[extractor/stacommu] Add extractors (#7432)

Authored by: urectanc
2024-11-15 10:22:15 +01:00 · 2023-07-01 03:27:07 +09:00 · 2023-07-01 03:27:07 +09:00 · af1fd12f67
commit af1fd12f67
parent fcbc9ed760
4 changed files with 173 additions and 14 deletions
--- a/README.md
+++ b/README.md
@ -1855,7 +1855,7 @@ The following extractors use this feature:
 #### twitter
 * `legacy_api`: Force usage of the legacy Twitter API instead of the GraphQL API for tweet extraction. Has no effect if login cookies are passed

-#### wrestleuniverse
+#### stacommu, wrestleuniverse
 * `device_id`: UUID value assigned by the website and used to enforce device limits for paid livestream content. Can be found in browser local storage

 #### twitch
--- a/yt_dlp/extractor/_extractors.py
+++ b/yt_dlp/extractor/_extractors.py
@ -1855,6 +1855,10 @@ from .srgssr import (
    SRGSSRPlayIE,
 )
 from .srmediathek import SRMediathekIE
+from .stacommu import (
+    StacommuLiveIE,
+    StacommuVODIE,
+)
 from .stanfordoc import StanfordOpenClassroomIE
 from .startv import StarTVIE
 from .steam import (
--- a/yt_dlp/extractor/stacommu.py
+++ b/yt_dlp/extractor/stacommu.py
@ -0,0 +1,148 @@
+import time
+
+from .wrestleuniverse import WrestleUniverseBaseIE
+from ..utils import (
+    int_or_none,
+    traverse_obj,
+    url_or_none,
+)
+
+
+class StacommuBaseIE(WrestleUniverseBaseIE):
+    _NETRC_MACHINE = 'stacommu'
+    _API_HOST = 'api.stacommu.jp'
+    _LOGIN_QUERY = {'key': 'AIzaSyCR9czxhH2eWuijEhTNWBZ5MCcOYEUTAhg'}
+    _LOGIN_HEADERS = {
+        'Accept': '*/*',
+        'Content-Type': 'application/json',
+        'X-Client-Version': 'Chrome/JsCore/9.9.4/FirebaseCore-web',
+        'Referer': 'https://www.stacommu.jp/',
+        'Origin': 'https://www.stacommu.jp',
+    }
+
+    @WrestleUniverseBaseIE._TOKEN.getter
+    def _TOKEN(self):
+        if self._REAL_TOKEN and self._TOKEN_EXPIRY <= int(time.time()):
+            self._refresh_token()
+
+        return self._REAL_TOKEN
+
+    def _get_formats(self, data, path, video_id=None):
+        if not traverse_obj(data, path) and not data.get('canWatch') and not self._TOKEN:
+            self.raise_login_required(method='password')
+        return super()._get_formats(data, path, video_id)
+
+    def _extract_hls_key(self, data, path, decrypt):
+        encryption_data = traverse_obj(data, path)
+        if traverse_obj(encryption_data, ('encryptType', {int})) == 0:
+            return None
+        return traverse_obj(encryption_data, {'key': ('key', {decrypt}), 'iv': ('iv', {decrypt})})
+
+
+class StacommuVODIE(StacommuBaseIE):
+    _VALID_URL = r'https?://www\.stacommu\.jp/videos/episodes/(?P<id>[\da-zA-Z]+)'
+    _TESTS = [{
+        # not encrypted
+        'url': 'https://www.stacommu.jp/videos/episodes/aXcVKjHyAENEjard61soZZ',
+        'info_dict': {
+            'id': 'aXcVKjHyAENEjard61soZZ',
+            'ext': 'mp4',
+            'title': 'スタコミュAWARDの裏側、ほぼ全部見せます！〜晴れ舞台の直前ドキドキ編〜',
+            'description': 'md5:6400275c57ae75c06da36b06f96beb1c',
+            'timestamp': 1679652000,
+            'upload_date': '20230324',
+            'thumbnail': 'https://image.stacommu.jp/6eLobQan8PFtBoU4RL4uGg/6eLobQan8PFtBoU4RL4uGg',
+            'cast': 'count:11',
+            'duration': 250,
+        },
+        'params': {
+            'skip_download': 'm3u8',
+        },
+    }, {
+        # encrypted; requires a premium account
+        'url': 'https://www.stacommu.jp/videos/episodes/3hybMByUvzMEqndSeu5LpD',
+        'info_dict': {
