[franceculture] Fix extraction (Closes #10324)

youtube_dl/extractor/extractors.py

@@ -272,10 +272,7 @@ from .fox import FOXIE
 from .foxgay import FoxgayIE
 from .foxnews import FoxNewsIE
 from .foxsports import FoxSportsIE
-from .franceculture import (
+from .franceculture import FranceCultureIE
-    FranceCultureIE,
-    FranceCultureEmissionIE,
-)
 from .franceinter import FranceInterIE
 from .francetv import (
     PluzzIE,

youtube_dl/extractor/franceculture.py

@@ -2,104 +2,56 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urlparse,
-)
 from ..utils import (
     determine_ext,
-    int_or_none,
+    unified_strdate,
-    ExtractorError,
 )
 
 
 class FranceCultureIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?franceculture\.fr/player/reecouter\?play=(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?franceculture\.fr/emissions/(?:[^/]+/)*(?P<id>[^/?#&]+)'
     _TEST = {
-        'url': 'http://www.franceculture.fr/player/reecouter?play=4795174',
+        'url': 'http://www.franceculture.fr/emissions/carnet-nomade/rendez-vous-au-pays-des-geeks',
         'info_dict': {
-            'id': '4795174',
+            'id': 'rendez-vous-au-pays-des-geeks',
+            'display_id': 'rendez-vous-au-pays-des-geeks',
             'ext': 'mp3',
             'title': 'Rendez-vous au pays des geeks',
-            'alt_title': 'Carnet nomade | 13-14',
+            'thumbnail': 're:^https?://.*\\.jpg$',
-            'vcodec': 'none',
             'upload_date': '20140301',
-            'thumbnail': r're:^http://static\.franceculture\.fr/.*/images/player/Carnet-nomade\.jpg$',
+            'vcodec': 'none',
-            'description': 'startswith:Avec :Jean-Baptiste Péretié pour son documentaire sur Arte "La revanche',
-            'timestamp': 1393700400,
         }
     }
 
-    def _extract_from_player(self, url, video_id):
+    def _real_extract(self, url):
-        webpage = self._download_webpage(url, video_id)
+        display_id = self._match_id(url)
 
-        video_path = self._search_regex(
+        webpage = self._download_webpage(url, display_id)
-            r'<a id="player".*?href="([^"]+)"', webpage, 'video path')
-        video_url = compat_urlparse.urljoin(url, video_path)
-        timestamp = int_or_none(self._search_regex(
-            r'<a id="player".*?data-date="([0-9]+)"',
-            webpage, 'upload date', fatal=False))
-        thumbnail = self._search_regex(
-            r'<a id="player".*?>\s+<img src="([^"]+)"',
-            webpage, 'thumbnail', fatal=False)
 
-        display_id = self._search_regex(
+        video_url = self._search_regex(
-            r'<span class="path-diffusion">emission-(.*?)</span>', webpage, 'display_id')
+            r'(?s)<div[^>]+class="[^"]*?title-zone-diffusion[^"]*?"[^>]*>.*?<a[^>]+href="([^"]+)"',
+            webpage, 'video path')
 
-        title = self._html_search_regex(
+        title = self._og_search_title(webpage)
-            r'<span class="title-diffusion">(.*?)</span>', webpage, 'title')
-        alt_title = self._html_search_regex(
-            r'<span class="title">(.*?)</span>',
-            webpage, 'alt_title', fatal=False)
-        description = self._html_search_regex(
-            r'<span class="description">(.*?)</span>',
-            webpage, 'description', fatal=False)
 
+        upload_date = unified_strdate(self._search_regex(
+            '(?s)<div[^>]+class="date"[^>]*>.*?<span[^>]+class="inner"[^>]*>([^<]+)<',
+            webpage, 'upload date', fatal=False))
+        thumbnail = self._search_regex(
+            r'(?s)<figure[^>]+itemtype="https://schema.org/ImageObject"[^>]*>.*?<img[^>]+data-pagespeed-(?:lazy|high-res)-src="([^"]+)"',
+            webpage, 'thumbnail', fatal=False)
         uploader = self._html_search_regex(
             r'(?s)<div id="emission".*?<span class="author">(.*?)</span>',
             webpage, 'uploader', default=None)
         vcodec = 'none' if determine_ext(video_url.lower()) == 'mp3' else None
 
         return {
-            'id': video_id,
+            'id': display_id,
+            'display_id': display_id,
             'url': video_url,
-            'vcodec': vcodec,
-            'uploader': uploader,
-            'timestamp': timestamp,
             'title': title,
-            'alt_title': alt_title,
             'thumbnail': thumbnail,
-            'description': description,
+            'vcodec': vcodec,
-            'display_id': display_id,
+            'uploader': uploader,
+            'upload_date': upload_date,
         }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        return self._extract_from_player(url, video_id)
-
-
-class FranceCultureEmissionIE(FranceCultureIE):
-    _VALID_URL = r'https?://(?:www\.)?franceculture\.fr/emission-(?P<id>[^?#]+)'
-    _TEST = {
-        'url': 'http://www.franceculture.fr/emission-les-carnets-de-la-creation-jean-gabriel-periot-cineaste-2015-10-13',
-        'info_dict': {
-            'title': 'Jean-Gabriel Périot, cinéaste',
-            'alt_title': 'Les Carnets de la création',
-            'id': '5093239',
-            'display_id': 'les-carnets-de-la-creation-jean-gabriel-periot-cineaste-2015-10-13',
-            'ext': 'mp3',
-            'timestamp': 1444762500,
-            'upload_date': '20151013',
-            'description': 'startswith:Aujourd\'hui dans "Les carnets de la création", le cinéaste',
-        },
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-        video_path = self._html_search_regex(
-            r'<a class="rf-player-open".*?href="([^"]+)"', webpage, 'video path', 'no_path_player')
-        if video_path == 'no_path_player':
-            raise ExtractorError('no player : no sound in this page.', expected=True)
-        new_id = self._search_regex('play=(?P<id>[0-9]+)', video_path, 'new_id', group='id')
-        video_url = compat_urlparse.urljoin(url, video_path)
-        return self._extract_from_player(video_url, new_id)