1
0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-23 17:48:00 +01:00

[YoutubeDL] Force redirect URL to unicode on python 2

This commit is contained in:
Sergey M․ 2020-02-29 19:08:44 +07:00
parent e2f8bf5888
commit fca6dba8b8
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
2 changed files with 12 additions and 1 deletions

View File

@ -92,6 +92,7 @@ from .utils import (
YoutubeDLCookieJar,
YoutubeDLCookieProcessor,
YoutubeDLHandler,
YoutubeDLRedirectHandler,
)
from .cache import Cache
from .extractor import get_info_extractor, gen_extractor_classes, _LAZY_LOADER
@ -2343,6 +2344,7 @@ class YoutubeDL(object):
debuglevel = 1 if self.params.get('debug_printtraffic') else 0
https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
redirect_handler = YoutubeDLRedirectHandler()
data_handler = compat_urllib_request_DataHandler()
# When passing our own FileHandler instance, build_opener won't add the
@ -2356,7 +2358,7 @@ class YoutubeDL(object):
file_handler.file_open = file_open
opener = compat_urllib_request.build_opener(
proxy_handler, https_handler, cookie_processor, ydlh, data_handler, file_handler)
proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
# Delete the default user-agent header, which would otherwise apply in
# cases where our custom HTTP handler doesn't come into play

View File

@ -2795,6 +2795,15 @@ class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
https_response = http_response
class YoutubeDLRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
if sys.version_info[0] < 3:
def redirect_request(self, req, fp, code, msg, headers, newurl):
# On python 2 urlh.geturl() may sometimes return redirect URL
# as byte string instead of unicode. This workaround allows
# to force it always return unicode.
return compat_urllib_request.HTTPRedirectHandler.redirect_request(self, req, fp, code, msg, headers, compat_str(newurl))
def extract_timezone(date_str):
m = re.search(
r'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',