X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/f46044c66663049e286c20ee015db99d47d9dd8a..4d073ced87f1fff7fd6fdfd82462444769b9f90b:/youtube_dl/extractor/lynda.py diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py index 5dc22da..f5c7abc 100644 --- a/youtube_dl/extractor/lynda.py +++ b/youtube_dl/extractor/lynda.py @@ -1,35 +1,114 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor from ..compat import ( + compat_HTTPError, compat_str, - compat_urllib_parse, - compat_urllib_request, + compat_urlparse, ) from ..utils import ( ExtractorError, int_or_none, + urlencode_postdata, ) -class LyndaIE(InfoExtractor): +class LyndaBaseIE(InfoExtractor): + _SIGNIN_URL = 'https://www.lynda.com/signin' + _PASSWORD_URL = 'https://www.lynda.com/signin/password' + _USER_URL = 'https://www.lynda.com/signin/user' + _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.' + _NETRC_MACHINE = 'lynda' + + def _real_initialize(self): + self._login() + + @staticmethod + def _check_error(json_string, key_or_keys): + keys = [key_or_keys] if isinstance(key_or_keys, compat_str) else key_or_keys + for key in keys: + error = json_string.get(key) + if error: + raise ExtractorError('Unable to login: %s' % error, expected=True) + + def _login_step(self, form_html, fallback_action_url, extra_form_data, note, referrer_url): + action_url = self._search_regex( + r']+action=(["\'])(?P.+?)\1', form_html, + 'post url', default=fallback_action_url, group='url') + + if not action_url.startswith('http'): + action_url = compat_urlparse.urljoin(self._SIGNIN_URL, action_url) + + form_data = self._hidden_inputs(form_html) + form_data.update(extra_form_data) + + try: + response = self._download_json( + action_url, None, note, + data=urlencode_postdata(form_data), + headers={ + 'Referer': referrer_url, + 'X-Requested-With': 'XMLHttpRequest', + }) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500: + response = self._parse_json(e.cause.read().decode('utf-8'), None) + self._check_error(response, ('email', 'password')) + raise + + self._check_error(response, 'ErrorMessage') + + return response, action_url + + def _login(self): + username, password = self._get_login_info() + if username is None: + return + + # Step 1: download signin page + signin_page = self._download_webpage( + self._SIGNIN_URL, None, 'Downloading signin page') + + # Already logged in + if any(re.search(p, signin_page) for p in ( + r'isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')): + return + + # Step 2: submit email + signin_form = self._search_regex( + r'(?s)(]+data-form-name=["\']signin["\'][^>]*>.+?)', + signin_page, 'signin form') + signin_page, signin_url = self._login_step( + signin_form, self._PASSWORD_URL, {'email': username}, + 'Submitting email', self._SIGNIN_URL) + + # Step 3: submit password + password_form = signin_page['body'] + self._login_step( + password_form, self._USER_URL, {'email': username, 'password': password}, + 'Submitting password', signin_url) + + +class LyndaIE(LyndaBaseIE): IE_NAME = 'lynda' IE_DESC = 'lynda.com videos' - _VALID_URL = r'https?://www\.lynda\.com/(?:[^/]+/[^/]+/\d+|player/embed)/(\d+)' - _LOGIN_URL = 'https://www.lynda.com/login/login.aspx' - _NETRC_MACHINE = 'lynda' + _VALID_URL = r'''(?x) + https?:// + (?:www\.)?(?:lynda\.com|educourse\.ga)/ + (?: + (?:[^/]+/){2,3}(?P\d+)| + player/embed + )/ + (?P\d+) + ''' - _SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true' _TIMECODE_REGEX = r'\[(?P\d+:\d+:\d+[\.,]\d+)\]' - ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.' - _TESTS = [{ - 'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html', - 'md5': 'ecfc6862da89489161fb9cd5f5a6fac1', + 'url': 'https://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html', + # md5 is unstable 'info_dict': { 'id': '114408', 'ext': 'mp4', @@ -39,58 +118,110 @@ class LyndaIE(InfoExtractor): }, { 'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0', 'only_matching': True, + }, { + 'url': 'https://educourse.ga/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html', + 'only_matching': True, + }, { + 'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Willkommen-Grundlagen-guten-Gestaltung/393570/393572-4.html', + 'only_matching': True, }] - def _real_initialize(self): - self._login() + def _raise_unavailable(self, video_id): + self.raise_login_required( + 'Video %s is only available for members' % video_id) def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id = mobj.group(1) + video_id = mobj.group('id') + course_id = mobj.group('course_id') + + query = { + 'videoId': video_id, + 'type': 