[mediasite] Add extractor, subsume sandia and collegerama extractors
authorfelix <felix.von.s@posteo.de>
Sat, 12 Nov 2016 20:52:02 +0000 (21:52 +0100)
committerSergey M․ <dstftw@gmail.com>
Sat, 30 Dec 2017 00:23:41 +0000 (07:23 +0700)
youtube_dl/extractor/collegerama.py [deleted file]
youtube_dl/extractor/extractors.py
youtube_dl/extractor/generic.py
youtube_dl/extractor/mediasite.py [new file with mode: 0644]
youtube_dl/extractor/sandia.py [deleted file]

diff --git a/youtube_dl/extractor/collegerama.py b/youtube_dl/extractor/collegerama.py
deleted file mode 100644 (file)
index 6a41db8..0000000
+++ /dev/null
@@ -1,93 +0,0 @@
-from __future__ import unicode_literals
-
-import json
-
-from .common import InfoExtractor
-from ..utils import (
-    float_or_none,
-    int_or_none,
-    sanitized_Request,
-)
-
-
-class CollegeRamaIE(InfoExtractor):
-    _VALID_URL = r'https?://collegerama\.tudelft\.nl/Mediasite/Play/(?P<id>[\da-f]+)'
-    _TESTS = [
-        {
-            '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',
-            },
-        },
-    ]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        player_options_request = {
-            'getPlayerOptionsRequest': {
-                'ResourceId': video_id,
-                'QueryString': '',
-            }
-        }
-
-        request = sanitized_Request(
-            'http://collegerama.tudelft.nl/Mediasite/PlayerService/PlayerService.svc/json/GetPlayerOptions',
-            json.dumps(player_options_request))
-        request.add_header('Content-Type', 'application/json')
-
-        player_options = self._download_json(request, video_id)
-
-        presentation = player_options['d']['Presentation']
-        title = presentation['Title']
-        description = presentation.get('Description')
-        thumbnail = None
-        duration = float_or_none(presentation.get('Duration'), 1000)
-        timestamp = int_or_none(presentation.get('UnixTime'), 1000)
-
-        formats = []
-        for stream in presentation['Streams']:
-            for video in stream['VideoUrls']:
-                thumbnail_url = stream.get('ThumbnailUrl')
-                if thumbnail_url:
-                    thumbnail = 'http://collegerama.tudelft.nl' + thumbnail_url
-                format_id = video['MediaType']
-                if format_id == 'SS':
-                    continue
-                formats.append({
-                    'url': video['Location'],
-                    'format_id': format_id,
-                })
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'timestamp': timestamp,
-            'formats': formats,
-        }
index 5ed031833b874bb33e4711c5425d1acd324f2c29..e64defe624dd9158b4cb36320b76d2a867a2d602 100644 (file)
@@ -205,7 +205,6 @@ from .cnn import (
     CNNArticleIE,
 )
 from .coub import CoubIE
-from .collegerama import CollegeRamaIE
 from .comedycentral import (
     ComedyCentralFullEpisodesIE,
     ComedyCentralIE,
@@ -576,6 +575,7 @@ from .massengeschmacktv import MassengeschmackTVIE
 from .matchtv import MatchTVIE
 from .mdr import MDRIE
 from .mediaset import MediasetIE
+from .mediasite import MediasiteIE
 from .medici import MediciIE
 from .megaphone import MegaphoneIE
 from .meipai import MeipaiIE
@@ -912,7 +912,6 @@ from .rutube import (
 from .rutv import RUTVIE
 from .ruutu import RuutuIE
 from .ruv import RuvIE
-from .sandia import SandiaIE
 from .safari import (
     SafariIE,
     SafariApiIE,
index c7b6092153a47a905e260b6d652dfc25863bbcaf..d5622c8230f3a916f3450159b0782f2505cdb490 100644 (file)
@@ -2883,6 +2883,16 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 vshare_urls, video_id, video_title, ie=VShareIE.ie_key())
 
+        # Look for Mediasite embeds
+        mobj = re.search(r'''(?xi)
+                <iframe[^>]+src="((?:https?://[a-z0-9\-\.:\[\]]+)?
+                    /Mediasite/Play/[0-9a-f]{32,34}(?:\?.*?)?)"
+            ''', webpage)
+        if mobj is not None:
+            return self.url_result(smuggle_url(
+                compat_urlparse.urljoin(url, unescapeHTML(mobj.group(1))),
+                { 'UrlReferrer': url }), 'Livestream')
+
         def merge_dicts(dict1, dict2):
             merged = {}
             for k, v in dict1.items():
diff --git a/youtube_dl/extractor/mediasite.py b/youtube_dl/extractor/mediasite.py
new file mode 100644 (file)
index 0000000..5d28168
--- /dev/null
@@ -0,0 +1,184 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..compat import compat_urlparse
+from ..utils import (
+    ExtractorError,
+    unsmuggle_url,
+    mimetype2ext,
+    float_or_none,
+)
+
+
+class MediasiteIE(InfoExtractor):
+    _VALID_URL = r'''(?xi)
+        https?://[a-z0-9\-\.:\[\]]+/Mediasite/Play/
+        (?P<id>[0-9a-f]{32,34})
+        (?P<QueryString>\?[^#]+|)
+    '''
+    _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,
+            }
+        }
+    ]
+
+    # look in Mediasite.Core.js (Mediasite.ContentStreamType[*])
+    _STREAM_TYPES = {
+        0: 'video1', # the main video
+        2: 'slide',
+        3: 'presentation',
+        4: 'video2', # screencast?
