[radiocanada] switch to the new media requests(closes #19115)

youtube_dl/extractor/radiocanada.py

@@ -5,14 +5,10 @@
 
 from .common import InfoExtractor
 from ..utils import (
-    xpath_text,
-    find_xpath_attr,
     determine_ext,
+    ExtractorError,
     int_or_none,
     unified_strdate,
-    xpath_element,
-    ExtractorError,
-    determine_protocol,
     unsmuggle_url,
 )
 
@@ -61,107 +57,53 @@ class RadioCanadaIE(InfoExtractor):
             'only_matching': True,
         }
     ]
+    _GEO_COUNTRIES = ['CA']
+
+    def _call_api(self, path, video_id, app_code, query):
+        query.update({
+            'appCode': app_code,
+            'idMedia': video_id,
+            'output': 'json',
+        })
+        return self._download_json(
+            'https://services.radio-canada.ca/media/' + path, video_id, headers={
+                'Authorization': 'Client-Key 773aea60-0e80-41bb-9c7f-e6d7c3ad17fb'
+            }, query=query)
 
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
         app_code, video_id = re.match(self._VALID_URL, url).groups()
 
-        metadata = self._download_xml(
+        metas = self._call_api('meta/v1/index.ashx', video_id, app_code, {})['Metas']
-            'http://api.radio-canada.ca/metaMedia/v1/index.ashx',
-            video_id, note='Downloading metadata XML', query={
-                'appCode': app_code,
-                'idMedia': video_id,
-            })
 
         def get_meta(name):
-            el = find_xpath_attr(metadata, './/Meta', 'name', name)
+            for meta in metas:
-            return el.text if el is not None else None
+                if meta.get('name') == name:
+                    text = meta.get('text')
+                    if text:
+                        return text
 
         # protectionType does not necessarily mean the video is DRM protected (see
         # https://github.com/rg3/youtube-dl/pull/18609).
         if get_meta('protectionType'):
             self.report_warning('This video is probably DRM protected.')
 
-        device_types = ['ipad']
+        query = {
-        if not smuggled_data:
+            'connectionType': 'hd',
-            device_types.append('flash')
+            'deviceType': 'ipad',
-            device_types.append('android')
+            'multibitrate': 'true',
-
+        }
-        formats = []
+        if smuggled_data:
-        error = None
+            query.update(smuggled_data)
-        # TODO: extract f4m formats
+        v_data = self._call_api('validation/v2/', video_id, app_code, query)
-        # f4m formats can be extracted using flashhd device_type but they produce unplayable file
+        v_url = v_data.get('url')
-        for device_type in device_types:
+        if not v_url:
-            validation_url = 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx'
+            error = v_data['message']
-            query = {
+            if error == "Le contenu sélectionné n'est pas disponible dans votre pays":
-                'appCode': app_code,
+                raise self.raise_geo_restricted(error, self._GEO_COUNTRIES)
-                'idMedia': video_id,
-                'connectionType': 'broadband',
-                'multibitrate': 'true',
-                'deviceType': device_type,
-            }
-            if smuggled_data:
-                validation_url = 'https://services.radio-canada.ca/media/validation/v2/'
-                query.update(smuggled_data)
-            else:
-                query.update({
-                    # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
-                    'paysJ391wsHjbOJwvCs26toz': 'CA',
-                    'bypasslock': 'NZt5K62gRqfc',
-                })
-            v_data = self._download_xml(validation_url, video_id, note='Downloading %s XML' % device_type, query=query, fatal=False)
-            v_url = xpath_text(v_data, 'url')
-            if not v_url:
-                continue
-            if v_url == 'null':
-                error = xpath_text(v_data, 'message')
-                continue
-            ext = determine_ext(v_url)
-            if ext == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
-                    v_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
-            elif ext == 'f4m':
-                formats.extend(self._extract_f4m_formats(
-                    v_url, video_id, f4m_id='hds', fatal=False))
-            else:
-                ext = determine_ext(v_url)
-                bitrates = xpath_element(v_data, 'bitrates')
-                for url_e in bitrates.findall('url'):
-                    tbr = int_or_none(url_e.get('bitrate'))
-                    if not tbr:
-                        continue
-                    f_url = re.sub(r'\d+\.%s' % ext, '%d.%s' % (tbr, ext), v_url)
-                    protocol = determine_protocol({'url': f_url})
-                    f = {
-                        'format_id': '%s-%d' % (protocol, tbr),
-                        'url': f_url,
-                        'ext': 'flv' if protocol == 'rtmp' else ext,
-                        'protocol': protocol,
-                        'width': int_or_none(url_e.get('width')),
-                        'height': int_or_none(url_e.get('height')),
-                        'tbr': tbr,
-                    }
-                    mobj = re.match(r'(?P<url>rtmp://[^/]+/[^/]+)/(?P<playpath>[^?]+)(?P<auth>\?.+)', f_url)
-                    if mobj:
-                        f.update({
-                            'url': mobj.group('url') + mobj.group('auth'),
-                            'play_path': mobj.group('playpath'),
-                        })
-                    formats.append(f)
-                    if protocol == 'rtsp':
-                        base_url = self._search_regex(
-                            r'rtsp://([^?]+)', f_url, 'base url', default=None)
-                        if base_url:
-                            base_url = 'http://' + base_url
-                            formats.extend(self._extract_m3u8_formats(
-                                base_url + '/playlist.m3u8', video_id, 'mp4',
-                                'm3u8_native', m3u8_id='hls', fatal=False))
-                            formats.extend(self._extract_f4m_formats(
-                                base_url + '/manifest.f4m', video_id,
-                                f4m_id='hds', fatal=False))
-        if not formats and error:
             raise ExtractorError(
                 '%s said: %s' % (self.IE_NAME, error), expected=True)
+        formats = self._extract_m3u8_formats(v_url, video_id, 'mp4')
         self._sort_formats(formats)
 
         subtitles = {}
@@ -189,8 +131,8 @@ def get_meta(name):
 
 class RadioCanadaAudioVideoIE(InfoExtractor):
     'radiocanada:audiovideo'
-    _VALID_URL = r'https?://ici\.radio-canada\.ca/audio-video/media-(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://ici\.radio-canada\.ca/([^/]+/)*media-(?P<id>[0-9]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam',
         'info_dict': {
             'id': '7527184',
@@ -203,7 +145,10 @@ class RadioCanadaAudioVideoIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         },
-    }
+    }, {
+        'url': 'https://ici.radio-canada.ca/info/videos/media-7527184/barack-obama-au-vietnam',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         return self.url_result('radiocanada:medianet:%s' % self._match_id(url))