[ie/MuseAI] Add extractor (#7614)

Closes #7543
Authored by: bashonly

diff --git a/yt_dlp/extractor/_extractors.py b/yt_dlp/extractor/_extractors.py
index bcd8dbe00697cc7f8a3bec9b7b6e785f8fbb4a90..ae73a9f96088480861c9f6738aafd8bf2fee79bb 100644 (file)
--- a/yt_dlp/extractor/_extractors.py
+++ b/yt_dlp/extractor/_extractors.py
@@ -1142,6 +1142,7 @@
 )
 from .muenchentv import MuenchenTVIE
 from .murrtube import MurrtubeIE, MurrtubeUserIE
+from .museai import MuseAIIE
 from .musescore import MuseScoreIE
 from .musicdex import (
     MusicdexSongIE,

diff --git a/yt_dlp/extractor/museai.py b/yt_dlp/extractor/museai.py
new file mode 100644 (file)
index 0000000..7f66928
--- /dev/null
+++ b/yt_dlp/extractor/museai.py
@@ -0,0 +1,112 @@
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    determine_ext,
+    float_or_none,
+    int_or_none,
+    js_to_json,
+    traverse_obj,
+    url_or_none,
+)
+
+
+class MuseAIIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?muse\.ai/(?:v|embed)/(?P<id>\w+)'
+    _TESTS = [{
+        'url': 'https://muse.ai/embed/YdTWvUW',
+        'md5': 'f994f9a38be1c3aaf9e37cbd7d76fe7c',
+        'info_dict': {
+            'id': 'YdTWvUW',
+            'ext': 'mp4',
+            'title': '2023-05-28-Grabien-1941111 (1)',
+            'description': '',
+            'uploader': 'Today News Africa',
+            'uploader_id': 'TodayNewsAfrica',
+            'upload_date': '20230528',
+            'timestamp': 1685285044,
+            'duration': 1291.3,
+            'view_count': int,
+            'availability': 'public',
+        },
+    }, {
+        'url': 'https://muse.ai/v/gQ4gGAA-0756',
+        'md5': '52dbfc78e865e56dc19a1715badc35e8',
+        'info_dict': {
+            'id': 'gQ4gGAA',
+            'ext': 'mp4',
+            'title': '0756',
+            'description': 'md5:0ca1483f9aac423e9a96ad00bb3a0785',
+            'uploader': 'Aerial.ie',
+            'uploader_id': 'aerial',
+            'upload_date': '20210306',
+            'timestamp': 1615072842,
+            'duration': 21.4,
+            'view_count': int,
+            'availability': 'public',
+        },
+    }]
+    _WEBPAGE_TESTS = [{
+        'url': 'https://muse.ai/docs',
+        'playlist_mincount': 4,
+        'info_dict': {
+            'id': 'docs',
+            'title': 'muse.ai | docs',
+            'description': 'md5:6c0293431481582739c82ee8902687fa',
+            'age_limit': 0,
+            'thumbnail': 'https://muse.ai/static/imgs/poster-img-docs.png',
+        },
+        'params': {'allowed_extractors': ['all', '-html5']},
+    }]
+    _EMBED_REGEX = [r'<iframe[^>]*\bsrc=["\'](?P<url>https://muse\.ai/embed/\w+)']
+
+    @classmethod
+    def _extract_embed_urls(cls, url, webpage):
+        yield from super()._extract_embed_urls(url, webpage)
+        for embed_id in re.findall(r'<script>[^<]*\bMusePlayer\(\{[^}<]*\bvideo:\s*["\'](\w+)["\']', webpage):
+            yield f'https://muse.ai/embed/{embed_id}'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(f'https://muse.ai/embed/{video_id}', video_id)
+        data = self._search_json(
+            r'player\.setData\(', webpage, 'player data', video_id, transform_source=js_to_json)
+
+        source_url = data['url']
+        if not url_or_none(source_url):
+            raise ExtractorError('Unable to extract video URL')
+
+        formats = [{
+            'url': source_url,
+            'format_id': 'source',
+            'quality': 1,
+            **traverse_obj(data, {
+                'ext': ('filename', {determine_ext}),
+                'width': ('width', {int_or_none}),
+                'height': ('height', {int_or_none}),
+                'filesize': ('size', {int_or_none}),
+            }),
+        }]
+        if source_url.endswith('/data'):
+            base_url = f'{source_url[:-5]}/videos'
+            formats.extend(self._extract_m3u8_formats(
+                f'{base_url}/hls.m3u8', video_id, m3u8_id='hls', fatal=False))
+            formats.extend(self._extract_mpd_formats(
+                f'{base_url}/dash.mpd', video_id, mpd_id='dash', fatal=False))
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            **traverse_obj(data, {
+                'title': ('title', {str}),
+                'description': ('description', {str}),
+                'duration': ('duration', {float_or_none}),
+                'timestamp': ('tcreated', {int_or_none}),
+                'uploader': ('owner_name', {str}),
+                'uploader_id': ('owner_username', {str}),
+                'view_count': ('views', {int_or_none}),
+                'age_limit': ('mature', {lambda x: 18 if x else None}),
+                'availability': ('visibility', {lambda x: x if x in ('private', 'unlisted') else 'public'}),
+            }),
+        }