[yt-dlp.git] / youtube_dl / extractor / udemy.py

from __future__ import unicode_literals

from .common import InfoExtractor
from ..compat import (
    compat_HTTPError,
    compat_urllib_parse,
    compat_urllib_request,
)
from ..utils import (
    ExtractorError,
    float_or_none,
    int_or_none,
    sanitized_Request,
    unescapeHTML,
)


class UdemyIE(InfoExtractor):
    IE_NAME = 'udemy'
    _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P<id>\d+)'
    _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1'
    _ORIGIN_URL = 'https://www.udemy.com'
    _NETRC_MACHINE = 'udemy'

    _TESTS = [{
        'url': 'https://www.udemy.com/java-tutorial/#/lecture/172757',
        'md5': '98eda5b657e752cf945d8445e261b5c5',
        'info_dict': {
            'id': '160614',
            'ext': 'mp4',
            'title': 'Introduction and Installation',
            'description': 'md5:c0d51f6f21ef4ec65f091055a5eef876',
            'duration': 579.29,
        },
        'skip': 'Requires udemy account credentials',
    }]

    def _enroll_course(self, webpage, course_id):
        checkout_url = unescapeHTML(self._search_regex(
            r'href=(["\'])(?P<url>https?://(?:www\.)?udemy\.com/payment/checkout/.+?)\1',
            webpage, 'checkout url', group='url', default=None))
        if checkout_url:
            raise ExtractorError(
                'Course %s is not free. You have to pay for it before you can download. '
                'Use this URL to confirm purchase: %s' % (course_id, checkout_url), expected=True)

        enroll_url = unescapeHTML(self._search_regex(
            r'href=(["\'])(?P<url>https?://(?:www\.)?udemy\.com/course/subscribe/.+?)\1',
            webpage, 'enroll url', group='url', default=None))
        if enroll_url:
            webpage = self._download_webpage(enroll_url, course_id, 'Enrolling in the course')
            if '>You have enrolled in' in webpage:
                self.to_screen('%s: Successfully enrolled in the course' % 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')

    def _handle_error(self, response):
        if not isinstance(response, dict):
            return
        error = response.get('error')
        if error:
            error_str = 'Udemy returned error #%s: %s' % (error.get('code'), error.get('message'))
            error_data = error.get('data')
            if error_data:
                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'):
        headers = {
            'X-Udemy-Snail-Case': 'true',
            'X-Requested-With': 'XMLHttpRequest',
        }
        for cookie in self._downloader.cookiejar:
            if cookie.name == 'client_id':
                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():
                url_or_request.add_header(header, value)
        else:
            url_or_request = sanitized_Request(url_or_request, headers=headers)

        response = super(UdemyIE, self)._download_json(url_or_request, video_id, note)
        self._handle_error(response)
        return response

    def _real_initialize(self):
        self._login()

    def _login(self):
        (username, password) = self._get_login_info()
        if username is None:
            return

        login_popup = self._download_webpage(
            self._LOGIN_URL, None, 'Downloading login popup')

        def is_logged(webpage):
            return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<'])

        # already logged in
        if is_logged(login_popup):
            return

        login_form = self._form_hidden_inputs('login-form', login_popup)

        login_form.update({
            'email': username.encode('utf-8'),
            'password': password.encode('utf-8'),
        })

        request = sanitized_Request(
            self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
        request.add_header('Referer', self._ORIGIN_URL)
        request.add_header('Origin', self._ORIGIN_URL)

        response = self._download_webpage(
            request, None, 'Logging in as %s' % username)

        if not is_logged(response):
            error = self._html_search_regex(
                r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
                response, 'error message', default=None)
            if error:
                raise ExtractorError('Unable to login: %s' % error, expected=True)
            raise ExtractorError('Unable to log in')

    def _real_extract(self, url):
        lecture_id = self._match_id(url)

        webpage = self._download_webpage(url, lecture_id)

        course_id = self._search_regex(
            (r'data-course-id=["\'](\d+)', r'&quot;id&quot;\s*:\s*(\d+)'),
            webpage, 'course id')

        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 = 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)

        stream_url = asset.get('streamUrl') or asset.get('stream_url')
        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 = 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)

        return {
            'id': video_id,
            'title': title,
            'description': description,
            'thumbnail': thumbnail,
            'duration': duration,
            'formats': formats
        }


class UdemyCourseIE(UdemyIE):
    IE_NAME = 'udemy:course'
    _VALID_URL = r'https?://www\.udemy\.com/(?P<id>[\da-z-]+)'
    _TESTS = []