+            'id': '3hybMByUvzMEqndSeu5LpD',
+            'ext': 'mp4',
+            'title': 'スタプラフェス2023〜裏側ほぼ全部見せます〜＃10',
+            'description': 'md5:85494488ccf1dfa1934accdeadd7b340',
+            'timestamp': 1682506800,
+            'upload_date': '20230426',
+            'thumbnail': 'https://image.stacommu.jp/eMdXtEefR4kEyJJMpAFi7x/eMdXtEefR4kEyJJMpAFi7x',
+            'cast': 'count:55',
+            'duration': 312,
+            'hls_aes': {
+                'key': '6bbaf241b8e1fd9f59ecf546a70e4ae7',
+                'iv': '1fc9002a23166c3bb1d240b953d09de9',
+            },
+        },
+        'params': {
+            'skip_download': 'm3u8',
+        },
+    }]
+
+    _API_PATH = 'videoEpisodes'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        video_info = self._download_metadata(
+            url, video_id, 'ja', ('dehydratedState', 'queries', 0, 'state', 'data'))
+        hls_info, decrypt = self._call_encrypted_api(
+            video_id, ':watch', 'stream information', data={'method': 1})
+
+        return {
+            'id': video_id,
+            'formats': self._get_formats(hls_info, ('protocolHls', 'url', {url_or_none}), video_id),
+            'hls_aes': self._extract_hls_key(hls_info, 'protocolHls', decrypt),
+            **traverse_obj(video_info, {
+                'title': ('displayName', {str}),
+                'description': ('description', {str}),
+                'timestamp': ('watchStartTime', {int_or_none}),
+                'thumbnail': ('keyVisualUrl', {url_or_none}),
+                'cast': ('casts', ..., 'displayName', {str}),
+                'duration': ('duration', {int}),
+            }),
+        }
+
+
+class StacommuLiveIE(StacommuBaseIE):
+    _VALID_URL = r'https?://www\.stacommu\.jp/live/(?P<id>[\da-zA-Z]+)'
+    _TESTS = [{
+        'url': 'https://www.stacommu.jp/live/d2FJ3zLnndegZJCAEzGM3m',
+        'info_dict': {
+            'id': 'd2FJ3zLnndegZJCAEzGM3m',
+            'ext': 'mp4',
+            'title': '仲村悠菜 2023/05/04',
+            'timestamp': 1683195647,
+            'upload_date': '20230504',
+            'thumbnail': 'https://image.stacommu.jp/pHGF57SPEHE2ke83FS92FN/pHGF57SPEHE2ke83FS92FN',
+            'duration': 5322,
+            'hls_aes': {
+                'key': 'efbb3ec0b8246f61adf1764c5a51213a',
+                'iv': '80621d19a1f19167b64cedb415b05d1c',
+            },
+        },
+        'params': {
+            'skip_download': 'm3u8',
+        },
+    }]
+
+    _API_PATH = 'events'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        video_info = self._call_api(video_id, msg='video information', query={'al': 'ja'}, auth=False)
+        hls_info, decrypt = self._call_encrypted_api(
+            video_id, ':watchArchive', 'stream information', data={'method': 1})
+
+        return {
+            'id': video_id,
+            'formats': self._get_formats(hls_info, ('hls', 'urls', ..., {url_or_none}), video_id),
+            'hls_aes': self._extract_hls_key(hls_info, 'hls', decrypt),
+            **traverse_obj(video_info, {
+                'title': ('displayName', {str}),
+                'timestamp': ('startTime', {int_or_none}),
+                'thumbnail': ('keyVisualUrl', {url_or_none}),
+                'duration': ('duration', {int_or_none}),
+            }),
+        }
--- a/yt_dlp/extractor/wrestleuniverse.py
+++ b/yt_dlp/extractor/wrestleuniverse.py
@ -14,12 +14,14 @@ from ..utils import (
    try_call,
    url_or_none,
    urlencode_postdata,
+    variadic,
 )


