1	# coding: utf-8
	2	from __future__ import unicode_literals
	3
	4	import re
	5
	6	from .common import InfoExtractor
	7	from ..utils import (
	8	clean_html,
	9	ExtractorError,
	10	js_to_json,
	11	base_url,
	12	url_basename,
	13	urljoin,
	14	)
	15
	16
	17	class RCSIE(InfoExtractor):
	18	_ALL_REPLACE = {
	19	'media2vam.corriere.it.edgesuite.net':
	20	'media2vam-corriere-it.akamaized.net',
	21	'media.youreporter.it.edgesuite.net':
	22	'media-youreporter-it.akamaized.net',
	23	'corrierepmd.corriere.it.edgesuite.net':
	24	'corrierepmd-corriere-it.akamaized.net',
	25	'media2vam-corriere-it.akamaized.net/fcs.quotidiani/vr/videos/':
	26	'video.corriere.it/vr360/videos/',
	27	'.net//': '.net/',
	28	}
	29	_MP4_REPLACE = {
	30	'media2vam.corbologna.corriere.it.edgesuite.net':
	31	'media2vam-bologna-corriere-it.akamaized.net',
	32	'media2vam.corfiorentino.corriere.it.edgesuite.net':
	33	'media2vam-fiorentino-corriere-it.akamaized.net',
	34	'media2vam.cormezzogiorno.corriere.it.edgesuite.net':
	35	'media2vam-mezzogiorno-corriere-it.akamaized.net',
	36	'media2vam.corveneto.corriere.it.edgesuite.net':
	37	'media2vam-veneto-corriere-it.akamaized.net',
	38	'media2.oggi.it.edgesuite.net':
	39	'media2-oggi-it.akamaized.net',
	40	'media2.quimamme.it.edgesuite.net':
	41	'media2-quimamme-it.akamaized.net',
	42	'media2.amica.it.edgesuite.net':
	43	'media2-amica-it.akamaized.net',
	44	'media2.living.corriere.it.edgesuite.net':
	45	'media2-living-corriere-it.akamaized.net',
	46	'media2.style.corriere.it.edgesuite.net':
	47	'media2-style-corriere-it.akamaized.net',
	48	'media2.iodonna.it.edgesuite.net':
	49	'media2-iodonna-it.akamaized.net',
	50	'media2.leitv.it.edgesuite.net':
	51	'media2-leitv-it.akamaized.net',
	52	}
	53	_MIGRATION_MAP = {
	54	'videoamica-vh.akamaihd': 'amica',
	55	'media2-amica-it.akamaized': 'amica',
	56	'corrierevam-vh.akamaihd': 'corriere',
	57	'media2vam-corriere-it.akamaized': 'corriere',
	58	'cormezzogiorno-vh.akamaihd': 'corrieredelmezzogiorno',
	59	'media2vam-mezzogiorno-corriere-it.akamaized': 'corrieredelmezzogiorno',
	60	'corveneto-vh.akamaihd': 'corrieredelveneto',
	61	'media2vam-veneto-corriere-it.akamaized': 'corrieredelveneto',
	62	'corbologna-vh.akamaihd': 'corrieredibologna',
	63	'media2vam-bologna-corriere-it.akamaized': 'corrieredibologna',
	64	'corfiorentino-vh.akamaihd': 'corrierefiorentino',
65	'media2vam-fiorentino-corriere-it.akamaized': 'corrierefiorentino',
66	'corinnovazione-vh.akamaihd': 'corriereinnovazione',
67	'media2-gazzanet-gazzetta-it.akamaized': 'gazzanet',
68	'videogazzanet-vh.akamaihd': 'gazzanet',
69	'videogazzaworld-vh.akamaihd': 'gazzaworld',
70	'gazzettavam-vh.akamaihd': 'gazzetta',
71	'media2vam-gazzetta-it.akamaized': 'gazzetta',
72	'videoiodonna-vh.akamaihd': 'iodonna',
73	'media2-leitv-it.akamaized': 'leitv',
74	'videoleitv-vh.akamaihd': 'leitv',
75	'videoliving-vh.akamaihd': 'living',
76	'media2-living-corriere-it.akamaized': 'living',
77	'media2-oggi-it.akamaized': 'oggi',
78	'videooggi-vh.akamaihd': 'oggi',
