Madeorsk
/
youtube-dl
mirror of https://codeberg.org/polarisfm/youtube-dl
1
0
Fork 0
Code Issues Releases Wiki Activity

Merge 0d4e58840b into 416da574ec

This commit is contained in:
TheZ3ro 2020-10-23 11:20:01 -04:00 committed by GitHub
parent 416da574ec 0d4e58840b
commit 776e31f7aa
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
5 changed files with 126 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
youtube_dl/downloader/hls.py
View File

@ -64,7 +64,7 @@ class HlsFD(FragmentFD):
s = urlh.read().decode('utf-8', 'ignore')
if not self.can_download(s, info_dict):
if info_dict.get('extra_param_to_segment_url') or info_dict.get('_decryption_key_url'):
if info_dict.get('extra_param_to_segment_url') or info_dict.get('extra_param_to_key_url'):
self.report_error('pycrypto not found. Please install it.')
return False
self.report_warning(
@ -113,10 +113,19 @@ class HlsFD(FragmentFD):
skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
test = self.params.get('test', False)
extra_query = None
extra_segment_query = None
extra_key_query = None
extra_key_url = None
extra_param_to_segment_url = info_dict.get('extra_param_to_segment_url')
extra_param_to_key_url = info_dict.get('extra_param_to_key_url')
if extra_param_to_segment_url:
extra_query = compat_urlparse.parse_qs(extra_param_to_segment_url)
extra_segment_query = compat_urlparse.parse_qs(extra_param_to_segment_url)
extra_key_query = compat_urlparse.parse_qs(extra_param_to_segment_url)
if extra_param_to_key_url:
if extra_param_to_key_url.startswith('http'):
extra_key_url = extra_param_to_key_url
else:
extra_key_query = compat_urlparse.parse_qs(extra_param_to_key_url)
i = 0
media_sequence = 0
decrypt_info = {'METHOD': 'NONE'}
@ -136,8 +145,8 @@ class HlsFD(FragmentFD):
line
if re.match(r'^https?://', line)
else compat_urlparse.urljoin(man_url, line))
if extra_query:
frag_url = update_url_query(frag_url, extra_query)
if extra_segment_query:
frag_url = update_url_query(frag_url, extra_segment_query)
count = 0
headers = info_dict.get('http_headers', {})
if byte_range:
@ -169,9 +178,13 @@ class HlsFD(FragmentFD):
if decrypt_info['METHOD'] == 'AES-128':
iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', media_sequence)
decrypt_info['KEY'] = decrypt_info.get('KEY') or self.ydl.urlopen(
self._prepare_url(info_dict, info_dict.get('_decryption_key_url') or decrypt_info['URI'])).read()
frag_content = AES.new(
decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
self._prepare_url(info_dict, decrypt_info['URI'])).read()
# Since "self._TEST_FILE_SIZE" is set to 10241 bytes, only those will be downloaded for the first fragment
# In case a fragment is bigger then 10241 bytes, the fragment will be cropped so AES-CBC decryption will fail.
# For this reason we can't decrypt fragments during the tests.
if not test:
frag_content = AES.new(
decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
self._append_fragment(ctx, frag_content)
# We only download the first fragment during the test
if test:
@ -187,8 +200,10 @@ class HlsFD(FragmentFD):
if not re.match(r'^https?://', decrypt_info['URI']):
decrypt_info['URI'] = compat_urlparse.urljoin(
man_url, decrypt_info['URI'])
if extra_query:
decrypt_info['URI'] = update_url_query(decrypt_info['URI'], extra_query)
if extra_key_query:
decrypt_info['URI'] = update_url_query(decrypt_info['URI'], extra_key_query)
elif extra_key_url:
decrypt_info['URI'] = extra_key_url
if decrypt_url != decrypt_info['URI']:
decrypt_info['KEY'] = None
elif line.startswith('#EXT-X-MEDIA-SEQUENCE'):

1
youtube_dl/extractor/extractors.py
View File

@ -1065,6 +1065,7 @@ from .sportbox import SportBoxIE
from .sportdeutschland import SportDeutschlandIE
from .springboardplatform import SpringboardPlatformIE
from .sprout import SproutIE
from .sproutvideo import SproutVideoIE
from .srgssr import (
SRGSSRIE,
SRGSSRPlayIE,

