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

from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..compat import (
    compat_HTTPError,
    compat_str,
    compat_urlparse,
)
from ..utils import (
    ExtractorError,
    int_or_none,
    urlencode_postdata,
)


class LyndaBaseIE(InfoExtractor):
    _SIGNIN_URL = 'https://www.lynda.com/signin'
    _PASSWORD_URL = 'https://www.lynda.com/signin/password'
    _USER_URL = 'https://www.lynda.com/signin/user'
    _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
    _NETRC_MACHINE = 'lynda'

    def _real_initialize(self):
        self._login()

    @staticmethod
    def _check_error(json_string, key_or_keys):
        keys = [key_or_keys] if isinstance(key_or_keys, compat_str) else key_or_keys
        for key in keys:
            error = json_string.get(key)
            if error:
                raise ExtractorError('Unable to login: %s' % error, expected=True)

    def _login_step(self, form_html, fallback_action_url, extra_form_data, note, referrer_url):
        action_url = self._search_regex(
            r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_html,
            'post url', default=fallback_action_url, group='url')

        if not action_url.startswith('http'):
            action_url = compat_urlparse.urljoin(self._SIGNIN_URL, action_url)

        form_data = self._hidden_inputs(form_html)
        form_data.update(extra_form_data)

        try:
            response = self._download_json(
                action_url, None, note,
                data=urlencode_postdata(form_data),
                headers={
                    'Referer': referrer_url,
                    'X-Requested-With': 'XMLHttpRequest',
                })
        except ExtractorError as e:
            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
                response = self._parse_json(e.cause.read().decode('utf-8'), None)
                self._check_error(response, ('email', 'password'))
            raise

        self._check_error(response, 'ErrorMessage')

        return response, action_url

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

        # Step 1: download signin page
        signin_page = self._download_webpage(
            self._SIGNIN_URL, None, 'Downloading signin page')

        # Already logged in
        if any(re.search(p, signin_page) for p in (
                r'isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')):
            return

        # Step 2: submit email
        signin_form = self._search_regex(
            r'(?s)(<form[^>]+data-form-name=["\']signin["\'][^>]*>.+?</form>)',
            signin_page, 'signin form')
        signin_page, signin_url = self._login_step(
            signin_form, self._PASSWORD_URL, {'email': username},
            'Submitting email', self._SIGNIN_URL)

        # Step 3: submit password
        password_form = signin_page['body']
        self._login_step(
            password_form, self._USER_URL, {'email': username, 'password': password},
            'Submitting password', signin_url)


class LyndaIE(LyndaBaseIE):
    IE_NAME = 'lynda'
    IE_DESC = 'lynda.com videos'
    _VALID_URL = r'https?://(?:www\.)?lynda\.com/(?:[^/]+/[^/]+/(?P<course_id>\d+)|player/embed)/(?P<id>\d+)'

    _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'

    _TESTS = [{
        'url': 'https://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
        # md5 is unstable
        '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 _raise_unavailable(self, video_id):
        self.raise_login_required(
            'Video %s is only available for members' % video_id)

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        video_id = mobj.group('id')
        course_id = mobj.group('course_id')

        query = {
            'videoId': video_id,
            'type': 'video',
        }

        video = self._download_json(
            'https://www.lynda.com/ajax/player', video_id,
            'Downloading video JSON', fatal=False, query=query)

        # Fallback scenario
        if not video:
            query['courseId'] = course_id

            play = self._download_json(
                'https://www.lynda.com/ajax/course/%s/%s/play'
                % (course_id, video_id), video_id, 'Downloading play JSON')

            if not play:
                self._raise_unavailable(video_id)

            formats = []
            for formats_dict in play:
                urls = formats_dict.get('urls')
                if not isinstance(urls, dict):
                    continue
                cdn = formats_dict.get('name')
                for format_id, format_url in urls.items():
                    if not format_url:
                        continue
                    formats.append({
                        'url': format_url,
                        'format_id': '%s-%s' % (cdn, format_id) if cdn else format_id,
                        'height': int_or_none(format_id),
                    })
            self._sort_formats(formats)