 class WrestleUniverseBaseIE(InfoExtractor):
    _NETRC_MACHINE = 'wrestleuniverse'
    _VALID_URL_TMPL = r'https?://(?:www\.)?wrestle-universe\.com/(?:(?P<lang>\w{2})/)?%s/(?P<id>\w+)'
+    _API_HOST = 'api.wrestle-universe.com'
    _API_PATH = None
    _REAL_TOKEN = None
    _TOKEN_EXPIRY = None
@ -67,24 +69,28 @@ class WrestleUniverseBaseIE(InfoExtractor):
                'returnSecureToken': True,
                'email': username,
                'password': password,
-            }, separators=(',', ':')).encode())
+            }, separators=(',', ':')).encode(), expected_status=400)
+        token = traverse_obj(login, ('idToken', {str}))
+        if not token:
+            raise ExtractorError(
+                f'Unable to log in: {traverse_obj(login, ("error", "message"))}', expected=True)
        self._REFRESH_TOKEN = traverse_obj(login, ('refreshToken', {str}))
        if not self._REFRESH_TOKEN:
            self.report_warning('No refresh token was granted')
-        self._TOKEN = traverse_obj(login, ('idToken', {str}))
+        self._TOKEN = token

    def _real_initialize(self):
-        if WrestleUniverseBaseIE._DEVICE_ID:
+        if self._DEVICE_ID:
            return

-        WrestleUniverseBaseIE._DEVICE_ID = self._configuration_arg('device_id', [None], ie_key='WrestleUniverse')[0]
-        if not WrestleUniverseBaseIE._DEVICE_ID:
-            WrestleUniverseBaseIE._DEVICE_ID = self.cache.load(self._NETRC_MACHINE, 'device_id')
-            if WrestleUniverseBaseIE._DEVICE_ID:
+        self._DEVICE_ID = self._configuration_arg('device_id', [None], ie_key=self._NETRC_MACHINE)[0]
+        if not self._DEVICE_ID:
+            self._DEVICE_ID = self.cache.load(self._NETRC_MACHINE, 'device_id')
+            if self._DEVICE_ID:
                return
-            WrestleUniverseBaseIE._DEVICE_ID = str(uuid.uuid4())
+            self._DEVICE_ID = str(uuid.uuid4())

-        self.cache.store(self._NETRC_MACHINE, 'device_id', WrestleUniverseBaseIE._DEVICE_ID)
+        self.cache.store(self._NETRC_MACHINE, 'device_id', self._DEVICE_ID)

    def _refresh_token(self):
        refresh = self._download_json(
@ -108,10 +114,10 @@ class WrestleUniverseBaseIE(InfoExtractor):
        if data:
            headers['Content-Type'] = 'application/json;charset=utf-8'
            data = json.dumps(data, separators=(',', ':')).encode()
-        if auth:
+        if auth and self._TOKEN:
            headers['Authorization'] = f'Bearer {self._TOKEN}'
        return self._download_json(
-            f'https://api.wrestle-universe.com/v1/{self._API_PATH}/{video_id}{param}', video_id,
+            f'https://{self._API_HOST}/v1/{self._API_PATH}/{video_id}{param}', video_id,
            note=f'Downloading {msg} JSON', errnote=f'Failed to download {msg} JSON',
            data=data, headers=headers, query=query, fatal=fatal)

@ -137,12 +143,13 @@ class WrestleUniverseBaseIE(InfoExtractor):
        }, query=query, fatal=fatal)
        return api_json, decrypt

-    def _download_metadata(self, url, video_id, lang, props_key):
+    def _download_metadata(self, url, video_id, lang, props_keys):
        metadata = self._call_api(video_id, msg='metadata', query={'al': lang or 'ja'}, auth=False, fatal=False)
        if not metadata:
            webpage = self._download_webpage(url, video_id)
            nextjs_data = self._search_nextjs_data(webpage, video_id)
-            metadata = traverse_obj(nextjs_data, ('props', 'pageProps', props_key, {dict})) or {}
+            metadata = traverse_obj(nextjs_data, (
+                'props', 'pageProps', *variadic(props_keys, (str, bytes, dict, set)), {dict})) or {}
        return metadata

    def _get_formats(self, data, path, video_id=None):