2 from __future__ import unicode_literals
10 from .common import InfoExtractor
11 from ..compat import (
24 class GloboIE(InfoExtractor):
25 _VALID_URL = r'(?:globo:|https?://.+?\.globo\.com/(?:[^/]+/)*(?:v/(?:[^/]+/)?|videos/))(?P<id>\d{7,})'
26 _NETRC_MACHINE = 'globo'
28 'url': 'http://g1.globo.com/carros/autoesporte/videos/t/exclusivos-do-g1/v/mercedes-benz-gla-passa-por-teste-de-colisao-na-europa/3607726/',
29 'md5': 'b3ccc801f75cd04a914d51dadb83a78d',
33 'title': 'Mercedes-Benz GLA passa por teste de colisão na Europa',
35 'uploader': 'Globo.com',
39 'url': 'http://globoplay.globo.com/v/4581987/',
40 'md5': 'f36a1ecd6a50da1577eee6dd17f67eff',
44 'title': 'Acidentes de trânsito estão entre as maiores causas de queda de energia em SP',
46 'uploader': 'Rede Globo',
50 'url': 'http://canalbrasil.globo.com/programas/sangue-latino/videos/3928201.html',
51 'only_matching': True,
53 'url': 'http://globosatplay.globo.com/globonews/v/4472924/',
54 'only_matching': True,
56 'url': 'http://globotv.globo.com/t/programa/v/clipe-sexo-e-as-negas-adeus/3836166/',
57 'only_matching': True,
59 '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/',
60 'only_matching': True,
62 'url': 'http://canaloff.globo.com/programas/desejar-profundo/videos/4518560.html',
63 'only_matching': True,
65 'url': 'globo:3607726',
66 'only_matching': True,
69 def _real_initialize(self):
70 email, password = self._get_login_info()
75 glb_id = (self._download_json(
76 'https://login.globo.com/api/authentication', None, data=json.dumps({
82 }).encode(), headers={
83 'Content-Type': 'application/json; charset=utf-8',
84 }) or {}).get('glbId')
86 self._set_cookie('.globo.com', 'GLBID', glb_id)
87 except ExtractorError as e:
88 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
89 resp = self._parse_json(e.cause.read(), None)
90 raise ExtractorError(resp.get('userMessage') or resp['id'], expected=True)
93 def _real_extract(self, url):
94 video_id = self._match_id(url)
96 video = self._download_json(
97 'http://api.globovideos.com/videos/%s/playlist' % video_id,
98 video_id)['videos'][0]
99 if video.get('encrypted') is True:
100 raise ExtractorError('This video is DRM protected.', expected=True)
102 title = video['title']
105 for resource in video['resources']:
106 resource_id = resource.get('_id')
107 resource_url = resource.get('url')
108 if not resource_id or not resource_url:
111 security = self._download_json(
112 'http://security.video.globo.com/videos/%s/hash' % video_id,
113 video_id, 'Downloading security hash for %s' % resource_id, query={
116 'resource_id': resource_id,
119 security_hash = security.get('hash')
120 if not security_hash:
121 message = security.get('message')
123 raise ExtractorError(
124 '%s returned error: %s' % (self.IE_NAME, message), expected=True)
127 assert security_hash[:2] in ('04', '14')
128 received_time = security_hash[3:13]
129 received_md5 = security_hash[24:]
131 sign_time = compat_str(int(received_time) + 86400)
132 padding = '%010d' % random.randint(1, 10000000000)
134 md5_data = (received_md5 + sign_time + padding + '0xAC10FD').encode()
135 signed_md5 = base64.urlsafe_b64encode(hashlib.md5(md5_data).digest()).decode().strip('=')
136 signed_hash = security_hash[:23] + sign_time + padding + signed_md5
138 signed_url = '%s?h=%s&k=html5&a=%s&u=%s' % (resource_url, signed_hash, 'F' if video.get('subscriber_only') else 'A', security.get('user') or '')
139 if resource_id.endswith('m3u8') or resource_url.endswith('.m3u8'):
140 formats.extend(self._extract_m3u8_formats(
141 signed_url, resource_id, 'mp4', entry_protocol='m3u8_native',
142 m3u8_id='hls', fatal=False))
143 elif resource_id.endswith('mpd') or resource_url.endswith('.mpd'):
144 formats.extend(self._extract_mpd_formats(
145 signed_url, resource_id, mpd_id='dash', fatal=False))
146 elif resource_id.endswith('manifest') or resource_url.endswith('/manifest'):
147 formats.extend(self._extract_ism_formats(
148 signed_url, resource_id, ism_id='mss', fatal=False))
152 'format_id': 'http-%s' % resource_id,
153 'height': int_or_none(resource.get('height')),
156 self._sort_formats(formats)
158 duration = float_or_none(video.get('duration'), 1000)
159 uploader = video.get('channel')
160 uploader_id = str_or_none(video.get('channel_id'))
165 'duration': duration,
166 'uploader': uploader,
167 'uploader_id': uploader_id,
172 class GloboArticleIE(InfoExtractor):
173 _VALID_URL = r'https?://.+?\.globo\.com/(?:[^/]+/)*(?P<id>[^/.]+)(?:\.html)?'
176 r'\bdata-video-id=["\'](\d{7,})',
177 r'\bdata-player-videosids=["\'](\d{7,})',
178 r'\bvideosIDs\s*:\s*["\']?(\d{7,})',
179 r'\bdata-id=["\'](\d{7,})',
180 r'<div[^>]+\bid=["\'](\d{7,})',
184 'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html',
186 'id': 'novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes',
187 'title': 'Novidade na fiscalização de bagagem pela Receita provoca discussões',
188 'description': 'md5:c3c4b4d4c30c32fce460040b1ac46b12',
192 'url': 'http://g1.globo.com/pr/parana/noticia/2016/09/mpf-denuncia-lula-marisa-e-mais-seis-na-operacao-lava-jato.html',
194 'id': 'mpf-denuncia-lula-marisa-e-mais-seis-na-operacao-lava-jato',
195 'title': "Lula era o 'comandante máximo' do esquema da Lava Jato, diz MPF",
196 'description': 'md5:8aa7cc8beda4dc71cc8553e00b77c54c',
200 'url': 'http://gq.globo.com/Prazeres/Poder/noticia/2015/10/all-o-desafio-assista-ao-segundo-capitulo-da-serie.html',
201 'only_matching': True,
203 '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',
204 'only_matching': True,
206 'url': 'http://oglobo.globo.com/rio/a-amizade-entre-um-entregador-de-farmacia-um-piano-19946271',
207 'only_matching': True,
211 def suitable(cls, url):
212 return False if GloboIE.suitable(url) else super(GloboArticleIE, cls).suitable(url)
214 def _real_extract(self, url):
215 display_id = self._match_id(url)
216 webpage = self._download_webpage(url, display_id)
218 for video_regex in self._VIDEOID_REGEXES:
219 video_ids.extend(re.findall(video_regex, webpage))
221 self.url_result('globo:%s' % video_id, GloboIE.ie_key())
222 for video_id in orderedSet(video_ids)]
223 title = self._og_search_title(webpage, fatal=False)
224 description = self._html_search_meta('description', webpage)
225 return self.playlist_result(entries, display_id, title, description)