[yt-dlp.git] / yt_dlp / extractor / mediasite.py

import re
import json

from .common import InfoExtractor
from ..compat import (
    compat_str,
    compat_urlparse,
)
from ..utils import (
    ExtractorError,
    float_or_none,
    mimetype2ext,
    str_or_none,
    try_call,
    try_get,
    smuggle_url,
    unsmuggle_url,
    url_or_none,
    urljoin,
)


_ID_RE = r'(?:[0-9a-f]{32,34}|[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12,14})'


class MediasiteIE(InfoExtractor):
    _VALID_URL = r'(?xi)https?://[^/]+/Mediasite/(?:Play|Showcase/[^/#?]+/Presentation)/(?P<id>%s)(?P<query>\?[^#]+|)' % _ID_RE
    _EMBED_REGEX = [r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/%s(?:\?.*?)?)\1' % _ID_RE]
    _TESTS = [
        {
            'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271681e4f199af3c60d1f82869b1d',
            'info_dict': {
                'id': '2db6c271681e4f199af3c60d1f82869b1d',
                'ext': 'mp4',
                'title': 'Lecture: Tuesday, September 20, 2016 - Sir Andrew Wiles',
                'description': 'Sir Andrew Wiles: “Equations in arithmetic”\\n\\nI will describe some of the interactions between modern number theory and the problem of solving equations in rational numbers or integers\\u0027.',
                'timestamp': 1474268400.0,
                'upload_date': '20160919',
            },
        },
        {
            'url': 'http://mediasite.uib.no/Mediasite/Play/90bb363295d945d6b548c867d01181361d?catalog=a452b7df-9ae1-46b7-a3ba-aceeb285f3eb',
            'info_dict': {
                'id': '90bb363295d945d6b548c867d01181361d',
                'ext': 'mp4',
                'upload_date': '20150429',
                'title': '5) IT-forum 2015-Dag 1  - Dungbeetle -  How and why Rain created a tiny bug tracker for Unity',
                'timestamp': 1430311380.0,
            },
        },
        {
            'url': 'https://collegerama.tudelft.nl/Mediasite/Play/585a43626e544bdd97aeb71a0ec907a01d',
            'md5': '481fda1c11f67588c0d9d8fbdced4e39',
            'info_dict': {
                'id': '585a43626e544bdd97aeb71a0ec907a01d',
                'ext': 'mp4',
                'title': 'Een nieuwe wereld: waarden, bewustzijn en techniek van de mensheid 2.0.',
                'description': '',
                'thumbnail': r're:^https?://.*\.jpg(?:\?.*)?$',
                'duration': 7713.088,
                'timestamp': 1413309600,
                'upload_date': '20141014',
            },
        },
        {
            'url': 'https://collegerama.tudelft.nl/Mediasite/Play/86a9ea9f53e149079fbdb4202b521ed21d?catalog=fd32fd35-6c99-466c-89d4-cd3c431bc8a4',
            'md5': 'ef1fdded95bdf19b12c5999949419c92',
            'info_dict': {
                'id': '86a9ea9f53e149079fbdb4202b521ed21d',
                'ext': 'wmv',
                'title': '64ste Vakantiecursus: Afvalwater',
                'description': 'md5:7fd774865cc69d972f542b157c328305',
                'thumbnail': r're:^https?://.*\.jpg(?:\?.*?)?$',
                'duration': 10853,
                'timestamp': 1326446400,
                'upload_date': '20120113',
            },
        },
        {
            'url': 'http://digitalops.sandia.gov/Mediasite/Play/24aace4429fc450fb5b38cdbf424a66e1d',
            'md5': '9422edc9b9a60151727e4b6d8bef393d',
            'info_dict': {
                'id': '24aace4429fc450fb5b38cdbf424a66e1d',
                'ext': 'mp4',
                'title': 'Xyce Software Training - Section 1',
                'description': r're:(?s)SAND Number: SAND 2013-7800.{200,}',
                'upload_date': '20120409',
                'timestamp': 1333983600,
                'duration': 7794,
            }
        },
        {
            'url': 'https://collegerama.tudelft.nl/Mediasite/Showcase/livebroadcast/Presentation/ada7020854f743c49fbb45c9ec7dbb351d',
            'only_matching': True,
        },
        {
            'url': 'https://mediasite.ntnu.no/Mediasite/Showcase/default/Presentation/7d8b913259334b688986e970fae6fcb31d',
            'only_matching': True,
        },
        {
            # dashed id
            'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271-681e-4f19-9af3-c60d1f82869b1d',
            'only_matching': True,
        }
    ]

