[ign] add support for pcmag and extract all formats and more metadata

youtube_dl/extractor/__init__.py

@@ -231,7 +231,11 @@
 from .huffpost import HuffPostIE
 from .hypem import HypemIE
 from .iconosquare import IconosquareIE
-from .ign import IGNIE, OneUPIE
+from .ign import (
+    IGNIE,
+    OneUPIE,
+    PCMagIE,
+)
 from .imdb import (
     ImdbIE,
     ImdbListIE

youtube_dl/extractor/ign.py

@@ -3,6 +3,10 @@
 import re
 
 from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_iso8601,
+)
 
 
 class IGNIE(InfoExtractor):
@@ -11,25 +15,23 @@ class IGNIE(InfoExtractor):
     Some videos of it.ign.com are also supported
     """
 
-    _VALID_URL = r'https?://.+?\.ign\.com/(?P<type>videos|show_videos|articles|(?:[^/]*/feature))(/.+)?/(?P<name_or_id>.+)'
+    _VALID_URL = r'https?://.+?\.ign\.com/(?:[^/]+/)?(?P<type>videos|show_videos|articles|feature|(?:[^/]+/\d+/video))(/.+)?/(?P<name_or_id>.+)'
     IE_NAME = 'ign.com'
 
-    _CONFIG_URL_TEMPLATE = 'http://www.ign.com/videos/configs/id/%s.config'
+    _API_URL_TEMPLATE = 'http://apis.ign.com/video/v3/videos/%s'
-    _DESCRIPTION_RE = [
+    _EMBED_RE = r'<iframe[^>]+?["\']((?:https?:)?//.+?\.ign\.com.+?/embed.+?)["\']'
-        r'<span class="page-object-description">(.+?)</span>',
-        r'id="my_show_video">.*?<p>(.*?)</p>',
-        r'<meta name="description" content="(.*?)"',
-    ]
 
     _TESTS = [
         {
             'url': 'http://www.ign.com/videos/2013/06/05/the-last-of-us-review',
-            'md5': 'eac8bdc1890980122c3b66f14bdd02e9',
+            'md5': 'febda82c4bafecd2d44b6e1a18a595f8',
             'info_dict': {
                 'id': '8f862beef863986b2785559b9e1aa599',
                 'ext': 'mp4',
                 'title': 'The Last of Us Review',
                 'description': 'md5:c8946d4260a4d43a00d5ae8ed998870c',
+                'timestamp': 1370440800,
+                'upload_date': '20130605',
             }
         },
         {
@@ -44,6 +46,8 @@ class IGNIE(InfoExtractor):
                         'ext': 'mp4',
                         'title': 'GTA 5 Video Review',
                         'description': 'Rockstar drops the mic on this generation of games. Watch our review of the masterly Grand Theft Auto V.',
+                        'timestamp': 1379339880,
+                        'upload_date': '20130916',
                     },
                 },
                 {
@@ -52,6 +56,8 @@ class IGNIE(InfoExtractor):
                         'ext': 'mp4',
                         'title': '26 Twisted Moments from GTA 5 in Slow Motion',
                         'description': 'The twisted beauty of GTA 5 in stunning slow motion.',
+                        'timestamp': 1386878820,
+                        'upload_date': '20131212',
                     },
                 },
             ],
@@ -66,10 +72,9 @@ class IGNIE(InfoExtractor):
                 'id': '078fdd005f6d3c02f63d795faa1b984f',
                 'ext': 'mp4',
                 'title': 'Rewind Theater - Wild Trailer Gamescom 2014',
-                'description': (
+                'description': 'Brian and Jared explore Michel Ancel\'s captivating new preview.',
-                    'Giant skeletons, bloody hunts, and captivating'
+                'timestamp': 1408047180,
-                    ' natural beauty take our breath away.'
+                'upload_date': '20140814',
-                ),
             },
         },
     ]
@@ -82,7 +87,7 @@ def _find_video_id(self, webpage):
             r'<object id="vid_(.+?)"',
             r'<meta name="og:image" content=".*/(.+?)-(.+?)/.+.jpg"',
         ]
-        return self._search_regex(res_id, webpage, 'video id')
+        return self._search_regex(res_id, webpage, 'video id', default=None)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -102,22 +107,45 @@ def _real_extract(self, url):
                 }
 
         video_id = self._find_video_id(webpage)
-        result = self._get_video_info(video_id)
+        if not video_id:
-        description = self._html_search_regex(self._DESCRIPTION_RE,
+            return self.url_result(self._search_regex(self._EMBED_RE, webpage, 'embed url'))
-                                              webpage, 'video description', flags=re.DOTALL)
+        return self._get_video_info(video_id)
-        result['description'] = description
-        return result
 
