Merge tag 'upstream/2016.08.17'

[youtubedl] / youtube_dl / extractor / nrk.py

diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py
index f6de260222c678e2233b668d4b557e22e51d224c..6ded5bd456fa86bf16e1762601889b46f2d68fe9 100644 (file)
--- a/youtube_dl/extractor/nrk.py
+++ b/youtube_dl/extractor/nrk.py
@@ -4,257 +4,290 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

+from ..compat import compat_urllib_parse_unquote

 from ..utils import (
     ExtractorError,
 from ..utils import (
     ExtractorError,

-    float_or_none,
+    int_or_none,
+    parse_age_limit,

     parse_duration,
     parse_duration,

-    unified_strdate,

 )
 )

-from .subtitles import SubtitlesInfoExtractor

 
 
 
 

-class NRKIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?nrk\.no/(?:video|lyd)/[^/]+/(?P<id>[\dA-F]{16})'
-
-    _TESTS = [
-        {
-            'url': 'http://www.nrk.no/video/dompap_og_andre_fugler_i_piip_show/D0FA54B5C8B6CE59/emne/piipshow/',
-            'md5': 'a6eac35052f3b242bb6bb7f43aed5886',
-            'info_dict': {
-                'id': '150533',
-                'ext': 'flv',
-                'title': 'Dompap og andre fugler i Piip-Show',
-                'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f'
-            }
-        },
-        {
-            'url': 'http://www.nrk.no/lyd/lyd_av_oppleser_for_blinde/AEFDDD5473BA0198/',
-            'md5': '3471f2a51718195164e88f46bf427668',
-            'info_dict': {
-                'id': '154915',
-                'ext': 'flv',
-                'title': 'Slik høres internett ut når du er blind',
-                'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
-            }
-        },
-    ]
+class NRKBaseIE(InfoExtractor):
+    def _extract_formats(self, manifest_url, video_id, fatal=True):
+        formats = []
+        formats.extend(self._extract_f4m_formats(
+            manifest_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81',
+            video_id, f4m_id='hds', fatal=fatal))
+        formats.extend(self._extract_m3u8_formats(manifest_url.replace(
+            'akamaihd.net/z/', 'akamaihd.net/i/').replace('/manifest.f4m', '/master.m3u8'),
+            video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=fatal))
+        return formats

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

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

 
 

-        page = self._download_webpage(url, video_id)
+        data = self._download_json(
+            'http://%s/mediaelement/%s' % (self._API_HOST, video_id),
+            video_id, 'Downloading mediaelement JSON')

 
 

-        video_id = self._html_search_regex(r'<div class="nrk-video" data-nrk-id="(\d+)">', page, 'video id')
+        title = data.get('fullTitle') or data.get('mainTitle') or data['title']
+        video_id = data.get('id') or video_id

 
 

-        data = self._download_json(
-            'http://v7.psapi.nrk.no/mediaelement/%s' % video_id, video_id, 'Downloading media JSON')
+        entries = []
+
+        media_assets = data.get('mediaAssets')
+        if media_assets and isinstance(media_assets, list):
+            def video_id_and_title(idx):
+                return ((video_id, title) if len(media_assets) == 1
+                        else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
+            for num, asset in enumerate(media_assets, 1):
+                asset_url = asset.get('url')
+                if not asset_url:
+                    continue
+                formats = self._extract_formats(asset_url, video_id, fatal=False)
+                if not formats:
+                    continue
+                self._sort_formats(formats)
+                entry_id, entry_title = video_id_and_title(num)
+                duration = parse_duration(asset.get('duration'))
+                subtitles = {}
+                for subtitle in ('webVtt', 'timedText'):
+                    subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
+                    if subtitle_url:
+                        subtitles.setdefault('no', []).append({
+                            'url': compat_urllib_parse_unquote(subtitle_url)
+                        })
+                entries.append({
+                    'id': asset.get('carrierId') or entry_id,
+                    'title': entry_title,
+                    'duration': duration,
+                    'subtitles': subtitles,
+                    'formats': formats,
+                })

 
 