    # look in Mediasite.Core.js (Mediasite.ContentStreamType[*])
    _STREAM_TYPES = {
        0: 'video1',  # the main video
        2: 'slide',
        3: 'presentation',
        4: 'video2',  # screencast?
        5: 'video3',
    }

    @classmethod
    def _extract_embed_urls(cls, url, webpage):
        for embed_url in super()._extract_embed_urls(url, webpage):
            yield smuggle_url(embed_url, {'UrlReferrer': url})

    def __extract_slides(self, *, stream_id, snum, Stream, duration, images):
        slide_base_url = Stream['SlideBaseUrl']

        fname_template = Stream['SlideImageFileNameTemplate']
        if fname_template != 'slide_{0:D4}.jpg':
            self.report_warning('Unusual slide file name template; report a bug if slide downloading fails')
        fname_template = re.sub(r'\{0:D([0-9]+)\}', r'{0:0\1}', fname_template)

        fragments = []
        for i, slide in enumerate(Stream['Slides']):
            if i == 0:
                if slide['Time'] > 0:
                    default_slide = images.get('DefaultSlide')
                    if default_slide is None:
                        default_slide = images.get('DefaultStreamImage')
                    if default_slide is not None:
                        default_slide = default_slide['ImageFilename']
                    if default_slide is not None:
                        fragments.append({
                            'path': default_slide,
                            'duration': slide['Time'] / 1000,
                        })

            next_time = try_call(
                lambda: Stream['Slides'][i + 1]['Time'],
                lambda: duration,
                lambda: slide['Time'],
                expected_type=(int, float))

            fragments.append({
                'path': fname_template.format(slide.get('Number', i + 1)),
                'duration': (next_time - slide['Time']) / 1000
            })

        return {
            'format_id': '%s-%u.slides' % (stream_id, snum),
            'ext': 'mhtml',
            'url': slide_base_url,
            'protocol': 'mhtml',
            'acodec': 'none',
            'vcodec': 'none',
            'format_note': 'Slides',
            'fragments': fragments,
            'fragment_base_url': slide_base_url,
        }

    def _real_extract(self, url):
        url, data = unsmuggle_url(url, {})
        mobj = self._match_valid_url(url)
        resource_id = mobj.group('id')
        query = mobj.group('query')

        webpage, urlh = self._download_webpage_handle(url, resource_id)  # XXX: add UrlReferrer?
        redirect_url = urlh.url

        # XXX: might have also extracted UrlReferrer and QueryString from the html
        service_path = compat_urlparse.urljoin(redirect_url, self._html_search_regex(
            r'<div[^>]+\bid=["\']ServicePath[^>]+>(.+?)</div>', webpage, resource_id,
            default='/Mediasite/PlayerService/PlayerService.svc/json'))

        player_options = self._download_json(
            '%s/GetPlayerOptions' % service_path, resource_id,
            headers={
                'Content-type': 'application/json; charset=utf-8',
                'X-Requested-With': 'XMLHttpRequest',
            },
            data=json.dumps({
                'getPlayerOptionsRequest': {
                    'ResourceId': resource_id,
                    'QueryString': query,
                    'UrlReferrer': data.get('UrlReferrer', ''),
                    'UseScreenReader': False,
                }
            }).encode('utf-8'))['d']

        presentation = player_options['Presentation']
        title = presentation['Title']

        if presentation is None:
            raise ExtractorError(
                'Mediasite says: %s' % player_options['PlayerPresentationStatusMessage'],
                expected=True)

        thumbnails = []
        formats = []
        for snum, Stream in enumerate(presentation['Streams']):
            stream_type = Stream.get('StreamType')
            if stream_type is None:
                continue

            video_urls = Stream.get('VideoUrls')
            if not isinstance(video_urls, list):
                video_urls = []

            stream_id = self._STREAM_TYPES.get(
                stream_type, 'type%u' % stream_type)

            stream_formats = []
            for unum, VideoUrl in enumerate(video_urls):
                video_url = url_or_none(VideoUrl.get('Location'))
                if not video_url:
                    continue
                # XXX: if Stream.get('CanChangeScheme', False), switch scheme to HTTP/HTTPS

