from .subtitles import SubtitlesInfoExtractor
from ..utils import (
- RegexNotFoundError,
+ compat_str,
)
class TEDIE(SubtitlesInfoExtractor):
- _VALID_URL=r'''(?x)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"
- '''
- _TEST = {
+ _VALID_URL = r'''(?x)
+ (?P<proto>https?://)
+ (?P<type>www|embed)(?P<urlmain>\.ted\.com/
+ (
+ (?P<type_playlist>playlists(?:/\d+)?) # We have a playlist
+ |
+ ((?P<type_talk>talks)) # We have a simple talk
+ |
+ (?P<type_watch>watch)/[^/]+/[^/]+
+ )
+ (/lang/(.*?))? # The url may contain the language
+ /(?P<name>[\w-]+) # Here goes the name and then ".html"
+ .*)$
+ '''
+ _TESTS = [{
'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',
- 'file': '102.mp4',
'md5': '4ea1dada91e4174b53dac2bb8ace429d',
'info_dict': {
+ 'id': '102',
+ 'ext': 'mp4',
'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.',
+ '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',
}
- }
+ }, {
+ 'url': 'http://www.ted.com/watch/ted-institute/ted-bcg/vishal-sikka-the-beauty-and-power-of-algorithms',
+ 'md5': '226f4fb9c62380d11b7995efa4c87994',
+ 'info_dict': {
+ 'id': 'vishal-sikka-the-beauty-and-power-of-algorithms',
+ 'ext': 'mp4',
+ 'title': 'Vishal Sikka: The beauty and power of algorithms',
+ '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.',
+ }
+ }]
_FORMATS_PREFERENCE = {
'low': 1,
'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)
+ if m.group('type') == 'embed':
+ desktop_url = m.group('proto') + 'www' + m.group('urlmain')
+ return self.url_result(desktop_url, 'TED')
+ 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)
+ elif m.group('type_watch'):
+ return self._watch_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)
- 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)
- info_json = self._search_regex(r'"talkPage.init",({.+})\)</script>', webpage, 'info json')
- info = json.loads(info_json)
- talk_info = info['talks'][0]
+ talk_info = self._extract_info(webpage)['talks'][0]
formats = [{
'ext': 'mp4',
} for (format_id, format_url) in talk_info['nativeDownloads'].items()]
self._sort_formats(formats)
- video_id = talk_info['id']
+ video_id = compat_str(talk_info['id'])
# subtitles
video_subtitles = self.extract_subtitles(video_id, talk_info)
if self._downloader.params.get('listsubtitles', False):
self._list_available_subtitles(video_id, talk_info)
return
+ thumbnail = talk_info['thumb']
+ if not thumbnail.startswith('http'):
+ thumbnail = 'http://' + thumbnail
return {
'id': video_id,
'title': talk_info['title'],
'uploader': talk_info['speaker'],
- 'thumbnail': talk_info['thumb'],
+ 'thumbnail': thumbnail,
'description': self._og_search_description(webpage),
'subtitles': video_subtitles,
'formats': formats,
else:
self._downloader.report_warning(u'video doesn\'t have subtitles')
return {}
+
+ def _watch_info(self, url, name):
+ webpage = self._download_webpage(url, name)
+
+ config_json = self._html_search_regex(
+ r"data-config='([^']+)", webpage, 'config')
+ config = json.loads(config_json)
+ video_url = config['video']['url']
+ thumbnail = config.get('image', {}).get('url')
+
+ title = self._html_search_regex(
+ r"(?s)<h1(?:\s+class='[^']+')?>(.+?)</h1>", webpage, 'title')
+ description = self._html_search_regex(
+ r'(?s)<h4 class="[^"]+" id="h3--about-this-talk">.*?</h4>(.*?)</div>',
+ webpage, 'description', fatal=False)
+
+ return {
+ 'id': name,
+ 'url': video_url,
+ 'title': title,
+ 'thumbnail': thumbnail,
+ 'description': description,
+ }
projects / youtube-dl / blobdiff