-        if data['usageRights']['isGeoBlocked']:
-            raise ExtractorError('NRK har ikke rettig-heter til å vise dette programmet utenfor Norge', expected=True)
+        if not entries:
+            media_url = data.get('mediaUrl')
+            if media_url:
+                formats = self._extract_formats(media_url, video_id)
+                self._sort_formats(formats)
+                duration = parse_duration(data.get('duration'))
+                entries = [{
+                    'id': video_id,
+                    'title': title,
+                    'duration': duration,
+                    'formats': formats,
+                }]
+
+        if not entries:
+            if data.get('usageRights', {}).get('isGeoBlocked'):
+                raise ExtractorError(
+                    'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
+                    expected=True)

 
 

-        video_url = data['mediaUrl'] + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124'
+        conviva = data.get('convivaStatistics') or {}
+        series = conviva.get('seriesName') or data.get('seriesTitle')
+        episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')

 
 

+        thumbnails = None

         images = data.get('images')
         images = data.get('images')

-        if images:
-            thumbnails = images['webImages']
-            thumbnails.sort(key=lambda image: image['pixelWidth'])
-            thumbnail = thumbnails[-1]['imageUrl']
-        else:
-            thumbnail = None
-
-        return {
-            'id': video_id,
-            'url': video_url,
+        if images and isinstance(images, dict):
+            web_images = images.get('webImages')
+            if isinstance(web_images, list):
+                thumbnails = [{
+                    'url': image['imageUrl'],
+                    'width': int_or_none(image.get('width')),
+                    'height': int_or_none(image.get('height')),
+                } for image in web_images if image.get('imageUrl')]
+
+        description = data.get('description')
+
+        common_info = {
+            'description': description,
+            'series': series,
+            'episode': episode,
+            'age_limit': parse_age_limit(data.get('legalAge')),
+            'thumbnails': thumbnails,
+        }
+
+        vcodec = 'none' if data.get('mediaType') == 'Audio' else None
+
+        # TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged
+
+        for entry in entries:
+            entry.update(common_info)
+            for f in entry['formats']:
+                f['vcodec'] = vcodec
+
+        return self.playlist_result(entries, video_id, title, description)
+
+
+class NRKIE(NRKBaseIE):
+    _VALID_URL = r'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
+    _API_HOST = 'v8.psapi.nrk.no'
+    _TESTS = [{
+        # video
+        'url': 'http://www.nrk.no/video/PS*150533',
+        'md5': '2f7f6eeb2aacdd99885f355428715cfa',
+        'info_dict': {
+            'id': '150533',
+            'ext': 'mp4',
+            'title': 'Dompap og andre fugler i Piip-Show',
+            'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
+            'duration': 263,
+        }
+    }, {
+        # audio
+        'url': 'http://www.nrk.no/video/PS*154915',
+        # MD5 is unstable
+        'info_dict': {
+            'id': '154915',

             'ext': 'flv',
             'ext': 'flv',

-            'title': data['title'],
-            'description': data['description'],
-            'thumbnail': thumbnail,
+            'title': 'Slik høres internett ut når du er blind',
+            'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
+            'duration': 20,

         }
         }

+    }]

 
 
 
 

-class NRKTVIE(SubtitlesInfoExtractor):
-    _VALID_URL = r'(?P<baseurl>http://tv\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
+class NRKTVIE(NRKBaseIE):
+    IE_DESC = 'NRK TV and NRK Radio'
+    _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
+    _API_HOST = 'psapi-we.nrk.no'

 
 

-    _TESTS = [
-        {
-            'url': 'http://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
-            'md5': 'adf2c5454fa2bf032f47a9f8fb351342',
-            'info_dict': {
-                'id': 'MUHH48000314',
-                'ext': 'flv',
-                'title': '20 spørsmål',
-                'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
-                'upload_date': '20140523',
-                'duration': 1741.52,
-            },
+    _TESTS = [{
+        'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
+        'md5': '4e9ca6629f09e588ed240fb11619922a',
+        'info_dict': {
+            'id': 'MUHH48000314AA',
+            'ext': 'mp4',
+            'title': '20 spørsmål 23.05.2014',
+            'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
+            'duration': 1741,