                media_type = VideoUrl.get('MediaType')
                if media_type == 'SS':
                    stream_formats.extend(self._extract_ism_formats(
                        video_url, resource_id,
                        ism_id='%s-%u.%u' % (stream_id, snum, unum),
                        fatal=False))
                elif media_type == 'Dash':
                    stream_formats.extend(self._extract_mpd_formats(
                        video_url, resource_id,
                        mpd_id='%s-%u.%u' % (stream_id, snum, unum),
                        fatal=False))
                else:
                    stream_formats.append({
                        'format_id': '%s-%u.%u' % (stream_id, snum, unum),
                        'url': video_url,
                        'ext': mimetype2ext(VideoUrl.get('MimeType')),
                    })

            if Stream.get('HasSlideContent', False):
                images = player_options['PlayerLayoutOptions']['Images']
                stream_formats.append(self.__extract_slides(
                    stream_id=stream_id,
                    snum=snum,
                    Stream=Stream,
                    duration=presentation.get('Duration'),
                    images=images,
                ))

            # disprefer 'secondary' streams
            if stream_type != 0:
                for fmt in stream_formats:
                    fmt['quality'] = -10

            thumbnail_url = Stream.get('ThumbnailUrl')
            if thumbnail_url:
                thumbnails.append({
                    'id': '%s-%u' % (stream_id, snum),
                    'url': urljoin(redirect_url, thumbnail_url),
                    'preference': -1 if stream_type != 0 else 0,
                })
            formats.extend(stream_formats)

        # XXX: Presentation['Presenters']
        # XXX: Presentation['Transcript']

        return {
            'id': resource_id,
            'title': title,
            'description': presentation.get('Description'),
            'duration': float_or_none(presentation.get('Duration'), 1000),
            'timestamp': float_or_none(presentation.get('UnixTime'), 1000),
            'formats': formats,
            'thumbnails': thumbnails,
        }


class MediasiteCatalogIE(InfoExtractor):
    _VALID_URL = r'''(?xi)
                        (?P<url>https?://[^/]+/Mediasite)
                        /Catalog/Full/
                        (?P<catalog_id>{0})
                        (?:
                            /(?P<current_folder_id>{0})
                            /(?P<root_dynamic_folder_id>{0})
                        )?
                    '''.format(_ID_RE)
    _TESTS = [{
        'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48530d454381549f955d08c75e21',
        'info_dict': {
            'id': '631f9e48530d454381549f955d08c75e21',
            'title': 'WCET Summit: Adaptive Learning in Higher Ed: Improving Outcomes Dynamically',
        },
        'playlist_count': 6,
        'expected_warnings': ['is not a supported codec'],
    }, {
        # with CurrentFolderId and RootDynamicFolderId
        'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
        'info_dict': {
            'id': '9518c4a6c5cf4993b21cbd53e828a92521',
            'title': 'IUSM Family and Friends Sessions',
        },
        'playlist_count': 2,
    }, {
        'url': 'http://uipsyc.mediasite.com/mediasite/Catalog/Full/d5d79287c75243c58c50fef50174ec1b21',
        'only_matching': True,
    }, {
        # no AntiForgeryToken
        'url': 'https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21',
        'only_matching': True,
    }, {
        'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
        'only_matching': True,
    }, {
        # dashed id
        'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48-530d-4543-8154-9f955d08c75e',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        mobj = self._match_valid_url(url)
        mediasite_url = mobj.group('url')
        catalog_id = mobj.group('catalog_id')
        current_folder_id = mobj.group('current_folder_id') or catalog_id
        root_dynamic_folder_id = mobj.group('root_dynamic_folder_id')

        webpage = self._download_webpage(url, catalog_id)

        # AntiForgeryToken is optional (e.g. [1])
        # 1. https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21
        anti_forgery_token = self._search_regex(
            r'AntiForgeryToken\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
            webpage, 'anti forgery token', default=None, group='value')
        if anti_forgery_token:
            anti_forgery_header = self._search_regex(
                r'AntiForgeryHeaderName\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1',
                webpage, 'anti forgery header name',
                default='X-SOFO-AntiForgeryHeader', group='value')

