diff options
Diffstat (limited to 'youtube_dl/extractor/infoq.py')
| -rw-r--r-- | youtube_dl/extractor/infoq.py | 75 |
1 files changed, 60 insertions, 15 deletions
diff --git a/youtube_dl/extractor/infoq.py b/youtube_dl/extractor/infoq.py index cca0b8a93..c3e892feb 100644 --- a/youtube_dl/extractor/infoq.py +++ b/youtube_dl/extractor/infoq.py @@ -4,8 +4,14 @@ from __future__ import unicode_literals import base64 -from ..compat import compat_urllib_parse_unquote -from ..utils import determine_ext +from ..compat import ( + compat_urllib_parse_unquote, + compat_urlparse, +) +from ..utils import ( + determine_ext, + update_url_query, +) from .bokecc import BokeCCBaseIE @@ -33,9 +39,21 @@ class InfoQIE(BokeCCBaseIE): 'ext': 'flv', 'description': 'md5:308d981fb28fa42f49f9568322c683ff', }, + }, { + 'url': 'https://www.infoq.com/presentations/Simple-Made-Easy', + 'md5': '0e34642d4d9ef44bf86f66f6399672db', + 'info_dict': { + 'id': 'Simple-Made-Easy', + 'title': 'Simple Made Easy', + 'ext': 'mp3', + 'description': 'md5:3e0e213a8bbd074796ef89ea35ada25b', + }, + 'params': { + 'format': 'bestaudio', + }, }] - def _extract_rtmp_videos(self, webpage): + def _extract_rtmp_video(self, webpage): # The server URL is hardcoded video_url = 'rtmpe://video.infoq.com/cfx/st/' @@ -47,26 +65,50 @@ class InfoQIE(BokeCCBaseIE): playpath = 'mp4:' + real_id return [{ - 'format_id': 'rtmp', + 'format_id': 'rtmp_video', 'url': video_url, 'ext': determine_ext(playpath), 'play_path': playpath, }] - def _extract_http_videos(self, webpage): + 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, + }] - 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') + def _extract_http_audio(self, webpage, video_id): + fields = self._hidden_inputs(webpage) + http_audio_url = fields.get('filename') + if not http_audio_url: + return [] + + # 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): + return [] return [{ - 'format_id': 'http', - 'url': http_video_url, - 'http_headers': { - 'Cookie': 'CloudFront-Policy=%s; CloudFront-Signature=%s; CloudFront-Key-Pair-Id=%s' % ( - policy, signature, key_pair_id), - }, + 'format_id': 'http_audio', + 'url': http_audio_url, + 'vcodec': 'none', }] def _real_extract(self, url): @@ -80,7 +122,10 @@ class InfoQIE(BokeCCBaseIE): # for China videos, HTTP video URL exists but always fails with 403 formats = self._extract_bokecc_formats(webpage, video_id) else: - formats = self._extract_rtmp_videos(webpage) + self._extract_http_videos(webpage) + formats = ( + self._extract_rtmp_video(webpage) + + self._extract_http_video(webpage) + + self._extract_http_audio(webpage, video_id)) self._sort_formats(formats) |