Prepare to release.

[youtubedl] / youtube_dl / extractor / rai.py

diff --git a/youtube_dl/extractor/rai.py b/youtube_dl/extractor/rai.py
index 7ff1d06c46aacfa67a0134c878d635dce598028f..81eb9db85b9127e292894e22c9dd6272b134832a 100644 (file)
--- a/youtube_dl/extractor/rai.py
+++ b/youtube_dl/extractor/rai.py
@@ -4,176 +4,417 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
 
 from .common import InfoExtractor
 from ..compat import (

-    compat_urllib_parse,

     compat_urlparse,
     compat_urlparse,

+    compat_str,

 )
 from ..utils import (
 )
 from ..utils import (

+    ExtractorError,
+    determine_ext,
+    find_xpath_attr,
+    fix_xml_ampersands,
+    GeoRestrictedError,
+    int_or_none,

     parse_duration,
     parse_duration,

+    strip_or_none,
+    try_get,

     unified_strdate,
     unified_strdate,

+    unified_timestamp,
+    update_url_query,
+    urljoin,
+    xpath_text,

 )
 
 
 )
 
 

-class RaiIE(InfoExtractor):
-    _VALID_URL = r'(?P<url>(?P<host>http://(?:.+?\.)?(?: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.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
-            'md5': 'c064c0b2d09c278fb293116ef5d0a32d',
-            'info_dict': {
-                'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
-                'ext': 'mp4',
-                'title': 'Report del 07/04/2014',
-                'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
-                'upload_date': '20140407',
-                'duration': 6160,
-            }
+class RaiBaseIE(InfoExtractor):
+    _UUID_RE = r'[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}'
+    _GEO_COUNTRIES = ['IT']
+    _GEO_BYPASS = False
+
+    def _extract_relinker_info(self, relinker_url, video_id):
+        formats = []
+        geoprotection = None
+        is_live = None
+        duration = None
+
+        for platform in ('mon', 'flash', 'native'):
+            relinker = self._download_xml(
+                relinker_url, video_id,
+                note='Downloading XML metadata for platform %s' % platform,
+                transform_source=fix_xml_ampersands,
+                query={'output': 45, 'pl': platform},
+                headers=self.geo_verification_headers())
+
+            if not geoprotection:
+                geoprotection = xpath_text(
+                    relinker, './geoprotection', default=None) == 'Y'
+
+            if not is_live:
+                is_live = xpath_text(
+                    relinker, './is_live', default=None) == 'Y'
+            if not duration:
+                duration = parse_duration(xpath_text(
+                    relinker, './duration', default=None))
+
+            url_elem = find_xpath_attr(relinker, './url', 'type', 'content')
+            if url_elem is None:
+                continue
+
+            media_url = url_elem.text
+
+            # This does not imply geo restriction (e.g.
+            # http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html)
+            if media_url == 'http://download.rai.it/video_no_available.mp4':
+                continue
+
+            ext = determine_ext(media_url)
+            if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'):
+                continue
+
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    media_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
+            elif ext == 'f4m':
+                manifest_url = update_url_query(
+                    media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'),
+                    {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'})
+                formats.extend(self._extract_f4m_formats(
+                    manifest_url, video_id, f4m_id='hds', fatal=False))
+            else:
+                bitrate = int_or_none(xpath_text(relinker, 'bitrate'))
+                formats.append({
+                    'url': media_url,
+                    'tbr': bitrate if bitrate > 0 else None,
+                    'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http',
+                })
+
+        if not formats and geoprotection is True:
+            self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
+
+        return dict((k, v) for k, v in {
+            'is_live': is_live,
+            'duration': duration,
+            'formats': formats,
+        }.items() if v is not None)
+
+    @staticmethod
+    def _extract_subtitles(url, subtitle_url):
+        subtitles = {}
+        if subtitle_url and isinstance(subtitle_url, compat_str):
+            subtitle_url = urljoin(url, subtitle_url)
+            STL_EXT = '.stl'
+            SRT_EXT = '.srt'
+            subtitles['it'] = [{
+                'ext': 'stl',
+                'url': subtitle_url,
+            }]
+            if subtitle_url.endswith(STL_EXT):
+                srt_url = subtitle_url[:-len(STL_EXT)] + SRT_EXT
+                subtitles['it'].append({
+                    'ext': 'srt',
+                    'url': srt_url,
+                })
+        return subtitles
+
+
+class RaiPlayIE(RaiBaseIE):
+    _VALID_URL = r'(?P<url>https?://(?:www\.)?raiplay\.it/.+?-(?P<id>%s)\.html)' % RaiBaseIE._UUID_RE
+    _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',
+            'alt_title': 'S2016 - Puntata del 23/10/2016',
+            'description': 'md5:a09d45890850458077d1f68bb036e0a5',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'Rai 3',
+            'creator': 'Rai 3',
+            'duration': 3278,
+            'timestamp': 1477764300,
+            'upload_date': '20161029',
+            'series': 'La Casa Bianca',
+            'season': '2016',

