X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/b8d8e13c1f9e4d3cdd7d41c5c9d711a36dd5f9c3..a7fd25993be110e616bb9bf6d602949b54df1a7d:/youtube_dl/extractor/infoq.py diff --git a/youtube_dl/extractor/infoq.py b/youtube_dl/extractor/infoq.py index 9fb71e8..391c2f5 100644 --- a/youtube_dl/extractor/infoq.py +++ b/youtube_dl/extractor/infoq.py @@ -2,13 +2,15 @@ from __future__ import unicode_literals -import base64 - from ..compat import ( + compat_b64decode, compat_urllib_parse_unquote, compat_urlparse, ) -from ..utils import determine_ext +from ..utils import ( + determine_ext, + update_url_query, +) from .bokecc import BokeCCBaseIE @@ -58,7 +60,7 @@ class InfoQIE(BokeCCBaseIE): encoded_id = self._search_regex( r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id', default=None) - real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8')) + real_id = compat_urllib_parse_unquote(compat_b64decode(encoded_id).decode('utf-8')) playpath = 'mp4:' + real_id return [{ @@ -68,45 +70,44 @@ class InfoQIE(BokeCCBaseIE): 'play_path': playpath, }] - def _extract_cookies(self, webpage): - policy = self._search_regex(r'InfoQConstants.scp\s*=\s*\'([^\']+)\'', webpage, 'policy') - signature = self._search_regex(r'InfoQConstants.scs\s*=\s*\'([^\']+)\'', webpage, 'signature') - key_pair_id = self._search_regex(r'InfoQConstants.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id') - return 'CloudFront-Policy=%s; CloudFront-Signature=%s; CloudFront-Key-Pair-Id=%s' % ( - policy, signature, key_pair_id) + def _extract_cf_auth(self, webpage): + policy = self._search_regex(r'InfoQConstants\.scp\s*=\s*\'([^\']+)\'', webpage, 'policy') + signature = self._search_regex(r'InfoQConstants\.scs\s*=\s*\'([^\']+)\'', webpage, 'signature') + key_pair_id = self._search_regex(r'InfoQConstants\.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id') + return { + 'Policy': policy, + 'Signature': signature, + 'Key-Pair-Id': key_pair_id, + } def _extract_http_video(self, webpage): http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL') + http_video_url = update_url_query(http_video_url, self._extract_cf_auth(webpage)) return [{ 'format_id': 'http_video', 'url': http_video_url, - 'http_headers': { - 'Cookie': self._extract_cookies(webpage) - }, }] def _extract_http_audio(self, webpage, video_id): fields = self._hidden_inputs(webpage) - http_audio_url = fields['filename'] - if http_audio_url is None: + http_audio_url = fields.get('filename') + if not http_audio_url: return [] - cookies_header = {'Cookie': self._extract_cookies(webpage)} - # base URL is found in the Location header in the response returned by # GET https://www.infoq.com/mp3download.action?filename=... when logged in. http_audio_url = compat_urlparse.urljoin('http://res.infoq.com/downloads/mp3downloads/', http_audio_url) + http_audio_url = update_url_query(http_audio_url, self._extract_cf_auth(webpage)) # audio file seem to be missing some times even if there is a download link # so probe URL to make sure - if not self._is_valid_url(http_audio_url, video_id, headers=cookies_header): + if not self._is_valid_url(http_audio_url, video_id): return [] return [{ 'format_id': 'http_audio', 'url': http_audio_url, 'vcodec': 'none', - 'http_headers': cookies_header, }] def _real_extract(self, url):