    @classmethod
    def suitable(cls, url):
        return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)

    def _real_extract(self, url):
        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 = response['id']
        course_title = response.get('title')

        self._enroll_course(webpage, course_id)

        response = self._download_json(
            'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
            course_id, 'Downloading course curriculum')

        entries = []
        chapter, chapter_number = None, None
        for asset in response:
            asset_type = asset.get('assetType') or asset.get('asset_type')
            if asset_type == 'Video':
                asset_id = asset.get('id')
                if asset_id:
                    entry = {
                        '_type': 'url_transparent',
                        'url': 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']),
                        'ie_key': UdemyIE.ie_key(),
                    }
                    if chapter_number:
                        entry['chapter_number'] = chapter_number
                    if chapter:
                        entry['chapter'] = chapter
                    entries.append(entry)
            elif asset.get('type') == 'chapter':
                chapter_number = asset.get('index') or asset.get('object_index')
                chapter = asset.get('title')

        return self.playlist_result(entries, course_id, course_title)
Commit	Line	Data
e5de3f6c S	1	from __future__ import unicode_literals
e5de3f6c S	2
e5de3f6c	3	from .common import InfoExtractor
1cc79574	4	from ..compat import (
328f82d5	5	compat_HTTPError,
e5de3f6c S	6	compat_urllib_parse,
e5de3f6c S	7	compat_urllib_request,
1cc79574 PH	8	)
1cc79574 PH	9	from ..utils import (
e5de3f6c	10	ExtractorError,
328f82d5	11	float_or_none,
3b35c342	12	int_or_none,
5c2266df	13	sanitized_Request,
17b2d7ca	14	unescapeHTML,
e5de3f6c S	15	)
	16
	17
	18	class UdemyIE(InfoExtractor):
	19	IE_NAME = 'udemy'
	20	_VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/\|lecture/view/?\?lectureId=)(?P<id>\d+)'
dcd4d95c S	21	_LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1'
dcd4d95c S	22	_ORIGIN_URL = 'https://www.udemy.com'
e5de3f6c S	23	_NETRC_MACHINE = 'udemy'
e5de3f6c S	24
6563837e	25	_TESTS = [{
e5de3f6c S	26	'url': 'https://www.udemy.com/java-tutorial/#/lecture/172757',
	27	'md5': '98eda5b657e752cf945d8445e261b5c5',
	28	'info_dict': {
	29	'id': '160614',
	30	'ext': 'mp4',
	31	'title': 'Introduction and Installation',
	32	'description': 'md5:c0d51f6f21ef4ec65f091055a5eef876',
	33	'duration': 579.29,
	34	},
	35	'skip': 'Requires udemy account credentials',
6563837e	36	}]
e5de3f6c	37
328f82d5	38	def _enroll_course(self, webpage, course_id):
17b2d7ca S	39	checkout_url = unescapeHTML(self._search_regex(
	40	r'href=(["\'])(?P<url>https?://(?:www\.)?udemy\.com/payment/checkout/.+?)\1',
	41	webpage, 'checkout url', group='url', default=None))
	42	if checkout_url:
	43	raise ExtractorError(
f20756fb	44	'Course %s is not free. You have to pay for it before you can download. '
17b2d7ca S	45	'Use this URL to confirm purchase: %s' % (course_id, checkout_url), expected=True)
	46
	47	enroll_url = unescapeHTML(self._search_regex(
328f82d5	48	r'href=(["\'])(?P<url>https?://(?:www\.)?udemy\.com/course/subscribe/.+?)\1',
17b2d7ca S	49	webpage, 'enroll url', group='url', default=None))
	50	if enroll_url:
	51	webpage = self._download_webpage(enroll_url, course_id, 'Enrolling in the course')
	52	if '>You have enrolled in' in webpage:
	53	self.to_screen('%s: Successfully enrolled in the course' % course_id)
328f82d5 S	54
	55	def _download_lecture(self, course_id, lecture_id):
	56	return self._download_json(
	57	'https://www.udemy.com/api-2.0/users/me/subscribed-courses/%s/lectures/%s?%s' % (
	58	course_id, lecture_id, compat_urllib_parse.urlencode({
	59	'video_only': '',
	60	'auto_play': '',
	61	'fields[lecture]': 'title,description,asset',
