X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/c512650955de0b16d37e7fa7fb29ea0985e415bb..5a9d00a5190582368d533f12db7638e735d7f7c8:/youtube_dl/extractor/udemy.py?ds=inline diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py index 054f427..195f5ce 100644 --- a/youtube_dl/extractor/udemy.py +++ b/youtube_dl/extractor/udemy.py @@ -3,17 +3,39 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import ( - compat_urllib_parse, +from ..compat import ( + compat_HTTPError, + compat_str, compat_urllib_request, + compat_urlparse, +) +from ..utils import ( + determine_ext, + extract_attributes, ExtractorError, + float_or_none, + int_or_none, + js_to_json, + 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 +49,60 @@ 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, + }, { + # no url in outputs format entry + 'url': 'https://www.udemy.com/learn-web-development-complete-step-by-step-guide-to-success/learn/v4/t/lecture/4125812', + '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'data-course-id=["\'](\d+)', webpage, 'course id') + return course_id, course.get('title') + + 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|cart)/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', + 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( + '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 @@ -40,8 +114,25 @@ 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, *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: + 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 @@ -51,68 +142,216 @@ 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(re.search(p, webpage) for p in ( + r'href=["\'](?:https://www\.udemy\.com)?/user/logout/', + r'>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 = self._download_json( - 'https://www.udemy.com/api-1.1/lectures/%s' % lecture_id, lecture_id, 'Downloading lecture JSON') + lecture_id = self._match_id(url) - if lecture['assetType'] != 'Video': - raise ExtractorError('Lecture %s is not a video' % lecture_id, expected=True) - - asset = lecture['asset'] + webpage = self._download_webpage(url, lecture_id) - 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') + course_id, _ = self._extract_course_info(webpage, lecture_id) - video_id = asset['id'] - thumbnail = asset['thumbnailUrl'] - duration = asset['data']['duration'] + 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 - download_url = asset['downloadUrl'] + title = lecture['title'] + description = lecture.get('description') - formats = [ - { - 'url': download_url['Video480p'][0], - 'format_id': '360p', - }, - { - 'url': download_url['Video'][0], - 'format_id': '720p', - }, - ] + asset = lecture['asset'] - title = lecture['title'] - description = lecture['description'] + 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) + + 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 = compat_str(asset['id']) + thumbnail = asset.get('thumbnail_url') or asset.get('thumbnailUrl') + duration = float_or_none(asset.get('data', {}).get('duration')) + + subtitles = {} + automatic_captions = {} + + formats = [] + + def extract_output_format(src, f_id): + return { + 'url': src.get('url'), + 'format_id': '%sp' % (src.get('height') or f_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, key) + output_format.update(f) + return output_format + return f + + def extract_formats(source_list): + if not isinstance(source_list, list): + return + for source in source_list: + video_url = source.get('file') or source.get('src') + if not video_url or not isinstance(video_url, compat_str): + continue + if source.get('type') == 'application/x-mpegURL' or determine_ext(video_url) == 'm3u8': + formats.extend(self._extract_m3u8_formats( + video_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) + continue + format_id = source.get('label') + f = { + 'url': video_url, + '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) + + def extract_subtitles(track_list): + if not isinstance(track_list, list): + return + for track in track_list: + if not isinstance(track, dict): + continue + if track.get('kind') != 'captions': + continue + src = track.get('src') + if not src or not isinstance(src, compat_str): + continue + lang = track.get('language') or track.get( + 'srclang') or track.get('label') + sub_dict = automatic_captions if track.get( + 'autogenerated') is True else subtitles + sub_dict.setdefault(lang, []).append({ + 'url': src, + }) + + download_urls = asset.get('download_urls') + if isinstance(download_urls, dict): + extract_formats(download_urls.get('Video')) + + 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)) + + # react rendition since 2017.04.15 (see + # https://github.com/rg3/youtube-dl/issues/12744) + data = self._parse_json( + self._search_regex( + r'videojs-setup-data=(["\'])(?P{.+?})\1', view_html, + 'setup data', default='{}', group='data'), video_id, + transform_source=unescapeHTML, fatal=False) + if data and isinstance(data, dict): + extract_formats(data.get('sources')) + if not duration: + duration = int_or_none(data.get('duration')) + extract_subtitles(data.get('tracks')) + + if not subtitles and not automatic_captions: + text_tracks = self._parse_json( + self._search_regex( + r'text-tracks=(["\'])(?P\[.+?\])\1', view_html, + 'text tracks', default='{}', group='data'), video_id, + transform_source=lambda s: js_to_json(unescapeHTML(s)), + fatal=False) + extract_subtitles(text_tracks) + + self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id')) return { 'id': video_id, @@ -120,15 +359,15 @@ class UdemyIE(InfoExtractor): 'description': description, 'thumbnail': thumbnail, 'duration': duration, - 'formats': formats + 'formats': formats, + 'subtitles': subtitles, + 'automatic_captions': automatic_captions, } 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[^/?#&]+)' _TESTS = [] @classmethod @@ -136,29 +375,47 @@ 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') + course_id, title = self._extract_course_info(webpage, course_path) - 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) + self._enroll_course(url, webpage, course_id) - response = self._download_json('https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id, - course_id, 'Downloading course curriculum') - - 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' - ] - - return self.playlist_result(entries, course_id, course_title) \ No newline at end of file + 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', + 'fields[lecture]': 'title,asset', + 'page_size': '1000', + }) + + entries = [] + chapter, chapter_number = [None] * 2 + for entry in response['results']: + clazz = entry.get('_class') + if clazz == 'lecture': + 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: + entry = { + '_type': 'url_transparent', + 'url': 'https://www.udemy.com/%s/learn/v4/t/lecture/%s' % (course_path, entry['id']), + 'title': entry.get('title'), + 'ie_key': UdemyIE.ie_key(), + } + if chapter_number: + entry['chapter_number'] = chapter_number + 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)