+        5: 'video3',
+    }
+
+    def _real_extract(self, url):
+        url, data = unsmuggle_url(url, {})
+        mobj = re.match(self._VALID_URL, url)
+        ResourceId = mobj.group('id')
+        QueryString = mobj.group('QueryString')
+
+        webpage = self._download_webpage(url, ResourceId) # XXX: add UrlReferrer?
+
+        # XXX: might have also extracted UrlReferrer and QueryString from the html
+        ServicePath = compat_urlparse.urljoin(url, self._html_search_regex(
+            r'<div id="ServicePath">(.+?)</div>', webpage, ResourceId,
+            default='/Mediasite/PlayerService/PlayerService.svc/json'))
+
+        PlayerOptions = self._download_json(
+            '%s/GetPlayerOptions' % (ServicePath), ResourceId,
+            headers={
+                'Content-type': 'application/json; charset=utf-8',
+                'X-Requested-With': 'XMLHttpRequest',
+            },
+            data=json.dumps({
+                'getPlayerOptionsRequest': {
+                    'ResourceId': ResourceId,
+                    'QueryString': QueryString,
+                    'UrlReferrer': data.get('UrlReferrer', ''),
+                    'UseScreenReader': False,
+                }
+            }).encode('utf-8'))
+        Presentation = PlayerOptions['d']['Presentation']
+        if Presentation is None:
+            raise ExtractorError('Mediasite says: %s' %
+                (PlayerOptions['d']['PlayerPresentationStatusMessage'],),
+                expected=True)
+
+        thumbnails = []
+        formats = []
+        for snum, Stream in enumerate(Presentation['Streams']):
+            stream_type = self._STREAM_TYPES.get(
+                Stream['StreamType'], 'type%u' % Stream['StreamType'])
+
+            stream_formats = []
+            for unum, VideoUrl in enumerate(Stream['VideoUrls']):
+                url = VideoUrl['Location']
+                # XXX: if Stream.get('CanChangeScheme', False), switch scheme to HTTP/HTTPS
+
+                if VideoUrl['MediaType'] == 'SS':
+                    stream_formats.extend(self._extract_ism_formats(
+                        url, ResourceId, ism_id='%s-%u.%u' % (stream_type, snum, unum)))
+                    continue
+
+                stream_formats.append({
+                    'format_id': '%s-%u.%u' % (stream_type, snum, unum),
+                    'url': url,
+                    'ext': mimetype2ext(VideoUrl['MimeType']),
+                })
+
+            # TODO: if Stream['HasSlideContent']:
+            # synthesise an MJPEG video stream '%s-%u.slides' % (stream_type, snum)
+            # from Stream['Slides']
+            # this will require writing a custom downloader...
+
+            # disprefer 'secondary' streams
+            if Stream['StreamType'] != 0:
+                for fmt in stream_formats:
+                    fmt['preference'] = -1
+
+            ThumbnailUrl = Stream.get('ThumbnailUrl')
+            if ThumbnailUrl:
+                thumbnails.append({
+                    'id': '%s-%u' % (stream_type, snum),
+                    'url': compat_urlparse.urljoin(url, ThumbnailUrl),
+                    'preference': -1 if Stream['StreamType'] != 0 else 0,
+                })
+            formats.extend(stream_formats)
+
+        self._sort_formats(formats)
+
+        # XXX: Presentation['Presenters']
+        # XXX: Presentation['Transcript']
+
+        return {
+            'id': ResourceId,
+            'title': Presentation['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,
+        }
diff --git a/youtube_dl/extractor/sandia.py b/youtube_dl/extractor/sandia.py
deleted file mode 100644 (file)
index 96e43af..0000000
+++ /dev/null
@@ -1,65 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import json
-
-from .common import InfoExtractor
-from ..utils import (
-    int_or_none,
-    mimetype2ext,
-)
-
-
-class SandiaIE(InfoExtractor):
-    IE_DESC = 'Sandia National Laboratories'
-    _VALID_URL = r'https?://digitalops\.sandia\.gov/Mediasite/Play/(?P<id>[0-9a-f]+)'
-    _TEST = {
-        'url': 'http://digitalops.sandia.gov/Mediasite/Play/24aace4429fc450fb5b38cdbf424a66e1d',
-        'md5': '9422edc9b9a60151727e4b6d8bef393d',
-        'info_dict': {
-            'id': '24aace4429fc450fb5b38cdbf424a66e1d',
-            'ext': 'mp4',
-            'title': 'Xyce Software Training - Section 1',
-            'description': 're:(?s)SAND Number: SAND 2013-7800.{200,}',
-            'upload_date': '20120409',
-            'timestamp': 1333983600,
-            'duration': 7794,
-        }
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        presentation_data = self._download_json(
-            'http://digitalops.sandia.gov/Mediasite/PlayerService/PlayerService.svc/json/GetPlayerOptions',
-            video_id, data=json.dumps({
-                'getPlayerOptionsRequest': {
-                    'ResourceId': video_id,
-                    'QueryString': '',
-                }
-            }), headers={
-                'Content-Type': 'application/json; charset=utf-8',
-            })['d']['Presentation']
-
-        title = presentation_data['Title']
-
-        formats = []
-        for stream in presentation_data.get('Streams', []):
-            for fd in stream.get('VideoUrls', []):
-                formats.append({
-                    'format_id': fd['MediaType'],
-                    'format_note': fd['MimeType'].partition('/')[2],
-                    'ext': mimetype2ext(fd['MimeType']),
-                    'url': fd['Location'],
-                    'protocol': 'f4m' if fd['MimeType'] == 'video/x-mp4-fragmented' else None,
-                })
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': presentation_data.get('Description'),
-            'formats': formats,
-            'timestamp': int_or_none(presentation_data.get('UnixTime'), 1000),
-            'duration': int_or_none(presentation_data.get('Duration'), 1000),
-        }