	62	'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data',
	63	'instructorPreviewMode': 'False',
	64	})),
24121bc7	65	lecture_id, 'Downloading lecture JSON')
328f82d5	66
e5de3f6c S	67	def _handle_error(self, response):
	68	if not isinstance(response, dict):
	69	return
	70	error = response.get('error')
	71	if error:
	72	error_str = 'Udemy returned error #%s: %s' % (error.get('code'), error.get('message'))
	73	error_data = error.get('data')
	74	if error_data:
	75	error_str += ' - %s' % error_data.get('formErrors')
	76	raise ExtractorError(error_str, expected=True)
	77
24121bc7	78	def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
e2937118 S	79	headers = {
	80	'X-Udemy-Snail-Case': 'true',
	81	'X-Requested-With': 'XMLHttpRequest',
	82	}
	83	for cookie in self._downloader.cookiejar:
	84	if cookie.name == 'client_id':
	85	headers['X-Udemy-Client-Id'] = cookie.value
	86	elif cookie.name == 'access_token':
	87	headers['X-Udemy-Bearer-Token'] = cookie.value
328f82d5	88	headers['X-Udemy-Authorization'] = 'Bearer %s' % cookie.value
61c0663c S	89
	90	if isinstance(url_or_request, compat_urllib_request.Request):
	91	for header, value in headers.items():
	92	url_or_request.add_header(header, value)
	93	else:
5c2266df	94	url_or_request = sanitized_Request(url_or_request, headers=headers)
61c0663c	95
24121bc7	96	response = super(UdemyIE, self)._download_json(url_or_request, video_id, note)
61c0663c S	97	self._handle_error(response)
61c0663c S	98	return response
e2937118	99
e5de3f6c S	100	def _real_initialize(self):
	101	self._login()
	102
	103	def _login(self):
	104	(username, password) = self._get_login_info()
	105	if username is None:
78717fc3	106	return
e5de3f6c S	107
e5de3f6c S	108	login_popup = self._download_webpage(
dcd4d95c	109	self._LOGIN_URL, None, 'Downloading login popup')
e5de3f6c	110
d609edf4 S	111	def is_logged(webpage):
	112	return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<'])
	113
	114	# already logged in
	115	if is_logged(login_popup):
e5de3f6c S	116	return
e5de3f6c S	117
dcd4d95c	118	login_form = self._form_hidden_inputs('login-form', login_popup)
e5de3f6c	119
dcd4d95c	120	login_form.update({
dcd4d95c S	121	'email': username.encode('utf-8'),
	122	'password': password.encode('utf-8'),
	123	})
	124
5c2266df	125	request = sanitized_Request(
61c0663c	126	self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
dcd4d95c S	127	request.add_header('Referer', self._ORIGIN_URL)
	128	request.add_header('Origin', self._ORIGIN_URL)
	129
	130	response = self._download_webpage(
e2937118	131	request, None, 'Logging in as %s' % username)
e5de3f6c	132
d609edf4	133	if not is_logged(response):
dcd4d95c S	134	error = self._html_search_regex(
	135	r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
	136	response, 'error message', default=None)
	137	if error:
	138	raise ExtractorError('Unable to login: %s' % error, expected=True)
e5de3f6c S	139	raise ExtractorError('Unable to log in')
	140
	141	def _real_extract(self, url):
810fb84d	142	lecture_id = self._match_id(url)
e5de3f6c	143
328f82d5 S	144	webpage = self._download_webpage(url, lecture_id)
	145
	146	course_id = self._search_regex(
70cab344	147	(r'data-course-id=["\'](\d+)', r'"id"\s:\s(\d+)'),
a7ba57dc	148	webpage, 'course id')
e5de3f6c	149
328f82d5 S	150	try:
	151	lecture = self._download_lecture(course_id, lecture_id)
	152	except ExtractorError as e:
	153	# Error could possibly mean we are not enrolled in the course
	154	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
	155	self._enroll_course(webpage, course_id)
3092fc40	156	lecture = self._download_lecture(course_id, lecture_id)
328f82d5 S	157	else:
	158	raise
	159
	160	title = lecture['title']
	161	description = lecture.get('description')
	162
	163	asset = lecture['asset']
	164
	165	asset_type = asset.get('assetType') or asset.get('asset_type')
e2937118 S	166	if asset_type != 'Video':
	167	raise ExtractorError(
