Merge 00f0acfade into b83ca24eb7

Authored by: seproDev

.github/workflows/build.yml

@@ -504,7 +504,8 @@ jobs:
       - windows32
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/download-artifact@v4
+      - name: Download artifacts
+        uses: actions/download-artifact@v4
         with:
           path: artifact
           pattern: build-bin-*

.github/workflows/release-master.yml

@@ -28,3 +28,20 @@ jobs:
       actions: write  # For cleaning up cache
       id-token: write  # mandatory for trusted publishing
     secrets: inherit
+
+  publish_pypi:
+    needs: [release]
+    if: vars.MASTER_PYPI_PROJECT != ''
+    runs-on: ubuntu-latest
+    permissions:
+      id-token: write  # mandatory for trusted publishing
+    steps:
+      - name: Download artifacts
+        uses: actions/download-artifact@v4
+        with:
+          path: dist
+          name: build-pypi
+      - name: Publish to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1
+        with:
+          verbose: true

.github/workflows/release-nightly.yml

@@ -41,3 +41,20 @@ jobs:
       actions: write  # For cleaning up cache
       id-token: write  # mandatory for trusted publishing
     secrets: inherit
+
+  publish_pypi:
+    needs: [release]
+    if: vars.NIGHTLY_PYPI_PROJECT != ''
+    runs-on: ubuntu-latest
+    permissions:
+      id-token: write  # mandatory for trusted publishing
+    steps:
+      - name: Download artifacts
+        uses: actions/download-artifact@v4
+        with:
+          path: dist
+          name: build-pypi
+      - name: Publish to PyPI
+        uses: pypa/gh-action-pypi-publish@release/v1
+        with:
+          verbose: true

.github/workflows/release.yml

@@ -2,10 +2,6 @@ name: Release
 on:
   workflow_call:
     inputs:
-      prerelease:
-        required: false
-        default: true
-        type: boolean
       source:
         required: false
         default: ''
@@ -18,6 +14,10 @@ on:
         required: false
         default: ''
         type: string
+      prerelease:
+        required: false
+        default: true
+        type: boolean
   workflow_dispatch:
     inputs:
       source:
@@ -278,11 +278,20 @@ jobs:
           make clean-cache
           python -m build --no-isolation .
 
+      - name: Upload artifacts
+        if: github.event_name != 'workflow_dispatch'
+        uses: actions/upload-artifact@v4
+        with:
+          name: build-pypi
+          path: |
+            dist/*
+          compression-level: 0
+
       - name: Publish to PyPI
+        if: github.event_name == 'workflow_dispatch'
         uses: pypa/gh-action-pypi-publish@release/v1
         with:
           verbose: true
-          attestations: false  # Currently doesn't work w/ reusable workflows (breaks nightly)
 
   publish:
     needs: [prepare, build]

pyproject.toml

@@ -52,7 +52,7 @@ default = [
     "pycryptodomex",
     "requests>=2.32.2,<3",
     "urllib3>=1.26.17,<3",
-    "websockets>=13.0",
+    "websockets>=13.0,<14",
 ]
 curl-cffi = [
     "curl-cffi==0.5.10; os_name=='nt' and implementation_name=='cpython'",

yt_dlp/dependencies/Cryptodome.py

@@ -24,7 +24,7 @@ try:
         from Crypto.Cipher import AES, PKCS1_OAEP, Blowfish, PKCS1_v1_5  # noqa: F401
         from Crypto.Hash import CMAC, SHA1  # noqa: F401
         from Crypto.PublicKey import RSA  # noqa: F401
-except ImportError:
+except (ImportError, OSError):
     __version__ = f'broken {__version__}'.strip()
 
 

yt_dlp/extractor/_extractors.py

@@ -2392,6 +2392,7 @@ from .vrt import (
     DagelijkseKostIE,
     KetnetIE,
     Radio1BeIE,
+    VRTMaxRadioIE,
     VrtNUIE,
 )
 from .vtm import VTMIE

