X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fphilharmoniedeparis.py;h=03da64b116128f01893f047e747a1fb40af1e2df;hb=adc13b0748784db87faef580929b1adadaff732e;hp=7fc2f32ab8ce86649162895202d918c6ecbc56c3;hpb=06d07c400088d35096df1406a3434c996795250d;p=youtube-dl

diff --git a/youtube_dl/extractor/philharmoniedeparis.py b/youtube_dl/extractor/philharmoniedeparis.py
index 7fc2f32ab..03da64b11 100644
--- a/youtube_dl/extractor/philharmoniedeparis.py
+++ b/youtube_dl/extractor/philharmoniedeparis.py
@@ -2,76 +2,105 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
-    determine_ext,
-    int_or_none,
-    parse_iso8601,
-    unified_strdate,
+    try_get,
+    urljoin,
 )
 
+
 class PhilharmonieDeParisIE(InfoExtractor):
-    _VALID_URL = r'http://live\.philharmoniedeparis\.fr/concert/(?P<id>\d+)(?:/|\.html)'
+    IE_DESC = 'Philharmonie de Paris'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:
+                            live\.philharmoniedeparis\.fr/(?:[Cc]oncert/|embed(?:app)?/|misc/Playlist\.ashx\?id=)|
+                            pad\.philharmoniedeparis\.fr/doc/CIMU/
+                        )
+                        (?P<id>\d+)
+                    '''
     _TESTS = [{
+        'url': 'http://pad.philharmoniedeparis.fr/doc/CIMU/1086697/jazz-a-la-villette-knower',
+        'md5': 'a0a4b195f544645073631cbec166a2c2',
+        'info_dict': {
+            'id': '1086697',
+            'ext': 'mp4',
+            'title': 'Jazz à la Villette : Knower',
+        },
+    }, {
         'url': 'http://live.philharmoniedeparis.fr/concert/1032066.html',
         'info_dict': {
             'id': '1032066',
-            'ext': 'mp4',
-            'title': "Week-end Bach. Passion selon saint Jean. Akademie für alte Musik Berlin, Rias Kammerchor, René Jacobs",
-            'upload_date': '20150404',
-        }
+            'title': 'md5:0a031b81807b3593cffa3c9a87a167a0',
+        },
+        'playlist_mincount': 2,
+    }, {
+        'url': 'http://live.philharmoniedeparis.fr/Concert/1030324.html',
+        'only_matching': True,
+    }, {
+        'url': 'http://live.philharmoniedeparis.fr/misc/Playlist.ashx?id=1030324&track=&lang=fr',
+        'only_matching': True,
+    }, {
+        'url': 'https://live.philharmoniedeparis.fr/embedapp/1098406/berlioz-fantastique-lelio-les-siecles-national-youth-choir-of.html?lang=fr-FR',
+        'only_matching': True,
+    }, {
+        'url': 'https://live.philharmoniedeparis.fr/embed/1098406/berlioz-fantastique-lelio-les-siecles-national-youth-choir-of.html?lang=fr-FR',
+        'only_matching': True,
     }]
+    _LIVE_URL = 'https://live.philharmoniedeparis.fr'
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
-        fichier_nom = self._html_search_regex(r'\sflashvars\s*:\s*\{\s*fichier_nom\s*:\s*\'(.*?)\'\s*,', webpage, 'fichier_nom')
-
-        playlist = self._download_xml('http://live.philharmoniedeparis.fr' + fichier_nom, video_id)
-
-        concert = playlist.find('.//concert')
-
-        formats = []
-        info_dict = {
-            'id': video_id,
-            'title': concert.find('./titre').text,
-            'formats': formats,
-        }
+        config = self._download_json(
+            '%s/otoPlayer/config.ashx' % self._LIVE_URL, video_id, query={
+                'id': video_id,
+                'lang': 'fr-FR',
+            })
 
-        if concert.attrib.get('heure'):
-            info_dict['timestamp'] = parse_iso8601(('%s-%s-%s%s') % (
-                concert.attrib['date'][0:4],
-                concert.attrib['date'][4:6],
-                concert.attrib['date'][6:8],
-                concert.attrib['heure']
-            ))
-        else:
-            info_dict['upload_date'] = concert.attrib['date']
+        def extract_entry(source):
+            if not isinstance(source, dict):
+                return
+            title = source.get('title')
+            if not title:
+                return
+            files = source.get('files')
+            if not isinstance(files, dict):
+                return
+            format_urls = set()
+            formats = []
+            for format_id in ('mobile', 'desktop'):
+                format_url = try_get(
+                    files, lambda x: x[format_id]['file'], compat_str)
+                if not format_url or format_url in format_urls:
+                    continue
+                format_urls.add(format_url)
+                m3u8_url = urljoin(self._LIVE_URL, format_url)
+                formats.extend(self._extract_m3u8_formats(
+                    m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            if not formats:
+                return
+            self._sort_formats(formats)
+            return {
+                'title': title,
+                'formats': formats,
+            }
 
-        fichiers = concert.find('./fichiers')
-        for fichier in fichiers.findall('./fichier'):
-            # Sometimes <ficher>s have no attributes at all. Skip them.
-            if 'url' not in fichier.attrib:
-                continue
+        thumbnail = urljoin(self._LIVE_URL, config.get('image'))
 
-            formats.append({
-                'format_id': 'lq',
-                'url': fichiers.attrib['serveurstream'],
-                'ext': determine_ext(fichier.attrib['url']),
-                'play_path': fichier.attrib['url'],
-                'width': int_or_none(concert.attrib['largeur']),
-                'height': int_or_none(concert.attrib['hauteur']),
-                'quality': 1,
+        info = extract_entry(config)
+        if info:
+            info.update({
+                'id': video_id,
+                'thumbnail': thumbnail,
             })
+            return info
 
-            formats.append({
-                'format_id': 'hq',
-                'url': fichiers.attrib['serveurstream'],
-                'ext': determine_ext(fichier.attrib['url_hd']),
-                'play_path': fichier.attrib['url_hd'],
-                'width': int_or_none(concert.attrib['largeur_hd']),
-                'height': int_or_none(concert.attrib['hauteur_hd']),
-                'quality': 2,
-            })
+        entries = []
+        for num, chapter in enumerate(config['chapters'], start=1):
+            entry = extract_entry(chapter)
+            entry['id'] = '%s-%d' % (video_id, num)
+            entries.append(entry)
 
-        return info_dict
+        return self.playlist_result(entries, video_id, config.get('title'))