Imported Upstream version 2016.08.17

[youtubedl] / youtube_dl / extractor / radiocanada.py

diff --git a/youtube_dl/extractor/radiocanada.py b/youtube_dl/extractor/radiocanada.py
index 4f05bbddc6d35fc8ec709fcb005638abcbe77779..8ec402646767a22f8a1e7cedbc89eb8576b59804 100644 (file)
--- a/youtube_dl/extractor/radiocanada.py
+++ b/youtube_dl/extractor/radiocanada.py
@@ -12,6 +12,7 @@ from ..utils import (
     unified_strdate,
     xpath_element,
     ExtractorError,
     unified_strdate,
     xpath_element,
     ExtractorError,

+    determine_protocol,

 )
 
 
 )
 
 


@@ -22,13 +23,13 @@ class RadioCanadaIE(InfoExtractor):
         'url': 'http://ici.radio-canada.ca/widgets/mediaconsole/medianet/7184272',
         'info_dict': {
             'id': '7184272',
         'url': 'http://ici.radio-canada.ca/widgets/mediaconsole/medianet/7184272',
         'info_dict': {
             'id': '7184272',

-            'ext': 'flv',
+            'ext': 'mp4',

             'title': 'Le parcours du tireur capté sur vidéo',
             'description': 'Images des caméras de surveillance fournies par la GRC montrant le parcours du tireur d\'Ottawa',
             'upload_date': '20141023',
         },
         'params': {
             'title': 'Le parcours du tireur capté sur vidéo',
             'description': 'Images des caméras de surveillance fournies par la GRC montrant le parcours du tireur d\'Ottawa',
             'upload_date': '20141023',
         },
         'params': {

-            # rtmp download
+            # m3u8 download

             'skip_download': True,
         },
     }
             'skip_download': True,
         },
     }


@@ -36,11 +37,14 @@ class RadioCanadaIE(InfoExtractor):
     def _real_extract(self, url):
         app_code, video_id = re.match(self._VALID_URL, url).groups()
 
     def _real_extract(self, url):
         app_code, video_id = re.match(self._VALID_URL, url).groups()
 

+        device_types = ['ipad', 'android']
+        if app_code != 'toutv':
+            device_types.append('flash')
+

         formats = []
         formats = []

-        # TODO: extract m3u8 and f4m formats
-        # m3u8 formats can be extracted using ipad device_type return 403 error code when ffmpeg try to download segements
+        # TODO: extract f4m formats

         # f4m formats can be extracted using flashhd device_type but they produce unplayable file
         # f4m formats can be extracted using flashhd device_type but they produce unplayable file

-        for device_type in ('flash',):
+        for device_type in device_types:

             v_data = self._download_xml(
                 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx',
                 video_id, note='Downloading %s XML' % device_type, query={
             v_data = self._download_xml(
                 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx',
                 video_id, note='Downloading %s XML' % device_type, query={


@@ -52,7 +56,7 @@ class RadioCanadaIE(InfoExtractor):
                     # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
                     'paysJ391wsHjbOJwvCs26toz': 'CA',
                     'bypasslock': 'NZt5K62gRqfc',
                     # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
                     'paysJ391wsHjbOJwvCs26toz': 'CA',
                     'bypasslock': 'NZt5K62gRqfc',

-                })
+                }, fatal=False)

             v_url = xpath_text(v_data, 'url')
             if not v_url:
                 continue
             v_url = xpath_text(v_data, 'url')
             if not v_url:
                 continue


@@ -64,7 +68,8 @@ class RadioCanadaIE(InfoExtractor):
                 formats.extend(self._extract_m3u8_formats(
                     v_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
             elif ext == 'f4m':
                 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))
+                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')
             else:
                 ext = determine_ext(v_url)
                 bitrates = xpath_element(v_data, 'bitrates')


@@ -72,15 +77,28 @@ class RadioCanadaIE(InfoExtractor):
                     tbr = int_or_none(url_e.get('bitrate'))
                     if not tbr:
                         continue
                     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})

                     formats.append({
                     formats.append({

-                        'format_id': 'rtmp-%d' % tbr,
-                        'url': re.sub(r'\d+\.%s' % ext, '%d.%s' % (tbr, ext), v_url),
-                        'ext': 'flv',
-                        'protocol': 'rtmp',
+                        '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,
                     })
                         'width': int_or_none(url_e.get('width')),
                         'height': int_or_none(url_e.get('height')),
                         'tbr': tbr,
                     })

+                    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))

         self._sort_formats(formats)
 
         metadata = self._download_xml(
         self._sort_formats(formats)
 
         metadata = self._download_xml(


@@ -115,13 +133,13 @@ class RadioCanadaAudioVideoIE(InfoExtractor):
         'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam',
         'info_dict': {
             'id': '7527184',
         'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam',
         'info_dict': {
             'id': '7527184',

-            'ext': 'flv',
+            'ext': 'mp4',

             'title': 'Barack Obama au Vietnam',
             'description': 'Les États-Unis lèvent l\'embargo sur la vente d\'armes qui datait de la guerre du Vietnam',
             'upload_date': '20160523',
         },
         'params': {
             'title': 'Barack Obama au Vietnam',
             'description': 'Les États-Unis lèvent l\'embargo sur la vente d\'armes qui datait de la guerre du Vietnam',
             'upload_date': '20160523',
         },
         'params': {

-            # rtmp download
+            # m3u8 download

             'skip_download': True,
         },
     }
             'skip_download': True,
         },
     }