     def _get_video_info(self, video_id):
-        config_url = self._CONFIG_URL_TEMPLATE % video_id
+        api_data = self._download_json(self._API_URL_TEMPLATE % video_id, video_id)
-        config = self._download_json(config_url, video_id)
+
-        media = config['playlist']['media']
+        formats = []
+        m3u8_url = api_data['refs'].get('m3uUrl')
+        if m3u8_url:
+            formats.extend(self._extract_m3u8_formats(m3u8_url, video_id))
+        f4m_url = api_data['refs'].get('f4mUrl')
+        if f4m_url:
+            formats.extend(self._extract_f4m_formats(f4m_url, video_id))
+        for asset in api_data['assets']:
+            formats.append({
+                'url': asset['url'],
+                'tbr': asset.get('actual_bitrate_kbps'),
+                'fps': asset.get('frame_rate'),
+                'height': int_or_none(asset.get('height')),
+                'width': int_or_none(asset.get('width')),
+            })
+        self._sort_formats(formats)
+
+        thumbnails = []
+        for thumbnail in api_data['thumbnails']:
+            thumbnails.append({'url': thumbnail['url']})
+
+        metadata = api_data['metadata']
 
         return {
-            'id': media['metadata']['videoId'],
+            'id': api_data.get('videoId') or video_id,
-            'url': media['url'],
+            'title': metadata.get('longTitle') or metadata.get('name') or metadata.get['title'],
-            'title': media['metadata']['title'],
+            'description': metadata.get('description'),
-            'thumbnail': media['poster'][0]['url'].replace('{size}', 'grande'),
+            'timestamp': parse_iso8601(metadata.get('publishDate')),
+            'duration': int_or_none(metadata.get('duration')),
+            'display_id': metadata.get('slug') or video_id,
+            'thumbnails': thumbnails,
+            'formats': formats,
         }
 
 
@@ -125,16 +153,16 @@ class OneUPIE(IGNIE):
     _VALID_URL = r'https?://gamevideos\.1up\.com/(?P<type>video)/id/(?P<name_or_id>.+)\.html'
     IE_NAME = '1up.com'
 
-    _DESCRIPTION_RE = r'<div id="vid_summary">(.+?)</div>'
-
     _TESTS = [{
         'url': 'http://gamevideos.1up.com/video/id/34976.html',
-        'md5': '68a54ce4ebc772e4b71e3123d413163d',
+        'md5': 'c9cc69e07acb675c31a16719f909e347',
         'info_dict': {
             'id': '34976',
             'ext': 'mp4',
             'title': 'Sniper Elite V2 - Trailer',
-            'description': 'md5:5d289b722f5a6d940ca3136e9dae89cf',
+            'description': 'md5:bf0516c5ee32a3217aa703e9b1bc7826',
+            'timestamp': 1313099220,
+            'upload_date': '20110811',
         }
     }]
 
@@ -143,3 +171,34 @@ def _real_extract(self, url):
         result = super(OneUPIE, self)._real_extract(url)
         result['id'] = mobj.group('name_or_id')
         return result
+
+
+class PCMagIE(IGNIE):
+    _VALID_URL = r'https?://(?:www\.)?pcmag\.com/(?P<type>videos|article2)(/.+)?/(?P<name_or_id>.+)'
+    IE_NAME = 'pcmag'
+
+    _EMBED_RE = r'iframe.setAttribute\("src",\s*__util.objToUrlString\("http://widgets\.ign\.com/video/embed/content.html?[^"]*url=([^"]+)["&]'
+
+    _TESTS = [{
+        'url': 'http://www.pcmag.com/videos/2015/01/06/010615-whats-new-now-is-gogo-snooping-on-your-data',
+        'md5': '212d6154fd0361a2781075f1febbe9ad',
+        'info_dict': {
+            'id': 'ee10d774b508c9b8ec07e763b9125b91',
+            'ext': 'mp4',
+            'title': '010615_What\'s New Now: Is GoGo Snooping on Your Data?',
+            'description': 'md5:a7071ae64d2f68cc821c729d4ded6bb3',
+            'timestamp': 1420571160,
+            'upload_date': '20150106',
+        }
+    },{
+        'url': 'http://www.pcmag.com/article2/0,2817,2470156,00.asp',
+        'md5': '94130c1ca07ba0adb6088350681f16c1',
+        'info_dict': {
+            'id': '042e560ba94823d43afcb12ddf7142ca',
+            'ext': 'mp4',
+            'title': 'HTC\'s Weird New Re Camera - What\'s New Now',
+            'description': 'md5:53433c45df96d2ea5d0fda18be2ca908',
+            'timestamp': 1412953920,
+            'upload_date': '20141010',
+        }
+    }]