         },
         },

-        {
-            'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
-            'md5': '8bb9c151924ce241b74dd52ef29ceafa',
-            'info_dict': {
-                'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
-                'ext': 'mp4',
-                'title': 'TG PRIMO TEMPO',
-                'description': '',
-                'upload_date': '20140612',
-                'duration': 1758,
-            },
-            'skip': 'Error 404',
+    }, {
+        '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 del 07/04/2014',
+            'alt_title': 'S2013/14 - Puntata del 07/04/2014',
+            'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'Rai 5',
+            'creator': 'Rai 5',
+            'duration': 6160,
+            'series': 'Report',
+            'season_number': 5,
+            'season': '2013/14',

         },
         },

-        {
-            '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',
+        'params': {
+            'skip_download': True,

         },
         },

-        {
-            'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-b4a49761-e0cc-4b14-8736-2729f6f73132-tg2.html',
-            'md5': '35694f062977fe6619943f08ed935730',
-            'info_dict': {
-                'id': 'b4a49761-e0cc-4b14-8736-2729f6f73132',
-                'ext': 'mp4',
-                'title': 'Alluvione in Sardegna e dissesto idrogeologico',
-                'description': 'Edizione delle ore 20:30 ',
-            }
+    }, {
+        'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        url, video_id = mobj.group('url', 'id')
+
+        media = self._download_json(
+            '%s?json' % url, video_id, 'Downloading video JSON')
+
+        title = media['name']
+
+        video = media['video']
+
+        relinker_info = self._extract_relinker_info(video['contentUrl'], video_id)
+        self._sort_formats(relinker_info['formats'])
+
+        thumbnails = []
+        if 'images' in media:
+            for _, value in media.get('images').items():
+                if value:
+                    thumbnails.append({
+                        'url': value.replace('[RESOLUTION]', '600x400')
+                    })
+
+        timestamp = unified_timestamp(try_get(
+            media, lambda x: x['availabilities'][0]['start'], compat_str))
+
+        subtitles = self._extract_subtitles(url, video.get('subtitles'))
+
+        info = {
+            'id': video_id,
+            'title': title,
+            'alt_title': media.get('subtitle'),
+            'description': media.get('description'),
+            'uploader': media.get('channel'),
+            'creator': media.get('editor'),
+            'duration': parse_duration(video.get('duration')),
+            'timestamp': timestamp,
+            'thumbnails': thumbnails,
+            'series': try_get(
+                media, lambda x: x['isPartOf']['name'], compat_str),
+            'season_number': int_or_none(try_get(
+                media, lambda x: x['isPartOf']['numeroStagioni'])),
+            'season': media.get('stagione') or None,
+            'subtitles': subtitles,
+        }
+
+        info.update(relinker_info)
+
+        return info
+
+
+class RaiIE(RaiBaseIE):
+    _VALID_URL = r'https?://[^/]+\.(?:rai\.(?:it|tv)|rainews\.it)/dl/.+?-(?P<id>%s)(?:-.+?)?\.html' % RaiBaseIE._UUID_RE
+    _TESTS = [{
+        # var uniquename = "ContentItem-..."
+        # data-id="ContentItem-..."
+        'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
+        'info_dict': {
+            'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
+            'ext': 'mp4',
+            'title': 'TG PRIMO TEMPO',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 1758,
+            'upload_date': '20140612',
+        }
+    }, {
+        # with ContentItem in many metas
+        'url': 'http://www.rainews.it/dl/rainews/media/Weekend-al-cinema-da-Hollywood-arriva-il-thriller-di-Tate-Taylor-La-ragazza-del-treno-1632c009-c843-4836-bb65-80c33084a64b.html',
+        'info_dict': {
+            'id': '1632c009-c843-4836-bb65-80c33084a64b',
+            'ext': 'mp4',
+            'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
+            'description': 'I film in uscita questa settimana.',
+            'thumbnail': r're:^https?://.*\.png$',
+            'duration': 833,
+            'upload_date': '20161103',
+        }
+    }, {
+        # with ContentItem in og:url
+        'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
+        'md5': '11959b4e44fa74de47011b5799490adf',
+        'info_dict': {
+            'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
+            'ext': 'mp4',
+            'title': 'TG1 ore 20:00 del 03/11/2016',
+            'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2214,
+            'upload_date': '20161103',
+        }
+    }, {
+        # drawMediaRaiTV(...)
+        'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
+        'md5': '2dd727e61114e1ee9c47f0da6914e178',
+        'info_dict': {
+            'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
+            'ext': 'mp4',
+            'title': 'Il pacco',
+            'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'upload_date': '20141221',

