From 177662e0f24bfd54e57b87698739d7a518321bac Mon Sep 17 00:00:00 2001
From: sam <mail@samueljenks.me>
Date: Tue, 4 Oct 2022 02:52:30 +1300
Subject: [PATCH] [extractor/MicrosoftEmbed] Add extractor (#5082)

Closes #2638
Authored by: DoubleCouponDay
---
 yt_dlp/YoutubeDL.py                |  2 +-
 yt_dlp/extractor/_extractors.py    |  1 +
 yt_dlp/extractor/microsoftembed.py | 70 ++++++++++++++++++++++++++++++
 3 files changed, 72 insertions(+), 1 deletion(-)
 create mode 100644 yt_dlp/extractor/microsoftembed.py

diff --git a/yt_dlp/YoutubeDL.py b/yt_dlp/YoutubeDL.py
index 4fcf1f5cc..bc6de4926 100644
--- a/yt_dlp/YoutubeDL.py
+++ b/yt_dlp/YoutubeDL.py
@@ -3640,7 +3640,7 @@ class YoutubeDL:
             return None
         return render_table(
             self._list_format_headers('ID', 'Width', 'Height', 'URL'),
-            [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
+            [[t.get('id'), t.get('width') or 'unknown', t.get('height') or 'unknown', t['url']] for t in thumbnails])
 
     def render_subtitles_table(self, video_id, subtitles):
         def _row(lang, formats):
diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py
index f4d7c3ab5..3a92c1d02 100644
--- a/yt_dlp/extractor/_extractors.py
+++ b/yt_dlp/extractor/_extractors.py
@@ -960,6 +960,7 @@ from .microsoftvirtualacademy import (
     MicrosoftVirtualAcademyIE,
     MicrosoftVirtualAcademyCourseIE,
 )
+from .microsoftembed import MicrosoftEmbedIE
 from .mildom import (
     MildomIE,
     MildomVodIE,
diff --git a/yt_dlp/extractor/microsoftembed.py b/yt_dlp/extractor/microsoftembed.py
new file mode 100644
index 000000000..8cdf66778
--- /dev/null
+++ b/yt_dlp/extractor/microsoftembed.py
@@ -0,0 +1,70 @@
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    traverse_obj,
+    unified_timestamp,
+)
+
+
+class MicrosoftEmbedIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?microsoft\.com/(?:[^/]+/)?videoplayer/embed/(?P<id>[a-z0-9A-Z]+)'
+
+    _TESTS = [{
+        'url': 'https://www.microsoft.com/en-us/videoplayer/embed/RWL07e',
+        'md5': 'eb0ae9007f9b305f9acd0a03e74cb1a9',
+        'info_dict': {
+            'id': 'RWL07e',
+            'title': 'Microsoft for Public Health and Social Services',
+            'ext': 'mp4',
+            'thumbnail': 'http://img-prod-cms-rt-microsoft-com.akamaized.net/cms/api/am/imageFileData/RWL7Ju?ver=cae5',
+            'age_limit': 0,
+            'timestamp': 1631658316,
+            'upload_date': '20210914'
+        }
+    }]
+    _API_URL = 'https://prod-video-cms-rt-microsoft-com.akamaized.net/vhs/api/videos/'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        metadata = self._download_json(self._API_URL + video_id, video_id)
+
+        formats = []
+        for source_type, source in metadata['streams'].items():
+            if source_type == 'smooth_Streaming':
+                formats.extend(self._extract_ism_formats(source['url'], video_id, 'mss'))
+            elif source_type == 'apple_HTTP_Live_Streaming':
+                formats.extend(self._extract_m3u8_formats(source['url'], video_id, 'mp4'))
+            elif source_type == 'mPEG_DASH':
+                formats.extend(self._extract_mpd_formats(source['url'], video_id))
+            else:
+                formats.append({
+                    'format_id': source_type,
+                    'url': source['url'],
+                    'height': source.get('heightPixels'),
+                    'width': source.get('widthPixels'),
+                })
+        self._sort_formats(formats)
+
+        subtitles = {
+            lang: [{
+                'url': data.get('url'),
+                'ext': 'vtt',
+            }] for lang, data in traverse_obj(metadata, 'captions', default={}).items()
+        }
+
+        thumbnails = [{
+            'url': thumb.get('url'),
+            'width': thumb.get('width') or None,
+            'height': thumb.get('height') or None,
+        } for thumb in traverse_obj(metadata, ('snippet', 'thumbnails', ...))]
+        self._remove_duplicate_formats(thumbnails)
+
+        return {
+            'id': video_id,
+            'title': traverse_obj(metadata, ('snippet', 'title')),
+            'timestamp': unified_timestamp(traverse_obj(metadata, ('snippet', 'activeStartDate'))),
+            'age_limit': int_or_none(traverse_obj(metadata, ('snippet', 'minimumAge'))) or 0,
+            'formats': formats,
+            'subtitles': subtitles,
+            'thumbnails': thumbnails,
+        }