            conviva = self._download_json(
                'https://www.lynda.com/ajax/player/conviva', video_id,
                'Downloading conviva JSON', query=query)

            return {
                'id': video_id,
                'title': conviva['VideoTitle'],
                'description': conviva.get('VideoDescription'),
                'release_year': int_or_none(conviva.get('ReleaseYear')),
                'duration': int_or_none(conviva.get('Duration')),
                'creator': conviva.get('Author'),
                'formats': formats,
            }

        if 'Status' in video:
            raise ExtractorError(
                'lynda returned error: %s' % video['Message'], expected=True)

        if video.get('HasAccess') is False:
            self._raise_unavailable(video_id)

        video_id = compat_str(video.get('ID') or video_id)
        duration = int_or_none(video.get('DurationInSeconds'))
        title = video['Title']

        formats = []

        fmts = video.get('Formats')
        if fmts:
            formats.extend([{
                'url': f['Url'],
                'ext': f.get('Extension'),
                'width': int_or_none(f.get('Width')),
                'height': int_or_none(f.get('Height')),
                'filesize': int_or_none(f.get('FileSize')),
                'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None,
            } for f in fmts if f.get('Url')])

        prioritized_streams = video.get('PrioritizedStreams')
        if prioritized_streams:
            for prioritized_stream_id, prioritized_stream in prioritized_streams.items():
                formats.extend([{
                    'url': video_url,
                    'height': int_or_none(format_id),
                    'format_id': '%s-%s' % (prioritized_stream_id, format_id),
                } for format_id, video_url in prioritized_stream.items()])

        self._check_formats(formats, video_id)
        self._sort_formats(formats)

        subtitles = self.extract_subtitles(video_id)

        return {
            'id': video_id,
            'title': title,
            'duration': duration,
            'subtitles': subtitles,
            'formats': formats
        }

    def _fix_subtitles(self, subs):
        srt = ''
        seq_counter = 0
        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'].strip()
            if text:
                seq_counter += 1
                srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (seq_counter, appear_time, disappear_time, text)
        if srt:
            return srt

    def _get_subtitles(self, video_id):
        url = 'https://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
        subs = self._download_json(url, None, False)
        if subs:
            return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]}
        else:
            return {}


class LyndaCourseIE(LyndaBaseIE):
    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
    _VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        course_path = mobj.group('coursepath')
        course_id = mobj.group('courseid')

        course = self._download_json(
            'https://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
            course_id, 'Downloading course JSON')

        if course.get('Status') == 'NotFound':
            raise ExtractorError(
                'Course %s does not exist' % course_id, expected=True)

        unaccessible_videos = 0
        entries = []

        # 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['Chapters']:
            for video in chapter.get('Videos', []):
                if video.get('HasAccess') is False:
                    unaccessible_videos += 1
                    continue
                video_id = video.get('ID')
                if video_id:
                    entries.append({
                        '_type': 'url_transparent',
                        'url': 'https://www.lynda.com/%s/%s-4.html' % (course_path, video_id),
                        'ie_key': LyndaIE.ie_key(),
                        'chapter': chapter.get('Title'),
                        'chapter_number': int_or_none(chapter.get('ChapterIndex')),
                        'chapter_id': compat_str(chapter.get('ID')),
                    })

        if unaccessible_videos > 0:
            self._downloader.report_warning(
                '%s videos are only available for members (or paid members) and will not be downloaded. '
                % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT)

        course_title = course.get('Title')
        course_description = course.get('Description')

