Initiate new release.

[youtubedl] / youtube_dl / extractor / openload.py

diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py
index 4e80ca9ff03100e244b157f70c8ff6897256cd55..d8036b54acd8defc75992e912af90a28c99f8d87 100644 (file)
--- a/youtube_dl/extractor/openload.py
+++ b/youtube_dl/extractor/openload.py
@@ -1,19 +1,18 @@
 # coding: utf-8
 # coding: utf-8

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

 
 

-import math
+import re

 
 from .common import InfoExtractor
 from ..compat import compat_chr
 from ..utils import (
 
 from .common import InfoExtractor
 from ..compat import compat_chr
 from ..utils import (

-    decode_png,

     determine_ext,
     ExtractorError,
 )
 
 
 class OpenloadIE(InfoExtractor):
     determine_ext,
     ExtractorError,
 )
 
 
 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)|oload\.tv)/(?:f|embed)/(?P<id>[a-zA-Z0-9-_]+)'

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


@@ -22,7 +21,23 @@ class OpenloadIE(InfoExtractor):
             'id': 'kUEfGclsU9o',
             'ext': 'mp4',
             'title': 'skyrim_no-audio_1080.mp4',
             'id': 'kUEfGclsU9o',
             'ext': 'mp4',
             'title': 'skyrim_no-audio_1080.mp4',

-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',
+        },
+    }, {
+        'url': 'https://openload.co/embed/rjC09fkPLYs',
+        'info_dict': {
+            'id': 'rjC09fkPLYs',
+            'ext': 'mp4',
+            'title': 'movie.mp4',
+            'thumbnail': r'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',
         },
     }, {
         'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',


@@ -38,86 +53,76 @@ class OpenloadIE(InfoExtractor):
         # for title and ext
         'url': 'https://openload.co/embed/Sxz5sADo82g/',
         'only_matching': True,
         # for title and ext
         'url': 'https://openload.co/embed/Sxz5sADo82g/',
         'only_matching': True,

+    }, {
+        'url': 'https://oload.tv/embed/KnG-kKZdcfY/',
+        'only_matching': True,

     }]
 
     }]
 

+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe[^>]+src=["\']((?:https?://)?(?:openload\.(?:co|io)|oload\.tv)/embed/[a-zA-Z0-9-_]+)',
+            webpage)
+

     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)
 

-        # The following extraction logic is proposed by @Belderak and @gdkchan
-        # and declared to be used freely in youtube-dl
-        # See https://github.com/rg3/youtube-dl/issues/9706
-
-        numbers_js = self._download_webpage(
-            'https://openload.co/assets/js/obfuscator/n.js', video_id,
-            note='Downloading signature numbers')
-        signums = self._search_regex(
-            r'window\.signatureNumbers\s*=\s*[\'"](?P<data>[a-z]+)[\'"]',
-            numbers_js, 'signature numbers', group='data')
-
-        linkimg_uri = self._search_regex(
-            r'<img[^>]+id="linkimg"[^>]+src="([^"]+)"', webpage, 'link image')
-        linkimg = self._request_webpage(
-            linkimg_uri, video_id, note=False).read()
-
-        width, height, pixels = decode_png(linkimg)
-
-        output = ''
-        for y in range(height):
-            for x in range(width):
-                r, g, b = pixels[y][3 * x:3 * x + 3]
-                if r == 0 and g == 0 and b == 0:
-                    break
-                else:
-                    output += compat_chr(r)
-                    output += compat_chr(g)
-                    output += compat_chr(b)
-
-        img_str_length = len(output) // 200
-        img_str = [[0 for x in range(img_str_length)] for y in range(10)]
-
-        sig_str_length = len(signums) // 260
-        sig_str = [[0 for x in range(sig_str_length)] for y in range(10)]
-
-        for i in range(10):
-            for j in range(img_str_length):
-                begin = i * img_str_length * 20 + j * 20
-                img_str[i][j] = output[begin:begin + 20]
-            for j in range(sig_str_length):
-                begin = i * sig_str_length * 26 + j * 26
-                sig_str[i][j] = signums[begin:begin + 26]
-
-        parts = []
-        # TODO: find better names for str_, chr_ and sum_
-        str_ = ''
-        for i in [2, 3, 5, 7]:
-            str_ = ''
-            sum_ = float(99)
-            for j in range(len(sig_str[i])):
-                for chr_idx in range(len(img_str[i][j])):
-                    if sum_ > float(122):
-                        sum_ = float(98)
-                    chr_ = compat_chr(int(math.floor(sum_)))
-                    if sig_str[i][j][chr_idx] == chr_ and j >= len(str_):
-                        sum_ += float(2.5)
-                        str_ += img_str[i][j][chr_idx]
-            parts.append(str_.replace(',', ''))
-
-        video_url = 'https://openload.co/stream/%s~%s~%s~%s' % (parts[3], parts[1], parts[2], parts[0])
+        ol_id = self._search_regex(
+            '<span[^>]+id="[^"]+"[^>]*>([0-9A-Za-z]+)</span>',
+            webpage, 'openload ID')
+
+        decoded = ''
+        a = ol_id[0:24]
+        b = []
+        for i in range(0, len(a), 8):
+            b.append(int(a[i:i + 8] or '0', 16))
+        ol_id = ol_id[24:]
+        j = 0
+        k = 0
+        while j < len(ol_id):
+            c = 128
+            d = 0
+            e = 0
+            f = 0
+            _more = True
+            while _more:
+                if j + 1 >= len(ol_id):
+                    c = 143
+                f = int(ol_id[j:j + 2] or '0', 16)
+                j += 2
+                d += (f & 127) << e
+                e += 7
+                _more = f >= c
+            g = d ^ b[k % 3]
+            for i in range(4):
+                char_dec = (g >> 8 * i) & (c + 127)
+                char = compat_chr(char_dec)
+                if char != '#':
+                    decoded += char
+            k += 1
+
+        video_url = 'https://openload.co/stream/%s?mime=true'
+        video_url = video_url % decoded

 
         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)
 

-        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
             '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),
+            'ext': determine_ext(title, 'mp4'),
+            'subtitles': subtitles,

         }
         }

+        return info_dict