Prepare for release.

[youtubedl] / youtube_dl / extractor / openload.py

diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py
index 6415b8fdcb7451e6b499d1d4d3515af2775c66c2..7f19b1ba5c3c355977c694334694b51f71a9840c 100644 (file)
--- a/youtube_dl/extractor/openload.py
+++ b/youtube_dl/extractor/openload.py
@@ -1,20 +1,25 @@
 # coding: utf-8
 # coding: utf-8

-from __future__ import unicode_literals
+from __future__ import unicode_literals, division

 
 import re
 
 from .common import InfoExtractor
 
 import re
 
 from .common import InfoExtractor

-from ..compat import compat_chr
+from ..compat import (
+    compat_chr,
+    compat_ord,
+)

 from ..utils import (
     determine_ext,
 from ..utils import (
     determine_ext,

-    encode_base_n,

     ExtractorError,
     ExtractorError,

-    mimetype2ext,
+)
+from ..jsinterp import (
+    JSInterpreter,
+    _NAME_RE

 )
 
 
 class OpenloadIE(InfoExtractor):
 )
 
 
 class OpenloadIE(InfoExtractor):

-    _VALID_URL = r'https://openload.(?:co|io)/(?:f|embed)/(?P<id>[a-zA-Z0-9-_]+)'
+    _VALID_URL = r'https?://openload\.(?:co|io)/(?:f|embed)/(?P<id>[a-zA-Z0-9-_]+)'

 
     _TESTS = [{
         'url': 'https://openload.co/f/kUEfGclsU9o',
 
     _TESTS = [{
         'url': 'https://openload.co/f/kUEfGclsU9o',


@@ -25,6 +30,22 @@ class OpenloadIE(InfoExtractor):
             'title': 'skyrim_no-audio_1080.mp4',
             'thumbnail': 're:^https?://.*\.jpg$',
         },
             'title': 'skyrim_no-audio_1080.mp4',
             'thumbnail': 're:^https?://.*\.jpg$',
         },

+    }, {
+        'url': 'https://openload.co/embed/rjC09fkPLYs',
+        'info_dict': {
+            'id': 'rjC09fkPLYs',
+            'ext': 'mp4',
+            'title': 'movie.mp4',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'subtitles': {
+                'en': [{
+                    'ext': 'vtt',
+                }],
+            },
+        },
+        'params': {
+            'skip_download': True,  # test subtitles only
+        },

     }, {
         'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',
         'only_matching': True,
     }, {
         'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',
         'only_matching': True,


@@ -41,90 +62,98 @@ class OpenloadIE(InfoExtractor):
         'only_matching': True,
     }]
 
         'only_matching': True,
     }]
 

-    @staticmethod
-    def openload_level2_debase(m):
-        radix, num = int(m.group(1)) + 27, int(m.group(2))
-        return '"' + encode_base_n(num, radix) + '"'
-
-    @classmethod
-    def openload_level2(cls, txt):
-        # The function name is ǃ \u01c3
-        # Using escaped unicode literals does not work in Python 3.2
-        return re.sub(r'ǃ\((\d+),(\d+)\)', cls.openload_level2_debase, txt, re.UNICODE).replace('"+"', '')
-
-    # Openload uses a variant of aadecode
-    # openload_decode and related functions are originally written by
-    # vitas@matfyz.cz and released with public domain
-    # See https://github.com/rg3/youtube-dl/issues/8489
-    @classmethod
-    def openload_decode(cls, txt):
-        symbol_table = [
-            ('_', '(ﾟДﾟ) [ﾟΘﾟ]'),
-            ('a', '(ﾟДﾟ) [ﾟωﾟﾉ]'),
-            ('b', '(ﾟДﾟ) [ﾟΘﾟﾉ]'),
-            ('c', '(ﾟДﾟ) [\'c\']'),
-            ('d', '(ﾟДﾟ) [ﾟｰﾟﾉ]'),
-            ('e', '(ﾟДﾟ) [ﾟДﾟﾉ]'),
-            ('f', '(ﾟДﾟ) [1]'),
-
-            ('o', '(ﾟДﾟ) [\'o\']'),
-            ('u', '(oﾟｰﾟo)'),
-            ('c', '(ﾟДﾟ) [\'c\']'),
-
-            ('7', '((ﾟｰﾟ) + (o^_^o))'),
-            ('6', '((o^_^o) +(o^_^o) +(c^_^o))'),
-            ('5', '((ﾟｰﾟ) + (ﾟΘﾟ))'),
-            ('4', '(-~3)'),
-            ('3', '(-~-~1)'),
-            ('2', '(-~1)'),
-            ('1', '(-~0)'),
-            ('0', '((c^_^o)-(c^_^o))'),
-        ]
+    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]+'
         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 = ''
         ret = ''