        return self.playlist_result(entries, course_id, course_title, course_description)
Commit	Line	Data
a7c26e73 PH	1	from __future__ import unicode_literals
a7c26e73 PH	2
c7f8537d	3	import re
c7f8537d	4
c7f8537d	5	from .common import InfoExtractor
bdf16f81 S	6	from ..compat import (
	7	compat_HTTPError,
	8	compat_str,
	9	compat_urlparse,
	10	)
1cc79574	11	from ..utils import (
16ff7ebc S	12	ExtractorError,
16ff7ebc S	13	int_or_none,
6e6bc8da	14	urlencode_postdata,
7ee40b5d	15	)
c7f8537d	16
c7f8537d	17
30cbd4e0	18	class LyndaBaseIE(InfoExtractor):
bdf16f81 S	19	_SIGNIN_URL = 'https://www.lynda.com/signin'
	20	_PASSWORD_URL = 'https://www.lynda.com/signin/password'
	21	_USER_URL = 'https://www.lynda.com/signin/user'
30cbd4e0	22	_ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
499bfcbf	23	_NETRC_MACHINE = 'lynda'
30cbd4e0 S	24
	25	def _real_initialize(self):
	26	self._login()
	27
bdf16f81 S	28	@staticmethod
	29	def _check_error(json_string, key_or_keys):
	30	keys = [key_or_keys] if isinstance(key_or_keys, compat_str) else key_or_keys
	31	for key in keys:
	32	error = json_string.get(key)
	33	if error:
	34	raise ExtractorError('Unable to login: %s' % error, expected=True)
	35
	36	def _login_step(self, form_html, fallback_action_url, extra_form_data, note, referrer_url):
	37	action_url = self._search_regex(
	38	r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_html,
	39	'post url', default=fallback_action_url, group='url')
	40
	41	if not action_url.startswith('http'):
	42	action_url = compat_urlparse.urljoin(self._SIGNIN_URL, action_url)
	43
	44	form_data = self._hidden_inputs(form_html)
	45	form_data.update(extra_form_data)
	46
	47	try:
	48	response = self._download_json(
	49	action_url, None, note,
	50	data=urlencode_postdata(form_data),
	51	headers={
	52	'Referer': referrer_url,
	53	'X-Requested-With': 'XMLHttpRequest',
	54	})
	55	except ExtractorError as e:
	56	if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
	57	response = self._parse_json(e.cause.read().decode('utf-8'), None)
	58	self._check_error(response, ('email', 'password'))
	59	raise
	60
	61	self._check_error(response, 'ErrorMessage')
	62
	63	return response, action_url
	64
30cbd4e0	65	def _login(self):
0d85c3a7	66	username, password = self._get_login_info()
30cbd4e0 S	67	if username is None:
	68	return
	69
bdf16f81 S	70	# Step 1: download signin page
	71	signin_page = self._download_webpage(
	72	self._SIGNIN_URL, None, 'Downloading signin page')
	73
3841256c S	74	# Already logged in
3841256c S	75	if any(re.search(p, signin_page) for p in (
ec85ded8	76	r'isLoggedIn\s:\strue', r'logout\.aspx', r'>Log out<')):
3841256c S	77	return
3841256c S	78
bdf16f81 S	79	# Step 2: submit email
	80	signin_form = self._search_regex(
	81	r'(?s)(<form[^>]+data-form-name=["\']signin["\'][^>]*>.+?</form>)',
	82	signin_page, 'signin form')
	83	signin_page, signin_url = self._login_step(
	84	signin_form, self._PASSWORD_URL, {'email': username},
	85	'Submitting email', self._SIGNIN_URL)
	86
	87	# Step 3: submit password
	88	password_form = signin_page['body']
	89	self._login_step(
	90	password_form, self._USER_URL, {'email': username, 'password': password},
	91	'Submitting password', signin_url)
30cbd4e0 S	92
	93
	94	class LyndaIE(LyndaBaseIE):
a7c26e73 PH	95	IE_NAME = 'lynda'
a7c26e73 PH	96	IE_DESC = 'lynda.com videos'
6edfc40a	97	_VALID_URL = r'https?://(?:www\.)?lynda\.com/(?:[^/]+/[^/]+/(?P<course_id>\d+)\|player/embed)/(?P<id>\d+)'
7ee40b5d	98
7ee40b5d	99	_TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
7ee40b5d	100
a0d64613	101	_TESTS = [{