        data = {
            'IsViewPage': True,
            'IsNewFolder': True,
            'AuthTicket': None,
            'CatalogId': catalog_id,
            'CurrentFolderId': current_folder_id,
            'RootDynamicFolderId': root_dynamic_folder_id,
            'ItemsPerPage': 1000,
            'PageIndex': 0,
            'PermissionMask': 'Execute',
            'CatalogSearchType': 'SearchInFolder',
            'SortBy': 'Date',
            'SortDirection': 'Descending',
            'StartDate': None,
            'EndDate': None,
            'StatusFilterList': None,
            'PreviewKey': None,
            'Tags': [],
        }

        headers = {
            'Content-Type': 'application/json; charset=UTF-8',
            'Referer': url,
            'X-Requested-With': 'XMLHttpRequest',
        }
        if anti_forgery_token:
            headers[anti_forgery_header] = anti_forgery_token

        catalog = self._download_json(
            '%s/Catalog/Data/GetPresentationsForFolder' % mediasite_url,
            catalog_id, data=json.dumps(data).encode(), headers=headers)

        entries = []
        for video in catalog['PresentationDetailsList']:
            if not isinstance(video, dict):
                continue
            video_id = str_or_none(video.get('Id'))
            if not video_id:
                continue
            entries.append(self.url_result(
                '%s/Play/%s' % (mediasite_url, video_id),
                ie=MediasiteIE.ie_key(), video_id=video_id))

        title = try_get(
            catalog, lambda x: x['CurrentFolder']['Name'], compat_str)

        return self.playlist_result(entries, catalog_id, title,)


class MediasiteNamedCatalogIE(InfoExtractor):
    _VALID_URL = r'(?xi)(?P<url>https?://[^/]+/Mediasite)/Catalog/catalogs/(?P<catalog_name>[^/?#&]+)'
    _TESTS = [{
        'url': 'https://msite.misis.ru/Mediasite/Catalog/catalogs/2016-industrial-management-skriabin-o-o',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        mobj = self._match_valid_url(url)
        mediasite_url = mobj.group('url')
        catalog_name = mobj.group('catalog_name')

        webpage = self._download_webpage(url, catalog_name)

        catalog_id = self._search_regex(
            r'CatalogId\s*:\s*["\'](%s)' % _ID_RE, webpage, 'catalog id')

