Prepare new version

[youtubedl] / youtube_dl / extractor / ivi.py

diff --git a/youtube_dl/extractor/ivi.py b/youtube_dl/extractor/ivi.py
index 98d1d272a6bb5085dadae9c8233a0e2592571636..86c014b076bfaa1d7da143bdd7cb29c10438d2f3 100644 (file)
--- a/youtube_dl/extractor/ivi.py
+++ b/youtube_dl/extractor/ivi.py
@@ -1,156 +1,220 @@
-# encoding: utf-8
+# coding: utf-8
+from __future__ import unicode_literals

 
 import re
 import json
 
 from .common import InfoExtractor
 from ..utils import (
 
 import re
 import json
 
 from .common import InfoExtractor
 from ..utils import (

-    compat_urllib_request,

     ExtractorError,
     ExtractorError,

+    int_or_none,
+    qualities,

 )
 
 
 class IviIE(InfoExtractor):
 )
 
 
 class IviIE(InfoExtractor):

-    IE_DESC = u'ivi.ru'
-    IE_NAME = u'ivi'
-    _VALID_URL = r'^https?://(?:www\.)?ivi\.ru/watch(?:/(?P<compilationid>[^/]+))?/(?P<videoid>\d+)'
+    IE_DESC = 'ivi.ru'
+    IE_NAME = 'ivi'
+    _VALID_URL = r'https?://(?:www\.)?ivi\.(?:ru|tv)/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)'
+    _GEO_BYPASS = False
+    _GEO_COUNTRIES = ['RU']

 
     _TESTS = [
         # Single movie
         {
 
     _TESTS = [
         # Single movie
         {

-            u'url': u'http://www.ivi.ru/watch/53141',
-            u'file': u'53141.mp4',
-            u'md5': u'6ff5be2254e796ed346251d117196cf4',
-            u'info_dict': {
-                u'title': u'Иван Васильевич меняет профессию',
-                u'description': u'md5:14d8eda24e9d93d29b5857012c6d6346',
-                u'duration': 5498,
-                u'thumbnail': u'http://thumbs.ivi.ru/f20.vcp.digitalaccess.ru/contents/d/1/c3c885163a082c29bceeb7b5a267a6.jpg',
+            'url': 'http://www.ivi.ru/watch/53141',
+            'md5': '6ff5be2254e796ed346251d117196cf4',
+            'info_dict': {
+                'id': '53141',
+                'ext': 'mp4',
+                'title': 'Иван Васильевич меняет профессию',
+                'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
+                'duration': 5498,
+                'thumbnail': r're:^https?://.*\.jpg$',

             },
             },

-            u'skip': u'Only works from Russia',
+            'skip': 'Only works from Russia',

         },
         },

-        # Serial's serie
+        # Serial's series

         {
         {

-            u'url': u'http://www.ivi.ru/watch/dezhurnyi_angel/74791',
-            u'file': u'74791.mp4',
-            u'md5': u'3e6cc9a848c1d2ebcc6476444967baa9',
-            u'info_dict': {
-                u'title': u'Дежурный ангел - 1 серия',
-                u'duration': 2490,
-                u'thumbnail': u'http://thumbs.ivi.ru/f7.vcp.digitalaccess.ru/contents/8/e/bc2f6c2b6e5d291152fdd32c059141.jpg',
+            'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/9549',
+            'md5': '221f56b35e3ed815fde2df71032f4b3e',
+            'info_dict': {
+                'id': '9549',
+                'ext': 'mp4',
+                'title': 'Двое из ларца - Дело Гольдберга (1 часть)',
+                'series': 'Двое из ларца',
+                'season': 'Сезон 1',
+                'season_number': 1,
+                'episode': 'Дело Гольдберга (1 часть)',
+                'episode_number': 1,
+                'duration': 2655,
+                'thumbnail': r're:^https?://.*\.jpg$',

             },
             },

-            u'skip': u'Only works from Russia',
-         }
+            'skip': 'Only works from Russia',
+        },
+        {
+            # with MP4-HD720 format
+            'url': 'http://www.ivi.ru/watch/146500',
+            'md5': 'd63d35cdbfa1ea61a5eafec7cc523e1e',
+            'info_dict': {
+                'id': '146500',
+                'ext': 'mp4',
+                'title': 'Кукла',
+                'description': 'md5:ffca9372399976a2d260a407cc74cce6',
+                'duration': 5599,
+                'thumbnail': r're:^https?://.*\.jpg$',
+            },
+            'skip': 'Only works from Russia',
+        },
+        {
+            'url': 'https://www.ivi.tv/watch/33560/',
+            'only_matching': True,
+        },

