[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/allocine.py b/youtube_dl/extractor/allocine.py
index b292ffdd90e3d3b46922523b460b4b1a5b51b7dd..cd533acfc7908076e85a2456c2ffab4413b85b75 100644 (file)
--- a/youtube_dl/extractor/allocine.py
+++ b/youtube_dl/extractor/allocine.py
@@ -2,8 +2,13 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
+    int_or_none,
     qualities,
+    remove_end,
+    try_get,
+    unified_timestamp,
     url_basename,
 )
 
@@ -20,7 +25,11 @@ class AllocineIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Astérix - Le Domaine des Dieux Teaser VF',
             'description': 'md5:4a754271d9c6f16c72629a8a993ee884',
-            'thumbnail': 're:http://.*\.jpg',
+            'thumbnail': r're:http://.*\.jpg',
+            'duration': 39,
+            'timestamp': 1404273600,
+            'upload_date': '20140702',
+            'view_count': int,
         },
     }, {
         'url': 'http://www.allocine.fr/video/player_gen_cmedia=19540403&cfilm=222257.html',
@@ -31,7 +40,11 @@ class AllocineIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Planes 2 Bande-annonce VF',
             'description': 'Regardez la bande annonce du film Planes 2 (Planes 2 Bande-annonce VF). Planes 2, un film de Roberts Gannaway',
-            'thumbnail': 're:http://.*\.jpg',
+            'thumbnail': r're:http://.*\.jpg',
+            'duration': 69,
+            'timestamp': 1385659800,
+            'upload_date': '20131128',
+            'view_count': int,
         },
     }, {
         'url': 'http://www.allocine.fr/video/player_gen_cmedia=19544709&cfilm=181290.html',
@@ -42,11 +55,22 @@ class AllocineIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Dragons 2 - Bande annonce finale VF',
             'description': 'md5:6cdd2d7c2687d4c6aafe80a35e17267a',
-            'thumbnail': 're:http://.*\.jpg',
+            'thumbnail': r're:http://.*\.jpg',
+            'duration': 144,
+            'timestamp': 1397589900,
+            'upload_date': '20140415',
+            'view_count': int,
         },
     }, {
         'url': 'http://www.allocine.fr/video/video-19550147/',
-        'only_matching': True,
+        'md5': '3566c0668c0235e2d224fd8edb389f67',
+        'info_dict': {
+            'id': '19550147',
+            'ext': 'mp4',
+            'title': 'Faux Raccord N°123 - Les gaffes de Cliffhanger',
+            'description': 'md5:bc734b83ffa2d8a12188d9eb48bb6354',
+            'thumbnail': r're:http://.*\.jpg',
+        },
     }]
 
     def _real_extract(self, url):
@@ -54,27 +78,55 @@ class AllocineIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
-        model = self._html_search_regex(
-            r'data-model="([^"]+)"', webpage, 'data model')
-        model_data = self._parse_json(model, display_id)
-
+        formats = []
         quality = qualities(['ld', 'md', 'hd'])
 
-        formats = []
-        for video_url in model_data['sources'].values():
-            video_id, format_id = url_basename(video_url).split('_')[:2]
-            formats.append({
-                'format_id': format_id,
-                'quality': quality(format_id),
-                'url': video_url,
-            })
+        model = self._html_search_regex(
+            r'data-model="([^"]+)"', webpage, 'data model', default=None)
+        if model:
+            model_data = self._parse_json(model, display_id)
+            video = model_data['videos'][0]
+            title = video['title']
+            for video_url in video['sources'].values():
+                video_id, format_id = url_basename(video_url).split('_')[:2]
+                formats.append({
+                    'format_id': format_id,
+                    'quality': quality(format_id),
+                    'url': video_url,
+                })
+            duration = int_or_none(video.get('duration'))
+            view_count = int_or_none(video.get('view_count'))
+            timestamp = unified_timestamp(try_get(
+                video, lambda x: x['added_at']['date'], compat_str))
+        else:
+            video_id = display_id
+            media_data = self._download_json(
+                'http://www.allocine.fr/ws/AcVisiondataV5.ashx?media=%s' % video_id, display_id)
+            title = remove_end(
+                self._html_search_regex(
+                    r'(?s)<title>(.+?)</title>', webpage, 'title').strip(),
+                ' - AlloCiné')
+            for key, value in media_data['video'].items():
+                if not key.endswith('Path'):
+                    continue
+                format_id = key[:-len('Path')]
+                formats.append({
+                    'format_id': format_id,
+                    'quality': quality(format_id),
+                    'url': value,
+                })
+            duration, view_count, timestamp = [None] * 3
+
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'display_id': display_id,
-            'title': model_data['title'],
+            'title': title,
+            'description': self._og_search_description(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),
+            'duration': duration,
+            'timestamp': timestamp,
+            'view_count': view_count,
             'formats': formats,
-            'description': self._og_search_description(webpage),
         }