        return self.url_result(
            '%s/Catalog/Full/%s' % (mediasite_url, catalog_id),
            ie=MediasiteCatalogIE.ie_key(), video_id=catalog_id)
Commit	Line	Data
8056c854	1	import re
	2	import json
	3
	4	from .common import InfoExtractor
2ca7ed41 S	5	from ..compat import (
	6	compat_str,
	7	compat_urlparse,
	8	)
8056c854	9	from ..utils import (
8056c854	10	ExtractorError,
8056c854	11	float_or_none,
2ca7ed41	12	mimetype2ext,
4014a486	13	str_or_none,
c4f60dd7	14	try_call,
4014a486	15	try_get,
bfd973ec	16	smuggle_url,
2ca7ed41	17	unsmuggle_url,
3052a30d	18	url_or_none,
2ca7ed41	19	urljoin,
8056c854	20	)
	21
	22
efee62ac	23	_ID_RE = r'(?:[0-9a-f]{32,34}\|[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12,14})'
4014a486 S	24
4014a486 S	25
8056c854	26	class MediasiteIE(InfoExtractor):
c72967d5	27	_VALID_URL = r'(?xi)https?://[^/]+/Mediasite/(?:Play\|Showcase/[^/#?]+/Presentation)/(?P<id>%s)(?P<query>\?[^#]+\|)' % _ID_RE
bfd973ec	28	_EMBED_REGEX = [r'(?xi)<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:(?:https?:)?//[^/]+)?/Mediasite/Play/%s(?:\?.*?)?)\1' % _ID_RE]
8056c854	29	_TESTS = [
	30	{
	31	'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271681e4f199af3c60d1f82869b1d',
	32	'info_dict': {
	33	'id': '2db6c271681e4f199af3c60d1f82869b1d',
	34	'ext': 'mp4',
	35	'title': 'Lecture: Tuesday, September 20, 2016 - Sir Andrew Wiles',
	36	'description': 'Sir Andrew Wiles: “Equations in arithmetic”\\n\\nI will describe some of the interactions between modern number theory and the problem of solving equations in rational numbers or integers\\u0027.',
	37	'timestamp': 1474268400.0,
	38	'upload_date': '20160919',
	39	},
	40	},
	41	{
	42	'url': 'http://mediasite.uib.no/Mediasite/Play/90bb363295d945d6b548c867d01181361d?catalog=a452b7df-9ae1-46b7-a3ba-aceeb285f3eb',
	43	'info_dict': {
	44	'id': '90bb363295d945d6b548c867d01181361d',
	45	'ext': 'mp4',
	46	'upload_date': '20150429',
	47	'title': '5) IT-forum 2015-Dag 1 - Dungbeetle - How and why Rain created a tiny bug tracker for Unity',
	48	'timestamp': 1430311380.0,
	49	},
	50	},
	51	{
	52	'url': 'https://collegerama.tudelft.nl/Mediasite/Play/585a43626e544bdd97aeb71a0ec907a01d',
	53	'md5': '481fda1c11f67588c0d9d8fbdced4e39',
	54	'info_dict': {
	55	'id': '585a43626e544bdd97aeb71a0ec907a01d',
	56	'ext': 'mp4',
	57	'title': 'Een nieuwe wereld: waarden, bewustzijn en techniek van de mensheid 2.0.',
	58	'description': '',
	59	'thumbnail': r're:^https?://.\.jpg(?:\?.)?$',
	60	'duration': 7713.088,
	61	'timestamp': 1413309600,
	62	'upload_date': '20141014',
	63	},
	64	},
	65	{
	66	'url': 'https://collegerama.tudelft.nl/Mediasite/Play/86a9ea9f53e149079fbdb4202b521ed21d?catalog=fd32fd35-6c99-466c-89d4-cd3c431bc8a4',
	67	'md5': 'ef1fdded95bdf19b12c5999949419c92',
	68	'info_dict': {
	69	'id': '86a9ea9f53e149079fbdb4202b521ed21d',
	70	'ext': 'wmv',
	71	'title': '64ste Vakantiecursus: Afvalwater',
	72	'description': 'md5:7fd774865cc69d972f542b157c328305',
	73	'thumbnail': r're:^https?://.\.jpg(?:\?.?)?$',
	74	'duration': 10853,
	75	'timestamp': 1326446400,
	76	'upload_date': '20120113',
	77	},
	78	},
	79	{
	80	'url': 'http://digitalops.sandia.gov/Mediasite/Play/24aace4429fc450fb5b38cdbf424a66e1d',
	81	'md5': '9422edc9b9a60151727e4b6d8bef393d',
	82	'info_dict': {
	83	'id': '24aace4429fc450fb5b38cdbf424a66e1d',
	84	'ext': 'mp4',
	85	'title': 'Xyce Software Training - Section 1',
	86	'description': r're:(?s)SAND Number: SAND 2013-7800.{200,}',
	87	'upload_date': '20120409',
	88	'timestamp': 1333983600,
	89	'duration': 7794,
	90	}
cfd13c4c DV	91	},
	92	{
	93	'url': 'https://collegerama.tudelft.nl/Mediasite/Showcase/livebroadcast/Presentation/ada7020854f743c49fbb45c9ec7dbb351d',
	94	'only_matching': True,
	95	},
4ee18454 S	96	{