18
youtube_dl/extractor/generic.py
View File

@ -119,6 +119,7 @@ from .expressen import ExpressenIE
from .zype import ZypeIE
from .odnoklassniki import OdnoklassnikiIE
from .kinja import KinjaEmbedIE
from .sproutvideo import SproutVideoIE
class GenericIE(InfoExtractor):
@ -2151,6 +2152,18 @@ class GenericIE(InfoExtractor):
'skip_download': True,
},
},
{
# SproutVideo iframe in page
'url': 'https://www.solidarum.org/vivre-ensemble/adrien-labaeye-berlin-des-communautes-aux-communs',
'info_dict': {
'id': '4c9dddb01910e3c9c4',
'ext': 'mp4',
'title': 'Adrien Labaeye : Berlin, des communautés aux communs',
},
'params': {
'skip_download': True,
},
},
# {
# # TODO: find another test
# # http://schema.org/VideoObject
@ -3213,6 +3226,11 @@ class GenericIE(InfoExtractor):
return self.playlist_from_matches(
zype_urls, video_id, video_title, ie=ZypeIE.ie_key())
sproutvideo_urls = SproutVideoIE._extract_urls(webpage)
if sproutvideo_urls:
return self.playlist_from_matches(
sproutvideo_urls, video_id, video_title, ie=SproutVideoIE.ie_key())
# Look for HTML5 media
entries = self._parse_html5_media_entries(url, webpage, video_id, m3u8_id='hls')
if entries:

81
youtube_dl/extractor/sproutvideo.py Normal file
View File

@ -0,0 +1,81 @@
# coding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import (
compat_b64decode,
compat_urllib_parse_urlencode,
)
class SproutVideoIE(InfoExtractor):
_NOSCHEMA_URL = r'//videos\.sproutvideo\.com/embed/(?P<id>[a-f0-9]+)/[a-f0-9]+'
_VALID_URL = r'https?:%s' % _NOSCHEMA_URL
_TEST = {
'url': 'https://videos.sproutvideo.com/embed/4c9dddb01910e3c9c4/0fc24387c4f24ee3',
'md5': 'fbc675bb97437e797d11d14d99563f50',
'info_dict': {
'id': '4c9dddb01910e3c9c4',
'ext': 'mp4',
'title': 'Adrien Labaeye : Berlin, des communautés aux communs',
}
}
@staticmethod
def _extract_urls(webpage):
# Fix the video URL if the iframe doesn't have a defined schema
return [sprout.group('url') for sprout in re.finditer(
r'<iframe[^>]+src=[\'"](?P<url>(?:https?:|)%s[^\'"]+)[\'"]' % SproutVideoIE._NOSCHEMA_URL,
webpage)]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
data = self._search_regex(r"var\s+dat\s+=\s+'([^']+)';", webpage, 'data')
data_decoded = compat_b64decode(data).decode('utf-8')
parsed_data = self._parse_json(data_decoded, video_id)
# https://github.com/ytdl-org/youtube-dl/issues/16996#issuecomment-406901324
# signature->m for manifests
# signature->k for keys
# signature->t for segments
m_sign = SproutVideoIE._policy_to_qs(parsed_data, 'm')
k_sign = SproutVideoIE._policy_to_qs(parsed_data, 'k')
t_sign = SproutVideoIE._policy_to_qs(parsed_data, 't')
resource_url = 'https://{0}.videos.sproutvideo.com/{1}/{2}/video/index.m3u8?{3}'.format(
parsed_data['base'], parsed_data['s3_user_hash'], parsed_data['s3_video_hash'], m_sign)
formats = self._extract_m3u8_formats(
resource_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
self._sort_formats(formats)
for entry in formats:
entry.update({
'url': '{0}?{1}'.format(entry['url'], m_sign),
'extra_param_to_segment_url': t_sign,
'extra_param_to_key_url': k_sign,
})
return {
'id': video_id,
'title': parsed_data['title'],
'formats': formats,
}
@staticmethod
def _format_qsdata(qs_data):
parsed_dict = dict()
for key in qs_data:
parsed_dict[key.replace('CloudFront-', '')] = qs_data[key]
return parsed_dict
@staticmethod
def _policy_to_qs(policy, key):
sign = SproutVideoIE._format_qsdata(policy['signatures'][key])
sign['sessionID'] = policy['sessionID']
return compat_urllib_parse_urlencode(sign, doseq=True)

2
youtube_dl/extractor/vzaar.py
View File

@ -97,7 +97,7 @@ class VzaarIE(InfoExtractor):
m3u8_id='hls', fatal=False)
if hls_aes:
for f in m3u8_formats:
f['_decryption_key_url'] = url_templ % ('goose', '') + qs
f['extra_param_to_key_url'] = url_templ % ('goose', '') + qs
formats.extend(m3u8_formats)
self._sort_formats(formats)