From 328f82d59a0d3b743280c3ab532949c9220383ed Mon Sep 17 00:00:00 2001 From: =?utf8?q?Sergey=20M=E2=80=A4?= Date: Wed, 2 Dec 2015 00:48:27 +0600 Subject: [PATCH] [udemy] Semi-switch to api 2.0 (Closes #7704) * Use api 2.0 to get lectures since it provides more formats * Fix authorization for api 2.0 * Autotry enrolling in the course for single lectures * Extract additional metadata rom asset['data']['outputs'] --- youtube_dl/extractor/udemy.py | 162 ++++++++++++++++++++-------------- 1 file changed, 96 insertions(+), 66 deletions(-) diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py index e743c0262..26ae8cc36 100644 --- a/youtube_dl/extractor/udemy.py +++ b/youtube_dl/extractor/udemy.py @@ -1,14 +1,15 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..compat import ( + compat_HTTPError, compat_urllib_parse, compat_urllib_request, + compat_urlparse, ) from ..utils import ( ExtractorError, + float_or_none, int_or_none, sanitized_Request, ) @@ -19,6 +20,8 @@ class UdemyIE(InfoExtractor): _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P\d+)' _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1' _ORIGIN_URL = 'https://www.udemy.com' + _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<' + _ALREADY_ENROLLED = '>You are already taking this course.<' _NETRC_MACHINE = 'udemy' _TESTS = [{ @@ -34,6 +37,29 @@ class UdemyIE(InfoExtractor): 'skip': 'Requires udemy account credentials', }] + def _enroll_course(self, webpage, course_id): + enroll_url = self._search_regex( + r'href=(["\'])(?Phttps?://(?:www\.)?udemy\.com/course/subscribe/.+?)\1', + webpage, 'enroll url', group='url', + default='https://www.udemy.com/course/subscribe/?courseId=%s' % course_id) + webpage = self._download_webpage(enroll_url, course_id, 'Enrolling in the course') + 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) + + 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?%s' % ( + course_id, lecture_id, compat_urllib_parse.urlencode({ + 'video_only': '', + 'auto_play': '', + 'fields[lecture]': 'title,description,asset', + 'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data', + 'instructorPreviewMode': 'False', + })), + lecture_id, 'Downloading lecture JSON', fatal=False) + def _handle_error(self, response): if not isinstance(response, dict): return @@ -45,7 +71,7 @@ class UdemyIE(InfoExtractor): error_str += ' - %s' % error_data.get('formErrors') raise ExtractorError(error_str, expected=True) - def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'): + def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', *args, **kwargs): headers = { 'X-Udemy-Snail-Case': 'true', 'X-Requested-With': 'XMLHttpRequest', @@ -55,6 +81,7 @@ class UdemyIE(InfoExtractor): 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(): @@ -62,7 +89,7 @@ class UdemyIE(InfoExtractor): else: url_or_request = sanitized_Request(url_or_request, headers=headers) - response = super(UdemyIE, self)._download_json(url_or_request, video_id, note) + response = super(UdemyIE, self)._download_json(url_or_request, video_id, note, *args, **kwargs) self._handle_error(response) return response @@ -110,66 +137,77 @@ class UdemyIE(InfoExtractor): def _real_extract(self, url): lecture_id = self._match_id(url) - lecture = self._download_json( - 'https://www.udemy.com/api-1.1/lectures/%s' % lecture_id, - lecture_id, 'Downloading lecture JSON') + webpage = self._download_webpage(url, lecture_id) + + course_id = self._search_regex( + r'data-course-id=["\'](\d+)', webpage, 'course id') - asset_type = lecture.get('assetType') or lecture.get('asset_type') + 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(webpage, course_id) + lecture_id = self._download_lecture(course_id, lecture_id) + else: + raise + + title = lecture['title'] + description = lecture.get('description') + + asset = lecture['asset'] + + asset_type = asset.get('assetType') or asset.get('asset_type') if