	97	'url': 'https://mediasite.ntnu.no/Mediasite/Showcase/default/Presentation/7d8b913259334b688986e970fae6fcb31d',
	98	'only_matching': True,
	99	},
efee62ac S	100	{
	101	# dashed id
	102	'url': 'https://hitsmediaweb.h-its.org/mediasite/Play/2db6c271-681e-4f19-9af3-c60d1f82869b1d',
	103	'only_matching': True,
	104	}
8056c854	105	]
	106
	107	# look in Mediasite.Core.js (Mediasite.ContentStreamType[*])
	108	_STREAM_TYPES = {
2ca7ed41	109	0: 'video1', # the main video
8056c854	110	2: 'slide',
8056c854	111	3: 'presentation',
2ca7ed41	112	4: 'video2', # screencast?
8056c854	113	5: 'video3',
	114	}
	115
bfd973ec	116	@classmethod
	117	def _extract_embed_urls(cls, url, webpage):
	118	for embed_url in super()._extract_embed_urls(url, webpage):
	119	yield smuggle_url(embed_url, {'UrlReferrer': url})
2ca7ed41	120
3fd4c2a5	121	def __extract_slides(self, *, stream_id, snum, Stream, duration, images):
	122	slide_base_url = Stream['SlideBaseUrl']
	123
	124	fname_template = Stream['SlideImageFileNameTemplate']
	125	if fname_template != 'slide_{0:D4}.jpg':
	126	self.report_warning('Unusual slide file name template; report a bug if slide downloading fails')
	127	fname_template = re.sub(r'\{0:D([0-9]+)\}', r'{0:0\1}', fname_template)
	128
	129	fragments = []
	130	for i, slide in enumerate(Stream['Slides']):
	131	if i == 0:
	132	if slide['Time'] > 0:
	133	default_slide = images.get('DefaultSlide')
	134	if default_slide is None:
	135	default_slide = images.get('DefaultStreamImage')
	136	if default_slide is not None:
	137	default_slide = default_slide['ImageFilename']
	138	if default_slide is not None:
	139	fragments.append({
	140	'path': default_slide,
	141	'duration': slide['Time'] / 1000,
	142	})
	143
c4f60dd7	144	next_time = try_call(
	145	lambda: Stream['Slides'][i + 1]['Time'],
	146	lambda: duration,
	147	lambda: slide['Time'],
	148	expected_type=(int, float))
3fd4c2a5	149
	150	fragments.append({
	151	'path': fname_template.format(slide.get('Number', i + 1)),
	152	'duration': (next_time - slide['Time']) / 1000
	153	})
	154
	155	return {
	156	'format_id': '%s-%u.slides' % (stream_id, snum),
	157	'ext': 'mhtml',
	158	'url': slide_base_url,
	159	'protocol': 'mhtml',
	160	'acodec': 'none',
	161	'vcodec': 'none',
	162	'format_note': 'Slides',
	163	'fragments': fragments,
	164	'fragment_base_url': slide_base_url,
	165	}
	166
8056c854	167	def _real_extract(self, url):
8056c854	168	url, data = unsmuggle_url(url, {})
5ad28e7f	169	mobj = self._match_valid_url(url)
2ca7ed41 S	170	resource_id = mobj.group('id')
2ca7ed41 S	171	query = mobj.group('query')
8056c854	172
2ca7ed41	173	webpage, urlh = self._download_webpage_handle(url, resource_id) # XXX: add UrlReferrer?
3d2623a8	174	redirect_url = urlh.url
8056c854	175
8056c854	176	# XXX: might have also extracted UrlReferrer and QueryString from the html
2ca7ed41 S	177	service_path = compat_urlparse.urljoin(redirect_url, self._html_search_regex(
2ca7ed41 S	178	r'<div[^>]+\bid=["\']ServicePath[^>]+>(.+?)</div>', webpage, resource_id,
8056c854	179	default='/Mediasite/PlayerService/PlayerService.svc/json'))
8056c854	180
2ca7ed41 S	181	player_options = self._download_json(
2ca7ed41 S	182	'%s/GetPlayerOptions' % service_path, resource_id,
8056c854	183	headers={
	184	'Content-type': 'application/json; charset=utf-8',
	185	'X-Requested-With': 'XMLHttpRequest',
	186	},
	187	data=json.dumps({
	188	'getPlayerOptionsRequest': {
2ca7ed41 S	189	'ResourceId': resource_id,
2ca7ed41 S	190	'QueryString': query,
8056c854	191	'UrlReferrer': data.get('UrlReferrer', ''),
	192	'UseScreenReader': False,
	193	}
2ca7ed41 S	194	}).encode('utf-8'))['d']
	195
	196	presentation = player_options['Presentation']
	197	title = presentation['Title']
	198
	199	if presentation is None:
	200	raise ExtractorError(
