debian/README.source: Change 'markup'.

[youtubedl] / youtube_dl / extractor / ted.py

diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py
index 212ac80abb52ec98c887b9652234fdb50bb81717..db5a4f44e6b9be254729256fc741d6f3eb3693f7 100644 (file)
--- a/youtube_dl/extractor/ted.py
+++ b/youtube_dl/extractor/ted.py
@@ -5,8 +5,12 @@ import re
 
 from .common import InfoExtractor
 
 
 from .common import InfoExtractor
 

-from ..compat import compat_str
+from ..compat import (
+    compat_str,
+    compat_urlparse
+)

 from ..utils import (
 from ..utils import (

+    extract_attributes,

     float_or_none,
     int_or_none,
     try_get,
     float_or_none,
     int_or_none,
     try_get,


@@ -20,7 +24,7 @@ class TEDIE(InfoExtractor):
         (?P<proto>https?://)
         (?P<type>www|embed(?:-ssl)?)(?P<urlmain>\.ted\.com/
         (
         (?P<proto>https?://)
         (?P<type>www|embed(?:-ssl)?)(?P<urlmain>\.ted\.com/
         (

-            (?P<type_playlist>playlists(?:/\d+)?) # We have a playlist
+            (?P<type_playlist>playlists(?:/(?P<playlist_id>\d+))?) # We have a playlist

             |
             ((?P<type_talk>talks)) # We have a simple talk
             |
             |
             ((?P<type_talk>talks)) # We have a simple talk
             |


@@ -84,6 +88,7 @@ class TEDIE(InfoExtractor):
         'info_dict': {
             'id': '10',
             'title': 'Who are the hackers?',
         'info_dict': {
             'id': '10',
             'title': 'Who are the hackers?',

+            'description': 'md5:49a0dbe8fb76d81a0e64b4a80af7f15a'

         },
         'playlist_mincount': 6,
     }, {
         },
         'playlist_mincount': 6,
     }, {


@@ -128,7 +133,7 @@ class TEDIE(InfoExtractor):
 
     def _extract_info(self, webpage):
         info_json = self._search_regex(
 
     def _extract_info(self, webpage):
         info_json = self._search_regex(

-            r'(?s)q\(\s*"\w+.init"\s*,\s*({.+})\)\s*</script>',
+            r'(?s)q\(\s*"\w+.init"\s*,\s*({.+?})\)\s*</script>',

             webpage, 'info json')
         return json.loads(info_json)
 
             webpage, 'info json')
         return json.loads(info_json)
 


@@ -150,22 +155,22 @@ class TEDIE(InfoExtractor):
 
         webpage = self._download_webpage(url, name,
                                          'Downloading playlist webpage')
 
         webpage = self._download_webpage(url, name,
                                          'Downloading playlist webpage')

-        info = self._extract_info(webpage)

 
 

-        playlist_info = try_get(
-            info, lambda x: x['__INITIAL_DATA__']['playlist'],
-            dict) or info['playlist']
+        playlist_entries = []
+        for entry in re.findall(r'(?s)<[^>]+data-ga-context=["\']playlist["\'][^>]*>', webpage):
+            attrs = extract_attributes(entry)
+            entry_url = compat_urlparse.urljoin(url, attrs['href'])
+            playlist_entries.append(self.url_result(entry_url, self.ie_key()))
+
+        final_url = self._og_search_url(webpage, fatal=False)
+        playlist_id = (
+            re.match(self._VALID_URL, final_url).group('playlist_id')
+            if final_url else None)

 
 

-        playlist_entries = [
-            self.url_result('http://www.ted.com/talks/' + talk['slug'], self.ie_key())
-            for talk in try_get(
-                info, lambda x: x['__INITIAL_DATA__']['talks'],
-                dict) or info['talks']
-        ]

         return self.playlist_result(
         return self.playlist_result(

-            playlist_entries,
-            playlist_id=compat_str(playlist_info['id']),
-            playlist_title=playlist_info['title'])
+            playlist_entries, playlist_id=playlist_id,
+            playlist_title=self._og_search_title(webpage, fatal=False),
+            playlist_description=self._og_search_description(webpage))

 
     def _talk_info(self, url, video_name):
         webpage = self._download_webpage(url, video_name)
 
     def _talk_info(self, url, video_name):
         webpage = self._download_webpage(url, video_name)


@@ -203,17 +208,13 @@ class TEDIE(InfoExtractor):
                 ext_url = None
                 if service.lower() == 'youtube':
                     ext_url = external.get('code')
                 ext_url = None
                 if service.lower() == 'youtube':
                     ext_url = external.get('code')

-                return {
-                    '_type': 'url',
-                    'url': ext_url or external['uri'],
-                }
+
+                return self.url_result(ext_url or external['uri'])

 
         resources_ = player_talk.get('resources') or talk_info.get('resources')
 
         http_url = None
         for format_id, resources in resources_.items():
 
         resources_ = player_talk.get('resources') or talk_info.get('resources')
 
         http_url = None
         for format_id, resources in resources_.items():

-            if not isinstance(resources, dict):
-                continue

             if format_id == 'h264':
                 for resource in resources:
                     h264_url = resource.get('file')
             if format_id == 'h264':
                 for resource in resources:
                     h264_url = resource.get('file')


@@ -242,6 +243,8 @@ class TEDIE(InfoExtractor):
                         'tbr': int_or_none(resource.get('bitrate')),
                     })
             elif format_id == 'hls':
                         'tbr': int_or_none(resource.get('bitrate')),
                     })
             elif format_id == 'hls':

+                if not isinstance(resources, dict):
+                    continue

                 stream_url = url_or_none(resources.get('stream'))
                 if not stream_url:
                     continue
                 stream_url = url_or_none(resources.get('stream'))
                 if not stream_url:
                     continue


@@ -267,6 +270,8 @@ class TEDIE(InfoExtractor):
                     'format_id': m3u8_format['format_id'].replace('hls', 'http'),
                     'protocol': 'http',
                 })
                     'format_id': m3u8_format['format_id'].replace('hls', 'http'),
                     'protocol': 'http',
                 })

+                if f.get('acodec') == 'none':
+                    del f['acodec']

                 formats.append(f)
 
         audio_download = talk_info.get('audioDownload')
                 formats.append(f)
 
         audio_download = talk_info.get('audioDownload')