youtube-dl/youtube_dl/extractor/udemy.py

362 lines
14 KiB
Python
Raw Normal View History

from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import (
compat_HTTPError,
compat_urllib_request,
2016-03-24 20:26:46 +00:00
compat_urlparse,
)
from ..utils import (
determine_ext,
extract_attributes,
ExtractorError,
float_or_none,
int_or_none,
Switch codebase to use sanitized_Request instead of compat_urllib_request.Request [downloader/dash] Use sanitized_Request [downloader/http] Use sanitized_Request [atresplayer] Use sanitized_Request [bambuser] Use sanitized_Request [bliptv] Use sanitized_Request [brightcove] Use sanitized_Request [cbs] Use sanitized_Request [ceskatelevize] Use sanitized_Request [collegerama] Use sanitized_Request [extractor/common] Use sanitized_Request [crunchyroll] Use sanitized_Request [dailymotion] Use sanitized_Request [dcn] Use sanitized_Request [dramafever] Use sanitized_Request [dumpert] Use sanitized_Request [eitb] Use sanitized_Request [escapist] Use sanitized_Request [everyonesmixtape] Use sanitized_Request [extremetube] Use sanitized_Request [facebook] Use sanitized_Request [fc2] Use sanitized_Request [flickr] Use sanitized_Request [4tube] Use sanitized_Request [gdcvault] Use sanitized_Request [extractor/generic] Use sanitized_Request [hearthisat] Use sanitized_Request [hotnewhiphop] Use sanitized_Request [hypem] Use sanitized_Request [iprima] Use sanitized_Request [ivi] Use sanitized_Request [keezmovies] Use sanitized_Request [letv] Use sanitized_Request [lynda] Use sanitized_Request [metacafe] Use sanitized_Request [minhateca] Use sanitized_Request [miomio] Use sanitized_Request [meovideo] Use sanitized_Request [mofosex] Use sanitized_Request [moniker] Use sanitized_Request [mooshare] Use sanitized_Request [movieclips] Use sanitized_Request [mtv] Use sanitized_Request [myvideo] Use sanitized_Request [neteasemusic] Use sanitized_Request [nfb] Use sanitized_Request [niconico] Use sanitized_Request [noco] Use sanitized_Request [nosvideo] Use sanitized_Request [novamov] Use sanitized_Request [nowness] Use sanitized_Request [nuvid] Use sanitized_Request [played] Use sanitized_Request [pluralsight] Use sanitized_Request [pornhub] Use sanitized_Request [pornotube] Use sanitized_Request [primesharetv] Use sanitized_Request [promptfile] Use sanitized_Request [qqmusic] Use sanitized_Request [rtve] Use sanitized_Request [safari] Use sanitized_Request [sandia] Use sanitized_Request [shared] Use sanitized_Request [sharesix] Use sanitized_Request [sina] Use sanitized_Request [smotri] Use sanitized_Request [sohu] Use sanitized_Request [spankwire] Use sanitized_Request [sportdeutschland] Use sanitized_Request [streamcloud] Use sanitized_Request [streamcz] Use sanitized_Request [tapely] Use sanitized_Request [tube8] Use sanitized_Request [tubitv] Use sanitized_Request [twitch] Use sanitized_Request [twitter] Use sanitized_Request [udemy] Use sanitized_Request [vbox7] Use sanitized_Request [veoh] Use sanitized_Request [vessel] Use sanitized_Request [vevo] Use sanitized_Request [viddler] Use sanitized_Request [videomega] Use sanitized_Request [viewvster] Use sanitized_Request [viki] Use sanitized_Request [vk] Use sanitized_Request [vodlocker] Use sanitized_Request [voicerepublic] Use sanitized_Request [wistia] Use sanitized_Request [xfileshare] Use sanitized_Request [xtube] Use sanitized_Request [xvideos] Use sanitized_Request [yandexmusic] Use sanitized_Request [youku] Use sanitized_Request [youporn] Use sanitized_Request [youtube] Use sanitized_Request [patreon] Use sanitized_Request [extractor/common] Remove unused import [nfb] PEP 8