	201	'Mediasite says: %s' % player_options['PlayerPresentationStatusMessage'],
8056c854	202	expected=True)
	203
	204	thumbnails = []
	205	formats = []
2ca7ed41 S	206	for snum, Stream in enumerate(presentation['Streams']):
	207	stream_type = Stream.get('StreamType')
	208	if stream_type is None:
	209	continue
	210
	211	video_urls = Stream.get('VideoUrls')
	212	if not isinstance(video_urls, list):
	213	video_urls = []
	214
	215	stream_id = self._STREAM_TYPES.get(
	216	stream_type, 'type%u' % stream_type)
8056c854	217
8056c854	218	stream_formats = []
2ca7ed41	219	for unum, VideoUrl in enumerate(video_urls):
3052a30d S	220	video_url = url_or_none(VideoUrl.get('Location'))
3052a30d S	221	if not video_url:
2ca7ed41	222	continue
8056c854	223	# XXX: if Stream.get('CanChangeScheme', False), switch scheme to HTTP/HTTPS
8056c854	224
2ca7ed41 S	225	media_type = VideoUrl.get('MediaType')
2ca7ed41 S	226	if media_type == 'SS':
8056c854	227	stream_formats.extend(self._extract_ism_formats(
2ca7ed41 S	228	video_url, resource_id,
	229	ism_id='%s-%u.%u' % (stream_id, snum, unum),
	230	fatal=False))
	231	elif media_type == 'Dash':
	232	stream_formats.extend(self._extract_mpd_formats(
	233	video_url, resource_id,
	234	mpd_id='%s-%u.%u' % (stream_id, snum, unum),
	235	fatal=False))
	236	else:
	237	stream_formats.append({
	238	'format_id': '%s-%u.%u' % (stream_id, snum, unum),
	239	'url': video_url,
	240	'ext': mimetype2ext(VideoUrl.get('MimeType')),
	241	})
8056c854	242
3fd4c2a5	243	if Stream.get('HasSlideContent', False):
	244	images = player_options['PlayerLayoutOptions']['Images']
	245	stream_formats.append(self.__extract_slides(
	246	stream_id=stream_id,
	247	snum=snum,
	248	Stream=Stream,
	249	duration=presentation.get('Duration'),
	250	images=images,
	251	))
8056c854	252
8056c854	253	# disprefer 'secondary' streams
2ca7ed41	254	if stream_type != 0:
8056c854	255	for fmt in stream_formats:
f983b875	256	fmt['quality'] = -10
8056c854	257
2ca7ed41 S	258	thumbnail_url = Stream.get('ThumbnailUrl')
2ca7ed41 S	259	if thumbnail_url:
8056c854	260	thumbnails.append({
2ca7ed41 S	261	'id': '%s-%u' % (stream_id, snum),
	262	'url': urljoin(redirect_url, thumbnail_url),
	263	'preference': -1 if stream_type != 0 else 0,
8056c854	264	})
	265	formats.extend(stream_formats)
	266
8056c854	267	# XXX: Presentation['Presenters']
	268	# XXX: Presentation['Transcript']
	269
	270	return {
2ca7ed41 S	271	'id': resource_id,
	272	'title': title,
	273	'description': presentation.get('Description'),
	274	'duration': float_or_none(presentation.get('Duration'), 1000),
	275	'timestamp': float_or_none(presentation.get('UnixTime'), 1000),
8056c854	276	'formats': formats,
	277	'thumbnails': thumbnails,
	278	}
4014a486 S	279
	280
	281	class MediasiteCatalogIE(InfoExtractor):
	282	_VALID_URL = r'''(?xi)
	283	(?P<url>https?://[^/]+/Mediasite)
	284	/Catalog/Full/
	285	(?P<catalog_id>{0})
	286	(?:
	287	/(?P<current_folder_id>{0})
	288	/(?P<root_dynamic_folder_id>{0})
	289	)?
	290	'''.format(_ID_RE)
	291	_TESTS = [{
	292	'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48530d454381549f955d08c75e21',
	293	'info_dict': {
	294	'id': '631f9e48530d454381549f955d08c75e21',
	295	'title': 'WCET Summit: Adaptive Learning in Higher Ed: Improving Outcomes Dynamically',
	296	},
	297	'playlist_count': 6,
	298	'expected_warnings': ['is not a supported codec'],
	299	}, {
	300	# with CurrentFolderId and RootDynamicFolderId
	301	'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
	302	'info_dict': {
	303	'id': '9518c4a6c5cf4993b21cbd53e828a92521',
	304	'title': 'IUSM Family and Friends Sessions',
	305	},
	306	'playlist_count': 2,
	307	}, {
	308	'url': 'http://uipsyc.mediasite.com/mediasite/Catalog/Full/d5d79287c75243c58c50fef50174ec1b21',
	309	'only_matching': True,
