X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Flynda.py;h=2160d6cb08ae5b71584ffdc0d76982e2a9bdf0c0;hb=9e1a5b845586a0a5431fb72467142046d8571e6f;hp=6deed27d73f999a2e245e0c1ac19025391d474c4;hpb=7ee40b5d1cf0d4e79a049f3851262378c08d9e4c;p=youtube-dl diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py index 6deed27d7..2160d6cb0 100644 --- a/youtube_dl/extractor/lynda.py +++ b/youtube_dl/extractor/lynda.py @@ -8,7 +8,9 @@ from .common import InfoExtractor from ..utils import ( compat_urllib_parse, compat_urllib_request, - ExtractorError + ExtractorError, + int_or_none, + compat_str, ) @@ -19,16 +21,17 @@ class LyndaIE(SubtitlesInfoExtractor): _LOGIN_URL = 'https://www.lynda.com/login/login.aspx' _NETRC_MACHINE = 'lynda' - _SUCCESSFUL_LOGIN_REGEX = r'My account' + _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.' _TEST = { 'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html', - 'file': '114408.mp4', 'md5': 'ecfc6862da89489161fb9cd5f5a6fac1', 'info_dict': { + 'id': '114408', + 'ext': 'mp4', 'title': 'Using the exercise files', 'duration': 68 } @@ -41,27 +44,44 @@ class LyndaIE(SubtitlesInfoExtractor): mobj = re.match(self._VALID_URL, url) video_id = mobj.group(1) - page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id, - video_id, 'Downloading video JSON') + 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) if 'Status' in video_json: raise ExtractorError('lynda returned error: %s' % video_json['Message'], expected=True) if video_json['HasAccess'] is False: - raise ExtractorError('Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True) + raise ExtractorError( + 'Video %s is only available for members. ' % video_id + self.ACCOUNT_CREDENTIALS_HINT, expected=True) - video_id = video_json['ID'] + video_id = compat_str(video_json['ID']) duration = video_json['DurationInSeconds'] title = video_json['Title'] - formats = [{'url': fmt['Url'], + formats = [] + + fmts = video_json.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 video_json['Formats']] + } for fmt in fmts]) + + prioritized_streams = video_json.get('PrioritizedStreams') + if prioritized_streams: + 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() + ]) self._sort_formats(formats) @@ -89,15 +109,15 @@ class LyndaIE(SubtitlesInfoExtractor): '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, note='Logging in as %s' % username) + 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) + response_json = json.loads(response) state = response_json['state'] if state == 'notlogged': @@ -116,7 +136,7 @@ class LyndaIE(SubtitlesInfoExtractor): 'stayPut': 'false', } request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(confirm_form)) - login_page = self._download_webpage(request, None, note='Confirming log in and log out from another device') + 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') @@ -150,7 +170,7 @@ class LyndaIE(SubtitlesInfoExtractor): def _get_available_subtitles(self, video_id, webpage): url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id - sub = self._download_webpage(url, None, note=False) + sub = self._download_webpage(url, None, False) sub_json = json.loads(sub) return {'en': url} if len(sub_json) > 0 else {} @@ -167,7 +187,7 @@ class LyndaCourseIE(InfoExtractor): 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) @@ -179,6 +199,9 @@ class LyndaCourseIE(InfoExtractor): videos = [] (username, _) = self._get_login_info() + # 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: @@ -198,4 +221,4 @@ class LyndaCourseIE(InfoExtractor): course_title = course_json['Title'] - return self.playlist_result(entries, course_id, course_title) \ No newline at end of file + return self.playlist_result(entries, course_id, course_title)