'video', + } - page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id, video_id, - 'Downloading video JSON') - video_json = json.loads(page) + video = self._download_json( + 'https://www.lynda.com/ajax/player', video_id, + 'Downloading video JSON', fatal=False, query=query) - if 'Status' in video_json: - raise ExtractorError('lynda returned error: %s' % video_json['Message'], expected=True) + # Fallback scenario + if not video: + query['courseId'] = course_id - if video_json['HasAccess'] is False: + play = self._download_json( + 'https://www.lynda.com/ajax/course/%s/%s/play' + % (course_id, video_id), video_id, 'Downloading play JSON') + + if not play: + self._raise_unavailable(video_id) + + formats = [] + for formats_dict in play: + urls = formats_dict.get('urls') + if not isinstance(urls, dict): + continue + cdn = formats_dict.get('name') + for format_id, format_url in urls.items(): + if not format_url: + continue + formats.append({ + 'url': format_url, + 'format_id': '%s-%s' % (cdn, format_id) if cdn else format_id, + 'height': int_or_none(format_id), + }) + self._sort_formats(formats) + + conviva = self._download_json( + 'https://www.lynda.com/ajax/player/conviva', video_id, + 'Downloading conviva JSON', query=query) + + return { + 'id': video_id, + 'title': conviva['VideoTitle'], + 'description': conviva.get('VideoDescription'), + 'release_year': int_or_none(conviva.get('ReleaseYear')), + 'duration': int_or_none(conviva.get('Duration')), + 'creator': conviva.get('Author'), + 'formats': formats, + } + + if 'Status' in video: raise ExtractorError( - 'Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True) + 'lynda returned error: %s' % video['Message'], expected=True) - video_id = compat_str(video_json['ID']) - duration = video_json['DurationInSeconds'] - title = video_json['Title'] + if video.get('HasAccess') is False: + self._raise_unavailable(video_id) + + video_id = compat_str(video.get('ID') or video_id) + duration = int_or_none(video.get('DurationInSeconds')) + title = video['Title'] formats = [] - fmts = video_json.get('Formats') + fmts = video.get('Formats') if fmts: - formats.extend([ - { - 'url': fmt['Url'], - 'ext': fmt['Extension'], - 'width': fmt['Width'], - 'height': fmt['Height'], - 'filesize': fmt['FileSize'], - 'format_id': str(fmt['Resolution']) - } for fmt in fmts]) - - prioritized_streams = video_json.get('PrioritizedStreams') + formats.extend([{ + 'url': f['Url'], + 'ext': f.get('Extension'), + 'width': int_or_none(f.get('Width')), + 'height': int_or_none(f.get('Height')), + 'filesize': int_or_none(f.get('FileSize')), + 'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None, + } for f in fmts if f.get('Url')]) + + prioritized_streams = video.get('PrioritizedStreams') if prioritized_streams: - formats.extend([ - { + for prioritized_stream_id, prioritized_stream in prioritized_streams.items(): + formats.extend([{ 'url': video_url, - 'width': int_or_none(format_id), - 'format_id': format_id, - } for format_id, video_url in prioritized_streams['0'].items() - ]) + 'height': int_or_none(format_id), + 'format_id': '%s-%s' % (prioritized_stream_id, format_id), + } for format_id, video_url in prioritized_stream.items()]) self._check_formats(formats, video_id) self._sort_formats(formats) - subtitles = self.extract_subtitles(video_id, page) + subtitles = self.extract_subtitles(video_id) return { 'id': video_id, @@ -100,50 +231,9 @@ class LyndaIE(InfoExtractor): 'formats': formats } - def _login(self): - (username, password) = self._get_login_info() - if username is None: - return - - login_form = { - 'username': username, - 'password': password, - 'remember': 'false', - 'stayPut': 'false' - } - request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form)) - login_page = self._download_webpage(request, None, 'Logging in as %s' % username) - - # Not (yet) logged in - m = re.search(r'loginResultJson = \'(?P[^\']+)\';', login_page) - if m is not None: - response = m.group('json') - response_json = json.loads(response) - state = response_json['state'] - - if state == 'notlogged': - raise ExtractorError('Unable to login, incorrect username and/or password', expected=True) - - # This is when we get popup: - # > You're already logged in to lynda.com on two devices. - # > If you log in here, we'll log you out of another device. - # So, we need to confirm this. - if state == 'conflicted': - confirm_form = { - 'username': '', - 'password': '', - 'resolve': 'true', - 'remember': 'false', - 'stayPut': 'false', - } - request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form)) - login_page = self._download_webpage(request, None, 