[streamango] Fix extraction (closes #14160)
authorgfabiano <gfabiano40@gmail.com>
Mon, 18 Sep 2017 21:59:14 +0000 (23:59 +0200)
committerSergey M․ <dstftw@gmail.com>
Sun, 25 Feb 2018 11:26:48 +0000 (18:26 +0700)
youtube_dl/extractor/streamango.py

index a9e34c027504ff0f6585fa51cae4ba3a1ec5bfcd..58c03f11acba6c72d18c3e8ce09c992f87325d25 100644 (file)
@@ -38,6 +38,27 @@ class StreamangoIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
+        def decrypt_src(str_, val):
+            k = '=/+9876543210zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA'
+            str_ = re.sub(r'[^A-Za-z0-9+/=]', '', str_)
+            src = ''
+            sm = [None] * 4
+            i = 0
+            str_len = len(str_)
+            while i < str_len:
+                for j in range(4):
+                    sm[j % 4] = k.index(str_[i])
+                    i += 1
+                charCode = ((sm[0] << 0x2) | (sm[1] >> 0x4)) ^ val
+                src += chr(charCode)
+                if (sm[2] != 0x40):
+                    charCode = ((sm[1] & 0xf) << 0x4) | (sm[2] >> 0x2)
+                    src += chr(charCode)
+                if (sm[3] != 0x40):
+                    charCode = ((sm[2] & 0x3) << 0x6) | sm[3]
+                    src += chr(charCode)
+            return src
+
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
@@ -45,14 +66,22 @@ class StreamangoIE(InfoExtractor):
         title = self._og_search_title(webpage, default=video_id)
 
         formats = []
-        for format_ in re.findall(r'({[^}]*\bsrc\s*:\s*[^}]*})', webpage):
+        for format_ in re.findall(r'\(\s*({[^}]*\bsrc\s*:\s*[^}]*})', webpage):
+            mobj = re.search(r'(src\s*:\s*[^(]\(([^)]*)\)[\s,]*)', format_)
+            if mobj is None:
+                continue
+            format_ = format_.replace(mobj.group(0), '')
+
             video = self._parse_json(
                 format_, video_id, transform_source=js_to_json, fatal=False)
             if not video:
                 continue
-            src = video.get('src')
-            if not src:
+
+            mobj = re.search(r'[\'"](?P<src>[^\'"]+)[\'"]\s*,\s*(?P<val>\d+)', mobj.group(1))
+            if mobj is None:
                 continue
+
+            src = decrypt_src(mobj.group('src'), int_or_none(mobj.group('val')))
             ext = determine_ext(src, default_ext=None)
             if video.get('type') == 'application/dash+xml' or ext == 'mpd':
                 formats.extend(self._extract_mpd_formats(