         },
         },

-        {
-            'url': 'http://tv.nrk.no/program/mdfp15000514',
-            'md5': '383650ece2b25ecec996ad7b5bb2a384',
+    }, {
+        'url': 'https://tv.nrk.no/program/mdfp15000514',
+        'md5': '43d0be26663d380603a9cf0c24366531',
+        'info_dict': {
+            'id': 'MDFP15000514CA',
+            'ext': 'mp4',
+            'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
+            'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
+            'duration': 4605,
+        },
+    }, {
+        # single playlist video
+        'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
+        'md5': 'adbd1dbd813edaf532b0a253780719c2',
+        'info_dict': {
+            'id': 'MSPO40010515-part2',
+            'ext': 'flv',
+            'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
+            'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
+        },
+        'skip': 'Only works from Norway',
+    }, {
+        'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
+        'playlist': [{
+            'md5': '9480285eff92d64f06e02a5367970a7a',

             'info_dict': {
             'info_dict': {

-                'id': 'mdfp15000514',
+                'id': 'MSPO40010515-part1',

                 'ext': 'flv',
                 'ext': 'flv',

-                'title': 'Kunnskapskanalen: Grunnlovsjubiléet - Stor ståhei for ingenting',
-                'description': 'md5:654c12511f035aed1e42bdf5db3b206a',
-                'upload_date': '20140524',
-                'duration': 4605.0,
+                'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
+                'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',

             },
             },

-        },
-        {
-            # single playlist video
-            'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
+        }, {

             'md5': 'adbd1dbd813edaf532b0a253780719c2',
             'info_dict': {
                 'id': 'MSPO40010515-part2',
                 'ext': 'flv',
                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
             'md5': 'adbd1dbd813edaf532b0a253780719c2',
             'info_dict': {
                 'id': 'MSPO40010515-part2',
                 'ext': 'flv',
                 'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
                 'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',

-                'upload_date': '20150106',

             },
             },

-            'skip': 'Only works from Norway',
+        }],
+        'info_dict': {
+            'id': 'MSPO40010515',
+            'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
+            'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
+            'duration': 6947.52,

         },
         },

-        {
-            'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
-            'playlist': [
-                {
-                    'md5': '9480285eff92d64f06e02a5367970a7a',
-                    'info_dict': {
-                        'id': 'MSPO40010515-part1',
-                        'ext': 'flv',
-                        'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
-                        'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
-                        'upload_date': '20150106',
-                    },
-                },
-                {
-                    'md5': 'adbd1dbd813edaf532b0a253780719c2',
-                    'info_dict': {
-                        'id': 'MSPO40010515-part2',
-                        'ext': 'flv',
-                        'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
-                        'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
-                        'upload_date': '20150106',
-                    },
-                },
-            ],
-            'info_dict': {
-                'id': 'MSPO40010515',
-                'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
-                'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
-                'upload_date': '20150106',
-                'duration': 6947.5199999999995,
-            },
-            'skip': 'Only works from Norway',
-        }
-    ]
-
-    def _seconds2str(self, s):
-        return '%02d:%02d:%02d.%03d' % (s / 3600, (s % 3600) / 60, s % 60, (s % 1) * 1000)
-
-    def _debug_print(self, txt):
-        if self._downloader.params.get('verbose', False):
-            self.to_screen('[debug] %s' % txt)
-
-    def _extract_captions(self, subtitlesurl, video_id, baseurl):
-        url = "%s%s" % (baseurl, subtitlesurl)
-        self._debug_print('%s: Subtitle url: %s' % (video_id, url))
-        captions = self._download_xml(url, video_id, 'Downloading subtitles')
-        lang = captions.get('lang', 'no')
-        ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/ns/ttml}'))
-        srt = ''
-        for pos, p in enumerate(ps):
-            begin = parse_duration(p.get('begin'))
-            duration = parse_duration(p.get('dur'))
-            starttime = self._seconds2str(begin)
-            endtime = self._seconds2str(begin + duration)
-            text = '\n'.join(p.itertext())
-            srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), starttime, endtime, text)
-        return {lang: srt}
-
-    def _extract_f4m(self, manifest_url, video_id):
-        return self._extract_f4m_formats(manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id)
+        'skip': 'Only works from Norway',
+    }, {
+        'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
+        'only_matching': True,
+    }]
+
+
+class NRKPlaylistIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
+
+    _TESTS = [{
+        'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
+        'info_dict': {
+            'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
+            'title': 'Gjenopplev den historiske solformørkelsen',
+            'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
+        },
+        'playlist_count': 2,
+    }, {
+        'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
+        'info_dict': {
+            'id': 'rivertonprisen-til-karin-fossum-1.12266449',
+            'title': 'Rivertonprisen til Karin Fossum',
+            'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
+        },
+        'playlist_count': 5,
+    }]