79	'media2-quimamme-it.akamaized': 'quimamme',
80	'quimamme-vh.akamaihd': 'quimamme',
81	'videorunning-vh.akamaihd': 'running',
82	'media2-style-corriere-it.akamaized': 'style',
83	'style-vh.akamaihd': 'style',
84	'videostyle-vh.akamaihd': 'style',
85	'media2-stylepiccoli-it.akamaized': 'stylepiccoli',
86	'stylepiccoli-vh.akamaihd': 'stylepiccoli',
87	'doveviaggi-vh.akamaihd': 'viaggi',
88	'media2-doveviaggi-it.akamaized': 'viaggi',
89	'media2-vivimilano-corriere-it.akamaized': 'vivimilano',
90	'vivimilano-vh.akamaihd': 'vivimilano',
91	'media2-youreporter-it.akamaized': 'youreporter'
92	}
93	_MIGRATION_MEDIA = {
94	'advrcs-vh.akamaihd': '',
95	'corriere-f.akamaihd': '',
96	'corrierepmd-corriere-it.akamaized': '',
97	'corrprotetto-vh.akamaihd': '',
98	'gazzetta-f.akamaihd': '',
99	'gazzettapmd-gazzetta-it.akamaized': '',
100	'gazzprotetto-vh.akamaihd': '',
101	'periodici-f.akamaihd': '',
102	'periodicisecure-vh.akamaihd': '',
103	'videocoracademy-vh.akamaihd': ''
104	}
105
106	def _get_video_src(self, video):
107	mediaFiles = video['mediaProfile']['mediaFile']
108	src = {}
109	# audio
110	if video['mediaType'] == 'AUDIO':
111	for aud in mediaFiles:
112	# todo: check
113	src['mp3'] = aud['value']
114	# video
115	else:
116	for vid in mediaFiles:
117	if vid['mimeType'] == 'application/vnd.apple.mpegurl':
118	src['m3u8'] = vid['value']
119	if vid['mimeType'] == 'video/mp4':
120	src['mp4'] = vid['value']
121
122	# replace host
123	for t in src:
124	for s, r in self._ALL_REPLACE.items():
125	src[t] = src[t].replace(s, r)
126	for s, r in self._MP4_REPLACE.items():
127	src[t] = src[t].replace(s, r)
128
129	# switch cdn
130	if 'mp4' in src and 'm3u8' in src:
131	if '-lh.akamaihd' not in src['m3u8'] and 'akamai' in src['mp4']:
132	if 'm3u8' in src:
133	matches = re.search(r'(?:https:)?\/\/(?P<host>.)\.net\/i(?P<path>.*)$', src['m3u8'])
134	src['m3u8'] = 'https://vod.rcsobjects.it/hls/%s%s' % (
135	self._MIGRATION_MAP[matches.group('host')],
136	matches.group('path').replace(
137	'///', '/').replace(
138	'//', '/').replace(
139	'.csmil', '.urlset'
140	)
141	)
142	if 'mp4' in src:
143	matches = re.search(r'(?:https:)?\/\/(?P<host>.)\.net\/i(?P<path>.*)$', src['mp4'])
144	if matches:
145	if matches.group('host') in self._MIGRATION_MEDIA:
146	vh_stream = 'https://media2.corriereobjects.it'
147	if src['mp4'].find('fcs.quotidiani_!'):
148	vh_stream = 'https://media2-it.corriereobjects.it'
149	src['mp4'] = '%s%s' % (
150	vh_stream,
151	matches.group('path').replace(
152	'///', '/').replace(
153	'//', '/').replace(
154	'/fcs.quotidiani/mediacenter', '').replace(
155	'/fcs.quotidiani_!/mediacenter', '').replace(
156	'corriere/content/mediacenter/', '').replace(
157	'gazzetta/content/mediacenter/', '')
158	)
159	else:
160	src['mp4'] = 'https://vod.rcsobjects.it/%s%s' % (
161	self._MIGRATION_MAP[matches.group('host')],
162	matches.group('path').replace('///', '/').replace('//', '/')
163	)
164
165	if 'mp3' in src:
166	src['mp3'] = src['mp3'].replace(
