diff options
Diffstat (limited to 'youtube_dl/extractor/ted.py')
| -rw-r--r-- | youtube_dl/extractor/ted.py | 144 |
1 files changed, 70 insertions, 74 deletions
diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index 8b31caa92..b9e65447f 100644 --- a/youtube_dl/extractor/ted.py +++ b/youtube_dl/extractor/ted.py @@ -6,115 +6,111 @@ import re from .subtitles import SubtitlesInfoExtractor from ..utils import ( + compat_str, RegexNotFoundError, ) class TEDIE(SubtitlesInfoExtractor): - _VALID_URL=r'''http://www\.ted\.com/ - ( - ((?P<type_playlist>playlists)/(?P<playlist_id>\d+)) # We have a playlist - | - ((?P<type_talk>talks)) # We have a simple talk - ) - (/lang/(.*?))? # The url may contain the language - /(?P<name>\w+) # Here goes the name and then ".html" - ''' + _VALID_URL = r'''(?x)http://www\.ted\.com/ + ( + (?P<type_playlist>playlists(?:/\d+)?) # We have a playlist + | + ((?P<type_talk>talks)) # We have a simple talk + ) + (/lang/(.*?))? # The url may contain the language + /(?P<name>\w+) # Here goes the name and then ".html" + ''' _TEST = { 'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html', 'file': '102.mp4', 'md5': '4ea1dada91e4174b53dac2bb8ace429d', 'info_dict': { - "description": "md5:c6fa72e6eedbd938c9caf6b2702f5922", - "title": "Dan Dennett: The illusion of consciousness" + 'title': 'The illusion of consciousness', + 'description': ('Philosopher Dan Dennett makes a compelling ' + 'argument that not only don\'t we understand our own ' + 'consciousness, but that half the time our brains are ' + 'actively fooling us.'), + 'uploader': 'Dan Dennett', } } - @classmethod - def suitable(cls, url): - """Receives a URL and returns True if suitable for this IE.""" - return re.match(cls._VALID_URL, url, re.VERBOSE) is not None + _FORMATS_PREFERENCE = { + 'low': 1, + 'medium': 2, + 'high': 3, + } + + def _extract_info(self, webpage): + info_json = self._search_regex(r'q\("\w+.init",({.+})\)</script>', + webpage, 'info json') + return json.loads(info_json) def _real_extract(self, url): - m=re.match(self._VALID_URL, url, re.VERBOSE) + m = re.match(self._VALID_URL, url, re.VERBOSE) + name = m.group('name') if m.group('type_talk'): - return self._talk_info(url) - else : - playlist_id=m.group('playlist_id') - name=m.group('name') - self.to_screen(u'Getting info of playlist %s: "%s"' % (playlist_id,name)) - return [self._playlist_videos_info(url,name,playlist_id)] + return self._talk_info(url, name) + else: + return self._playlist_videos_info(url, name) - - def _playlist_videos_info(self, url, name, playlist_id): + def _playlist_videos_info(self, url, name): '''Returns the videos of the playlist''' - webpage = self._download_webpage( - url, playlist_id, 'Downloading playlist webpage') - matches = re.finditer( - r'<p\s+class="talk-title[^"]*"><a\s+href="(?P<talk_url>/talks/[^"]+\.html)">[^<]*</a></p>', - webpage) - - playlist_title = self._html_search_regex(r'div class="headline">\s*?<h1>\s*?<span>(.*?)</span>', - webpage, 'playlist title') + webpage = self._download_webpage(url, name, + 'Downloading playlist webpage') + info = self._extract_info(webpage) + playlist_info = info['playlist'] playlist_entries = [ - self.url_result(u'http://www.ted.com' + m.group('talk_url'), 'TED') - for m in matches + self.url_result(u'http://www.ted.com/talks/' + talk['slug'], self.ie_key()) + for talk in info['talks'] ] return self.playlist_result( - playlist_entries, playlist_id=playlist_id, playlist_title=playlist_title) + playlist_entries, + playlist_id=compat_str(playlist_info['id']), + playlist_title=playlist_info['title']) - def _talk_info(self, url, video_id=0): - """Return the video for the talk in the url""" - m = re.match(self._VALID_URL, url,re.VERBOSE) - video_name = m.group('name') - webpage = self._download_webpage(url, video_id, 'Downloading \"%s\" page' % video_name) + def _talk_info(self, url, video_name): + webpage = self._download_webpage(url, 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>', - webpage, 'title') - json_data = self._search_regex(r'<script.*?>var talkDetails = ({.*?})</script>', - webpage, 'json data') - info = json.loads(json_data) - desc = self._html_search_regex(r'<div class="talk-intro">.*?<p.*?>(.*?)</p>', - webpage, 'description', flags = re.DOTALL) - - 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'] + talk_info = self._extract_info(webpage)['talks'][0] + formats = [{ + 'ext': 'mp4', + 'url': format_url, + 'format_id': format_id, + 'format': format_id, + 'preference': self._FORMATS_PREFERENCE.get(format_id, -1), + } for (format_id, format_url) in talk_info['nativeDownloads'].items()] + self._sort_formats(formats) + + video_id = talk_info['id'] # subtitles - video_subtitles = self.extract_subtitles(video_id, webpage) + video_subtitles = self.extract_subtitles(video_id, talk_info) if self._downloader.params.get('listsubtitles', False): - self._list_available_subtitles(video_id, webpage) + self._list_available_subtitles(video_id, talk_info) return return { 'id': video_id, - 'title': title, - 'thumbnail': thumbnail, - 'description': desc, + 'title': talk_info['title'], + 'uploader': talk_info['speaker'], + 'thumbnail': talk_info['thumb'], + 'description': self._og_search_description(webpage), 'subtitles': video_subtitles, 'formats': formats, } - 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: + def _get_available_subtitles(self, video_id, talk_info): + languages = [lang['languageCode'] for lang in talk_info.get('languages', [])] + 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 + else: self._downloader.report_warning(u'video doesn\'t have subtitles') - return {} + return {} |