68d9561c	102	'url': 'https://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
61a98b86	103	# md5 is unstable
ac260dd8	104	'info_dict': {
136db788 S	105	'id': '114408',
136db788 S	106	'ext': 'mp4',
a7c26e73 PH	107	'title': 'Using the exercise files',
a7c26e73 PH	108	'duration': 68
c7f8537d	109	}
a0d64613 S	110	}, {
	111	'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0',
	112	'only_matching': True,
	113	}]
7ee40b5d	114
6edfc40a S	115	def _raise_unavailable(self, video_id):
	116	self.raise_login_required(
	117	'Video %s is only available for members' % video_id)
	118
c7f8537d	119	def _real_extract(self, url):
6edfc40a S	120	mobj = re.match(self._VALID_URL, url)
	121	video_id = mobj.group('id')
	122	course_id = mobj.group('course_id')
	123
	124	query = {
	125	'videoId': video_id,
	126	'type': 'video',
	127	}
c7f8537d	128
ea8ed40b	129	video = self._download_json(
6edfc40a S	130	'https://www.lynda.com/ajax/player', video_id,
	131	'Downloading video JSON', fatal=False, query=query)
	132
	133	# Fallback scenario
	134	if not video:
	135	query['courseId'] = course_id
	136
	137	play = self._download_json(
	138	'https://www.lynda.com/ajax/course/%s/%s/play'
	139	% (course_id, video_id), video_id, 'Downloading play JSON')
	140
	141	if not play:
	142	self._raise_unavailable(video_id)
	143
	144	formats = []
	145	for formats_dict in play:
	146	urls = formats_dict.get('urls')
	147	if not isinstance(urls, dict):
	148	continue
	149	cdn = formats_dict.get('name')
	150	for format_id, format_url in urls.items():
	151	if not format_url:
	152	continue
	153	formats.append({
	154	'url': format_url,
	155	'format_id': '%s-%s' % (cdn, format_id) if cdn else format_id,
	156	'height': int_or_none(format_id),
	157	})
	158	self._sort_formats(formats)
	159
	160	conviva = self._download_json(
	161	'https://www.lynda.com/ajax/player/conviva', video_id,
	162	'Downloading conviva JSON', query=query)
	163
	164	return {
	165	'id': video_id,
	166	'title': conviva['VideoTitle'],
	167	'description': conviva.get('VideoDescription'),
	168	'release_year': int_or_none(conviva.get('ReleaseYear')),
	169	'duration': int_or_none(conviva.get('Duration')),
	170	'creator': conviva.get('Author'),
	171	'formats': formats,
	172	}
c7f8537d	173
ea8ed40b	174	if 'Status' in video:
30cbd4e0	175	raise ExtractorError(
ea8ed40b	176	'lynda returned error: %s' % video['Message'], expected=True)
c7f8537d	177
ea8ed40b	178	if video.get('HasAccess') is False:
6edfc40a	179	self._raise_unavailable(video_id)
c7f8537d	180
ea8ed40b S	181	video_id = compat_str(video.get('ID') or video_id)
	182	duration = int_or_none(video.get('DurationInSeconds'))
	183	title = video['Title']
c7f8537d	184
16ff7ebc S	185	formats = []
16ff7ebc S	186
ea8ed40b	187	fmts = video.get('Formats')
16ff7ebc	188	if fmts:
ea8ed40b S	189	formats.extend([{
	190	'url': f['Url'],
	191	'ext': f.get('Extension'),
	192	'width': int_or_none(f.get('Width')),
	193	'height': int_or_none(f.get('Height')),
	194	'filesize': int_or_none(f.get('FileSize')),
	195	'format_id': compat_str(f.get('Resolution')) if f.get('Resolution') else None,
	196	} for f in fmts if f.get('Url')])
	197
	198	prioritized_streams = video.get('PrioritizedStreams')
16ff7ebc	199	if prioritized_streams:
5a11b793	200	for prioritized_stream_id, prioritized_stream in prioritized_streams.items():
ea8ed40b S	201	formats.extend([{
ea8ed40b S	202	'url': video_url,
ea8aefd1	203	'height': int_or_none(format_id),
ea8ed40b S	204	'format_id': '%s-%s' % (prioritized_stream_id, format_id),
ea8ed40b S	205	} for format_id, video_url in prioritized_stream.items()])
c7f8537d	206
a57e8ce6	207	self._check_formats(formats, video_id)
