[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,
    compat_urlparse,
)
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, base_url, 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:
            if not enroll_url.startswith('http'):
                enroll_url = compat_urlparse.urljoin(base_url, 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(url, 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(url, 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,
ff9d5d09	8	compat_urlparse,
1cc79574 PH	9	)
1cc79574 PH	10	from ..utils import (
e5de3f6c	11	ExtractorError,
328f82d5	12	float_or_none,
3b35c342	13	int_or_none,
5c2266df	14	sanitized_Request,
17b2d7ca	15	unescapeHTML,
e5de3f6c S	16	)
	17
	18
	19	class UdemyIE(InfoExtractor):
	20	IE_NAME = 'udemy'
	21	_VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/\|lecture/view/?\?lectureId=)(?P<id>\d+)'
dcd4d95c S	22	_LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1'
dcd4d95c S	23	_ORIGIN_URL = 'https://www.udemy.com'
e5de3f6c S	24	_NETRC_MACHINE = 'udemy'
e5de3f6c S	25
6563837e	26	_TESTS = [{
e5de3f6c S	27	'url': 'https://www.udemy.com/java-tutorial/#/lecture/172757',
	28	'md5': '98eda5b657e752cf945d8445e261b5c5',
	29	'info_dict': {
	30	'id': '160614',
	31	'ext': 'mp4',
	32	'title': 'Introduction and Installation',
	33	'description': 'md5:c0d51f6f21ef4ec65f091055a5eef876',
	34	'duration': 579.29,
	35	},
	36	'skip': 'Requires udemy account credentials',
6563837e	37	}]
e5de3f6c	38
ff9d5d09	39	def _enroll_course(self, base_url, webpage, course_id):
17b2d7ca S	40	checkout_url = unescapeHTML(self._search_regex(
	41	r'href=(["\'])(?P<url>https?://(?:www\.)?udemy\.com/payment/checkout/.+?)\1',
	42	webpage, 'checkout url', group='url', default=None))
	43	if checkout_url:
	44	raise ExtractorError(
f20756fb	45	'Course %s is not free. You have to pay for it before you can download. '
17b2d7ca S	46	'Use this URL to confirm purchase: %s' % (course_id, checkout_url), expected=True)
	47
	48	enroll_url = unescapeHTML(self._search_regex(
ff9d5d09	49	r'href=(["\'])(?P<url>(?:https?://(?:www\.)?udemy\.com)?/course/subscribe/.+?)\1',
17b2d7ca S	50	webpage, 'enroll url', group='url', default=None))
17b2d7ca S	51	if enroll_url:
ff9d5d09 S	52	if not enroll_url.startswith('http'):
ff9d5d09 S	53	enroll_url = compat_urlparse.urljoin(base_url, enroll_url)
17b2d7ca S	54	webpage = self._download_webpage(enroll_url, course_id, 'Enrolling in the course')
	55	if '>You have enrolled in' in webpage:
	56	self.to_screen('%s: Successfully enrolled in the course' % course_id)
328f82d5 S	57
	58	def _download_lecture(self, course_id, lecture_id):
	59	return self._download_json(
	60	'https://www.udemy.com/api-2.0/users/me/subscribed-courses/%s/lectures/%s?%s' % (
	61	course_id, lecture_id, compat_urllib_parse.urlencode({
	62	'video_only': '',
	63	'auto_play': '',
	64	'fields[lecture]': 'title,description,asset',
	65	'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data',
	66	'instructorPreviewMode': 'False',
	67	})),
24121bc7	68	lecture_id, 'Downloading lecture JSON')
328f82d5	69
e5de3f6c S	70	def _handle_error(self, response):
	71	if not isinstance(response, dict):
	72	return
	73	error = response.get('error')
	74	if error:
	75	error_str = 'Udemy returned error #%s: %s' % (error.get('code'), error.get('message'))
	76	error_data = error.get('data')
	77	if error_data:
	78	error_str += ' - %s' % error_data.get('formErrors')
	79	raise ExtractorError(error_str, expected=True)
	80
24121bc7	81	def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
e2937118 S	82	headers = {
	83	'X-Udemy-Snail-Case': 'true',
	84	'X-Requested-With': 'XMLHttpRequest',
	85	}
	86	for cookie in self._downloader.cookiejar:
	87	if cookie.name == 'client_id':
	88	headers['X-Udemy-Client-Id'] = cookie.value
	89	elif cookie.name == 'access_token':
	90	headers['X-Udemy-Bearer-Token'] = cookie.value
328f82d5	91	headers['X-Udemy-Authorization'] = 'Bearer %s' % cookie.value
