[ivi] fix format extraction(closes #21991)

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

diff --git a/youtube_dl/extractor/ivi.py b/youtube_dl/extractor/ivi.py
index 7c8cb21c2c5619b4809f5daf8605958a808eccb9..efdc3cc984fefe093fa18635c7f334ebbfd8dd55 100644 (file)
--- a/youtube_dl/extractor/ivi.py
+++ b/youtube_dl/extractor/ivi.py
@@ -15,7 +15,11 @@ from ..utils import (
 class IviIE(InfoExtractor):
     IE_DESC = 'ivi.ru'
     IE_NAME = 'ivi'
-    _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?ivi\.(?:ru|tv)/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
+    _GEO_BYPASS = False
+    _GEO_COUNTRIES = ['RU']
+    _LIGHT_KEY = b'\xf1\x02\x32\xb7\xbc\x5c\x7a\xe8\xf7\x96\xc1\x33\x2b\x27\xa1\x8c'
+    _LIGHT_URL = 'https://api.ivi.ru/light/'
 
     _TESTS = [
         # Single movie
@@ -28,7 +32,7 @@ class IviIE(InfoExtractor):
                 'title': 'Иван Васильевич меняет профессию',
                 'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
                 'duration': 5498,
-                'thumbnail': 're:^https?://.*\.jpg$',
+                'thumbnail': r're:^https?://.*\.jpg$',
             },
             'skip': 'Only works from Russia',
         },
@@ -46,7 +50,7 @@ class IviIE(InfoExtractor):
                 'episode': 'Дело Гольдберга (1 часть)',
                 'episode_number': 1,
                 'duration': 2655,
-                'thumbnail': 're:^https?://.*\.jpg$',
+                'thumbnail': r're:^https?://.*\.jpg$',
             },
             'skip': 'Only works from Russia',
         },
@@ -60,10 +64,14 @@ class IviIE(InfoExtractor):
                 'title': 'Кукла',
                 'description': 'md5:ffca9372399976a2d260a407cc74cce6',
                 'duration': 5599,
-                'thumbnail': 're:^https?://.*\.jpg$',
+                'thumbnail': r're:^https?://.*\.jpg$',
             },
             'skip': 'Only works from Russia',
-        }
+        },
+        {
+            'url': 'https://www.ivi.tv/watch/33560/',
+            'only_matching': True,
+        },
     ]
 
     # Sorted by quality
@@ -72,46 +80,70 @@ class IviIE(InfoExtractor):
         'MP4-SHQ', 'MP4-HD720', 'MP4-HD1080')
 
     def _real_extract(self, url):
+        try:
+            from Crypto.Cipher import Blowfish
+            from Crypto.Hash import CMAC
+        except ImportError:
+            raise ExtractorError('pycrypto not found. Please install it.', expected=True)
+
         video_id = self._match_id(url)
 
-        data = {
+        timestamp = self._download_json(
+            self._LIGHT_URL, video_id,
+            'Downloading timestamp JSON', data=json.dumps({
+                'method': 'da.timestamp.get',
+                'params': []
+            }).encode())['result']
+
+        data = json.dumps({
             'method': 'da.content.get',
             'params': [
                 video_id, {
-                    'site': 's183',
+                    'site': 's353',
                     'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
                     'contentid': video_id
                 }
             ]
-        }
+        }).encode()
 
         video_json = self._download_json(
-            'http://api.digitalaccess.ru/api/json/', video_id,
-            'Downloading video JSON', data=json.dumps(data))
-
-        if 'error' in video_json:
-            error = video_json['error']
-            if error['origin'] == 'NoRedisValidData':
+            self._LIGHT_URL, video_id,
+            'Downloading video JSON', data=data, query={
+                'ts': timestamp,
+                'sign': CMAC.new(self._LIGHT_KEY, timestamp.encode() + data, Blowfish).hexdigest(),
+            })
+
+        error = video_json.get('error')
+        if error:
+            origin = error['origin']
+            if origin == 'NotAllowedForLocation':
+                self.raise_geo_restricted(
+                    msg=error['message'], countries=self._GEO_COUNTRIES)
+            elif origin == 'NoRedisValidData':
                 raise ExtractorError('Video %s does not exist' % video_id, expected=True)
             raise ExtractorError(
                 'Unable to download video %s: %s' % (video_id, error['message']),
                 expected=True)
 
         result = video_json['result']
+        title = result['title']
 
         quality = qualities(self._KNOWN_FORMATS)
 
-        formats = [{
-            'url': x['url'],
-            'format_id': x.get('content_format'),
-            'quality': quality(x.get('content_format')),
-        } for x in result['files'] if x.get('url')]
-
+        formats = []
+        for f in result.get('files', []):
+            f_url = f.get('url')
+            content_format = f.get('content_format')
+            if not f_url or '-MDRM-' in content_format or '-FPS-' in content_format:
+                continue
+            formats.append({
+                'url': f_url,
+                'format_id': content_format,
+                'quality': quality(content_format),
+                'filesize': int_or_none(f.get('size_in_bytes')),
+            })
         self._sort_formats(formats)
 
-        title = result['title']
-
-        duration = int_or_none(result.get('duration'))
         compilation = result.get('compilation')
         episode = title if compilation else None
 
@@ -148,7 +180,7 @@ class IviIE(InfoExtractor):
             'episode_number': episode_number,
             'thumbnails': thumbnails,
             'description': description,
-            'duration': duration,
+            'duration': int_or_none(result.get('duration')),
             'formats': formats,
         }