	168	'Lecture %s is not a video' % lecture_id, expected=True)
e5de3f6c	169
e2937118	170	stream_url = asset.get('streamUrl') or asset.get('stream_url')
328f82d5 S	171	if stream_url:
	172	youtube_url = self._search_regex(
	173	r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url, 'youtube URL', default=None)
	174	if youtube_url:
	175	return self.url_result(youtube_url, 'Youtube')
e5de3f6c S	176
e5de3f6c S	177	video_id = asset['id']
e2937118	178	thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url')
328f82d5 S	179	duration = float_or_none(asset.get('data', {}).get('duration'))
	180	outputs = asset.get('data', {}).get('outputs', {})
	181
	182	formats = []
	183	for format_ in asset.get('download_urls', {}).get('Video', []):
	184	video_url = format_.get('file')
	185	if not video_url:
	186	continue
	187	format_id = format_.get('label')
	188	f = {
	189	'url': format_['file'],
	190	'height': int_or_none(format_id),
	191	}
	192	if format_id:
	193	# Some videos contain additional metadata (e.g.
	194	# https://www.udemy.com/ios9-swift/learn/#/lecture/3383208)
	195	output = outputs.get(format_id)
	196	if isinstance(output, dict):
	197	f.update({
	198	'format_id': '%sp' % (output.get('label') or format_id),
	199	'width': int_or_none(output.get('width')),
	200	'height': int_or_none(output.get('height')),
	201	'vbr': int_or_none(output.get('video_bitrate_in_kbps')),
	202	'vcodec': output.get('video_codec'),
	203	'fps': int_or_none(output.get('frame_rate')),
	204	'abr': int_or_none(output.get('audio_bitrate_in_kbps')),
	205	'acodec': output.get('audio_codec'),
	206	'asr': int_or_none(output.get('audio_sample_rate')),
	207	'tbr': int_or_none(output.get('total_bitrate_in_kbps')),
	208	'filesize': int_or_none(output.get('file_size_in_bytes')),
3b35c342	209	})
328f82d5 S	210	else:
	211	f['format_id'] = '%sp' % format_id
	212	formats.append(f)
3b35c342 S	213
3b35c342 S	214	self._sort_formats(formats)
e5de3f6c	215
e5de3f6c S	216	return {
	217	'id': video_id,
	218	'title': title,
	219	'description': description,
	220	'thumbnail': thumbnail,
	221	'duration': duration,
	222	'formats': formats
	223	}
	224
	225
	226	class UdemyCourseIE(UdemyIE):
	227	IE_NAME = 'udemy:course'
328f82d5	228	_VALID_URL = r'https?://www\.udemy\.com/(?P<id>[\da-z-]+)'
6563837e	229	_TESTS = []
e5de3f6c S	230
	231	@classmethod
	232	def suitable(cls, url):
	233	return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
	234
	235	def _real_extract(self, url):
328f82d5 S	236	course_path = self._match_id(url)
	237
	238	webpage = self._download_webpage(url, course_path)
e5de3f6c	239
61c0663c	240	response = self._download_json(
e2937118 S	241	'https://www.udemy.com/api-1.1/courses/%s' % course_path,
e2937118 S	242	course_path, 'Downloading course JSON')
e5de3f6c	243
328f82d5 S	244	course_id = response['id']
328f82d5 S	245	course_title = response.get('title')
e5de3f6c	246
328f82d5	247	self._enroll_course(webpage, course_id)
e5de3f6c	248
61c0663c	249	response = self._download_json(
e2937118 S	250	'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
e2937118 S	251	course_id, 'Downloading course curriculum')
e5de3f6c	252
4d402db5	253	entries = []
5bafcf65	254	chapter, chapter_number = None, None
4d402db5 S	255	for asset in response:
	256	asset_type = asset.get('assetType') or asset.get('asset_type')
	257	if asset_type == 'Video':
	258	asset_id = asset.get('id')
	259	if asset_id:
	260	entry = {
	261	'_type': 'url_transparent',
	262	'url': 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']),
	263	'ie_key': UdemyIE.ie_key(),
	264	}
5bafcf65 S	265	if chapter_number:
5bafcf65 S	266	entry['chapter_number'] = chapter_number
4d402db5 S	267	if chapter:
	268	entry['chapter'] = chapter
	269	entries.append(entry)
	270	elif asset.get('type') == 'chapter':
5bafcf65	271	chapter_number = asset.get('index') or asset.get('object_index')
4d402db5	272	chapter = asset.get('title')
e5de3f6c	273
5f6a1245	274	return self.playlist_result(entries, course_id, course_title)