	310	}, {
	311	# no AntiForgeryToken
	312	'url': 'https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21',
	313	'only_matching': True,
	314	}, {
	315	'url': 'https://medaudio.medicine.iu.edu/Mediasite/Catalog/Full/9518c4a6c5cf4993b21cbd53e828a92521/97a9db45f7ab47428c77cd2ed74bb98f14/9518c4a6c5cf4993b21cbd53e828a92521',
	316	'only_matching': True,
efee62ac S	317	}, {
	318	# dashed id
	319	'url': 'http://events7.mediasite.com/Mediasite/Catalog/Full/631f9e48-530d-4543-8154-9f955d08c75e',
	320	'only_matching': True,
4014a486 S	321	}]
	322
	323	def _real_extract(self, url):
5ad28e7f	324	mobj = self._match_valid_url(url)
4014a486 S	325	mediasite_url = mobj.group('url')
	326	catalog_id = mobj.group('catalog_id')
	327	current_folder_id = mobj.group('current_folder_id') or catalog_id
	328	root_dynamic_folder_id = mobj.group('root_dynamic_folder_id')
	329
	330	webpage = self._download_webpage(url, catalog_id)
	331
	332	# AntiForgeryToken is optional (e.g. [1])
	333	# 1. https://live.libraries.psu.edu/Mediasite/Catalog/Full/8376d4b24dd1457ea3bfe4cf9163feda21
	334	anti_forgery_token = self._search_regex(
	335	r'AntiForgeryToken\s:\s(["\'])(?P<value>(?:(?!\1).)+)\1',
	336	webpage, 'anti forgery token', default=None, group='value')
	337	if anti_forgery_token:
	338	anti_forgery_header = self._search_regex(
	339	r'AntiForgeryHeaderName\s:\s(["\'])(?P<value>(?:(?!\1).)+)\1',
	340	webpage, 'anti forgery header name',
	341	default='X-SOFO-AntiForgeryHeader', group='value')
	342
	343	data = {
	344	'IsViewPage': True,
	345	'IsNewFolder': True,
	346	'AuthTicket': None,
	347	'CatalogId': catalog_id,
	348	'CurrentFolderId': current_folder_id,
	349	'RootDynamicFolderId': root_dynamic_folder_id,
	350	'ItemsPerPage': 1000,
	351	'PageIndex': 0,
	352	'PermissionMask': 'Execute',
	353	'CatalogSearchType': 'SearchInFolder',
	354	'SortBy': 'Date',
	355	'SortDirection': 'Descending',
	356	'StartDate': None,
	357	'EndDate': None,
	358	'StatusFilterList': None,
	359	'PreviewKey': None,
	360	'Tags': [],
	361	}
	362
	363	headers = {
	364	'Content-Type': 'application/json; charset=UTF-8',
	365	'Referer': url,
	366	'X-Requested-With': 'XMLHttpRequest',
	367	}
	368	if anti_forgery_token:
	369	headers[anti_forgery_header] = anti_forgery_token
	370
	371	catalog = self._download_json(
	372	'%s/Catalog/Data/GetPresentationsForFolder' % mediasite_url,
	373	catalog_id, data=json.dumps(data).encode(), headers=headers)
	374
	375	entries = []
	376	for video in catalog['PresentationDetailsList']:
	377	if not isinstance(video, dict):
	378	continue
	379	video_id = str_or_none(video.get('Id'))
	380	if not video_id:
	381	continue
	382	entries.append(self.url_result(
	383	'%s/Play/%s' % (mediasite_url, video_id),
	384	ie=MediasiteIE.ie_key(), video_id=video_id))
	385
	386	title = try_get(
	387	catalog, lambda x: x['CurrentFolder']['Name'], compat_str)
	388
389	return self.playlist_result(entries, catalog_id, title,)
efee62ac S	390
	391
	392	class MediasiteNamedCatalogIE(InfoExtractor):
	393	_VALID_URL = r'(?xi)(?P<url>https?://[^/]+/Mediasite)/Catalog/catalogs/(?P<catalog_name>[^/?#&]+)'
	394	_TESTS = [{
	395	'url': 'https://msite.misis.ru/Mediasite/Catalog/catalogs/2016-industrial-management-skriabin-o-o',
	396	'only_matching': True,
	397	}]
	398
	399	def _real_extract(self, url):
5ad28e7f	400	mobj = self._match_valid_url(url)
efee62ac S	401	mediasite_url = mobj.group('url')
	402	catalog_name = mobj.group('catalog_name')
	403
	404	webpage = self._download_webpage(url, catalog_name)
	405
	406	catalog_id = self._search_regex(
	407	r'CatalogId\s:\s["\'](%s)' % _ID_RE, webpage, 'catalog id')
	408
	409	return self.url_result(
	410	'%s/Catalog/Full/%s' % (mediasite_url, catalog_id),
	411	ie=MediasiteCatalogIE.ie_key(), video_id=catalog_id)