         },
         },

-        {
-            'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html',
-            'md5': '02b64456f7cc09f96ff14e7dd489017e',
-            'info_dict': {
-                'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6',
-                'ext': 'flv',
-                'title': 'Il Candidato - Primo episodio: "Le Primarie"',
-                'description': 'Primo appuntamento con "Il candidato" con Filippo Timi, alias Piero Zucca presidente!',
-                'uploader': 'RaiTre',
-            }
+    }, {
+        # initEdizione('ContentItem-...'
+        'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
+        'info_dict': {
+            'id': 'c2187016-8484-4e3a-8ac8-35e475b07303',
+            'ext': 'mp4',
+            'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
+            'duration': 2274,
+            'upload_date': '20170401',
+        },
+        'skip': 'Changes daily',
+    }, {
+        # HDS live stream with only relinker URL
+        'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
+        'info_dict': {
+            'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
+            'ext': 'flv',
+            'title': 'EuroNews',

         },
         },

-        {
-            'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
-            'md5': '037104d2c14132887e5e4cf114569214',
-            'info_dict': {
-                'id': '0c7a664b-d0f4-4b2c-8835-3f82e46f433e',
-                'ext': 'flv',
-                'title': 'Il pacco',
-                'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
-                'uploader': 'RaiTre',
-                'upload_date': '20141221',
-            },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # HLS live stream with ContentItem in og:url
+        'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
+        'info_dict': {
+            'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
+            'ext': 'mp4',
+            'title': 'La diretta di Rainews24',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
+
+    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')
+
+        title = media['name'].strip()
+
+        media_type = media['type']
+        if 'Audio' in media_type:
+            relinker_info = {
+                'formats': {
+                    'format_id': media.get('formatoAudio'),
+                    'url': media['audioUrl'],
+                    'ext': media.get('formatoAudio'),
+                }
+            }
+        elif 'Video' in media_type:
+            relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
+        else:
+            raise ExtractorError('not a media file')
+
+        self._sort_formats(relinker_info['formats'])
+
+        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),
+                })
+
+        subtitles = self._extract_subtitles(url, media.get('subtitlesUrl'))
+
+        info = {
+            'id': content_id,
+            'title': title,
+            'description': strip_or_none(media.get('desc')),
+            'thumbnails': thumbnails,
+            'uploader': media.get('author'),
+            'upload_date': unified_strdate(media.get('date')),
+            'duration': parse_duration(media.get('length')),
+            'subtitles': subtitles,

         }
         }

-    ]

 
 

-    def _extract_relinker_url(self, webpage):
-        return self._proto_relative_url(self._search_regex(
-            [r'name="videourl" content="([^"]+)"', r'var\s+videoURL(?:_MP4)?\s*=\s*"([^"]+)"'],
-            webpage, 'relinker url', default=None))
+        info.update(relinker_info)
+
+        return info

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

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        host = mobj.group('host')
+        video_id = self._match_id(url)

 
         webpage = self._download_webpage(url, video_id)
 
 
         webpage = self._download_webpage(url, video_id)
 

-        relinker_url = self._extract_relinker_url(webpage)
-
-        if not relinker_url:
-            iframe_url = self._search_regex(
-                [r'<iframe[^>]+src="([^"]*/dl/[^"]+\?iframe\b[^"]*)"',
-                 r'drawMediaRaiTV\(["\'](.+?)["\']'],
-                webpage, 'iframe')
-            if not iframe_url.startswith('http'):
-                iframe_url = compat_urlparse.urljoin(url, iframe_url)
-            webpage = self._download_webpage(
-                iframe_url, video_id)
-            relinker_url = self._extract_relinker_url(webpage)
-
-        relinker = self._download_json(
-            '%s&output=47' % relinker_url, video_id)
-
-        media_url = relinker['video'][0]
-        ct = relinker.get('ct')
-        if ct == 'f4m':
-            formats = self._extract_f4m_formats(
-                media_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id)
-        else:
-            formats = [{
-                'url': media_url,
-                'format_id': ct,
-            }]
+        content_item_id = None

