X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/d1a5eac72c0b1116e8efbbf50766d8d00678f5c8..d2e1a98478a93d5d191b83a9aa545df270ebf323:/youtube_dl/extractor/lynda.py diff --git a/youtube_dl/extractor/lynda.py b/youtube_dl/extractor/lynda.py index 6deed27..5dc22da 100644 --- a/youtube_dl/extractor/lynda.py +++ b/youtube_dl/extractor/lynda.py @@ -3,36 +3,43 @@ from __future__ import unicode_literals import re import json -from .subtitles import SubtitlesInfoExtractor from .common import InfoExtractor -from ..utils import ( +from ..compat import ( + compat_str, compat_urllib_parse, compat_urllib_request, - ExtractorError +) +from ..utils import ( + ExtractorError, + int_or_none, ) -class LyndaIE(SubtitlesInfoExtractor): +class LyndaIE(InfoExtractor): IE_NAME = 'lynda' IE_DESC = 'lynda.com videos' - _VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html' + _VALID_URL = r'https?://www\.lynda\.com/(?:[^/]+/[^/]+/\d+|player/embed)/(\d+)' _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 = { + _TESTS = [{ '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 } - } + }, { + 'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0', + 'only_matching': True, + }] def _real_initialize(self): self._login() @@ -41,35 +48,49 @@ 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._check_formats(formats, video_id) self._sort_formats(formats) - if self._downloader.params.get('listsubtitles', False): - self._list_available_subtitles(video_id, page) - return - - subtitles = self._fix_subtitles(self.extract_subtitles(video_id, page)) + subtitles = self.extract_subtitles(video_id, page) return { 'id': video_id, @@ -89,15 +110,15 @@ 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': @@ -116,43 +137,36 @@ 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') - 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: + def _fix_subtitles(self, subs): + 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 - 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): + 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'].lstrip() + srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), 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 - sub = self._download_webpage(url, None, note=False) - sub_json = json.loads(sub) - return {'en': url} if len(sub_json) > 0 else {} + subs = self._download_json(url, None, False) + if subs: + return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]} + else: + return {} class LyndaCourseIE(InfoExtractor): @@ -167,7 +181,7 @@ 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 +193,9 @@ 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 +215,4 @@ 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)