[rai] Fix extraction

youtube_dl/extractor/extractors.py

@@ -802,7 +802,7 @@ from .radiojavan import RadioJavanIE
 from .radiobremen import RadioBremenIE
 from .radiofrance import RadioFranceIE
 from .rai import (
-    RaiTVIE,
+    RaiPlayIE,
     RaiIE,
 )
 from .rbmaradio import RBMARadioIE

youtube_dl/extractor/rai.py

@@ -3,8 +3,8 @@ from __future__ import unicode_literals
 from .common import InfoExtractor
 from ..compat import compat_urlparse
 from ..utils import (
-    determine_ext,
     ExtractorError,
+    determine_ext,
     find_xpath_attr,
     fix_xml_ampersands,
     int_or_none,
@@ -55,135 +55,120 @@ class RaiBaseIE(InfoExtractor):
 
         return formats
 
-    def _extract_from_content_id(self, content_id, base_url):
+
-        media = self._download_json(
+class RaiPlayIE(RaiBaseIE):
-            'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
+    _VALID_URL = r'https?://(?:www\.)?raiplay\.it/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})\.html'
-            content_id, 'Downloading video JSON')
+    _TESTS = [{
+        'url': 'http://www.raiplay.it/video/2016/10/La-Casa-Bianca-e06118bb-59a9-4636-b914-498e4cfd2c66.html?source=twitter',
+        'md5': '340aa3b7afb54bfd14a8c11786450d76',
+        'info_dict': {
+            'id': 'e06118bb-59a9-4636-b914-498e4cfd2c66',
+            'ext': 'mp4',
+            'title': 'La Casa Bianca',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': r're:^Rai.+',
+            'description': 're:^[A-Za-z]+'
+        }
+    }, {
+        'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
+        'md5': 'ed4da3d70ccf8129a33ab16b34d20ab8',
+        'info_dict': {
+            'id': 'efebe701-969c-4593-92f3-285f0d1ce750',
+            'ext': 'mp4',
+            'title': 'Gazebo - #gazebotraindesi',
+            'thumbnail': r're:^https?://.*\.png$',
+            'uploader': r're:^Rai.+',
+            'description': r're:^[A-Za-z]+'
+        }
+    }, {
+        'url': 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
+        'md5': '8970abf8caf8aef4696e7b1f2adfc696',
+        'info_dict': {
+            'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
+            'ext': 'mp4',
+            'title': 'Report - Report del 07/04/2014',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': r're:^Rai.+',
+            'description': r're:^[A-Za-z]+'
+        }
+    }]
+    _RESOLUTION = '600x400'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        # remove query and fragment part from url
+        canonical_url = compat_urlparse.urljoin(url, compat_urlparse.urlparse(url).path)
+        webpage = self._download_webpage(canonical_url, video_id)
+
+        media = self._download_json('%s?json' % canonical_url,
+                                    video_id, 'Downloading video JSON')
 
         thumbnails = []
-        for image_type in ('image', 'image_medium', 'image_300'):
+        if 'images' in media:
-            thumbnail_url = media.get(image_type)
+            for _, value in media.get('images').items():
-            if thumbnail_url:
+                if value:
                     thumbnails.append({
-                    'url': compat_urlparse.urljoin(base_url, thumbnail_url),
+                        'url': value.replace('[RESOLUTION]', self._RESOLUTION)
                     })
 
-        formats = []
+        if 'video' not in media:
-        media_type = media['type']
+            raise ExtractorError('No video found')
-        if 'Audio' in media_type:
-            formats.append({
-                'format_id': media.get('formatoAudio'),
-                'url': media['audioUrl'],
-                'ext': media.get('formatoAudio'),
-            })
-        elif 'Video' in media_type:
-            formats.extend(self._extract_relinker_formats(media['mediaUri'], content_id))
-            self._sort_formats(formats)
-        else:
-            raise ExtractorError('not a media file')
 