     ]
     ]

-    
-    # Sorted by quality
-    _known_formats = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
-
-    # Sorted by size
-    _known_thumbnails = ['Thumb-120x90', 'Thumb-160', 'Thumb-640x480']

 
 

-    def _extract_description(self, html):
-        m = re.search(r'<meta name="description" content="(?P<description>[^"]+)"/>', html)
-        return m.group('description') if m is not None else None
-
-    def _extract_comment_count(self, html):
-        m = re.search(u'(?s)<a href="#" id="view-comments" class="action-button dim gradient">\s*Комментарии:\s*(?P<commentcount>\d+)\s*</a>', html)
-        return int(m.group('commentcount')) if m is not None else 0
+    # Sorted by quality
+    _KNOWN_FORMATS = (
+        'MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi',
+        'MP4-SHQ', 'MP4-HD720', 'MP4-HD1080')

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('videoid')
-
-        api_url = 'http://api.digitalaccess.ru/api/json/'
-
-        data = {u'method': u'da.content.get',
-                u'params': [video_id, {u'site': u's183',
-                                       u'referrer': u'http://www.ivi.ru/watch/%s' % video_id,
-                                       u'contentid': video_id
-                                    }
-                            ]
+        video_id = self._match_id(url)
+
+        data = {
+            'method': 'da.content.get',
+            'params': [
+                video_id, {
+                    'site': 's183',
+                    'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
+                    'contentid': video_id

                 }
                 }

+            ]
+        }

 
 

-        request = compat_urllib_request.Request(api_url, json.dumps(data))
+        video_json = self._download_json(
+            'http://api.digitalaccess.ru/api/json/', video_id,
+            'Downloading video JSON', data=json.dumps(data))

 
 

-        video_json_page = self._download_webpage(request, video_id, u'Downloading video JSON')
-        video_json = json.loads(video_json_page)
+        if 'error' in video_json:
+            error = video_json['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)

 
 

-        if u'error' in video_json:
-            error = video_json[u'error']
-            if error[u'origin'] == u'NoRedisValidData':
-                raise ExtractorError(u'Video %s does not exist' % video_id, expected=True)
-            raise ExtractorError(u'Unable to download video %s: %s' % (video_id, error[u'message']), expected=True)
+        result = video_json['result']

 
 

-        result = video_json[u'result']
+        quality = qualities(self._KNOWN_FORMATS)

 
         formats = [{
 
         formats = [{

-            'url': x[u'url'],
-            'format_id': x[u'content_format'],
-            'preference': self._known_formats.index(x[u'content_format']),
-        } for x in result[u'files'] if x[u'content_format'] in self._known_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')]

 
         self._sort_formats(formats)
 
 
         self._sort_formats(formats)
 

-        if not formats:
-            raise ExtractorError(u'No media links available for %s' % video_id)
+        title = result['title']
+
+        duration = int_or_none(result.get('duration'))
+        compilation = result.get('compilation')
+        episode = title if compilation else None

 
 

-        duration = result[u'duration']
-        compilation = result[u'compilation']
-        title = result[u'title']
+        title = '%s - %s' % (compilation, title) if compilation is not None else title

 
 

-        title = '%s - %s' % (compilation, title) if compilation is not None else title  
+        thumbnails = [{
+            'url': preview['url'],
+            'id': preview.get('content_format'),
+        } for preview in result.get('preview', []) if preview.get('url')]

 
 

-        previews = result[u'preview']
-        previews.sort(key=lambda fmt: self._known_thumbnails.index(fmt['content_format']))
-        thumbnail = previews[-1][u'url'] if len(previews) > 0 else None
+        webpage = self._download_webpage(url, video_id)

 
 

-        video_page = self._download_webpage(url, video_id, u'Downloading video page')
-        description = self._extract_description(video_page)
-        comment_count = self._extract_comment_count(video_page)
+        season = self._search_regex(
+            r'<li[^>]+class="season active"[^>]*><a[^>]+>([^<]+)',
+            webpage, 'season', default=None)
+        season_number = int_or_none(self._search_regex(
+            r'<li[^>]+class="season active"[^>]*><a[^>]+data-season(?:-index)?="(\d+)"',
+            webpage, 'season number', default=None))
+
+        episode_number = int_or_none(self._search_regex(
+            r'[^>]+itemprop="episode"[^>]*>\s*<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
+            webpage, 'episode number', default=None))
+
+        description = self._og_search_description(webpage, default=None) or self._html_search_meta(
+            'description', webpage, 'description', default=None)

 
         return {
             'id': video_id,
             'title': title,
 
         return {
             'id': video_id,
             'title': title,

-            'thumbnail': thumbnail,
+            'series': compilation,
+            'season': season,
+            'season_number': season_number,
+            'episode': episode,
+            'episode_number': episode_number,
+            'thumbnails': thumbnails,

