X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/a070911bf98c751e4425008f1335cc57c86d30fc..bad87b6c6e31ea0056d795ca3e7aa81ceaad2291:/youtube_dl/extractor/nosvideo.py diff --git a/youtube_dl/extractor/nosvideo.py b/youtube_dl/extractor/nosvideo.py index f5ef856..53c500c 100644 --- a/youtube_dl/extractor/nosvideo.py +++ b/youtube_dl/extractor/nosvideo.py @@ -4,11 +4,9 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_request, -) from ..utils import ( ExtractorError, + sanitized_Request, urlencode_postdata, xpath_text, xpath_with_ns, @@ -19,7 +17,7 @@ _x = lambda p: xpath_with_ns(p, {'xspf': 'http://xspf.org/ns/0/'}) class NosVideoIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?nosvideo\.com/' + \ - '(?:embed/|\?v=)(?P[A-Za-z0-9]{12})/?' + r'(?:embed/|\?v=)(?P[A-Za-z0-9]{12})/?' _PLAYLIST_URL = 'http://nosvideo.com/xml/{xml_id:s}.xml' _FILE_DELETED_REGEX = r'File Not Found' _TEST = { @@ -29,7 +27,7 @@ class NosVideoIE(InfoExtractor): 'id': 'mu8fle7g7rpq', 'ext': 'mp4', 'title': 'big_buck_bunny_480p_surround-fix.avi.mp4', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', } } @@ -41,7 +39,7 @@ class NosVideoIE(InfoExtractor): 'op': 'download1', 'method_free': 'Continue to Video', } - req = compat_urllib_request.Request(url, urlencode_postdata(fields)) + req = sanitized_Request(url, urlencode_postdata(fields)) req.add_header('Content-type', 'application/x-www-form-urlencoded') webpage = self._download_webpage(req, video_id, 'Downloading download page')