-        subtitles = {}
+        video = media.get('video')
-        captions = media.get('subtitlesUrl')
+        duration = parse_duration(video.get('duration')),
-        if captions:
+        formats = self._extract_relinker_formats(video.get('contentUrl'), video_id)
-            STL_EXT = '.stl'
+        self._sort_formats(formats)
-            SRT_EXT = '.srt'
-            if captions.endswith(STL_EXT):
-                captions = captions[:-len(STL_EXT)] + SRT_EXT
-            subtitles['it'] = [{
-                'ext': 'srt',
-                'url': captions,
-            }]
 
         return {
-            'id': content_id,
+            'id': video_id,
-            'title': media['name'],
+            'title': self._og_search_title(webpage).replace(' - video - RaiPlay', ''),
-            'description': media.get('desc'),
+            'description': self._og_search_description(webpage),
+            'uploader': media.get('channel'),
+            'duration': duration,
             'thumbnails': thumbnails,
-            'uploader': media.get('author'),
+            'formats': formats
-            'upload_date': unified_strdate(media.get('date')),
-            'duration': parse_duration(media.get('length')),
-            'formats': formats,
-            'subtitles': subtitles,
         }
 
 
-class RaiTVIE(RaiBaseIE):
+class RaiIE(RaiBaseIE):
-    _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/(?:[^/]+/)+(?:media|ondemand)/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html'
+    _VALID_URL = r'https?://.+\.(?:rai|rainews)\.it/dl/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html'
-    _TESTS = [
+    _TESTS = [{
-        {
+        # subdomain test case
-            'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
-            'md5': '8970abf8caf8aef4696e7b1f2adfc696',
-            'info_dict': {
-                'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
-                'ext': 'mp4',
-                'title': 'Report del 07/04/2014',
-                'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
-                'upload_date': '20140407',
-                'duration': 6160,
-                'thumbnail': r're:^https?://.*\.jpg$',
-            }
-        },
-        {
-            # no m3u8 stream
         'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
-            # HDS download, MD5 is unstable
         'info_dict': {
             'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
-                'ext': 'flv',
+            'ext': 'mp4',
             'title': 'TG PRIMO TEMPO',
             'upload_date': '20140612',
             'duration': 1758,
-                'thumbnail': r're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$'
-            },
+        }
-            'skip': 'Geo-restricted to Italy',
+    }, {
-        },
+        # rainews test case
-        {
+        'url': 'http://www.rainews.it/dl/rainews/media/Weekend-al-cinema-da-Hollywood-arriva-il-thriller-di-Tate-Taylor-La-ragazza-del-treno-1632c009-c843-4836-bb65-80c33084a64b.html',
-            'url': 'http://www.rainews.it/dl/rainews/media/state-of-the-net-Antonella-La-Carpia-regole-virali-7aafdea9-0e5d-49d5-88a6-7e65da67ae13.html',
-            'md5': '35cf7c229f22eeef43e48b5cf923bef0',
-            'info_dict': {
-                'id': '7aafdea9-0e5d-49d5-88a6-7e65da67ae13',
-                'ext': 'mp4',
-                'title': 'State of the Net, Antonella La Carpia: regole virali',
-                'description': 'md5:b0ba04a324126903e3da7763272ae63c',
-                'upload_date': '20140613',
-            },
-            'skip': 'Error 404',
-        },
-        {
-            'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-b4a49761-e0cc-4b14-8736-2729f6f73132-tg2.html',
         'info_dict': {
-                'id': 'b4a49761-e0cc-4b14-8736-2729f6f73132',
+            'id': '1632c009-c843-4836-bb65-80c33084a64b',
             'ext': 'mp4',
-                'title': 'Alluvione in Sardegna e dissesto idrogeologico',
+            'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor \"La ragazza del treno\" ',
-                'description': 'Edizione delle ore 20:30 ',
+            'upload_date': '20161103',
-            },
+            'thumbnail': r're:^https?://.*\.png$',
-            'skip': 'invalid urls',
+            'description': r're:^[A-Za-z]+'
-        },
+        }
-        {
+    }, {
-            'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html',
+        # with media information
-            'md5': 'e57493e1cb8bc7c564663f363b171847',
+        'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
+        'md5': '11959b4e44fa74de47011b5799490adf',
         'info_dict': {
-                'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6',
+            'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
             'ext': 'mp4',
-                'title': 'Il Candidato - Primo episodio: "Le Primarie"',
+            'title': 'TG1 ore 20:00 del 03/11/2016',
-                'description': 'md5:364b604f7db50594678f483353164fb8',
-                'upload_date': '20140923',
-                'duration': 386,
             'thumbnail': r're:^https?://.*\.jpg$',
+            'upload_date': '20161103',
+            'description': r're:^[A-Za-z]+'
         }
-        },
+    }, {
-    ]
+        # drawMediaRaiTV test case
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        return self._extract_from_content_id(video_id, url)
-
-
-class RaiIE(RaiBaseIE):
-    _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html'
-    _TESTS = [
-        {
         'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
         'md5': '2dd727e61114e1ee9c47f0da6914e178',
         'info_dict': {
@@ -193,8 +178,7 @@ class RaiIE(RaiBaseIE):
             'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
             'upload_date': '20141221',
         },
-        },
+    }, {
-        {
         # Direct relinker URL
         'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
         # HDS live stream, MD5 is unstable
@@ -203,33 +187,25 @@ class RaiIE(RaiBaseIE):
             'ext': 'flv',
             'title': 'EuroNews',
         },
-            'skip': 'Geo-restricted to Italy',
+    }, {
-        },
-        {
         # Embedded content item ID
         'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
-            'md5': '84c1135ce960e8822ae63cec34441d63',
         'info_dict': {
-                'id': '0960e765-62c8-474a-ac4b-7eb3e2be39c8',
+            'id': 'd80d4b70-3812-4501-a888-92edec729f00',
             'ext': 'mp4',
-                'title': 'TG1 ore 20:00 del 02/07/2016',
+            'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
-                'upload_date': '20160702',
+            'upload_date': r're:\d{8}',
-            },
+            'description': r're:.+',
         },
-        {
+    }, {
         'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
         # HDS live stream, MD5 is unstable
         'info_dict': {
             'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
-                'ext': 'flv',
+            'ext': 'mp4',
             'title': 'La diretta di Rainews24',
         },
-        },
+    }]
-    ]
-
-    @classmethod
-    def suitable(cls, url):
-        return False if RaiTVIE.suitable(url) else super(RaiIE, cls).suitable(url)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -250,6 +226,12 @@ class RaiIE(RaiBaseIE):
         if content_item_id:
             return self._extract_from_content_id(content_item_id, url)
 