             'description': description,
             'duration': duration,
             'description': description,
             'duration': duration,

-            'comment_count': comment_count,

             'formats': formats,
         }
 
 
 class IviCompilationIE(InfoExtractor):
             'formats': formats,
         }
 
 
 class IviCompilationIE(InfoExtractor):

-    IE_DESC = u'ivi.ru compilations'
-    IE_NAME = u'ivi:compilation'
-    _VALID_URL = r'^https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
+    IE_DESC = 'ivi.ru compilations'
+    IE_NAME = 'ivi:compilation'
+    _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
+    _TESTS = [{
+        'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa',
+        'info_dict': {
+            'id': 'dvoe_iz_lartsa',
+            'title': 'Двое из ларца (2006 - 2008)',
+        },
+        'playlist_mincount': 24,
+    }, {
+        'url': 'http://www.ivi.ru/watch/dvoe_iz_lartsa/season1',
+        'info_dict': {
+            'id': 'dvoe_iz_lartsa/season1',
+            'title': 'Двое из ларца (2006 - 2008) 1 сезон',
+        },
+        'playlist_mincount': 12,
+    }]

 
     def _extract_entries(self, html, compilation_id):
 
     def _extract_entries(self, html, compilation_id):

-        return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi')
-                for serie in re.findall(r'<strong><a href="/watch/%s/(\d+)">(?:[^<]+)</a></strong>' % compilation_id, html)]
+        return [
+            self.url_result(
+                'http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), IviIE.ie_key())
+            for serie in re.findall(
+                r'<a href="/watch/%s/(\d+)"[^>]+data-id="\1"' % compilation_id, html)]

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         compilation_id = mobj.group('compilationid')
         season_id = mobj.group('seasonid')
 
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         compilation_id = mobj.group('compilationid')
         season_id = mobj.group('seasonid')
 

-        if season_id is not None: # Season link
-            season_page = self._download_webpage(url, compilation_id, u'Downloading season %s web page' % season_id)
+        if season_id is not None:  # Season link
+            season_page = self._download_webpage(
+                url, compilation_id, 'Downloading season %s web page' % season_id)

             playlist_id = '%s/season%s' % (compilation_id, season_id)
             playlist_id = '%s/season%s' % (compilation_id, season_id)

-            playlist_title = self._html_search_meta(u'title', season_page, u'title')
+            playlist_title = self._html_search_meta('title', season_page, 'title')

             entries = self._extract_entries(season_page, compilation_id)
             entries = self._extract_entries(season_page, compilation_id)

-        else: # Compilation link            
-            compilation_page = self._download_webpage(url, compilation_id, u'Downloading compilation web page')
+        else:  # Compilation link
+            compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')

             playlist_id = compilation_id
             playlist_id = compilation_id

-            playlist_title = self._html_search_meta(u'title', compilation_page, u'title')
-            seasons = re.findall(r'<a href="/watch/%s/season(\d+)">[^<]+</a>' % compilation_id, compilation_page)
-            if len(seasons) == 0: # No seasons in this compilation
+            playlist_title = self._html_search_meta('title', compilation_page, 'title')
+            seasons = re.findall(
+                r'<a href="/watch/%s/season(\d+)' % compilation_id, compilation_page)
+            if not seasons:  # No seasons in this compilation

                 entries = self._extract_entries(compilation_page, compilation_id)
             else:
                 entries = []
                 for season_id in seasons:
                 entries = self._extract_entries(compilation_page, compilation_id)
             else:
                 entries = []
                 for season_id in seasons:

-                    season_page = self._download_webpage('http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
-                                                         compilation_id, u'Downloading season %s web page' % season_id)
+                    season_page = self._download_webpage(
+                        'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
+                        compilation_id, 'Downloading season %s web page' % season_id)

                     entries.extend(self._extract_entries(season_page, compilation_id))
 
                     entries.extend(self._extract_entries(season_page, compilation_id))
 

-        return self.playlist_result(entries, playlist_id, playlist_title)
\ No newline at end of file
+        return self.playlist_result(entries, playlist_id, playlist_title)