X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Flynda.py;h=2160d6cb08ae5b71584ffdc0d76982e2a9bdf0c0;hb=9e1a5b845586a0a5431fb72467142046d8571e6f;hp=dd59aa3e6cb3be989a39aa56a95791e4165a8834;hpb=c7f8537dd9752e83661b69e46a4f113222c7a503;p=youtube-dl diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py index dd59aa3e6..2160d6cb0 100644 --- a/youtube_dl/extractor/lynda.py +++ b/youtube_dl/extractor/lynda.py @@ -1,64 +1,183 @@ +from __future__ import unicode_literals + import re import json +from .subtitles import SubtitlesInfoExtractor from .common import InfoExtractor -from ..utils import ExtractorError +from ..utils import ( + compat_urllib_parse, + compat_urllib_request, + ExtractorError, + int_or_none, + compat_str, +) + + +class LyndaIE(SubtitlesInfoExtractor): + IE_NAME = 'lynda' + IE_DESC = 'lynda.com videos' + _VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html' + _LOGIN_URL = 'https://www.lynda.com/login/login.aspx' + _NETRC_MACHINE = 'lynda' + _SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true' + _TIMECODE_REGEX = r'\[(?P\d+:\d+:\d+[\.,]\d+)\]' -class LyndaIE(InfoExtractor): - IE_NAME = u'lynda' - IE_DESC = u'lynda.com videos' - _VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html' + ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.' _TEST = { - u'url': u'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html', - u'file': u'114408.mp4', - u'md5': u'ecfc6862da89489161fb9cd5f5a6fac1', - u"info_dict": { - u'title': u'Using the exercise files', - u'duration': 68 + 'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html', + 'md5': 'ecfc6862da89489161fb9cd5f5a6fac1', + 'info_dict': { + 'id': '114408', + 'ext': 'mp4', + 'title': 'Using the exercise files', + 'duration': 68 } } + def _real_initialize(self): + self._login() + def _real_extract(self, url): 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, u'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 u'Status' in video_json and video_json[u'Status'] == u'NotFound': - raise ExtractorError(u'Video %s does not exist' % video_id, expected=True) + 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) + + video_id = compat_str(video_json['ID']) + duration = video_json['DurationInSeconds'] + title = video_json['Title'] + + 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 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() + ]) - if video_json[u'HasAccess'] is False: - raise ExtractorError(u'Video %s is only available for members' % video_id, expected=True) - - video_id = video_json[u'ID'] - duration = video_json[u'DurationInSeconds'] - title = video_json[u'Title'] + self._sort_formats(formats) - formats = [{'url': fmt[u'Url'], - 'ext': fmt[u'Extension'], - 'width': fmt[u'Width'], - 'height': fmt[u'Height'], - 'filesize': fmt[u'FileSize'], - 'format_id': fmt[u'Resolution'] - } for fmt in video_json[u'Formats']] + if self._downloader.params.get('listsubtitles', False): + self._list_available_subtitles(video_id, page) + return - self._sort_formats(formats) + subtitles = self._fix_subtitles(self.extract_subtitles(video_id, page)) return { 'id': video_id, 'title': title, 'duration': duration, + 'subtitles': subtitles, '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, subtitles): + if subtitles is None: + return subtitles # subtitles not requested + + fixed_subtitles = {} + for k, v in subtitles.items(): + subs = json.loads(v) + if len(subs) == 0: + continue + srt = '' + for pos in range(0, len(subs) - 1): + seq_current = subs[pos] + m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode']) + if m_current is None: + continue + seq_next = subs[pos + 1] + m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode']) + if m_next is None: + continue + appear_time = m_current.group('timecode') + disappear_time = m_next.group('timecode') + text = seq_current['Caption'] + srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text) + if srt: + fixed_subtitles[k] = srt + return fixed_subtitles + + 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, False) + sub_json = json.loads(sub) + return {'en': url} if len(sub_json) > 0 else {} + class LyndaCourseIE(InfoExtractor): - IE_NAME = u'lynda:course' - IE_DESC = u'lynda.com online courses' + 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 @@ -70,27 +189,36 @@ class LyndaCourseIE(InfoExtractor): course_id = mobj.group('courseid') page = self._download_webpage('http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id, - course_id, u'Downloading course JSON') + course_id, 'Downloading course JSON') course_json = json.loads(page) - if u'Status' in course_json and course_json[u'Status'] == u'NotFound': - raise ExtractorError(u'Course %s does not exist' % course_id, expected=True) + if 'Status' in course_json and course_json['Status'] == 'NotFound': + raise ExtractorError('Course %s does not exist' % course_id, expected=True) unaccessible_videos = 0 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[u'Chapters']: - for video in chapter[u'Videos']: - if video[u'HasAccess'] is not True: + for chapter in course_json['Chapters']: + for video in chapter['Videos']: + if username is None and video['HasAccess'] is False: unaccessible_videos += 1 continue - videos.append(video[u'ID']) + videos.append(video['ID']) if unaccessible_videos > 0: - self._downloader.report_warning(u'%s videos are only available for members and will not be downloaded' % unaccessible_videos) + 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] + entries = [ + self.url_result('http://www.lynda.com/%s/%s-4.html' % + (course_path, video_id), + 'Lynda') + for video_id in videos] - course_title = course_json[u'Title'] + 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)