X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fudemy.py;h=71bea5363ed77ddbf476bb92050e4d675c6f13a9;hb=c9c39c22c5740c1eedcc9ce7a10f5df199ea5c78;hp=1a7d01c6790e0caa8c408c388ebc5922dc379889;hpb=9e1a5b845586a0a5431fb72467142046d8571e6f;p=youtube-dl diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py index 1a7d01c67..71bea5363 100644 --- a/youtube_dl/extractor/udemy.py +++ b/youtube_dl/extractor/udemy.py @@ -3,17 +3,38 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import ( - compat_urllib_parse, +from ..compat import ( + compat_HTTPError, + compat_urllib_parse_urlencode, compat_urllib_request, + compat_urlparse, +) +from ..utils import ( + determine_ext, + extract_attributes, ExtractorError, + float_or_none, + int_or_none, + sanitized_Request, + unescapeHTML, + urlencode_postdata, ) class UdemyIE(InfoExtractor): IE_NAME = 'udemy' - _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P\d+)' - _LOGIN_URL = 'https://www.udemy.com/join/login-submit/' + _VALID_URL = r'''(?x) + https?:// + www\.udemy\.com/ + (?: + [^#]+\#/lecture/| + lecture/view/?\?lectureId=| + [^/]+/learn/v4/t/lecture/ + ) + (?P\d+) + ''' + _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1' + _ORIGIN_URL = 'https://www.udemy.com' _NETRC_MACHINE = 'udemy' _TESTS = [{ @@ -27,8 +48,45 @@ class UdemyIE(InfoExtractor): 'duration': 579.29, }, 'skip': 'Requires udemy account credentials', + }, { + # new URL schema + 'url': 'https://www.udemy.com/electric-bass-right-from-the-start/learn/v4/t/lecture/4580906', + 'only_matching': True, }] + def _enroll_course(self, base_url, webpage, course_id): + 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( + r'href=(["\'])(?P(?:https?://(?:www\.)?udemy\.com)?/payment/checkout/.+?)\1', + webpage, 'checkout url', group='url', default=None)) + if checkout_url: + raise ExtractorError( + 'Course %s is not free. You have to pay for it before you can download. ' + 'Use this URL to confirm purchase: %s' + % (course_id, combine_url(base_url, checkout_url)), + expected=True) + + enroll_url = unescapeHTML(self._search_regex( + r'href=(["\'])(?P(?:https?://(?:www\.)?udemy\.com)?/course/subscribe/.+?)\1', + webpage, 'enroll url', group='url', default=None)) + if enroll_url: + webpage = self._download_webpage( + combine_url(base_url, enroll_url), + course_id, 'Enrolling in the course') + 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( + 'https://www.udemy.com/api-2.0/users/me/subscribed-courses/%s/lectures/%s?%s' % ( + course_id, lecture_id, compat_urllib_parse_urlencode({ + 'fields[lecture]': 'title,description,view_html,asset', + 'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data', + })), + lecture_id, 'Downloading lecture JSON') + def _handle_error(self, response): if not isinstance(response, dict): return @@ -40,8 +98,25 @@ class UdemyIE(InfoExtractor): error_str += ' - %s' % error_data.get('formErrors') raise ExtractorError(error_str, expected=True) - def _download_json(self, url, video_id, note='Downloading JSON metadata'): - response = super(UdemyIE, self)._download_json(url, video_id, note) + def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'): + 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: + url_or_request = sanitized_Request(url_or_request, headers=headers) + + response = super(UdemyIE, self)._download_json(url_or_request, video_id, note) self._handle_error(response) return response @@ -51,68 +126,164 @@ class UdemyIE(InfoExtractor): def _login(self): (username, password) = self._get_login_info() if username is None: - raise ExtractorError( - 'Udemy account is required, use --username and --password options to provide account credentials.', - expected=True) + return login_popup = self._download_webpage( - 'https://www.udemy.com/join/login-popup?displayType=ajax&showSkipButton=1', None, - 'Downloading login popup') + self._LOGIN_URL, None, 'Downloading login popup') + + def is_logged(webpage): + return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<']) - if login_popup == '
': + # already logged in + if is_logged(login_popup): return - csrf = self._html_search_regex(r']+class="form-errors[^"]*">(.