1
0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-15 14:47:44 +01:00
yt-dlp/yt_dlp/extractor/globo.py

244 lines
9.8 KiB
Python
Raw Normal View History

import base64
import hashlib
import json
import random
import re
from .common import InfoExtractor
from ..networking import HEADRequest
from ..utils import (
ExtractorError,
float_or_none,
orderedSet,
2015-11-04 12:12:42 +01:00
str_or_none,
try_get,
)
class GloboIE(InfoExtractor):
2016-09-08 12:04:57 +02:00
_VALID_URL = r'(?:globo:|https?://.+?\.globo\.com/(?:[^/]+/)*(?:v/(?:[^/]+/)?|videos/))(?P<id>\d{7,})'
_NETRC_MACHINE = 'globo'
2015-11-04 11:46:26 +01:00
_TESTS = [{
'url': 'http://g1.globo.com/carros/autoesporte/videos/t/exclusivos-do-g1/v/mercedes-benz-gla-passa-por-teste-de-colisao-na-europa/3607726/',
'info_dict': {
'id': '3607726',
'ext': 'mp4',
'title': 'Mercedes-Benz GLA passa por teste de colisão na Europa',
'duration': 103.204,
'uploader': 'G1',
'uploader_id': '2015',
},
'params': {
'skip_download': True,
2015-11-04 12:10:45 +01:00
},
2015-11-04 11:46:26 +01:00
}, {
2015-11-04 12:10:45 +01:00
'url': 'http://globoplay.globo.com/v/4581987/',
2015-11-04 11:46:26 +01:00
'info_dict': {
2015-11-04 12:10:45 +01:00
'id': '4581987',
2015-11-04 11:46:26 +01:00
'ext': 'mp4',
2015-11-04 12:10:45 +01:00
'title': 'Acidentes de trânsito estão entre as maiores causas de queda de energia em SP',
'duration': 137.973,
'uploader': 'Rede Globo',
2015-11-04 12:12:42 +01:00
'uploader_id': '196',
2015-11-04 12:10:45 +01:00
},
'params': {
'skip_download': True,
},
2015-11-04 12:10:45 +01:00
}, {
'url': 'http://canalbrasil.globo.com/programas/sangue-latino/videos/3928201.html',
'only_matching': True,
}, {
'url': 'http://globosatplay.globo.com/globonews/v/4472924/',
'only_matching': True,
}, {
'url': 'http://globotv.globo.com/t/programa/v/clipe-sexo-e-as-negas-adeus/3836166/',
'only_matching': True,
}, {
'url': 'http://globotv.globo.com/canal-brasil/sangue-latino/t/todos-os-videos/v/ator-e-diretor-argentino-ricado-darin-fala-sobre-utopias-e-suas-perdas/3928201/',
'only_matching': True,
2015-11-04 12:42:11 +01:00
}, {
'url': 'http://canaloff.globo.com/programas/desejar-profundo/videos/4518560.html',
'only_matching': True,
}, {
'url': 'globo:3607726',
'only_matching': True,
}, {
'url': 'https://globoplay.globo.com/v/10248083/',
'info_dict': {
'id': '10248083',
'ext': 'mp4',
'title': 'Melhores momentos: Equador 1 x 1 Brasil pelas Eliminatórias da Copa do Mundo 2022',
'duration': 530.964,
'uploader': 'SporTV',
'uploader_id': '698',
},
'params': {
'skip_download': True,
},
2015-11-04 11:46:26 +01:00
}]
def _real_extract(self, url):
video_id = self._match_id(url)
self._request_webpage(
HEADRequest('https://globo-ab.globo.com/v2/selected-alternatives?experiments=player-isolated-experiment-02&skipImpressions=true'),
video_id, 'Getting cookies')
video = self._download_json(
f'http://api.globovideos.com/videos/{video_id}/playlist',
video_id)['videos'][0]
if not self.get_param('allow_unplayable_formats') and video.get('encrypted') is True:
self.report_drm(video_id)
title = video['title']
formats = []
security = self._download_json(
'https://playback.video.globo.com/v2/video-session', video_id, f'Downloading security hash for {video_id}',
headers={'content-type': 'application/json'}, data=json.dumps({
'player_type': 'desktop',
'video_id': video_id,
'quality': 'max',
'content_protection': 'widevine',
'vsid': '581b986b-4c40-71f0-5a58-803e579d5fa2',
'tz': '-3.0:00',
}).encode())
self._request_webpage(HEADRequest(security['sources'][0]['url_template']), video_id, 'Getting locksession cookie')
security_hash = security['sources'][0]['token']
if not security_hash:
message = security.get('message')
if message:
raise ExtractorError(
f'{self.IE_NAME} returned error: {message}', expected=True)
hash_code = security_hash[:2]
padding = '%010d' % random.randint(1, 10000000000)
if hash_code in ('04', '14'):
received_time = security_hash[3:13]
received_md5 = security_hash[24:]
hash_prefix = security_hash[:23]
elif hash_code in ('02', '12', '03', '13'):
received_time = security_hash[2:12]
received_md5 = security_hash[22:]
padding += '1'
hash_prefix = '05' + security_hash[:22]
padded_sign_time = str(int(received_time) + 86400) + padding
md5_data = (received_md5 + padded_sign_time + '0xAC10FD').encode()
signed_md5 = base64.urlsafe_b64encode(hashlib.md5(md5_data).digest()).decode().strip('=')