asset_type != 'Video': raise ExtractorError( 'Lecture %s is not a video' % lecture_id, expected=True) - asset = lecture['asset'] - stream_url = asset.get('streamUrl') or asset.get('stream_url') - mobj = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url) - if mobj: - return self.url_result(mobj.group(1), 'Youtube') + 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 = asset['id'] thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url') - duration = int_or_none(asset.get('data', {}).get('duration')) - - download_url = asset.get('downloadUrl') or asset.get('download_url') - - video = download_url.get('Video') or download_url.get('video') - video_480p = download_url.get('Video480p') or download_url.get('video_480p') - - formats = [{ - 'url': video_480p[0], - 'format_id': 'download-360p', - }, { - 'url': video[0], - 'format_id': 'download-720p', - }] - - # Some videos also contain formats in asset['data']['outputs'] (e.g. - # https://www.udemy.com/ios9-swift/learn/#/lecture/3383208) - outputs = asset.get('data', {}).get('outputs') - if isinstance(outputs, dict): - for format_id, f in outputs.items(): - video_url = f.get('url') - if video_url: - formats.append({ - 'url': video_url, - 'format_id': '%sp' % (f.get('labe1l') or format_id), - 'width': int_or_none(f.get('width')), - 'height': int_or_none(f.get('height')), - 'vbr': int_or_none(f.get('video_bitrate_in_kbps')), - 'vcodec': f.get('video_codec'), - 'fps': int_or_none(f.get('frame_rate')), - 'abr': int_or_none(f.get('audio_bitrate_in_kbps')), - 'acodec': f.get('audio_codec'), - 'asr': int_or_none(f.get('audio_sample_rate')), - 'tbr': int_or_none(f.get('total_bitrate_in_kbps')), - 'filesize': int_or_none(f.get('file_size_in_bytes')), + duration = float_or_none(asset.get('data', {}).get('duration')) + outputs = asset.get('data', {}).get('outputs', {}) + + formats = [] + for format_ in asset.get('download_urls', {}).get('Video', []): + video_url = format_.get('file') + if not video_url: + continue + format_id = format_.get('label') + f = { + 'url': format_['file'], + '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) + output = outputs.get(format_id) + if isinstance(output, dict): + f.update({ + 'format_id': '%sp' % (output.get('label') or format_id), + 'width': int_or_none(output.get('width')), + 'height': int_or_none(output.get('height')), + 'vbr': int_or_none(output.get('video_bitrate_in_kbps')), + 'vcodec': output.get('video_codec'), + 'fps': int_or_none(output.get('frame_rate')), + 'abr': int_or_none(output.get('audio_bitrate_in_kbps')), + 'acodec': output.get('audio_codec'), + 'asr': int_or_none(output.get('audio_sample_rate')), + 'tbr': int_or_none(output.get('total_bitrate_in_kbps')), + 'filesize': int_or_none(output.get('file_size_in_bytes')), }) + else: + f['format_id'] = '%sp' % format_id + formats.append(f) self._sort_formats(formats) - title = lecture['title'] - description = lecture.get('description') - return { 'id': video_id, 'title': title, @@ -182,9 +220,7 @@ class UdemyIE(InfoExtractor): 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[\da-z-]+)' _TESTS = [] @classmethod @@ -192,24 +228,18 @@ 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') + course_path = self._match_id(url) + + webpage = self._download_webpage(url, course_path) response = self._download_json( 'https://www.udemy.com/api-1.1/courses/%s' % course_path, course_path, 'Downloading course JSON') - course_id = int(response['id']) - course_title = response['title'] + course_id = response['id'] + course_title = response.get('title') - webpage = self._download_webpage( - 'https://www.udemy.com/course/subscribe/?courseId=%s' % course_id, - course_id, 'Enrolling in the course') - - 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(webpage, course_id) response = self._download_json( 'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id, -- 2.22.2