+?)', + 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): - mobj = re.match(self._VALID_URL, url) - lecture_id = mobj.group('id') + lecture_id = self._match_id(url) - lecture = self._download_json( - 'https://www.udemy.com/api-1.1/lectures/%s' % lecture_id, lecture_id, 'Downloading lecture JSON') + webpage = self._download_webpage(url, lecture_id) - if lecture['assetType'] != 'Video': - raise ExtractorError('Lecture %s is not a video' % lecture_id, expected=True) + course_id = self._search_regex( + (r'data-course-id=["\'](\d+)', r'"id"\s*:\s*(\d+)'), + webpage, 'course 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: + self._enroll_course(url, webpage, course_id) + lecture = self._download_lecture(course_id, lecture_id) + else: + raise + + title = lecture['title'] + description = lecture.get('description') asset = lecture['asset'] - stream_url = asset['streamUrl'] - mobj = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url) - if mobj: - return self.url_result(mobj.group(1), 'Youtube') + asset_type = asset.get('assetType') or asset.get('asset_type') + if asset_type != 'Video': + raise ExtractorError( + 'Lecture %s is not a video' % lecture_id, expected=True) - video_id = asset['id'] - thumbnail = asset['thumbnailUrl'] - duration = asset['data']['duration'] - - download_url = asset['downloadUrl'] - - formats = [ - { - 'url': download_url['Video480p'][0], - 'format_id': '360p', - }, - { - 'url': download_url['Video'][0], - 'format_id': '720p', - }, - ] + stream_url = asset.get('streamUrl') or asset.get('stream_url') + 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') - title = lecture['title'] - description = lecture['description'] + video_id = asset['id'] + thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url') + duration = float_or_none(asset.get('data', {}).get('duration')) + + formats = [] + + def extract_output_format(src): + return { + 'url': src['url'], + 'format_id': '%sp' % (src.get('height') or format_id), + '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')), + } + + outputs = asset.get('data', {}).get('outputs') + if not isinstance(outputs, dict): + outputs = {} + + 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 + return f + + 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'], + 'format_id': '%sp' % format_id, + '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) + f = add_output_format_meta(f, format_id) + formats.append(f) + + view_html = lecture.get('view_html') + if view_html: + view_html_urls = set() + for source in re.findall(r']+>', 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\d{3,4})_(?P\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, + 'format_id': '%dp' % height if height else None, + 'height': height, + }, res)) + + self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id')) return { 'id': video_id, @@ -126,9 +297,7 @@ class UdemyIE(InfoExtractor): class UdemyCourseIE(UdemyIE): IE_NAME = 'udemy:course' - _VALID_URL = r'https?://www\.udemy\.com/(?P[\da-z-]+)' - _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<' - _ALREADY_ENROLLED = '>You are already taking this course.<' + _VALID_URL = r'https?://www\.udemy\.com/(?P[\da-z-]+)' _TESTS = [] @classmethod @@ -136,29 +305,42 @@ class UdemyCourseIE(UdemyIE): return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url) def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - course_path = mobj.group('coursepath') - - response = self._download_json( - 'https://www.udemy.com/api-1.1/courses/%s' % course_path, course_path, 'Downloading course JSON') + course_path = self._match_id(url) - course_id = int(response['id']) - course_title = response['title'] + webpage = self._download_webpage(url, course_path) - webpage = self._download_webpage( - 'https://www.udemy.com/course/subscribe/?courseId=%s' % course_id, course_id, 'Enrolling in the course') + response = self._download_json( + 'https://www.udemy.com/api-1.1/courses/%s' % course_path, + course_path, 'Downloading course JSON') - if self._SUCCESSFULLY_ENROLLED in webpage: - self.to_screen('%s: Successfully enrolled in' % course_id) - elif self._ALREADY_ENROLLED in webpage: - self.to_screen('%s: Already enrolled in' % course_id) + course_id = response['id'] + course_title = response.get('title') - response = self._download_json('https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id, - course_id, 'Downloading course curriculum') + self._enroll_course(url, webpage, course_id) - entries = [ - self.url_result('https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']), 'Udemy') - for asset in response if asset.get('assetType') == 'Video' - ] + response = self._download_json( + 'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id, + course_id, 'Downloading course curriculum') + + entries = [] + chapter, chapter_number = None, None + for asset in response: + asset_type = asset.get('assetType') or asset.get('asset_type') + if asset_type == 'Video': + asset_id = asset.get('id') + if asset_id: + entry = { + '_type': 'url_transparent', + 'url': 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']), + 'ie_key': UdemyIE.ie_key(), + } + if chapter_number: + entry['chapter_number'] = chapter_number + if chapter: + entry['chapter'] = chapter + entries.append(entry) + elif asset.get('type') == 'chapter': + chapter_number = asset.get('index') or asset.get('object_index') + chapter = asset.get('title') return self.playlist_result(entries, course_id, course_title)