X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fted.py;h=1cca47771290beaa2d4090126e181afe4059f460;hb=cc7fec5818254f4679896823c7de9d17f50201ca;hp=be7a0f866cb0ce37d977e6ce504ff3d612af5bbb;hpb=0ba77818f39acbf59402d383e8082bb34e0dec5a;p=youtube-dl diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index be7a0f866..1cca47771 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 +49,22 @@ 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, }] _NATIVE_FORMATS = { @@ -84,7 +100,7 @@ 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( @@ -102,11 +118,23 @@ class TEDIE(SubtitlesInfoExtractor): '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']) @@ -138,7 +166,7 @@ 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 +181,10 @@ class TEDIE(SubtitlesInfoExtractor): title = self._html_search_regex( r"(?s)(.+?)", webpage, 'title') description = self._html_search_regex( - r'(?s)

.*?

(.*?)', + [ + r'(?s)

.*?

(.*?)', + r'(?s)

About this talk:\s+(.*?)

', + ], webpage, 'description', fatal=False) return {