Merge branch 'master' of https://github.com/DarkstaIkers/youtube-dl into DarkstaIkers...

[youtube-dl] / youtube_dl / extractor / openload.py

diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py
index c8dde7ae325c01363dce7b8c92a74506ba57558e..7f19b1ba5c3c355977c694334694b51f71a9840c 100644 (file)
--- 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,
@@ -10,10 +12,14 @@ from ..utils import (
     determine_ext,
     ExtractorError,
 )
+from ..jsinterp import (
+    JSInterpreter,
+    _NAME_RE
+)
 
 
 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',
@@ -24,6 +30,22 @@ class OpenloadIE(InfoExtractor):
             '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,
@@ -40,6 +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)
@@ -51,7 +111,20 @@ class OpenloadIE(InfoExtractor):
         # 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[^>]+id="hiddenurl"[^>]*>([^<]+)</span>', webpage, 'encrypted data')
+            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')
+
+        offset = jsi.call_function(m_offset_fun)
+        diff = jsi.call_function(m_diff_fun)
 
         video_url_chars = []
 
@@ -59,8 +132,8 @@ class OpenloadIE(InfoExtractor):
             j = compat_ord(c)
             if j >= 33 and j <= 126:
                 j = ((j + 14) % 94) + 33
-            if idx == len(enc_data) - 1:
-                j += 2
+            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)
@@ -70,11 +143,17 @@ class OpenloadIE(InfoExtractor):
             'title', default=None) or self._html_search_meta(
             'description', webpage, 'title', fatal=True)
 
-        return {
+        entries = self._parse_html5_media_entries(url, webpage, video_id)
+        subtitles = entries[0]['subtitles'] if entries else None
+
+        info_dict = {
             'id': video_id,
             'title': title,
             '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