]>
Commit | Line | Data |
---|---|---|
f47754f0 S |
1 | # coding: utf-8 |
2 | from __future__ import unicode_literals | |
3 | ||
db2058f6 RA |
4 | import base64 |
5 | import hashlib | |
6 | import json | |
f47754f0 | 7 | import random |
26394d02 | 8 | import re |
f47754f0 S |
9 | |
10 | from .common import InfoExtractor | |
e5187493 RA |
11 | from ..compat import ( |
12 | compat_HTTPError, | |
13 | compat_str, | |
14 | ) | |
8c25f81b PH |
15 | from ..utils import ( |
16 | ExtractorError, | |
17 | float_or_none, | |
fffccaaf | 18 | int_or_none, |
26394d02 | 19 | orderedSet, |
e7d34c03 | 20 | str_or_none, |
8c25f81b | 21 | ) |
f47754f0 S |
22 | |
23 | ||
24 | class GloboIE(InfoExtractor): | |
25042f73 | 25 | _VALID_URL = r'(?:globo:|https?://.+?\.globo\.com/(?:[^/]+/)*(?:v/(?:[^/]+/)?|videos/))(?P<id>\d{7,})' |
d81ffc3a | 26 | _NETRC_MACHINE = 'globo' |
ad607563 | 27 | _TESTS = [{ |
ad607563 S |
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', | |
30 | 'info_dict': { | |
31 | 'id': '3607726', | |
32 | 'ext': 'mp4', | |
33 | 'title': 'Mercedes-Benz GLA passa por teste de colisão na Europa', | |
34 | 'duration': 103.204, | |
35 | 'uploader': 'Globo.com', | |
e7d34c03 | 36 | 'uploader_id': '265', |
264cd00f | 37 | }, |
ad607563 | 38 | }, { |
264cd00f S |
39 | 'url': 'http://globoplay.globo.com/v/4581987/', |
40 | 'md5': 'f36a1ecd6a50da1577eee6dd17f67eff', | |
ad607563 | 41 | 'info_dict': { |
264cd00f | 42 | 'id': '4581987', |
ad607563 | 43 | 'ext': 'mp4', |
264cd00f S |
44 | 'title': 'Acidentes de trânsito estão entre as maiores causas de queda de energia em SP', |
45 | 'duration': 137.973, | |
46 | 'uploader': 'Rede Globo', | |
e7d34c03 | 47 | 'uploader_id': '196', |
264cd00f S |
48 | }, |
49 | }, { | |
50 | 'url': 'http://canalbrasil.globo.com/programas/sangue-latino/videos/3928201.html', | |
51 | 'only_matching': True, | |
52 | }, { | |
53 | 'url': 'http://globosatplay.globo.com/globonews/v/4472924/', | |
54 | 'only_matching': True, | |
55 | }, { | |
56 | 'url': 'http://globotv.globo.com/t/programa/v/clipe-sexo-e-as-negas-adeus/3836166/', | |
57 | 'only_matching': True, | |
58 | }, { | |
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, | |
5d501a09 S |
61 | }, { |
62 | 'url': 'http://canaloff.globo.com/programas/desejar-profundo/videos/4518560.html', | |
63 | 'only_matching': True, | |
26394d02 S |
64 | }, { |
65 | 'url': 'globo:3607726', | |
66 | 'only_matching': True, | |
ad607563 | 67 | }] |
f47754f0 | 68 | |
db2058f6 | 69 | def _real_initialize(self): |
db2058f6 RA |
70 | email, password = self._get_login_info() |
71 | if email is None: | |
72 | return | |
73 | ||
e5187493 | 74 | try: |
65615be3 | 75 | glb_id = (self._download_json( |
e5187493 RA |
76 | 'https://login.globo.com/api/authentication', None, data=json.dumps({ |
77 | 'payload': { | |
78 | 'email': email, | |
79 | 'password': password, | |
80 | 'serviceId': 4654, | |
81 | }, | |
82 | }).encode(), headers={ | |
83 | 'Content-Type': 'application/json; charset=utf-8', | |
65615be3 RA |
84 | }) or {}).get('glbId') |
85 | if glb_id: | |
86 | self._set_cookie('.globo.com', 'GLBID', glb_id) | |
e5187493 RA |
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) | |
91 | raise | |
f47754f0 S |
92 | |
93 | def _real_extract(self, url): | |
94 | video_id = self._match_id(url) | |
95 | ||
f47754f0 | 96 | video = self._download_json( |
db2058f6 RA |
