commit 6aeb64b67358176f6c29965d0d138be75c2ad972
parent 6cd64b6806e92b7246aebd89448189180d88db82
Author: Yen Chi Hsuan <yan12125@gmail.com>
Date: Sun, 19 Jun 2016 13:25:08 +0800
Merge pull request #8201 from remitamine/hls-aes
[downloader/hls] Add support for AES-128 encrypted segments in hlsnative downloader
Diffstat:
3 files changed, 68 insertions(+), 36 deletions(-)
diff --git a/youtube_dl/downloader/hls.py b/youtube_dl/downloader/hls.py
@@ -2,14 +2,24 @@ from __future__ import unicode_literals
import os.path
import re
+import binascii
+try:
+ from Crypto.Cipher import AES
+ can_decrypt_frag = True
+except ImportError:
+ can_decrypt_frag = False
from .fragment import FragmentFD
from .external import FFmpegFD
-from ..compat import compat_urlparse
+from ..compat import (
+ compat_urlparse,
+ compat_struct_pack,
+)
from ..utils import (
encodeFilename,
sanitize_open,
+ parse_m3u8_attributes,
)
@@ -21,7 +31,7 @@ class HlsFD(FragmentFD):
@staticmethod
def can_download(manifest):
UNSUPPORTED_FEATURES = (
- r'#EXT-X-KEY:METHOD=(?!NONE)', # encrypted streams [1]
+ r'#EXT-X-KEY:METHOD=(?!NONE|AES-128)', # encrypted streams [1]
r'#EXT-X-BYTERANGE', # playlists composed of byte ranges of media files [2]
# Live streams heuristic does not always work (e.g. geo restricted to Germany
@@ -39,7 +49,9 @@ class HlsFD(FragmentFD):
# 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.2
# 4. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.5
)
- return all(not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES)
+ check_results = [not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES]
+ check_results.append(not (re.search(r'#EXT-X-KEY:METHOD=AES-128', manifest) and not can_decrypt_frag))
+ return all(check_results)
def real_download(self, filename, info_dict):
man_url = info_dict['url']
@@ -57,36 +69,58 @@ class HlsFD(FragmentFD):
fd.add_progress_hook(ph)
return fd.real_download(filename, info_dict)
- fragment_urls = []
+ total_frags = 0
for line in s.splitlines():
line = line.strip()
if line and not line.startswith('#'):
- segment_url = (
- line
- if re.match(r'^https?://', line)
- else compat_urlparse.urljoin(man_url, line))
- fragment_urls.append(segment_url)
- # We only download the first fragment during the test
- if self.params.get('test', False):
- break
+ total_frags += 1
ctx = {
'filename': filename,
- 'total_frags': len(fragment_urls),
+ 'total_frags': total_frags,
}
self._prepare_and_start_frag_download(ctx)
+ i = 0
+ media_sequence = 0
+ decrypt_info = {'METHOD': 'NONE'}
frags_filenames = []
- for i, frag_url in enumerate(fragment_urls):
- frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i)
- success = ctx['dl'].download(frag_filename, {'url': frag_url})
- if not success:
- return False
- down, frag_sanitized = sanitize_open(frag_filename, 'rb')
- ctx['dest_stream'].write(down.read())
- down.close()
- frags_filenames.append(frag_sanitized)
+ for line in s.splitlines():
+ line = line.strip()
+ if line:
+ if not line.startswith('#'):
+ frag_url = (
+ line
+ if re.match(r'^https?://', line)
+ else compat_urlparse.urljoin(man_url, line))
+ frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i)
+ success = ctx['dl'].download(frag_filename, {'url': frag_url})
+ if not success:
+ return False
+ down, frag_sanitized = sanitize_open(frag_filename, 'rb')
+ frag_content = down.read()
+ down.close()
+ if decrypt_info['METHOD'] == 'AES-128':
+ iv = decrypt_info.get('IV') or compat_struct_pack(">8xq", media_sequence)
+ frag_content = AES.new(decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
+ ctx['dest_stream'].write(frag_content)
+ frags_filenames.append(frag_sanitized)
+ # We only download the first fragment during the test
+ if self.params.get('test', False):
+ break
+ i += 1
+ media_sequence += 1
+ elif line.startswith('#EXT-X-KEY'):
+ decrypt_info = parse_m3u8_attributes(line[11:])
+ if decrypt_info['METHOD'] == 'AES-128':
+ if 'IV' in decrypt_info:
+ decrypt_info['IV'] = binascii.unhexlify(decrypt_info['IV'][2:])
+ if not re.match(r'^https?://', decrypt_info['URI']):
+ decrypt_info['URI'] = compat_urlparse.urljoin(man_url, decrypt_info['URI'])
+ decrypt_info['KEY'] = self.ydl.urlopen(decrypt_info['URI']).read()
+ elif line.startswith('#EXT-X-MEDIA-SEQUENCE'):
+ media_sequence = int(line[22:])
self._finish_frag_download(ctx)
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
@@ -53,6 +53,7 @@ from ..utils import (
mimetype2ext,
update_Request,
update_url_query,
+ parse_m3u8_attributes,
)
@@ -1150,23 +1151,11 @@ class InfoExtractor(object):
}]
last_info = None
last_media = None
- kv_rex = re.compile(
- r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
for line in m3u8_doc.splitlines():
if line.startswith('#EXT-X-STREAM-INF:'):
- last_info = {}
- for m in kv_rex.finditer(line):
- v = m.group('val')
- if v.startswith('"'):
- v = v[1:-1]
- last_info[m.group('key')] = v
+ last_info = parse_m3u8_attributes(line)
elif line.startswith('#EXT-X-MEDIA:'):
- last_media = {}
- for m in kv_rex.finditer(line):
- v = m.group('val')
- if v.startswith('"'):
- v = v[1:-1]
- last_media[m.group('key')] = v
+ last_media = parse_m3u8_attributes(line)
elif line.startswith('#') or not line.strip():
continue
else:
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
@@ -2852,3 +2852,12 @@ def decode_packed_codes(code):
return re.sub(
r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)],
obfucasted_code)
+
+
+def parse_m3u8_attributes(attrib):
+ info = {}
+ for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib):
+ if val.startswith('"'):
+ val = val[1:-1]
+ info[key] = val
+ return info