2015-11-21 16:18:17 +00:00
sanitized_Request,
unescapeHTML,
urlencode_postdata,
)
class UdemyIE(InfoExtractor):
IE_NAME = 'udemy'
2016-03-24 20:28:39 +00:00
_VALID_URL = r'''(?x)
https?://
www\.udemy\.com/
(?:
[^#]+\#/lecture/|
lecture/view/?\?lectureId=|
[^/]+/learn/v4/t/lecture/
)
(?P<id>\d+)
'''
_LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1'
_ORIGIN_URL = 'https://www.udemy.com'
_NETRC_MACHINE = 'udemy'
_TESTS = [{
'url': 'https://www.udemy.com/java-tutorial/#/lecture/172757',
'md5': '98eda5b657e752cf945d8445e261b5c5',
'info_dict': {
'id': '160614',
'ext': 'mp4',
'title': 'Introduction and Installation',
'description': 'md5:c0d51f6f21ef4ec65f091055a5eef876',
'duration': 579.29,
},
'skip': 'Requires udemy account credentials',
2016-03-24 20:28:39 +00:00
}, {
# new URL schema
'url': 'https://www.udemy.com/electric-bass-right-from-the-start/learn/v4/t/lecture/4580906',
'only_matching': True,
}]
def _extract_course_info(self, webpage, video_id):
course = self._parse_json(
unescapeHTML(self._search_regex(
r'ng-init=["\'].*\bcourse=({.+?});', webpage, 'course', default='{}')),
video_id, fatal=False) or {}
course_id = course.get('id') or self._search_regex(
(r'&quot;id&quot;\s*:\s*(\d+)', r'data-course-id=["\'](\d+)'),
webpage, 'course id')
return course_id, course.get('title')
2016-03-24 20:26:46 +00:00
def _enroll_course(self, base_url, webpage, course_id):
2016-03-26 18:09:12 +00:00
def combine_url(base_url, url):
return compat_urlparse.urljoin(base_url, url) if not url.startswith('http') else url
checkout_url = unescapeHTML(self._search_regex(
2016-03-26 18:09:12 +00:00
r'href=(["\'])(?P<url>(?:https?://(?:www\.)?udemy\.com)?/payment/checkout/.+?)\1',
webpage, 'checkout url', group='url', default=None))
if checkout_url:
raise ExtractorError(
2016-01-05 18:03:39 +00:00
'Course %s is not free. You have to pay for it before you can download. '
2016-03-26 18:09:12 +00:00
'Use this URL to confirm purchase: %s'
% (course_id, combine_url(base_url, checkout_url)),
expected=True)
enroll_url = unescapeHTML(self._search_regex(
2016-03-24 20:26:46 +00:00
r'href=(["\'])(?P<url>(?:https?://(?:www\.)?udemy\.com)?/course/subscribe/.+?)\1',
webpage, 'enroll url', group='url', default=None))
if enroll_url:
2016-03-26 18:09:12 +00:00
webpage = self._download_webpage(
combine_url(base_url, enroll_url),
course_id, 'Enrolling in the course',
headers={'Referer': base_url})
if '>You have enrolled in' in webpage:
self.to_screen('%s: Successfully enrolled in the course' % course_id)
def _download_lecture(self, course_id, lecture_id):
return self._download_json(
2016-05-04 17:14:12 +00:00
'https://www.udemy.com/api-2.0/users/me/subscribed-courses/%s/lectures/%s?'