61c0663c S	92
	93	if isinstance(url_or_request, compat_urllib_request.Request):
	94	for header, value in headers.items():
	95	url_or_request.add_header(header, value)
	96	else:
5c2266df	97	url_or_request = sanitized_Request(url_or_request, headers=headers)
61c0663c	98
24121bc7	99	response = super(UdemyIE, self)._download_json(url_or_request, video_id, note)
61c0663c S	100	self._handle_error(response)
61c0663c S	101	return response
e2937118	102
e5de3f6c S	103	def _real_initialize(self):
	104	self._login()
	105
	106	def _login(self):
	107	(username, password) = self._get_login_info()
	108	if username is None:
78717fc3	109	return
e5de3f6c S	110
e5de3f6c S	111	login_popup = self._download_webpage(
dcd4d95c	112	self._LOGIN_URL, None, 'Downloading login popup')
e5de3f6c	113
d609edf4 S	114	def is_logged(webpage):
	115	return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<'])
	116
	117	# already logged in
	118	if is_logged(login_popup):
e5de3f6c S	119	return
e5de3f6c S	120
dcd4d95c	121	login_form = self._form_hidden_inputs('login-form', login_popup)
e5de3f6c	122
dcd4d95c	123	login_form.update({
dcd4d95c S	124	'email': username.encode('utf-8'),
	125	'password': password.encode('utf-8'),
	126	})
	127
5c2266df	128	request = sanitized_Request(
61c0663c	129	self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
dcd4d95c S	130	request.add_header('Referer', self._ORIGIN_URL)
	131	request.add_header('Origin', self._ORIGIN_URL)
	132
	133	response = self._download_webpage(
e2937118	134	request, None, 'Logging in as %s' % username)
e5de3f6c	135
d609edf4	136	if not is_logged(response):
dcd4d95c S	137	error = self._html_search_regex(
	138	r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
	139	response, 'error message', default=None)
	140	if error:
	141	raise ExtractorError('Unable to login: %s' % error, expected=True)
e5de3f6c S	142	raise ExtractorError('Unable to log in')
	143
	144	def _real_extract(self, url):
810fb84d	145	lecture_id = self._match_id(url)
e5de3f6c	146
328f82d5 S	147	webpage = self._download_webpage(url, lecture_id)
	148
	149	course_id = self._search_regex(
70cab344	150	(r'data-course-id=["\'](\d+)', r'"id"\s:\s(\d+)'),
a7ba57dc	151	webpage, 'course id')
e5de3f6c	152
328f82d5 S	153	try:
	154	lecture = self._download_lecture(course_id, lecture_id)
	155	except ExtractorError as e:
	156	# Error could possibly mean we are not enrolled in the course
	157	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
ff9d5d09	158	self._enroll_course(url, webpage, course_id)
3092fc40	159	lecture = self._download_lecture(course_id, lecture_id)
328f82d5 S	160	else:
	161	raise
	162
	163	title = lecture['title']
	164	description = lecture.get('description')
	165
	166	asset = lecture['asset']
	167
	168	asset_type = asset.get('assetType') or asset.get('asset_type')
e2937118 S	169	if asset_type != 'Video':
	170	raise ExtractorError(
	171	'Lecture %s is not a video' % lecture_id, expected=True)
e5de3f6c	172
e2937118	173	stream_url = asset.get('streamUrl') or asset.get('stream_url')
328f82d5 S	174	if stream_url:
	175	youtube_url = self._search_regex(
	176	r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url, 'youtube URL', default=None)
	177	if youtube_url:
	178	return self.url_result(youtube_url, 'Youtube')
e5de3f6c S	179
e5de3f6c S	180	video_id = asset['id']
e2937118	181	thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url')
328f82d5 S	182	duration = float_or_none(asset.get('data', {}).get('duration'))
	183	outputs = asset.get('data', {}).get('outputs', {})
	184
	185	formats = []
	186	for format_ in asset.get('download_urls', {}).get('Video', []):
	187	video_url = format_.get('file')
	188	if not video_url:
	189	continue
	190	format_id = format_.get('label')
	191	f = {
	192	'url': format_['file'],
	193	'height': int_or_none(format_id),
	194	}
	195	if format_id:
	196	# Some videos contain additional metadata (e.g.