-        for aachar in txt.split(delim):
-            for val, pat in symbol_table:
-                aachar = aachar.replace(pat, val)
-            aachar = aachar.replace('+ ', '')
-            m = re.match(r'^\d+', aachar)
-            if m:
-                ret += compat_chr(int(m.group(0), 8))
-            else:
-                m = re.match(r'^u([\da-f]+)', aachar)
-                if m:
-                    ret += compat_chr(int(m.group(1), 16))
-        return cls.openload_level2(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)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage('https://openload.co/embed/%s/' % video_id, video_id)

 
 

-        if 'File not found' in webpage:
+        if 'File not found' in webpage or 'deleted by the owner' in webpage:

             raise ExtractorError('File not found', expected=True)
 
             raise ExtractorError('File not found', expected=True)
 

-        code = self._search_regex(
-            r'</video>\s*</div>\s*<script[^>]+>[^>]+</script>\s*<script[^>]+>([^<]+)</script>',
-            webpage, 'JS code')
+        # The following decryption algorithm is written by @yokrysty and
+        # declared to be freely used in youtube-dl
+        # See https://github.com/rg3/youtube-dl/issues/10408
+        enc_data = self._html_search_regex(
+            r'<span[^>]*>([^<]+)</span>\s*<span[^>]*>[^<]+</span>\s*<span[^>]+id="streamurl"',
+            webpage, 'encrypted data')
+
+        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')

 
 

-        decoded = self.openload_decode(code)
+        offset = jsi.call_function(m_offset_fun)
+        diff = jsi.call_function(m_diff_fun)

 
 

-        video_url = self._search_regex(
-            r'return\s+"(https?://[^"]+)"', decoded, 'video URL')
+        video_url_chars = []
+
+        for idx, c in enumerate(enc_data):
+            j = compat_ord(c)
+            if j >= 33 and j <= 126:
+                j = ((j + 14) % 94) + 33
+            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)

 
         title = self._og_search_title(webpage, default=None) or self._search_regex(
             r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
             'title', default=None) or self._html_search_meta(
             'description', webpage, 'title', fatal=True)
 
 
         title = self._og_search_title(webpage, default=None) or self._search_regex(
             r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
             'title', default=None) or self._html_search_meta(
             'description', webpage, 'title', fatal=True)
 

-        ext = mimetype2ext(self._search_regex(
-            r'window\.vt\s*=\s*(["\'])(?P<mimetype>.+?)\1', decoded,
-            'mimetype', default=None, group='mimetype')) or determine_ext(
-            video_url, 'mp4')
+        entries = self._parse_html5_media_entries(url, webpage, video_id)
+        subtitles = entries[0]['subtitles'] if entries else None

 
 

-        return {
+        info_dict = {

             'id': video_id,
             'title': title,
             'id': video_id,
             'title': title,

-            'ext': ext,

             'thumbnail': self._og_search_thumbnail(webpage, default=None),
             'url': video_url,
             'thumbnail': self._og_search_thumbnail(webpage, default=None),
             'url': video_url,

+            # Seems all videos have extensions in their titles
+            'ext': determine_ext(title),
+            'subtitles': subtitles,

         }
         }

+
+        return info_dict