167	'media2vam-corriere-it.akamaized.net',
168	'vod.rcsobjects.it/corriere')
169	if 'mp4' in src:
170	if src['mp4'].find('fcs.quotidiani_!'):
171	src['mp4'] = src['mp4'].replace('vod.rcsobjects', 'vod-it.rcsobjects')
172	if 'm3u8' in src:
173	if src['m3u8'].find('fcs.quotidiani_!'):
174	src['m3u8'] = src['m3u8'].replace('vod.rcsobjects', 'vod-it.rcsobjects')
175
176	if 'geoblocking' in video['mediaProfile']:
177	if 'm3u8' in src:
178	src['m3u8'] = src['m3u8'].replace('vod.rcsobjects', 'vod-it.rcsobjects')
179	if 'mp4' in src:
180	src['mp4'] = src['mp4'].replace('vod.rcsobjects', 'vod-it.rcsobjects')
181	if 'm3u8' in src:
182	if src['m3u8'].find('csmil') and src['m3u8'].find('vod'):
183	src['m3u8'] = src['m3u8'].replace('.csmil', '.urlset')
184
185	return src
186
187	def _create_formats(self, urls, video_id):
188	formats = []
189	formats = self._extract_m3u8_formats(
190	urls['m3u8'], video_id, 'mp4', entry_protocol='m3u8_native',
191	m3u8_id='hls', fatal=False)
192
193	if not formats:
194	formats.append({
195	'format_id': 'http-mp4',
196	'url': urls['mp4']
197	})
198	self._sort_formats(formats)
199	return formats
200
201	def _real_extract(self, url):
202	video_id = self._match_id(url)
203	mobj = re.search(self._VALID_URL, url).groupdict()
204
205	if not mobj['cdn']:
206	raise ExtractorError('CDN not found in url: %s' % url)
207
208	# for leitv/youreporter/viaggi don't use the embed page
209	if (mobj['cdn'] not in ['leitv.it', 'youreporter.it']) and (mobj['vid'] == 'video'):
210	url = 'https://video.%s/video-embed/%s' % (mobj['cdn'], video_id)
211
212	page = self._download_webpage(url, video_id)
213
214	video_data = None
215	# look for json video data url
216	json = self._search_regex(
217	r'''var url\s=\s["']((?:https?:)?//video\.rcs\.it/fragment-includes/video-includes/.+?\.json)["'];''',
218	page, video_id, default=None)
219	if json:
220	if json.startswith('//'):
221	json = 'https:%s' % json
222	video_data = self._download_json(json, video_id)
223
224	# if url not found, look for json video data directly in the page
225	else:
226	json = self._search_regex(
227	r'[\s;]video\s=\s({[\s\S]+?})(?:;\|,playlist=)',
228	page, video_id, default=None)
229	if json:
230	video_data = self._parse_json(
231	json, video_id, transform_source=js_to_json)
232	else:
233	# if no video data found try search for iframes
234	emb = RCSEmbedsIE._extract_url(page)
235	if emb:
236	return self._real_extract(emb)
237
238	if not video_data:
239	raise ExtractorError('Video data not found in the page')
240
241	formats = self._create_formats(
242	self._get_video_src(video_data), video_id)
243
244	return {
245	'id': video_id,
246	'title': video_data['title'],
247	'description': video_data['description'] or clean_html(video_data['htmlDescription']),
248	'uploader': video_data['provider'] if video_data['provider'] else mobj['cdn'],
249	'formats': formats
250	}
251
252
253	class RCSEmbedsIE(RCSIE):
254	IE_NAME = 'rcs:rcs'
255	_VALID_URL = r'''(?x)
256	https?://(?P<vid>video)\.