 
 

-        json_link = self._html_search_meta(
-            'jsonlink', webpage, 'JSON link', default=None)
-        if json_link:
-            media = self._download_json(
-                host + json_link, video_id, 'Downloading video JSON')
-            title = media.get('name')
-            description = media.get('desc')
-            thumbnail = media.get('image_300') or media.get('image_medium') or media.get('image')
-            duration = parse_duration(media.get('length'))
-            uploader = media.get('author')
-            upload_date = unified_strdate(media.get('date'))
-        else:
-            title = (self._search_regex(
-                r'var\s+videoTitolo\s*=\s*"(.+?)";',
-                webpage, 'title', default=None) or self._og_search_title(webpage)).replace('\\"', '"')
-            description = self._og_search_description(webpage)
-            thumbnail = self._og_search_thumbnail(webpage)
-            duration = None
-            uploader = self._html_search_meta('Editore', webpage, 'uploader')
-            upload_date = unified_strdate(self._html_search_meta(
-                'item-date', webpage, 'upload date', default=None))
-
-        subtitles = self.extract_subtitles(video_id, webpage)
-
-        return {
+        content_item_url = self._html_search_meta(
+            ('og:url', 'og:video', 'og:video:secure_url', 'twitter:url',
+             'twitter:player', 'jsonlink'), webpage, default=None)
+        if content_item_url:
+            content_item_id = self._search_regex(
+                r'ContentItem-(%s)' % self._UUID_RE, content_item_url,
+                'content item id', default=None)
+
+        if not content_item_id:
+            content_item_id = self._search_regex(
+                r'''(?x)
+                    (?:
+                        (?:initEdizione|drawMediaRaiTV)\(|
+                        <(?:[^>]+\bdata-id|var\s+uniquename)=
+                    )
+                    (["\'])
+                    (?:(?!\1).)*\bContentItem-(?P<id>%s)
+                ''' % self._UUID_RE,
+                webpage, 'content item id', default=None, group='id')
+
+        content_item_ids = set()
+        if content_item_id:
+            content_item_ids.add(content_item_id)
+        if video_id not in content_item_ids:
+            content_item_ids.add(video_id)
+
+        for content_item_id in content_item_ids:
+            try:
+                return self._extract_from_content_id(content_item_id, url)
+            except GeoRestrictedError:
+                raise
+            except ExtractorError:
+                pass
+
+        relinker_url = self._search_regex(
+            r'''(?x)
+                (?:
+                    var\s+videoURL|
+                    mediaInfo\.mediaUri
+                )\s*=\s*
+                ([\'"])
+                (?P<url>
+                    (?:https?:)?
+                    //mediapolis(?:vod)?\.rai\.it/relinker/relinkerServlet\.htm\?
+                    (?:(?!\1).)*\bcont=(?:(?!\1).)+)\1
+            ''',
+            webpage, 'relinker URL', group='url')
+
+        relinker_info = self._extract_relinker_info(
+            urljoin(url, relinker_url), video_id)
+        self._sort_formats(relinker_info['formats'])
+
+        title = self._search_regex(
+            r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
+            webpage, 'title', group='title',
+            default=None) or self._og_search_title(webpage)
+
+        info = {

             'id': video_id,
             'title': title,
             'id': video_id,
             'title': title,

-            'description': description,
-            'thumbnail': thumbnail,
-            'uploader': uploader,
-            'upload_date': upload_date,
-            'duration': duration,
-            'formats': formats,
-            'subtitles': subtitles,

         }
 
         }
 

-    def _get_subtitles(self, video_id, webpage):
-        subtitles = {}
-        m = re.search(r'<meta name="closedcaption" content="(?P<captions>[^"]+)"', webpage)
-        if m:
-            captions = m.group('captions')
-            STL_EXT = '.stl'
-            SRT_EXT = '.srt'
-            if captions.endswith(STL_EXT):
-                captions = captions[:-len(STL_EXT)] + SRT_EXT
-            subtitles['it'] = [{
-                'ext': 'srt',
-                'url': 'http://www.rai.tv%s' % compat_urllib_parse.quote(captions),
-            }]
-        return subtitles
+        info.update(relinker_info)
+
+        return info