	197	# https://www.udemy.com/ios9-swift/learn/#/lecture/3383208)
	198	output = outputs.get(format_id)
	199	if isinstance(output, dict):
	200	f.update({
	201	'format_id': '%sp' % (output.get('label') or format_id),
	202	'width': int_or_none(output.get('width')),
	203	'height': int_or_none(output.get('height')),
	204	'vbr': int_or_none(output.get('video_bitrate_in_kbps')),
	205	'vcodec': output.get('video_codec'),
	206	'fps': int_or_none(output.get('frame_rate')),
	207	'abr': int_or_none(output.get('audio_bitrate_in_kbps')),
	208	'acodec': output.get('audio_codec'),
	209	'asr': int_or_none(output.get('audio_sample_rate')),
	210	'tbr': int_or_none(output.get('total_bitrate_in_kbps')),
	211	'filesize': int_or_none(output.get('file_size_in_bytes')),
3b35c342	212	})
328f82d5 S	213	else:
	214	f['format_id'] = '%sp' % format_id
	215	formats.append(f)
3b35c342 S	216
3b35c342 S	217	self._sort_formats(formats)
e5de3f6c	218
e5de3f6c S	219	return {
	220	'id': video_id,
	221	'title': title,
	222	'description': description,
	223	'thumbnail': thumbnail,
	224	'duration': duration,
	225	'formats': formats
	226	}
	227
	228
	229	class UdemyCourseIE(UdemyIE):
	230	IE_NAME = 'udemy:course'
328f82d5	231	_VALID_URL = r'https?://www\.udemy\.com/(?P<id>[\da-z-]+)'
6563837e	232	_TESTS = []
e5de3f6c S	233
	234	@classmethod
	235	def suitable(cls, url):
	236	return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
	237
	238	def _real_extract(self, url):
328f82d5 S	239	course_path = self._match_id(url)
	240
	241	webpage = self._download_webpage(url, course_path)
e5de3f6c	242
61c0663c	243	response = self._download_json(
e2937118 S	244	'https://www.udemy.com/api-1.1/courses/%s' % course_path,
e2937118 S	245	course_path, 'Downloading course JSON')
e5de3f6c	246
328f82d5 S	247	course_id = response['id']
328f82d5 S	248	course_title = response.get('title')
e5de3f6c	249
ff9d5d09	250	self._enroll_course(url, webpage, course_id)
e5de3f6c	251
61c0663c	252	response = self._download_json(
e2937118 S	253	'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
e2937118 S	254	course_id, 'Downloading course curriculum')
e5de3f6c	255
4d402db5	256	entries = []
5bafcf65	257	chapter, chapter_number = None, None
4d402db5 S	258	for asset in response:
	259	asset_type = asset.get('assetType') or asset.get('asset_type')
	260	if asset_type == 'Video':
	261	asset_id = asset.get('id')
	262	if asset_id:
	263	entry = {
	264	'_type': 'url_transparent',
	265	'url': 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']),
	266	'ie_key': UdemyIE.ie_key(),
	267	}
5bafcf65 S	268	if chapter_number:
5bafcf65 S	269	entry['chapter_number'] = chapter_number
4d402db5 S	270	if chapter:
	271	entry['chapter'] = chapter
	272	entries.append(entry)
	273	elif asset.get('type') == 'chapter':
5bafcf65	274	chapter_number = asset.get('index') or asset.get('object_index')
4d402db5	275	chapter = asset.get('title')
e5de3f6c	276
5f6a1245	277	return self.playlist_result(entries, course_id, course_title)