administration/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-10 11:20:41 +00:00
Code Issues Projects Releases Wiki Activity

[extractor/generic] Force Accept-Encoding to any for extraction pass

Browse Source
This commit is contained in:
Sergey M․ 2015-05-31 00:44:54 +06:00
parent 339516072b
commit 58bde34a23
1 changed files with 15 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
youtube_dl/extractor/generic.py
View File

@ -11,6 +11,7 @@ from ..compat import (
compat_urllib_parse,
compat_urlparse,
compat_xml_parse_error,
compat_urllib_request,
)
from ..utils import (
determine_ext,
@ -916,7 +917,9 @@ class GenericIE(InfoExtractor):
full_response = None
if head_response is False:
full_response = self._request_webpage(url, video_id)
request = compat_urllib_request.Request(url)
request.add_header('Accept-Encoding', '*')
full_response = self._request_webpage(request, video_id)
head_response = full_response
# Check for direct link to a video
@ -941,7 +944,17 @@ class GenericIE(InfoExtractor):
self._downloader.report_warning('Falling back on generic information extractor.')
if not full_response:
full_response = self._request_webpage(url, video_id)
request = compat_urllib_request.Request(url)
# Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
# making it impossible to download only chunk of the file (yet we need only 512kB to
# test whether it's HTML or not). According to youtube-dl default Accept-Encoding
# that will always result in downloading the whole file that is not desirable.
# Therefore for extraction pass we have to override Accept-Encoding to any in order
# to accept raw bytes and being able to download only a chunk.
# It may probably better to solve this by checking Content-Type for application/octet-stream
# after HEAD request finishes, but not sure if we can rely on this.
request.add_header('Accept-Encoding', '*')
full_response = self._request_webpage(request, video_id)
# Maybe it's a direct link to a video?
# Be careful not to download the whole thing!