% (course_id, lecture_id),
lecture_id, 'Downloading lecture JSON', query={
'fields[lecture]': 'title,description,view_html,asset',
'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data',
})
def _handle_error(self, response):
if not isinstance(response, dict):
return
error = response.get('error')
if error:
error_str = 'Udemy returned error #%s: %s' % (error.get('code'), error.get('message'))
error_data = error.get('data')
if error_data:
error_str += ' - %s' % error_data.get('formErrors')
raise ExtractorError(error_str, expected=True)
def _download_json(self, url_or_request, *args, **kwargs):
headers = {
'X-Udemy-Snail-Case': 'true',
'X-Requested-With': 'XMLHttpRequest',
}
for cookie in self._downloader.cookiejar:
if cookie.name == 'client_id':
headers['X-Udemy-Client-Id'] = cookie.value
elif cookie.name == 'access_token':
headers['X-Udemy-Bearer-Token'] = cookie.value
headers['X-Udemy-Authorization'] = 'Bearer %s' % cookie.value
if isinstance(url_or_request, compat_urllib_request.Request):
for header, value in headers.items():
url_or_request.add_header(header, value)
else:
Switch codebase to use sanitized_Request instead of compat_urllib_request.Request [downloader/dash] Use sanitized_Request [downloader/http] Use sanitized_Request [atresplayer] Use sanitized_Request [bambuser] Use sanitized_Request [bliptv] Use sanitized_Request [brightcove] Use sanitized_Request [cbs] Use sanitized_Request [ceskatelevize] Use sanitized_Request [collegerama] Use sanitized_Request [extractor/common] Use sanitized_Request [crunchyroll] Use sanitized_Request [dailymotion] Use sanitized_Request [dcn] Use sanitized_Request [dramafever] Use sanitized_Request [dumpert] Use sanitized_Request [eitb] Use sanitized_Request [escapist] Use sanitized_Request [everyonesmixtape] Use sanitized_Request [extremetube] Use sanitized_Request [facebook] Use sanitized_Request [fc2] Use sanitized_Request [flickr] Use sanitized_Request [4tube] Use sanitized_Request [gdcvault] Use sanitized_Request [extractor/generic] Use sanitized_Request [hearthisat] Use sanitized_Request [hotnewhiphop] Use sanitized_Request [hypem] Use sanitized_Request [iprima] Use sanitized_Request [ivi] Use sanitized_Request [keezmovies] Use sanitized_Request [letv] Use sanitized_Request [lynda] Use sanitized_Request [metacafe] Use sanitized_Request [minhateca] Use sanitized_Request [miomio] Use sanitized_Request [meovideo] Use sanitized_Request [mofosex] Use sanitized_Request [moniker] Use sanitized_Request [mooshare] Use sanitized_Request [movieclips] Use sanitized_Request [mtv] Use sanitized_Request [myvideo] Use sanitized_Request [neteasemusic] Use sanitized_Request [nfb] Use sanitized_Request [niconico] Use sanitized_Request [noco] Use sanitized_Request [nosvideo] Use sanitized_Request [novamov] Use sanitized_Request [nowness] Use sanitized_Request [nuvid] Use sanitized_Request [played] Use sanitized_Request [pluralsight] Use sanitized_Request [pornhub] Use sanitized_Request [pornotube] Use sanitized_Request [primesharetv] Use sanitized_Request [promptfile] Use sanitized_Request [qqmusic] Use sanitized_Request [rtve] Use sanitized_Request [safari] Use sanitized_Request [sandia] Use sanitized_Request [shared] Use sanitized_Request [sharesix] Use sanitized_Request [sina] Use sanitized_Request [smotri] Use sanitized_Request [sohu] Use sanitized_Request [spankwire] Use sanitized_Request [sportdeutschland] Use sanitized_Request [streamcloud] Use sanitized_Request [streamcz] Use sanitized_Request [tapely] Use sanitized_Request [tube8] Use sanitized_Request [tubitv] Use sanitized_Request [twitch] Use sanitized_Request [twitter] Use sanitized_Request [udemy] Use sanitized_Request [vbox7] Use sanitized_Request [veoh] Use sanitized_Request [vessel] Use sanitized_Request [vevo] Use sanitized_Request [viddler] Use sanitized_Request [videomega] Use sanitized_Request [viewvster] Use sanitized_Request [viki] Use sanitized_Request [vk] Use sanitized_Request [vodlocker] Use sanitized_Request [voicerepublic] Use sanitized_Request [wistia] Use sanitized_Request [xfileshare] Use sanitized_Request [xtube] Use sanitized_Request [xvideos] Use sanitized_Request [yandexmusic] Use sanitized_Request [youku] Use sanitized_Request [youporn] Use sanitized_Request [youtube] Use sanitized_Request [patreon] Use sanitized_Request [extractor/common] Remove unused import [nfb] PEP 8