'Confirming log in and log out from another device') - - if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None: - raise ExtractorError('Unable to log in') - def _fix_subtitles(self, subs): srt = '' + seq_counter = 0 for pos in range(0, len(subs) - 1): seq_current = subs[pos] m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode']) @@ -155,64 +245,95 @@ class LyndaIE(InfoExtractor): continue appear_time = m_current.group('timecode') disappear_time = m_next.group('timecode') - text = seq_current['Caption'].lstrip() - srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text) + text = seq_current['Caption'].strip() + if text: + seq_counter += 1 + srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (seq_counter, appear_time, disappear_time, text) if srt: return srt - def _get_subtitles(self, video_id, webpage): - url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id + def _get_subtitles(self, video_id): + url = 'https://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id subs = self._download_json(url, None, False) - if subs: - return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]} + fixed_subs = self._fix_subtitles(subs) + if fixed_subs: + return {'en': [{'ext': 'srt', 'data': fixed_subs}]} else: return {} -class LyndaCourseIE(InfoExtractor): +class LyndaCourseIE(LyndaBaseIE): IE_NAME = 'lynda:course' IE_DESC = 'lynda.com online courses' # Course link equals to welcome/introduction video link of same course # We will recognize it as course link - _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P[^/]+/[^/]+/(?P\d+))-\d\.html' + _VALID_URL = r'https?://(?:www|m)\.(?:lynda\.com|educourse\.ga)/(?P(?:[^/]+/){2,3}(?P\d+))-2\.html' + + _TESTS = [{ + 'url': 'https://www.lynda.com/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html', + 'only_matching': True, + }, { + 'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) course_path = mobj.group('coursepath') course_id = mobj.group('courseid') - page = self._download_webpage('http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id, - course_id, 'Downloading course JSON') - course_json = json.loads(page) - - if 'Status' in course_json and course_json['Status'] == 'NotFound': - raise ExtractorError('Course %s does not exist' % course_id, expected=True) + item_template = 'https://www.lynda.com/%s/%%s-4.html' % course_path + + course = self._download_json( + 'https://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id, + course_id, 'Downloading course JSON', fatal=False) + + if not course: + webpage = self._download_webpage(url, course_id) + entries = [ + self.url_result( + item_template % video_id, ie=LyndaIE.ie_key(), + video_id=video_id) + for video_id in re.findall( + r'data-video-id=["\'](\d+)', webpage)] + return self.playlist_result( + entries, course_id, + self._og_search_title(webpage, fatal=False), + self._og_search_description(webpage)) + + if course.get('Status') == 'NotFound': + raise ExtractorError( + 'Course %s does not exist' % course_id, expected=True) unaccessible_videos = 0 - videos = [] - (username, _) = self._get_login_info() + entries = [] # Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided # by single video API anymore - for chapter in course_json['Chapters']: - for video in chapter['Videos']: - if username is None and video['HasAccess'] is False: + for chapter in course['Chapters']: + for video in chapter.get('Videos', []): + if video.get('HasAccess') is False: unaccessible_videos += 1 continue - videos.append(video['ID']) + video_id = video.get('ID') + if video_id: + entries.append({ + '_type': 'url_transparent', + 'url': item_template % video_id, + 'ie_key': LyndaIE.ie_key(), + 'chapter': chapter.get('Title'), + 'chapter_number': int_or_none(chapter.get('ChapterIndex')), + 'chapter_id': compat_str(chapter.get('ID')), + }) if unaccessible_videos > 0: - self._downloader.report_warning('%s videos are only available for members and will not be downloaded. ' - % unaccessible_videos + LyndaIE.ACCOUNT_CREDENTIALS_HINT) - - entries = [ - self.url_result('http://www.lynda.com/%s/%s-4.html' % - (course_path, video_id), - 'Lynda') - for video_id in videos] + self._downloader.report_warning( + '%s videos are only available for members (or paid members) and will not be downloaded. ' + % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT) - course_title = course_json['Title'] + course_title = course.get('Title') + course_description = course.get('Description') - return self.playlist_result(entries, course_id, course_title) + return self.playlist_result(entries, course_id, course_title, course_description)