New upstream version 2020.03.24

[youtubedl] / youtube_dl / extractor / ina.py

diff --git a/youtube_dl/extractor/ina.py b/youtube_dl/extractor/ina.py
index 962c5921447e72a3f15ce20ca1f8e293acf26c44..12695af27236d02b383fd83fa7664afe16e94656 100644 (file)
--- a/youtube_dl/extractor/ina.py
+++ b/youtube_dl/extractor/ina.py
@@ -1,39 +1,83 @@
-import re
+# coding: utf-8
+from __future__ import unicode_literals

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

+from ..utils import (
+    determine_ext,
+    int_or_none,
+    strip_or_none,
+    xpath_attr,
+    xpath_text,
+)

 
 
 class InaIE(InfoExtractor):
 
 
 class InaIE(InfoExtractor):

-    """Information Extractor for Ina.fr"""
-    _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I[0-9]+)/.*'
-    _TEST = {
-        u'url': u'www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
-        u'file': u'I12055569.mp4',
-        u'md5': u'a667021bf2b41f8dc6049479d9bb38a3',
-        u'info_dict': {
-            u"title": u"Fran\u00e7ois Hollande \"Je crois que c'est clair\""
+    _VALID_URL = r'https?://(?:www\.)?ina\.fr/(?:video|audio)/(?P<id>[A-Z0-9_]+)'
+    _TESTS = [{
+        'url': 'http://www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
+        'md5': 'a667021bf2b41f8dc6049479d9bb38a3',
+        'info_dict': {
+            'id': 'I12055569',
+            'ext': 'mp4',
+            'title': 'François Hollande "Je crois que c\'est clair"',
+            'description': 'md5:3f09eb072a06cb286b8f7e4f77109663',

         }
         }

-    }
+    }, {
+        'url': 'https://www.ina.fr/video/S806544_001/don-d-organes-des-avancees-mais-d-importants-besoins-video.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.ina.fr/audio/P16173408',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.ina.fr/video/P16173408-video.html',
+        'only_matching': True,
+    }]

 
 

-    def _real_extract(self,url):
-        mobj = re.match(self._VALID_URL, url)
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        info_doc = self._download_xml(
+            'http://player.ina.fr/notices/%s.mrss' % video_id, video_id)
+        item = info_doc.find('channel/item')
+        title = xpath_text(item, 'title', fatal=True)
+        media_ns_xpath = lambda x: self._xpath_ns(x, 'http://search.yahoo.com/mrss/')
+        content = item.find(media_ns_xpath('content'))

 
 

-        video_id = mobj.group('id')
-        mrss_url='http://player.ina.fr/notices/%s.mrss' % video_id
-        video_extension = 'mp4'
-        webpage = self._download_webpage(mrss_url, video_id)
+        get_furl = lambda x: xpath_attr(content, media_ns_xpath(x), 'url')
+        formats = []
+        for q, w, h in (('bq', 400, 300), ('mq', 512, 384), ('hq', 768, 576)):
+            q_url = get_furl(q)
+            if not q_url:
+                continue
+            formats.append({
+                'format_id': q,
+                'url': q_url,
+                'width': w,
+                'height': h,
+            })
+        if not formats:
+            furl = get_furl('player') or content.attrib['url']
+            ext = determine_ext(furl)
+            formats = [{
+                'url': furl,
+                'vcodec': 'none' if ext == 'mp3' else None,
+                'ext': ext,
+            }]

 
 

-        self.report_extraction(video_id)
+        thumbnails = []
+        for thumbnail in content.findall(media_ns_xpath('thumbnail')):
+            thumbnail_url = thumbnail.get('url')
+            if not thumbnail_url:
+                continue
+            thumbnails.append({
+                'url': thumbnail_url,
+                'height': int_or_none(thumbnail.get('height')),
+                'width': int_or_none(thumbnail.get('width')),
+            })

 
 

-        video_url = self._html_search_regex(r'<media:player url="(?P<mp4url>http://mp4.ina.fr/[^"]+\.mp4)',
-            webpage, u'video URL')
-
-        video_title = self._search_regex(r'<title><!\[CDATA\[(?P<titre>.*?)]]></title>',
-            webpage, u'title')
-
-        return [{
-            'id':       video_id,
-            'url':      video_url,
-            'ext':      video_extension,
-            'title':    video_title,
-        }]
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': title,
+            'description': strip_or_none(xpath_text(item, 'description')),
+            'thumbnails': thumbnails,
+        }