2015-11-21 16:18:17 +00:00
url_or_request = sanitized_Request(url_or_request, headers=headers)
response = super(UdemyIE, self)._download_json(url_or_request, *args, **kwargs)
self._handle_error(response)
return response
def _real_initialize(self):
self._login()
def _login(self):
(username, password) = self._get_login_info()
if username is None:
return
login_popup = self._download_webpage(
self._LOGIN_URL, None, 'Downloading login popup')
def is_logged(webpage):
return any(re.search(p, webpage) for p in (
r'href=["\'](?:https://www\.udemy\.com)?/user/logout/',
r'>Logout<'))
# already logged in
if is_logged(login_popup):
return
login_form = self._form_hidden_inputs('login-form', login_popup)
login_form.update({
'email': username,
'password': password,
})
response = self._download_webpage(
2016-05-04 17:14:12 +00:00
self._LOGIN_URL, None, 'Logging in as %s' % username,
data=urlencode_postdata(login_form),
headers={
'Referer': self._ORIGIN_URL,
'Origin': self._ORIGIN_URL,
})
if not is_logged(response):
error = self._html_search_regex(
r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
response, 'error message', default=None)
if error:
raise ExtractorError('Unable to login: %s' % error, expected=True)
raise ExtractorError('Unable to log in')
def _real_extract(self, url):
lecture_id = self._match_id(url)
webpage = self._download_webpage(url, lecture_id)
course_id, _ = self._extract_course_info(webpage, lecture_id)
try:
lecture = self._download_lecture(course_id, lecture_id)
except ExtractorError as e:
# Error could possibly mean we are not enrolled in the course
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
2016-03-24 20:26:46 +00:00
self._enroll_course(url, webpage, course_id)
2015-12-30 19:09:21 +00:00
lecture = self._download_lecture(course_id, lecture_id)
else:
raise
title = lecture['title']
description = lecture.get('description')
asset = lecture['asset']
2016-04-01 16:38:56 +00:00
asset_type = asset.get('asset_type') or asset.get('assetType')
if asset_type != 'Video':
raise ExtractorError(
'Lecture %s is not a video' % lecture_id, expected=True)
2016-04-01 16:38:56 +00:00
stream_url = asset.get('stream_url') or asset.get('streamUrl')
if stream_url:
youtube_url = self._search_regex(
r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url, 'youtube URL', default=None)
if youtube_url:
return self.url_result(youtube_url, 'Youtube')
video_id = asset['id']
2016-04-01 16:38:56 +00:00
thumbnail = asset.get('thumbnail_url') or asset.get('thumbnailUrl')
duration = float_or_none(asset.get('data', {}).get('duration'))
formats = []
2016-03-24 20:27:13 +00:00
def extract_output_format(src):
return {
'url': src['url'],
2016-03-26 17:43:25 +00:00
'format_id': '%sp' % (src.get('height') or format_id),
2016-03-24 20:27:13 +00:00
'width': int_or_none(src.get('width')),
'height': int_or_none(src.get('height')),
'vbr': int_or_none(src.get('video_bitrate_in_kbps')),
'vcodec': src.get('video_codec'),
'fps': int_or_none(src.get('frame_rate')),
'abr': int_or_none(src.get('audio_bitrate_in_kbps')),
'acodec': src.get('audio_codec'),
'asr': int_or_none(src.get('audio_sample_rate')),
'tbr': int_or_none(src.get('total_bitrate_in_kbps')),
'filesize': int_or_none(src.get('file_size_in_bytes')),
}
2016-03-24 20:27:13 +00:00
outputs = asset.get('data', {}).get('outputs')
if not isinstance(outputs, dict):
outputs = {}
2016-03-26 18:02:52 +00:00
def add_output_format_meta(f, key):
output = outputs.get(key)
if isinstance(output, dict):
output_format = extract_output_format(output)
output_format.update(f)
return output_format
2016-03-26 18:02:52 +00:00
return f
2016-03-24 20:27:13 +00:00
download_urls = asset.get('download_urls')
if isinstance(download_urls, dict):
video = download_urls.get('Video')
if isinstance(video, list):
for format_ in video:
video_url = format_.get('file')
if not video_url:
continue
format_id = format_.get('label')
f = {
'url': format_['file'],
2016-03-26 18:02:52 +00:00
'format_id': '%sp' % format_id,
2016-03-24 20:27:13 +00:00
'height': int_or_none(format_id),
}
if format_id:
# Some videos contain additional metadata (e.g.
# https://www.udemy.com/ios9-swift/learn/#/lecture/3383208)
2016-03-26 18:02:52 +00:00
f = add_output_format_meta(f, format_id)
2016-03-24 20:27:13 +00:00
formats.append(f)
view_html = lecture.get('view_html')
if view_html:
view_html_urls = set()
for source in re.findall(r'<source[^>]+>', view_html):
attributes = extract_attributes(source)
src = attributes.get('src')
if not src:
continue
res = attributes.get('data-res')
height = int_or_none(res)
if src in view_html_urls:
continue
view_html_urls.add(src)
if attributes.get('type') == 'application/x-mpegURL' or determine_ext(src) == 'm3u8':
m3u8_formats = self._extract_m3u8_formats(
src, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False)
for f in m3u8_formats:
m = re.search(r'/hls_(?P<height>\d{3,4})_(?P<tbr>\d{2,})/', f['url'])
if m:
if not f.get('height'):
f['height'] = int(m.group('height'))
if not f.get('tbr'):
f['tbr'] = int(m.group('tbr'))
formats.extend(m3u8_formats)
else:
formats.append(add_output_format_meta({
'url': src,
2016-03-26 18:02:52 +00:00
'format_id': '%dp' % height if height else None,
'height': height,
2016-03-26 18:02:52 +00:00
}, res))
2016-03-26 17:42:46 +00:00
self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id'))
return {
'id': video_id,
'title': title,
'description': description,
'thumbnail': thumbnail,
'duration': duration,
'formats': formats
}
class UdemyCourseIE(UdemyIE):
IE_NAME = 'udemy:course'
2016-03-31 20:24:22 +00:00
_VALID_URL = r'https?://www\.udemy\.com/(?P<id>[^/?#&]+)'
_TESTS = []
@classmethod
def suitable(cls, url):
return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
def _real_extract(self, url):
course_path = self._match_id(url)
webpage = self._download_webpage(url, course_path)
course_id, title = self._extract_course_info(webpage, course_path)
2016-03-24 20:26:46 +00:00
self._enroll_course(url, webpage, course_id)
response = self._download_json(
'https://www.udemy.com/api-2.0/courses/%s/cached-subscriber-curriculum-items' % course_id,
course_id, 'Downloading course curriculum', query={
'fields[chapter]': 'title,object_index',
2016-04-01 16:38:56 +00:00
'fields[lecture]': 'title,asset',
'page_size': '1000',
})
2015-12-30 21:11:21 +00:00
entries = []
chapter, chapter_number = [None] * 2
for entry in response['results']:
clazz = entry.get('_class')
if clazz == 'lecture':
2016-04-01 16:38:56 +00:00
asset = entry.get('asset')
if isinstance(asset, dict):
asset_type = asset.get('asset_type') or asset.get('assetType')
if asset_type != 'Video':
continue
lecture_id = entry.get('id')
if lecture_id:
2015-12-30 21:11:21 +00:00
entry = {
'_type': 'url_transparent',
2016-03-31 20:24:22 +00:00
'url': 'https://www.udemy.com/%s/learn/v4/t/lecture/%s' % (course_path, entry['id']),
'title': entry.get('title'),
2015-12-30 21:11:21 +00:00
'ie_key': UdemyIE.ie_key(),
}
2016-01-01 14:34:29 +00:00
if chapter_number:
entry['chapter_number'] = chapter_number
2015-12-30 21:11:21 +00:00
if chapter:
entry['chapter'] = chapter
entries.append(entry)
elif clazz == 'chapter':
chapter_number = entry.get('object_index')
chapter = entry.get('title')
return self.playlist_result(entries, course_id, title)