X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Fextractor%2Fted.py;h=cd4af96fdb02b2f4ea392fe19482371b6a02acad;hb=7fc54e5262966370762ca7f186e9c709990b354d;hp=be7a0f866cb0ce37d977e6ce504ff3d612af5bbb;hpb=77477fa4c916599e7eaa236a3f3eb5703923cf91;p=youtube-dl.git diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index be7a0f866..cd4af96fd 100644 --- a/youtube_dl/extractor/ted.py +++ b/youtube_dl/extractor/ted.py @@ -27,7 +27,7 @@ class TEDIE(SubtitlesInfoExtractor): ''' _TESTS = [{ 'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html', - 'md5': '4ea1dada91e4174b53dac2bb8ace429d', + 'md5': 'fc94ac279feebbce69f21c0c6ee82810', 'info_dict': { 'id': '102', 'ext': 'mp4', @@ -49,6 +49,38 @@ class TEDIE(SubtitlesInfoExtractor): 'thumbnail': 're:^https?://.+\.jpg', 'description': 'Adaptive, intelligent, and consistent, algorithms are emerging as the ultimate app for everything from matching consumers to products to assessing medical diagnoses. Vishal Sikka shares his appreciation for the algorithm, charting both its inherent beauty and its growing power.', } + }, { + 'url': 'http://www.ted.com/talks/gabby_giffords_and_mark_kelly_be_passionate_be_courageous_be_your_best', + 'info_dict': { + 'id': '1972', + 'ext': 'mp4', + 'title': 'Be passionate. Be courageous. Be your best.', + 'uploader': 'Gabby Giffords and Mark Kelly', + 'description': 'md5:5174aed4d0f16021b704120360f72b92', + }, + }, { + 'url': 'http://www.ted.com/playlists/who_are_the_hackers', + 'info_dict': { + 'id': '10', + 'title': 'Who are the hackers?', + }, + 'playlist_mincount': 6, + }, { + # contains a youtube video + 'url': 'https://www.ted.com/talks/douglas_adams_parrots_the_universe_and_everything', + 'add_ie': ['Youtube'], + 'info_dict': { + 'id': '_ZG8HBuDjgc', + 'ext': 'mp4', + 'title': 'Douglas Adams: Parrots the Universe and Everything', + 'description': 'md5:01ad1e199c49ac640cb1196c0e9016af', + 'uploader': 'University of California Television (UCTV)', + 'uploader_id': 'UCtelevision', + 'upload_date': '20080522', + }, + 'params': { + 'skip_download': True, + }, }] _NATIVE_FORMATS = { @@ -84,7 +116,7 @@ class TEDIE(SubtitlesInfoExtractor): playlist_info = info['playlist'] playlist_entries = [ - self.url_result(u'http://www.ted.com/talks/' + talk['slug'], self.ie_key()) + self.url_result('http://www.ted.com/talks/' + talk['slug'], self.ie_key()) for talk in info['talks'] ] return self.playlist_result( @@ -98,15 +130,34 @@ class TEDIE(SubtitlesInfoExtractor): talk_info = self._extract_info(webpage)['talks'][0] + if talk_info.get('external') is not None: + self.to_screen('Found video from %s' % talk_info['external']['service']) + return { + '_type': 'url', + 'url': talk_info['external']['uri'], + } + formats = [{ 'url': format_url, 'format_id': format_id, 'format': format_id, - } for (format_id, format_url) in talk_info['nativeDownloads'].items()] - for f in formats: - finfo = self._NATIVE_FORMATS.get(f['format_id']) - if finfo: - f.update(finfo) + } for (format_id, format_url) in talk_info['nativeDownloads'].items() if format_url is not None] + if formats: + for f in formats: + finfo = self._NATIVE_FORMATS.get(f['format_id']) + if finfo: + f.update(finfo) + else: + # Use rtmp downloads + formats = [{ + 'format_id': f['name'], + 'url': talk_info['streamer'], + 'play_path': f['file'], + 'ext': 'flv', + 'width': f['width'], + 'height': f['height'], + 'tbr': f['bitrate'], + } for f in talk_info['resources']['rtmp']] self._sort_formats(formats) video_id = compat_str(talk_info['id']) @@ -121,7 +172,7 @@ class TEDIE(SubtitlesInfoExtractor): thumbnail = 'http://' + thumbnail return { 'id': video_id, - 'title': talk_info['title'], + 'title': talk_info['title'].strip(), 'uploader': talk_info['speaker'], 'thumbnail': thumbnail, 'description': self._og_search_description(webpage), @@ -138,7 +189,7 @@ class TEDIE(SubtitlesInfoExtractor): sub_lang_list[l] = url return sub_lang_list else: - self._downloader.report_warning(u'video doesn\'t have subtitles') + self._downloader.report_warning('video doesn\'t have subtitles') return {} def _watch_info(self, url, name): @@ -153,7 +204,10 @@ class TEDIE(SubtitlesInfoExtractor): title = self._html_search_regex( r"(?s)
About this talk:\s+(.*?)
', + ], webpage, 'description', fatal=False) return {