yt_dlp/extractor/vrt.py

@@ -21,6 +21,7 @@ from ..utils import (
     str_or_none,
     strip_or_none,
     traverse_obj,
+    try_call,
     url_or_none,
     urlencode_postdata,
 )
@@ -486,3 +487,124 @@ class Radio1BeIE(VRTBaseIE):
                     'description': self._html_search_meta(['description', 'og:description', 'twitter:description'], webpage),
                     'thumbnail': self._html_search_meta(['og:image', 'twitter:image'], webpage),
             }))
+
+
+class VRTMaxRadioIE(VRTBaseIE):
+    IE_DESC = 'VRT MAX (radio)'
+    _VALID_URL = r'https?://(?:www\.)?vrt\.be/(?:vrtmax|vrtnu)/luister/radio/(?P<show1l>[^/])/(?P<show>[^/]+)/(?P<episode_id>[^/?#&]+)/?'
+    _TESTS = [{
+        'url': 'https://www.vrt.be/vrtmax/luister/radio/d/duyster~11-177/duyster~11-27934-0/',
+        'md5': '14d002b1ebd8591ae360ff54a9b51515',  # ... of first Fragment.
+        'info_dict': {
+            'id': 'duyster~11-27934-0',
+            'ext': 'mp4',
+            'title': 'DUYSTER.',
+            'description': 'DUYSTER. - maandag 10 juni 2024 om 22:00 | VRT MAX',
+            'thumbnail': 'https://images.vrt.be/vrtmax_prog_radio1/2023/10/05/1266f9e6-633c-11ee-91d7-02b7b76bf47f.png',
+            'timestamp': 1718049600,
+            'upload_date': '20240610',
+            'duration': 7200,
+            'channel': 'Radio 1',
+            'channel_url': 'https://www.vrt.be/vrtmax/kanalen/radio-1/',
+            'display_id': 'duyster~11-27934-0',
+        },
+    }, {
+        'url': 'https://www.vrt.be/vrtnu/luister/radio/n/nachtbaders~31-182/nachtbaders~31-22512-0/',
+        'md5': '85e9b0edf133f6638aa8d56581ac8773',
+        'info_dict': {
+            'id': 'nachtbaders~31-22512-0',
+            'ext': 'mp4',
+            'title': 'Nachtbaders',
+            'description': 'Nachtbaders - vrijdag 5 januari 2024 om 22:00 | VRT MAX | Nachtbaders Tom Soetaert en Niels Van Paemel dompelen je onder in de dromerige wereld van de neoklassiek. In hun queeste naar het beste uit het genre leggen ze de luisteraar te week in een bad van bedwelmende muziek van onder meer Max Richter, Bon Iver, Sigur Rós, Nils Frahm en Jóhann Jóhannson.',
+            'thumbnail': 'https://images.vrt.be/orig/2021/10/20/fdd5992e-31ad-11ec-b07d-02b7b76bf47f.jpg',
+            'timestamp': 1704488400.0,
+            'upload_date': '20240105',
+            'duration': 7200.0,
+            'channel': 'Klara',
+            'channel_url': 'https://www.vrt.be/vrtmax/kanalen/klara/',
+            'display_id': 'nachtbaders~31-22512-0',
+        },
+    }]
+
+    _GRAPHQUERY = '''query RadioEpisodePage($pageId: ID!) {
+  page(id: $pageId) {
+    ... on RadioEpisodePage {
+      title
+      socialSharing {
+        title
+      }
+      player {
+        listenAction {
+          ... on RadioEpisodeListenAction {
+            streamId
+            startDate
+            endDate
+          }
+        }
+      }
+      radioEpisode {
+        richDescription {
+          text
+        }
+        brand
+        image {
+          templateUrl
+        }
+        actionItems {
+          action {
+            ... on LinkAction {
+              link
+              linkType
+            }
+          }
+        }
+      }
+    }
+  }
+}'''
+
+    def get_metadata(self, show1l, show, episode_id):
+        # Rather fragile, download responds with nothing but "400: Bad Request" if any
+        # GraphQL part or header is out of place.  Best to keep them minimal.
+        graphqlvar = f'/vrtnu/luister/radio/{show1l}/{show}/{episode_id}/'
+        postdata = json.dumps({
+            'query': self._GRAPHQUERY,
+            'variables': {'pageId': graphqlvar},
+        }).encode()
+        return self._download_json(
+            'https://www.vrt.be/vrtnu-api/graphql/public/v1', episode_id,
+            note='Downloading GraphQL metadata', data=postdata,
+            headers={
+                'content-type': 'application/json',
+                'Accept': 'application/graphql+json, application/json',
+                'Accept-Encoding': 'gzip, deflate, br',
+                'x-vrt-client-name': 'WEB',
+            })
+
+    def _real_extract(self, url):
+        (show1l, show, episode_id) = self._match_valid_url(url).groups()
+        metadata = self.get_metadata(show1l, show, episode_id)['data']['page']
+
+        audio_id = traverse_obj(metadata, ('player', 'listenAction', 'streamId'))
+        media_items = self._call_api(audio_id, 'vrtnu-web@PROD', version='v2')
+        formats, _ = self._extract_formats_and_subtitles(media_items, audio_id)
+
+        return {
+            'id': episode_id,
+            'formats': formats,
+            **traverse_obj(metadata, {
+                'title': 'title',
+                'description': ([('socialSharing', 'title'),
+                                 ('radioEpisode', 'richDescription', 'text')],
+                                all, {lambda txts: ' | '.join(txts) or None}),
+                'thumbnail': ('radioEpisode', 'image', 'templateUrl'),
+                'timestamp': ('player', 'listenAction', 'startDate', {lambda x: x / 1000}),
+                # Duration of original transmission, downloaded file is sometimes shorter:
+                'duration': ('player', 'listenAction',
+                             {lambda a: try_call(lambda: (a['endDate'] - a['startDate']) / 1000)}),
+                'channel': ('radioEpisode', 'brand'),
+                'channel_url': ('radioEpisode', 'actionItems', ..., 'action', all,
+                                lambda _, act: act.get('linkType') == 'channel', any,
+                                'link', {lambda link: 'https://www.vrt.be' + link}),
+            }),
+        }