257	(?P<cdn>
258	(?:
259	rcs\|
260	(?:corriere\w+\.)?corriere\|
261	(?:gazzanet\.)?gazzetta
262	)\.it)
263	/video-embed/(?P<id>[^/=&\?]+?)(?:$\|\?)'''
264	_TESTS = [{
265	'url': 'https://video.rcs.it/video-embed/iodonna-0001585037',
266	'md5': '623ecc8ffe7299b2d0c1046d8331a9df',
267	'info_dict': {
268	'id': 'iodonna-0001585037',
269	'ext': 'mp4',
270	'title': 'Sky Arte racconta Madonna nella serie "Artist to icon"',
271	'description': 'md5:65b09633df9ffee57f48b39e34c9e067',
272	'uploader': 'rcs.it',
273	}
274	}, {
275	'url': 'https://video.corriere.it/video-embed/b727632a-f9d0-11ea-91b0-38d50a849abb?player',
276	'match_only': True
277	}, {
278	'url': 'https://video.gazzetta.it/video-embed/49612410-00ca-11eb-bcd8-30d4253e0140',
279	'match_only': True
280	}]
281
282	@staticmethod
283	def _sanitize_urls(urls):
284	# add protocol if missing
285	for i, e in enumerate(urls):
286	if e.startswith('//'):
287	urls[i] = 'https:%s' % e
288	# clean iframes urls
289	for i, e in enumerate(urls):
290	urls[i] = urljoin(base_url(e), url_basename(e))
291	return urls
292
293	@staticmethod
294	def _extract_urls(webpage):
295	entries = [
296	mobj.group('url')
297	for mobj in re.finditer(r'''(?x)
298	(?:
299	data-frame-src=\|
300	<iframe[^\n]+src=
301	)
302	(["'])
303	(?P<url>(?:https?:)?//video\.
304	(?:
305	rcs\|
306	(?:corriere\w+\.)?corriere\|
307	(?:gazzanet\.)?gazzetta
308	)
309	\.it/video-embed/.+?)
310	\1''', webpage)]
311	return RCSEmbedsIE._sanitize_urls(entries)
312
313	@staticmethod
314	def _extract_url(webpage):
315	urls = RCSEmbedsIE._extract_urls(webpage)
316	return urls[0] if urls else None
317
318
319	class CorriereIE(RCSIE):
320	IE_NAME = 'rcs:corriere'
321	_VALID_URL = r'''(?x)https?://(?P<vid>video\|viaggi)\.
322	(?P<cdn>
323	(?:
324	corrieredelmezzogiorno\.\|
325	corrieredelveneto\.\|
326	corrieredibologna\.\|
327	corrierefiorentino\.
328	)?