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

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        part_id = mobj.group('part_id')
-        baseurl = mobj.group('baseurl')
-
-        webpage = self._download_webpage(url, video_id)
-
-        title = self._html_search_meta(
-            'title', webpage, 'title')
-        description = self._html_search_meta(
-            'description', webpage, 'description')
-
-        thumbnail = self._html_search_regex(
-            r'data-posterimage="([^"]+)"',
-            webpage, 'thumbnail', fatal=False)
-        upload_date = unified_strdate(self._html_search_meta(
-            'rightsfrom', webpage, 'upload date', fatal=False))
-        duration = float_or_none(self._html_search_regex(
-            r'data-duration="([^"]+)"',
-            webpage, 'duration', fatal=False))
-
-        # playlist
-        parts = re.findall(
-            r'<a href="#del=(\d+)"[^>]+data-argument="([^"]+)">([^<]+)</a>', webpage)
-        if parts:
-            entries = []
-            for current_part_id, stream_url, part_title in parts:
-                if part_id and current_part_id != part_id:
-                    continue
-                video_part_id = '%s-part%s' % (video_id, current_part_id)
-                formats = self._extract_f4m(stream_url, video_part_id)
-                entries.append({
-                    'id': video_part_id,
-                    'title': part_title,
-                    'description': description,
-                    'thumbnail': thumbnail,
-                    'upload_date': upload_date,
-                    'formats': formats,
-                })
-            if part_id:
-                if entries:
-                    return entries[0]
-            else:
-                playlist = self.playlist_result(entries, video_id, title, description)
-                playlist.update({
-                    'thumbnail': thumbnail,
-                    'upload_date': upload_date,
-                    'duration': duration,
-                })
-                return playlist
+        playlist_id = self._match_id(url)

 
 

-        formats = []
+        webpage = self._download_webpage(url, playlist_id)

 
 

-        f4m_url = re.search(r'data-media="([^"]+)"', webpage)
-        if f4m_url:
-            formats.extend(self._extract_f4m(f4m_url.group(1), video_id))
-
-        m3u8_url = re.search(r'data-hls-media="([^"]+)"', webpage)
-        if m3u8_url:
-            formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4'))
-        self._sort_formats(formats)
-
-        subtitles_url = self._html_search_regex(
-            r'data-subtitlesurl[ ]*=[ ]*"([^"]+)"',
-            webpage, 'subtitle URL', default=None)
-        subtitles = None
-        if subtitles_url:
-            subtitles = self._extract_captions(subtitles_url, video_id, baseurl)
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, subtitles)
-            return
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'upload_date': upload_date,
-            'duration': duration,
-            'formats': formats,
-            'subtitles': subtitles,
-        }
+        entries = [
+            self.url_result('nrk:%s' % video_id, 'NRK')
+            for video_id in re.findall(
+                r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"',
+                webpage)
+        ]
+
+        playlist_title = self._og_search_title(webpage)
+        playlist_description = self._og_search_description(webpage)
+
+        return self.playlist_result(
+            entries, playlist_id, playlist_title, playlist_description)
+
+
+class NRKSkoleIE(InfoExtractor):
+    IE_DESC = 'NRK Skole'
+    _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
+
+    _TESTS = [{
+        'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
+        'md5': '6bc936b01f9dd8ed45bc58b252b2d9b6',
+        'info_dict': {
+            'id': '6021',
+            'ext': 'mp4',
+            'title': 'Genetikk og eneggede tvillinger',
+            'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
+            'duration': 399,
+        },
+    }, {
+        'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
+            video_id)
+
+        nrk_id = self._parse_json(
+            self._search_regex(
+                r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
+                webpage, 'application json'),
+            video_id)['activeMedia']['psId']
+
+        return self.url_result('nrk:%s' % nrk_id)