+        try:
+            return self._extract_from_content_id(video_id, url)
+        except ExtractorError:
+            # no media data, only direct relinker
+            pass
+
         relinker_url = compat_urlparse.urljoin(url, self._search_regex(
             r'(?:var\s+videoURL|mediaInfo\.mediaUri)\s*=\s*(?P<q1>[\'"])(?P<url>(https?:)?//mediapolis\.rai\.it/relinker/relinkerServlet\.htm\?cont=\d+)(?P=q1)',
             webpage, 'relinker URL', group='url'))
@@ -265,3 +247,54 @@ class RaiIE(RaiBaseIE):
             'title': title,
             'formats': formats,
         }
+
+    def _extract_from_content_id(self, content_id, url):
+        media = self._download_json(
+            'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
+            content_id, 'Downloading video JSON')
+
+        thumbnails = []
+        for image_type in ('image', 'image_medium', 'image_300'):
+            thumbnail_url = media.get(image_type)
+            if thumbnail_url:
+                thumbnails.append({
+                    'url': compat_urlparse.urljoin(url, thumbnail_url),
+                })
+
+        formats = []
+        media_type = media['type']
+        if 'Audio' in media_type:
+            formats.append({
+                'format_id': media.get('formatoAudio'),
+                'url': media['audioUrl'],
+                'ext': media.get('formatoAudio'),
+            })
+        elif 'Video' in media_type:
+            formats.extend(self._extract_relinker_formats(media['mediaUri'], content_id))
+            self._sort_formats(formats)
+        else:
+            raise ExtractorError('not a media file')
+
+        subtitles = {}
+        captions = media.get('subtitlesUrl')
+        if captions:
+            STL_EXT = '.stl'
+            SRT_EXT = '.srt'
+            if captions.endswith(STL_EXT):
+                captions = captions[:-len(STL_EXT)] + SRT_EXT
+            subtitles['it'] = [{
+                'ext': 'srt',
+                'url': captions,
+            }]
+
+        return {
+            'id': content_id,
+            'title': media['name'],
+            'description': media.get('desc'),
+            'thumbnails': thumbnails,
+            'uploader': media.get('author'),
+            'upload_date': unified_strdate(media.get('date')),
+            'duration': parse_duration(media.get('length')),
+            'formats': formats,
+            'subtitles': subtitles,
+        }