329	corriere\.it)/.+?/(?P<id>[^/]+)(?=\?\|/$\|$)'''
330	_TESTS = [{
331	'url': 'https://video.corriere.it/sport/formula-1/vettel-guida-ferrari-sf90-mugello-suo-fianco-c-elecrerc-bendato-video-esilarante/b727632a-f9d0-11ea-91b0-38d50a849abb',
332	'md5': '0f4ededc202b0f00b6e509d831e2dcda',
333	'info_dict': {
334	'id': 'b727632a-f9d0-11ea-91b0-38d50a849abb',
335	'ext': 'mp4',
336	'title': 'Vettel guida la Ferrari SF90 al Mugello e al suo fianco c\'è Leclerc (bendato): il video è esilarante',
337	'description': 'md5:93b51c9161ac8a64fb2f997b054d0152',
338	'uploader': 'Corriere Tv',
339	}
340	}, {
341	'url': 'https://viaggi.corriere.it/video/norvegia-il-nuovo-ponte-spettacolare-sopra-la-cascata-di-voringsfossen/',
342	'md5': 'da378e4918d2afbf7d61c35abb948d4c',
343	'info_dict': {
344	'id': '5b7cd134-e2c1-11ea-89b3-b56dd0df2aa2',
345	'ext': 'mp4',
346	'title': 'La nuova spettacolare attrazione in Norvegia: il ponte sopra Vøringsfossen',
347	'description': 'md5:18b35a291f6746c0c8dacd16e5f5f4f8',
348	'uploader': 'DOVE Viaggi',
349	}
350	}, {
351	'url': 'https://video.corriere.it/video-embed/b727632a-f9d0-11ea-91b0-38d50a849abb?player',
352	'match_only': True
353	}, {
354	'url': 'https://video.corriere.it/video-360/metro-copenaghen-tutta-italiana/a248a7f0-e2db-11e9-9830-af2de6b1f945',
355	'match_only': True
356	}]
357
358
359	class GazzettaIE(RCSIE):
360	IE_NAME = 'rcs:gazzetta'
361	_VALID_URL = r'https?://(?P<vid>video)\.(?P<cdn>(?:gazzanet\.)?gazzetta\.it)/.+?/(?P<id>[^/]+?)(?:$\|\?)'
362	_TESTS = [{
363	'url': 'https://video.gazzetta.it/video-motogp-catalogna-cadute-dovizioso-vale-rossi/49612410-00ca-11eb-bcd8-30d4253e0140?vclk=Videobar',
364	'md5': 'eedc1b5defd18e67383afef51ff7bdf9',
365	'info_dict': {
366	'id': '49612410-00ca-11eb-bcd8-30d4253e0140',
367	'ext': 'mp4',
368	'title': 'Dovizioso, il contatto con Zarco e la caduta. E anche Vale finisce a terra',
369	'description': 'md5:8c6e905dc3b9413218beca11ebd69778',
370	'uploader': 'AMorici',
371	}
372	}, {
373	'url': 'https://video.gazzetta.it/video-embed/49612410-00ca-11eb-bcd8-30d4253e0140',
374	'match_only': True
375	}, {
376	'url': 'https://video.gazzanet.gazzetta.it/video-embed/gazzanet-mo05-0000260789',
377	'match_only': True
378	}]
379
380
381	class RCSVariousIE(RCSIE):
382	IE_NAME = 'rcs:various'
383	_VALID_URL = r'''(?x)https?://www\.
384	(?P<cdn>
385	leitv\.it\|
386	youreporter\.it
387	)/(?:video/)?(?P<id>[^/]+?)(?:$\|\?\|/)'''
388	_TESTS = [{
389	'url': 'https://www.leitv.it/video/marmellata-di-ciliegie-fatta-in-casa/',
390	'md5': '618aaabac32152199c1af86784d4d554',
391	'info_dict': {
392	'id': 'marmellata-di-ciliegie-fatta-in-casa',
393	'ext': 'mp4',
394	'title': 'Marmellata di ciliegie fatta in casa',
395	'description': 'md5:89133864d6aad456dbcf6e7a29f86263',
396	'uploader': 'leitv.it',
397	}
398	}, {
399	'url': 'https://www.youreporter.it/fiume-sesia-3-ottobre-2020/',
400	'md5': '8dccd436b47a830bab5b4a88232f391a',
401	'info_dict': {
402	'id': 'fiume-sesia-3-ottobre-2020',
403	'ext': 'mp4',
404	'title': 'Fiume Sesia 3 ottobre 2020',
405	'description': 'md5:0070eef1cc884d13c970a4125063de55',
406	'uploader': 'youreporter.it',
407	}
408	}]