aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rw-r--r--ChangeLog6
-rw-r--r--test/test_jsinterp.py7
-rw-r--r--youtube_dl/extractor/openload.py65
-rw-r--r--youtube_dl/jsinterp.py4
4 files changed, 73 insertions, 9 deletions
diff --git a/ChangeLog b/ChangeLog
index 577709c44..9e9f5cfc9 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,9 @@
+version <unreleased>
+
+Extractors
+* [openload] Fix extraction (#10408, #11122)
+
+
version 2016.11.14.1
Core
diff --git a/test/test_jsinterp.py b/test/test_jsinterp.py
index 63c350b8f..a9abae5f5 100644
--- a/test/test_jsinterp.py
+++ b/test/test_jsinterp.py
@@ -104,6 +104,13 @@ class TestJSInterpreter(unittest.TestCase):
}''')
self.assertEqual(jsi.call_function('x'), [20, 20, 30, 40, 50])
+ def test_call(self):
+ jsi = JSInterpreter('''
+ function x() { return 2; }
+ function y(a) { return x() + a; }
+ function z() { return y(3); }
+ ''')
+ self.assertEqual(jsi.call_function('z'), 5)
if __name__ == '__main__':
unittest.main()
diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py
index d3d4101de..7f19b1ba5 100644
--- a/youtube_dl/extractor/openload.py
+++ b/youtube_dl/extractor/openload.py
@@ -1,6 +1,8 @@
# coding: utf-8
from __future__ import unicode_literals, division
+import re
+
from .common import InfoExtractor
from ..compat import (
compat_chr,
@@ -10,6 +12,10 @@ from ..utils import (
determine_ext,
ExtractorError,
)
+from ..jsinterp import (
+ JSInterpreter,
+ _NAME_RE
+)
class OpenloadIE(InfoExtractor):
@@ -56,6 +62,44 @@ class OpenloadIE(InfoExtractor):
'only_matching': True,
}]
+ def openload_decode(self, txt):
+ symbol_dict = {
+ '(゚Д゚) [゚Θ゚]': '_',
+ '(゚Д゚) [゚ω゚ノ]': 'a',
+ '(゚Д゚) [゚Θ゚ノ]': 'b',
+ '(゚Д゚) [\'c\']': 'c',
+ '(゚Д゚) [゚ー゚ノ]': 'd',
+ '(゚Д゚) [゚Д゚ノ]': 'e',
+ '(゚Д゚) [1]': 'f',
+ '(゚Д゚) [\'o\']': 'o',
+ '(o゚ー゚o)': 'u',
+ '(゚Д゚) [\'c\']': 'c',
+ '((゚ー゚) + (o^_^o))': '7',
+ '((o^_^o) +(o^_^o) +(c^_^o))': '6',
+ '((゚ー゚) + (゚Θ゚))': '5',
+ '(-~3)': '4',
+ '(-~-~1)': '3',
+ '(-~1)': '2',
+ '(-~0)': '1',
+ '((c^_^o)-(c^_^o))': '0',
+ }
+ delim = '(゚Д゚)[゚ε゚]+'
+ end_token = '(゚Д゚)[゚o゚]'
+ symbols = '|'.join(map(re.escape, symbol_dict.keys()))
+ txt = re.sub('(%s)\+\s?' % symbols, lambda m: symbol_dict[m.group(1)], txt)
+ ret = ''
+ for aacode in re.findall(r'{0}\+\s?{1}(.*?){0}'.format(re.escape(end_token), re.escape(delim)), txt):
+ for aachar in aacode.split(delim):
+ if aachar.isdigit():
+ ret += compat_chr(int(aachar, 8))
+ else:
+ m = re.match(r'^u([\da-f]{4})$', aachar)
+ if m:
+ ret += compat_chr(int(m.group(1), 16))
+ else:
+ self.report_warning("Cannot decode: %s" % aachar)
+ return ret
+
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage('https://openload.co/embed/%s/' % video_id, video_id)
@@ -70,19 +114,26 @@ class OpenloadIE(InfoExtractor):
r'<span[^>]*>([^<]+)</span>\s*<span[^>]*>[^<]+</span>\s*<span[^>]+id="streamurl"',
webpage, 'encrypted data')
- magic = compat_ord(enc_data[-1])
+ enc_code = self._html_search_regex(r'<script[^>]+>(゚ω゚[^<]+)</script>',
+ webpage, 'encrypted code')
+
+ js_code = self.openload_decode(enc_code)
+ jsi = JSInterpreter(js_code)
+
+ m_offset_fun = self._search_regex(r'slice\(0\s*-\s*(%s)\(\)' % _NAME_RE, js_code, 'javascript offset function')
+ m_diff_fun = self._search_regex(r'charCodeAt\(0\)\s*\+\s*(%s)\(\)' % _NAME_RE, js_code, 'javascript diff function')
+
+ offset = jsi.call_function(m_offset_fun)
+ diff = jsi.call_function(m_diff_fun)
+
video_url_chars = []
for idx, c in enumerate(enc_data):
j = compat_ord(c)
- if j == magic:
- j -= 1
- elif j == magic - 1:
- j += 1
if j >= 33 and j <= 126:
j = ((j + 14) % 94) + 33
- if idx == len(enc_data) - 1:
- j += 3
+ if idx == len(enc_data) - offset:
+ j += diff
video_url_chars += compat_chr(j)
video_url = 'https://openload.co/stream/%s?mime=true' % ''.join(video_url_chars)
diff --git a/youtube_dl/jsinterp.py b/youtube_dl/jsinterp.py
index 9737f7002..a8df4aef0 100644
--- a/youtube_dl/jsinterp.py
+++ b/youtube_dl/jsinterp.py
@@ -198,12 +198,12 @@ class JSInterpreter(object):
return opfunc(x, y)
m = re.match(
- r'^(?P<func>%s)\((?P<args>[a-zA-Z0-9_$,]+)\)$' % _NAME_RE, expr)
+ r'^(?P<func>%s)\((?P<args>[a-zA-Z0-9_$,]*)\)$' % _NAME_RE, expr)
if m:
fname = m.group('func')
argvals = tuple([
int(v) if v.isdigit() else local_vars[v]
- for v in m.group('args').split(',')])
+ for v in m.group('args').split(',')]) if len(m.group('args')) > 0 else tuple()
if fname not in self._functions:
self._functions[fname] = self.extract_function(fname)
return self._functions[fname](argvals)