Merge tag 'upstream/2013.11.11'

[youtubedl] / youtube_dl / extractor / ted.py

diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py
index ccad1c7c2087b533712b62aedaed43abcfa0bf90..76cfdfb90c886a94e95cb60b144e93b3f0acbb1a 100644 (file)
--- a/youtube_dl/extractor/ted.py
+++ b/youtube_dl/extractor/ted.py
@@ -1,10 +1,14 @@
 import json
 import re
 
-from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
 
+from ..utils import (
+    compat_str,
+    RegexNotFoundError,
+)
 
-class TEDIE(InfoExtractor):
+class TEDIE(SubtitlesInfoExtractor):
     _VALID_URL=r'''http://www\.ted\.com/
                    (
                         ((?P<type_playlist>playlists)/(?P<playlist_id>\d+)) # We have a playlist
@@ -14,6 +18,15 @@ class TEDIE(InfoExtractor):
                    (/lang/(.*?))? # The url may contain the language
                    /(?P<name>\w+) # Here goes the name and then ".html"
                    '''
+    _TEST = {
+        u'url': u'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',
+        u'file': u'102.mp4',
+        u'md5': u'2d76ee1576672e0bd8f187513267adf6',
+        u'info_dict': {
+            u"description": u"md5:c6fa72e6eedbd938c9caf6b2702f5922", 
+            u"title": u"Dan Dennett: The illusion of consciousness"
+        }
+    }
 
     @classmethod
     def suitable(cls, url):
@@ -23,7 +36,7 @@ class TEDIE(InfoExtractor):
     def _real_extract(self, url):
         m=re.match(self._VALID_URL, url, re.VERBOSE)
         if m.group('type_talk'):
-            return [self._talk_info(url)]
+            return self._talk_info(url)
         else :
             playlist_id=m.group('playlist_id')
             name=m.group('name')
@@ -58,7 +71,7 @@ class TEDIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name)
         self.report_extraction(video_name)
         # If the url includes the language we get the title translated
-        title = self._html_search_regex(r'<span id="altHeadline" >(?P<title>.*)</span>',
+        title = self._html_search_regex(r'<span .*?id="altHeadline".+?>(?P<title>.*)</span>',
                                         webpage, 'title')
         json_data = self._search_regex(r'<script.*?>var talkDetails = ({.*?})</script>',
                                     webpage, 'json data')
@@ -68,12 +81,44 @@ class TEDIE(InfoExtractor):
         
         thumbnail = self._search_regex(r'</span>[\s.]*</div>[\s.]*<img src="(.*?)"',
                                        webpage, 'thumbnail')
+        formats = [{
+            'ext': 'mp4',
+            'url': stream['file'],
+            'format': stream['id']
+            } for stream in info['htmlStreams']]
+
+        video_id = info['id']
+
+        # subtitles
+        video_subtitles = self.extract_subtitles(video_id, webpage)
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(video_id, webpage)
+            return
+
         info = {
-                'id': info['id'],
-                'url': info['htmlStreams'][-1]['file'],
-                'ext': 'mp4',
-                'title': title,
-                'thumbnail': thumbnail,
-                'description': desc,
-                }
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'description': desc,
+            'subtitles': video_subtitles,
+            'formats': formats,
+        }
+
+        # TODO: Remove when #980 has been merged
+        info.update(info['formats'][-1])
+
         return info
+
+    def _get_available_subtitles(self, video_id, webpage):
+        try:
+            options = self._search_regex(r'(?:<select name="subtitles_language_select" id="subtitles_language_select">)(.*?)(?:</select>)', webpage, 'subtitles_language_select', flags=re.DOTALL)
+            languages = re.findall(r'(?:<option value=")(\S+)"', options)
+            if languages:
+                sub_lang_list = {}
+                for l in languages:
+                    url = 'http://www.ted.com/talks/subtitles/id/%s/lang/%s/format/srt' % (video_id, l)
+                    sub_lang_list[l] = url
+                return sub_lang_list
+        except RegexNotFoundError as err:
+            self._downloader.report_warning(u'video doesn\'t have subtitles')
+        return {}