97 | 'http://api.globovideos.com/videos/%s/playlist' % video_id, |
98 | video_id)['videos'][0] | |
a06916d9 | 99 | if not self.get_param('allow_unplayable_formats') and video.get('encrypted') is True: |
d4bb825b | 100 | raise ExtractorError('This video is DRM protected.', expected=True) |
f47754f0 S |
101 | |
102 | title = video['title'] | |
f47754f0 S |
103 | |
104 | formats = [] | |
30eb05cb | 105 | subtitles = {} |
f47754f0 S |
106 | for resource in video['resources']: |
107 | resource_id = resource.get('_id') | |
db2058f6 | 108 | resource_url = resource.get('url') |
30eb05cb RA |
109 | resource_type = resource.get('type') |
110 | if not resource_url or (resource_type == 'media' and not resource_id) or resource_type not in ('subtitle', 'media'): | |
111 | continue | |
112 | ||
113 | if resource_type == 'subtitle': | |
114 | subtitles.setdefault(resource.get('language') or 'por', []).append({ | |
115 | 'url': resource_url, | |
116 | }) | |
f47754f0 S |
117 | continue |
118 | ||
119 | security = self._download_json( | |
db2058f6 RA |
120 | 'http://security.video.globo.com/videos/%s/hash' % video_id, |
121 | video_id, 'Downloading security hash for %s' % resource_id, query={ | |
d4bb825b RA |
122 | 'player': 'desktop', |
123 | 'version': '5.19.1', | |
db2058f6 RA |
124 | 'resource_id': resource_id, |
125 | }) | |
f47754f0 S |
126 | |
127 | security_hash = security.get('hash') | |
128 | if not security_hash: | |
129 | message = security.get('message') | |
130 | if message: | |
131 | raise ExtractorError( | |
132 | '%s returned error: %s' % (self.IE_NAME, message), expected=True) | |
133 | continue | |
134 | ||
0c2d10d2 | 135 | hash_code = security_hash[:2] |
f47754f0 | 136 | padding = '%010d' % random.randint(1, 10000000000) |
0c2d10d2 RA |
137 | if hash_code in ('04', '14'): |
138 | received_time = security_hash[3:13] | |
139 | received_md5 = security_hash[24:] | |
140 | hash_prefix = security_hash[:23] | |
141 | elif hash_code in ('02', '12', '03', '13'): | |
142 | received_time = security_hash[2:12] | |
143 | received_md5 = security_hash[22:] | |
144 | padding += '1' | |
145 | hash_prefix = '05' + security_hash[:22] | |
146 | ||
147 | padded_sign_time = compat_str(int(received_time) + 86400) + padding | |
148 | md5_data = (received_md5 + padded_sign_time + '0xAC10FD').encode() | |
db2058f6 | 149 | signed_md5 = base64.urlsafe_b64encode(hashlib.md5(md5_data).digest()).decode().strip('=') |
0c2d10d2 | 150 | signed_hash = hash_prefix + padded_sign_time + signed_md5 |
d4bb825b | 151 | 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 '') |
0c2d10d2 | 152 | |
8c72beb2 | 153 | if resource_id.endswith('m3u8') or resource_url.endswith('.m3u8'): |
7e5edcfd | 154 | formats.extend(self._extract_m3u8_formats( |
5d235ca7 | 155 | signed_url, resource_id, 'mp4', entry_protocol='m3u8_native', |
7e5edcfd | 156 | m3u8_id='hls', fatal=False)) |
db2058f6 RA |
157 | elif resource_id.endswith('mpd') or resource_url.endswith('.mpd'): |
158 | formats.extend(self._extract_mpd_formats( | |
159 | signed_url, resource_id, mpd_id='dash', fatal=False)) | |
160 | elif resource_id.endswith('manifest') or resource_url.endswith('/manifest'): | |
161 | formats.extend(self._extract_ism_formats( | |
162 | signed_url, resource_id, ism_id='mss', fatal=False)) | |
8c72beb2 S |
163 | else: |
164 | formats.append({ | |
165 | 'url': signed_url, | |
