[openload] separate PhantomJS code from extractor

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

diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py
index 32289d8976dcf602839546d179c06ef224a79b20..ac5e0bb08afd84602720d9525a27e63aac84e4c4 100644 (file)
--- a/youtube_dl/extractor/openload.py
+++ b/youtube_dl/extractor/openload.py
@@ -4,10 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_chr
 from ..utils import (
     determine_ext,
     ExtractorError,
+    get_element_by_id,
+    PhantomJSwrapper,
 )
 
 
@@ -58,6 +59,8 @@ class OpenloadIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    _USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'
+
     @staticmethod
     def _extract_urls(webpage):
         return re.findall(
@@ -66,26 +69,20 @@ class OpenloadIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage('https://openload.co/embed/%s/' % video_id, video_id)
-
-        if 'File not found' in webpage or 'deleted by the owner' in webpage:
-            raise ExtractorError('File not found', expected=True)
+        url = 'https://openload.co/embed/%s/' % video_id
+        headers = {
+            'User-Agent': self._USER_AGENT,
+        }
 
-        ol_id = self._search_regex(
-            '<span[^>]+id="[^"]+"[^>]*>([0-9]+)</span>',
-            webpage, 'openload ID')
+        phantom = PhantomJSwrapper(self)
+        webpage, _ = phantom.get(url, video_id=video_id, headers=headers)
 
-        first_three_chars = int(float(ol_id[0:][:3]))
-        fifth_char = int(float(ol_id[3:5]))
-        urlcode = ''
-        num = 5
+        if 'File not found' in webpage or 'deleted by the owner' in webpage:
+            raise ExtractorError('File not found', expected=True, video_id=video_id)
 
-        while num < len(ol_id):
-            urlcode += compat_chr(int(float(ol_id[num:][:3])) +
-                                  first_three_chars - fifth_char * int(float(ol_id[num + 3:][:2])))
-            num += 5
+        decoded_id = get_element_by_id('streamurl', webpage)
 
-        video_url = 'https://openload.co/stream/' + urlcode
+        video_url = 'https://openload.co/stream/%s?mime=true' % decoded_id
 
         title = self._og_search_title(webpage, default=None) or self._search_regex(
             r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
@@ -93,15 +90,17 @@ class OpenloadIE(InfoExtractor):
             'description', webpage, 'title', fatal=True)
 
         entries = self._parse_html5_media_entries(url, webpage, video_id)
-        subtitles = entries[0]['subtitles'] if entries else None
+        entry = entries[0] if entries else {}
+        subtitles = entry.get('subtitles')
 
         info_dict = {
             'id': video_id,
             'title': title,
-            'thumbnail': self._og_search_thumbnail(webpage, default=None),
+            'thumbnail': entry.get('thumbnail') or self._og_search_thumbnail(webpage, default=None),
             'url': video_url,
             # Seems all videos have extensions in their titles
             'ext': determine_ext(title, 'mp4'),
             'subtitles': subtitles,
+            'http_headers': headers,
         }
         return info_dict