c7f8537d	208	self._sort_formats(formats)
7ee40b5d	209
ea8ed40b	210	subtitles = self.extract_subtitles(video_id)
7ee40b5d	211
c7f8537d	212	return {
	213	'id': video_id,
	214	'title': title,
	215	'duration': duration,
62bcfa8c	216	'subtitles': subtitles,
c7f8537d	217	'formats': formats
c7f8537d	218	}
7ee40b5d	219
311c3938 JMF	220	def _fix_subtitles(self, subs):
311c3938 JMF	221	srt = ''
7594be85	222	seq_counter = 0
311c3938 JMF	223	for pos in range(0, len(subs) - 1):
	224	seq_current = subs[pos]
	225	m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
	226	if m_current is None:
	227	continue
	228	seq_next = subs[pos + 1]
	229	m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
	230	if m_next is None:
62bcfa8c	231	continue
311c3938 JMF	232	appear_time = m_current.group('timecode')
311c3938 JMF	233	disappear_time = m_next.group('timecode')
7594be85 S	234	text = seq_current['Caption'].strip()
	235	if text:
	236	seq_counter += 1
	237	srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (seq_counter, appear_time, disappear_time, text)
311c3938 JMF	238	if srt:
	239	return srt
	240
ea8ed40b	241	def _get_subtitles(self, video_id):
68d9561c	242	url = 'https://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
311c3938 JMF	243	subs = self._download_json(url, None, False)
	244	if subs:
	245	return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]}
	246	else:
	247	return {}
c7f8537d	248
c7f8537d	249
30cbd4e0	250	class LyndaCourseIE(LyndaBaseIE):
a7c26e73 PH	251	IE_NAME = 'lynda:course'
a7c26e73 PH	252	IE_DESC = 'lynda.com online courses'
c7f8537d	253
	254	# Course link equals to welcome/introduction video link of same course
	255	# We will recognize it as course link
	256	_VALID_URL = r'https?://(?:www\|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
	257
	258	def _real_extract(self, url):
	259	mobj = re.match(self._VALID_URL, url)
	260	course_path = mobj.group('coursepath')
	261	course_id = mobj.group('courseid')
5f6a1245	262
71bb0161	263	course = self._download_json(
68d9561c	264	'https://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
30cbd4e0	265	course_id, 'Downloading course JSON')
c7f8537d	266
71bb0161	267	if course.get('Status') == 'NotFound':
30cbd4e0 S	268	raise ExtractorError(
30cbd4e0 S	269	'Course %s does not exist' % course_id, expected=True)
c7f8537d	270
c7f8537d	271	unaccessible_videos = 0
00322ad4	272	entries = []
c7f8537d	273
16ff7ebc S	274	# Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
	275	# by single video API anymore
	276
71bb0161 S	277	for chapter in course['Chapters']:
	278	for video in chapter.get('Videos', []):
	279	if video.get('HasAccess') is False:
c7f8537d	280	unaccessible_videos += 1
c7f8537d	281	continue
00322ad4 S	282	video_id = video.get('ID')
	283	if video_id:
	284	entries.append({
	285	'_type': 'url_transparent',
68d9561c	286	'url': 'https://www.lynda.com/%s/%s-4.html' % (course_path, video_id),
00322ad4 S	287	'ie_key': LyndaIE.ie_key(),
	288	'chapter': chapter.get('Title'),
	289	'chapter_number': int_or_none(chapter.get('ChapterIndex')),
	290	'chapter_id': compat_str(chapter.get('ID')),
	291	})
c7f8537d	292
c7f8537d	293	if unaccessible_videos > 0:
30cbd4e0 S	294	self._downloader.report_warning(
	295	'%s videos are only available for members (or paid members) and will not be downloaded. '
	296	% unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT)
c7f8537d	297
71bb0161	298	course_title = course.get('Title')
04343588	299	course_description = course.get('Description')
c7f8537d	300
04343588	301	return self.playlist_result(entries, course_id, course_title, course_description)