a4a6b7b8 S |
166 | 'format_id': 'http-%s' % resource_id, |
167 | 'height': int_or_none(resource.get('height')), | |
8c72beb2 | 168 | }) |
f47754f0 S |
169 | |
170 | self._sort_formats(formats) | |
171 | ||
fffccaaf | 172 | duration = float_or_none(video.get('duration'), 1000) |
fffccaaf | 173 | uploader = video.get('channel') |
e7d34c03 | 174 | uploader_id = str_or_none(video.get('channel_id')) |
fffccaaf | 175 | |
f47754f0 S |
176 | return { |
177 | 'id': video_id, | |
178 | 'title': title, | |
179 | 'duration': duration, | |
180 | 'uploader': uploader, | |
181 | 'uploader_id': uploader_id, | |
30eb05cb RA |
182 | 'formats': formats, |
183 | 'subtitles': subtitles, | |
5f6a1245 | 184 | } |
ad607563 S |
185 | |
186 | ||
187 | class GloboArticleIE(InfoExtractor): | |
26394d02 | 188 | _VALID_URL = r'https?://.+?\.globo\.com/(?:[^/]+/)*(?P<id>[^/.]+)(?:\.html)?' |
ad607563 S |
189 | |
190 | _VIDEOID_REGEXES = [ | |
191 | r'\bdata-video-id=["\'](\d{7,})', | |
192 | r'\bdata-player-videosids=["\'](\d{7,})', | |
9e5751b9 | 193 | r'\bvideosIDs\s*:\s*["\']?(\d{7,})', |
ad607563 S |
194 | r'\bdata-id=["\'](\d{7,})', |
195 | r'<div[^>]+\bid=["\'](\d{7,})', | |
196 | ] | |
197 | ||
5d501a09 | 198 | _TESTS = [{ |
ad607563 | 199 | 'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html', |
ad607563 | 200 | 'info_dict': { |
26394d02 S |
201 | 'id': 'novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes', |
202 | 'title': 'Novidade na fiscalização de bagagem pela Receita provoca discussões', | |
203 | 'description': 'md5:c3c4b4d4c30c32fce460040b1ac46b12', | |
204 | }, | |
205 | 'playlist_count': 1, | |
206 | }, { | |
207 | 'url': 'http://g1.globo.com/pr/parana/noticia/2016/09/mpf-denuncia-lula-marisa-e-mais-seis-na-operacao-lava-jato.html', | |
208 | 'info_dict': { | |
209 | 'id': 'mpf-denuncia-lula-marisa-e-mais-seis-na-operacao-lava-jato', | |
210 | 'title': "Lula era o 'comandante máximo' do esquema da Lava Jato, diz MPF", | |
211 | 'description': 'md5:8aa7cc8beda4dc71cc8553e00b77c54c', | |
212 | }, | |
213 | 'playlist_count': 6, | |
5d501a09 S |
214 | }, { |
215 | 'url': 'http://gq.globo.com/Prazeres/Poder/noticia/2015/10/all-o-desafio-assista-ao-segundo-capitulo-da-serie.html', | |
216 | 'only_matching': True, | |
217 | }, { | |
218 | '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', | |
219 | 'only_matching': True, | |
9e5751b9 S |
220 | }, { |
221 | 'url': 'http://oglobo.globo.com/rio/a-amizade-entre-um-entregador-de-farmacia-um-piano-19946271', | |
222 | 'only_matching': True, | |
5d501a09 | 223 | }] |
ad607563 S |
224 | |
225 | @classmethod | |
226 | def suitable(cls, url): | |
227 | return False if GloboIE.suitable(url) else super(GloboArticleIE, cls).suitable(url) | |
228 | ||
229 | def _real_extract(self, url): | |
230 | display_id = self._match_id(url) | |
231 | webpage = self._download_webpage(url, display_id) | |
26394d02 S |
232 | video_ids = [] |
233 | for video_regex in self._VIDEOID_REGEXES: | |
234 | video_ids.extend(re.findall(video_regex, webpage)) | |
235 | entries = [ | |
236 | self.url_result('globo:%s' % video_id, GloboIE.ie_key()) | |
237 | for video_id in orderedSet(video_ids)] | |
238 | title = self._og_search_title(webpage, fatal=False) | |
239 | description = self._html_search_meta('description', webpage) | |
240 | return self.playlist_result(entries, display_id, title, description) |