signed_hash = hash_prefix + padded_sign_time + signed_md5
source = security['sources'][0]['url_parts']
resource_url = source['scheme'] + '://' + source['domain'] + source['path']
signed_url = '{}?h={}&k=html5&a={}'.format(resource_url, signed_hash, 'F' if video.get('subscriber_only') else 'A')
fmts, subtitles = self._extract_m3u8_formats_and_subtitles(
signed_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
formats.extend(fmts)
for resource in video['resources']:
if resource.get('type') == 'subtitle':
subtitles.setdefault(resource.get('language') or 'por', []).append({
'url': resource.get('url'),
})
subs = try_get(security, lambda x: x['source']['subtitles'], expected_type=dict) or {}
for sub_lang, sub_url in subs.items():
if sub_url:
subtitles.setdefault(sub_lang or 'por', []).append({
'url': sub_url,
})
subs = try_get(security, lambda x: x['source']['subtitles_webvtt'], expected_type=dict) or {}
for sub_lang, sub_url in subs.items():
if sub_url:
subtitles.setdefault(sub_lang or 'por', []).append({
'url': sub_url,
})
duration = float_or_none(video.get('duration'), 1000)
uploader = video.get('channel')
2015-11-04 12:12:42 +01:00
uploader_id = str_or_none(video.get('channel_id'))
return {
'id': video_id,
'title': title,
'duration': duration,
'uploader': uploader,
'uploader_id': uploader_id,
'formats': formats,
'subtitles': subtitles,
2014-11-23 20:41:03 +01:00
}
2015-11-04 11:46:26 +01:00
class GloboArticleIE(InfoExtractor):
_VALID_URL = r'https?://.+?\.globo\.com/(?:[^/]+/)*(?P<id>[^/.]+)(?:\.html)?'
2015-11-04 11:46:26 +01:00
_VIDEOID_REGEXES = [
r'\bdata-video-id=["\'](\d{7,})["\']',
r'\bdata-player-videosids=["\'](\d{7,})["\']',
r'\bvideosIDs\s*:\s*["\']?(\d{7,})',
r'\bdata-id=["\'](\d{7,})["\']',
r'<div[^>]+\bid=["\'](\d{7,})["\']',
r'<bs-player[^>]+\bvideoid=["\'](\d{8,})["\']',
2015-11-04 11:46:26 +01:00
]
2015-11-04 12:42:11 +01:00
_TESTS = [{
2015-11-04 11:46:26 +01:00
'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html',
'info_dict': {
'id': 'novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes',
'title': 'Novidade na fiscalização de bagagem pela Receita provoca discussões',
'description': 'md5:c3c4b4d4c30c32fce460040b1ac46b12',
},
'playlist_count': 1,
}, {
'url': 'http://g1.globo.com/pr/parana/noticia/2016/09/mpf-denuncia-lula-marisa-e-mais-seis-na-operacao-lava-jato.html',
'info_dict': {
'id': 'mpf-denuncia-lula-marisa-e-mais-seis-na-operacao-lava-jato',
'title': "Lula era o 'comandante máximo' do esquema da Lava Jato, diz MPF",
'description': 'md5:8aa7cc8beda4dc71cc8553e00b77c54c',
},
'playlist_count': 6,
2015-11-04 12:42:11 +01:00
}, {
'url': 'http://gq.globo.com/Prazeres/Poder/noticia/2015/10/all-o-desafio-assista-ao-segundo-capitulo-da-serie.html',
'only_matching': True,
}, {
'url': 'http://gshow.globo.com/programas/tv-xuxa/O-Programa/noticia/2014/01/xuxa-e-junno-namoram-muuuito-em-luau-de-zeze-di-camargo-e-luciano.html',
'only_matching': True,
}, {
'url': 'http://oglobo.globo.com/rio/a-amizade-entre-um-entregador-de-farmacia-um-piano-19946271',
'only_matching': True,
}, {
'url': 'https://ge.globo.com/video/ta-na-area-como-foi-assistir-ao-jogo-do-palmeiras-que-a-globo-nao-passou-10287094.ghtml',
'info_dict': {
'id': 'ta-na-area-como-foi-assistir-ao-jogo-do-palmeiras-que-a-globo-nao-passou-10287094',
'title': 'Tá na Área: como foi assistir ao jogo do Palmeiras que a Globo não passou',
'description': 'md5:2d089d036c4c9675117d3a56f8c61739',
},
'playlist_count': 1,
}, {
'url': 'https://redeglobo.globo.com/rpc/meuparana/noticia/a-producao-de-chocolates-no-parana.ghtml',
'info_dict': {
'id': 'a-producao-de-chocolates-no-parana',
'title': 'A produção de chocolates no Paraná',
'description': 'md5:f2e3daf00ffd1dc0e9a8a6c7cfb0a89e',
},
'playlist_count': 2,
2015-11-04 12:42:11 +01:00
}]
2015-11-04 11:46:26 +01:00
@classmethod
def suitable(cls, url):
return False if GloboIE.suitable(url) else super().suitable(url)
2015-11-04 11:46:26 +01:00
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
video_ids = []
for video_regex in self._VIDEOID_REGEXES:
video_ids.extend(re.findall(video_regex, webpage))
entries = [
self.url_result(f'globo:{video_id}', GloboIE.ie_key())
for video_id in orderedSet(video_ids)]
title = self._og_search_title(webpage).strip()
description = self._html_search_meta('description', webpage)
return self.playlist_result(entries, display_id, title, description)