mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-10-31 06:35:12 +00:00
Code Issues Projects Releases Wiki Activity

[sohu] Handle encoding, and fix tests

Browse Source
This commit is contained in:
Philipp Hagemeister
2013-08-28 13:59:08 +02:00
parent f8b362739e
commit f143d86ad2
2 changed files with 72 additions and 70 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
youtube_dl/extractor/common.py
View File

@@ -145,12 +145,17 @@ class InfoExtractor(object):
urlh = self._request_webpage(url_or_request, video_id, note, errnote)
content_type = urlh.headers.get('Content-Type', '')
webpage_bytes = urlh.read()
m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
if m:
encoding = m.group(1)
else:
encoding = 'utf-8'
webpage_bytes = urlh.read()
m = re.search(br'<meta[^>]+charset="?([^"]+)[ /">]',
webpage_bytes[:1024])
if m:
encoding = m.group(1).decode('ascii')
else:
encoding = 'utf-8'
if self._downloader.params.get('dump_intermediate_pages', False):
try:
url = url_or_request.get_full_url()