X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/e76f531201cd41dfc0ce00be28bcc5c575c7acc5..4231113a7be907539052ce59df44e23c54d21d38:/youtube_dl/extractor/globo.py diff --git a/youtube_dl/extractor/globo.py b/youtube_dl/extractor/globo.py index c65ef6b..dc7b266 100644 --- a/youtube_dl/extractor/globo.py +++ b/youtube_dl/extractor/globo.py @@ -2,6 +2,7 @@ from __future__ import unicode_literals import random +import re import math from .common import InfoExtractor @@ -14,12 +15,13 @@ from ..utils import ( ExtractorError, float_or_none, int_or_none, + orderedSet, str_or_none, ) class GloboIE(InfoExtractor): - _VALID_URL = '(?:globo:|https?://.+?\.globo\.com/(?:[^/]+/)*(?:v/(?:[^/]+/)?|videos/))(?P\d{7,})' + _VALID_URL = r'(?:globo:|https?://.+?\.globo\.com/(?:[^/]+/)*(?:v/(?:[^/]+/)?|videos/))(?P\d{7,})' _API_URL_TEMPLATE = 'http://api.globovideos.com/videos/%s/playlist' _SECURITY_URL_TEMPLATE = 'http://security.video.globo.com/videos/%s/hash?player=flash&version=17.0.0.132&resource_id=%s' @@ -63,9 +65,12 @@ class GloboIE(InfoExtractor): }, { 'url': 'http://canaloff.globo.com/programas/desejar-profundo/videos/4518560.html', 'only_matching': True, + }, { + 'url': 'globo:3607726', + 'only_matching': True, }] - class MD5: + class MD5(object): HEX_FORMAT_LOWERCASE = 0 HEX_FORMAT_UPPERCASE = 1 BASE64_PAD_CHARACTER_DEFAULT_COMPLIANCE = '' @@ -369,11 +374,9 @@ class GloboIE(InfoExtractor): resource_url = resource['url'] signed_url = '%s?h=%s&k=%s' % (resource_url, signed_hash, 'flash') if resource_id.endswith('m3u8') or resource_url.endswith('.m3u8'): - m3u8_formats = self._extract_m3u8_formats( + formats.extend(self._extract_m3u8_formats( signed_url, resource_id, 'mp4', entry_protocol='m3u8_native', - m3u8_id='hls', fatal=False) - if m3u8_formats: - formats.extend(m3u8_formats) + m3u8_id='hls', fatal=False)) else: formats.append({ 'url': signed_url, @@ -398,33 +401,41 @@ class GloboIE(InfoExtractor): class GloboArticleIE(InfoExtractor): - _VALID_URL = 'https?://.+?\.globo\.com/(?:[^/]+/)*(?P[^/]+)\.html' + _VALID_URL = r'https?://.+?\.globo\.com/(?:[^/]+/)*(?P[^/.]+)(?:\.html)?' _VIDEOID_REGEXES = [ r'\bdata-video-id=["\'](\d{7,})', r'\bdata-player-videosids=["\'](\d{7,})', - r'\bvideosIDs\s*:\s*["\'](\d{7,})', + r'\bvideosIDs\s*:\s*["\']?(\d{7,})', r'\bdata-id=["\'](\d{7,})', r']+\bid=["\'](\d{7,})', ] _TESTS = [{ 'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html', - 'md5': '307fdeae4390ccfe6ba1aa198cf6e72b', 'info_dict': { - 'id': '3652183', - 'ext': 'mp4', - 'title': 'Receita Federal explica como vai fiscalizar bagagens de quem retorna ao Brasil de avião', - 'duration': 110.711, - 'uploader': 'Rede Globo', - 'uploader_id': '196', - } + '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, }, { '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, }] @classmethod @@ -434,5 +445,12 @@ class GloboArticleIE(InfoExtractor): def _real_extract(self, url): display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) - video_id = self._search_regex(self._VIDEOID_REGEXES, webpage, 'video id') - return self.url_result('globo:%s' % video_id, 'Globo') + video_ids = [] + for video_regex in self._VIDEOID_REGEXES: + video_ids.extend(re.findall(video_regex, webpage)) + entries = [ + self.url_result('globo:%s' % video_id, GloboIE.ie_key()) + for video_id in orderedSet(video_ids)] + title = self._og_search_title(webpage, fatal=False) + description = self._html_search_meta('description', webpage) + return self.playlist_result(entries, display_id, title, description)