youtube-dl/youtube_dl/extractor/ted.py

120 lines
4.0 KiB
Python
Raw Normal View History

2014-01-17 03:52:17 +01:00
from __future__ import unicode_literals
2013-06-23 21:55:53 +02:00
import json
import re
from .subtitles import SubtitlesInfoExtractor
2013-06-23 21:55:53 +02:00
from ..utils import (
compat_str,
)
2014-01-17 03:52:17 +01:00
class TEDIE(SubtitlesInfoExtractor):
2014-03-05 13:27:26 +01:00
_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"
'''
2013-06-27 20:46:46 +02:00
_TEST = {
2014-01-17 03:52:17 +01:00
'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',
2014-01-17 03:54:54 +01:00
'md5': '4ea1dada91e4174b53dac2bb8ace429d',
2014-01-17 03:52:17 +01:00
'info_dict': {
'id': '102',
'ext': 'mp4',
'title': 'The illusion of consciousness',
2014-03-05 13:27:26 +01:00
'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',
2013-06-27 20:46:46 +02:00
}
}
2013-06-23 21:55:53 +02:00
_FORMATS_PREFERENCE = {
'low': 1,
'medium': 2,
'high': 3,
}
2013-06-23 21:55:53 +02:00
def _extract_info(self, webpage):
2014-03-05 13:27:26 +01:00
info_json = self._search_regex(r'q\("\w+.init",({.+})\)</script>',
webpage, 'info json')
return json.loads(info_json)
2013-06-23 21:55:53 +02:00
def _real_extract(self, url):
2014-03-05 13:27:26 +01:00
m = re.match(self._VALID_URL, url, re.VERBOSE)
name = m.group('name')
2013-06-23 21:55:53 +02:00
if m.group('type_talk'):
2014-03-05 13:27:26 +01:00
return self._talk_info(url, name)
else:
return self._playlist_videos_info(url, name)
2013-06-23 21:55:53 +02:00
def _playlist_videos_info(self, url, name):
2013-06-23 21:55:53 +02:00
'''Returns the videos of the playlist'''
2013-11-15 14:33:51 +01:00
webpage = self._download_webpage(url, name,
'Downloading playlist webpage')
info = self._extract_info(webpage)
playlist_info = info['playlist']
2013-06-23 21:55:53 +02:00
2013-11-15 14:33:51 +01:00
playlist_entries = [
self.url_result(u'http://www.ted.com/talks/' + talk['slug'], self.ie_key())
for talk in info['talks']
2013-11-15 14:33:51 +01:00
]
return self.playlist_result(
playlist_entries,
playlist_id=compat_str(playlist_info['id']),
playlist_title=playlist_info['title'])
2013-06-23 21:55:53 +02:00
2014-03-05 13:27:26 +01:00
def _talk_info(self, url, video_name):
webpage = self._download_webpage(url, video_name)
2013-06-23 21:55:53 +02:00
self.report_extraction(video_name)
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 = 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
2013-11-15 14:06:38 +01:00
return {
'id': video_id,
'title': talk_info['title'],
'uploader': talk_info['speaker'],
'thumbnail': thumbnail,
'description': self._og_search_description(webpage),
'subtitles': video_subtitles,
2013-10-04 10:32:34 +02:00
'formats': formats,
}
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 {}