Merge ea7d7d85d3 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]

README.md

@@ -1869,6 +1869,9 @@ The following extractors use this feature:
 #### digitalconcerthall
 * `prefer_combined_hls`: Prefer extracting combined/pre-merged video and audio HLS formats. This will exclude 4K/HEVC video and lossless/FLAC audio formats, which are only available as split video/audio HLS formats
 
+#### bahamut
+* `device_id`: (optional) Device ID got from `https://ani.gamer.com.tw/ajax/getdeviceid.php` (bound to cookies and `User-Agent` HTTP header). The extractor will automatically fetch one if it is not present. E.g. `"bahamut:device_id=1234567890abcdef1234567890abcdef1234567890abcdef1234567890ab"`
+
 **Note**: These options may be changed/removed in the future without concern for backward compatibility
 
 <!-- MANPAGE: MOVE "INSTALLATION" SECTION HERE -->

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

@@ -196,6 +196,7 @@ from .awaan import (
 )
 from .axs import AxsIE
 from .azmedien import AZMedienIE
+from .bahamut import BahamutIE
 from .baidu import BaiduVideoIE
 from .banbye import (
     BanByeChannelIE,

yt_dlp/extractor/bahamut.py

@@ -0,0 +1,111 @@
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    smuggle_url,
+    unified_timestamp,
+    unsmuggle_url,
+)
+from ..utils.traversal import traverse_obj
+
+
+class BahamutIE(InfoExtractor):
+    _VALID_URL = r'https?://ani\.gamer\.com\.tw/animeVideo\.php\?sn=(?P<id>\d+)'
+
+    # see anime_player.js
+    RATING_TO_AGE_LIMIT = {
+        1: 0,
+        2: 6,
+        3: 12,
+        4: 15,
+        5: 18,
+        6: 18,  # age-gated, needs login
+    }
+
+    def _real_extract(self, url):
+        url, unsmuggled_data = unsmuggle_url(url, {})
+        video_id = self._match_id(url)
+        device_id = (
+            self._configuration_arg('device_id', [None], casesense=True)[0]
+            or unsmuggled_data.get('device_id')
+            or self._download_json(
+                'https://ani.gamer.com.tw/ajax/getdeviceid.php', video_id,
+                'Downloading device ID', 'Failed to download device ID',
+                impersonate=True, headers=self.geo_verification_headers())['deviceid'])
+
+        # TODO: extract metadata from webpage
+        metadata = {}
+        if api_result := self._download_json(
+                'https://api.gamer.com.tw/anime/v1/video.php', video_id,
+                'Downloading video info', 'Failed to download video info',
+                impersonate=True, query={'videoSn': video_id}).get('data'):
+
+            metadata.update(traverse_obj(api_result, ('anime', {
+                'description': 'content',
+                'thumbnail': 'cover',
+                'tags': 'tags',
+                'creators': ('director', {lambda x: [x]}),
+                'title': 'title',
+            })))
+            playlist_id = traverse_obj(api_result, ('video', 'animeSn')) or ''
+            if self._yes_playlist(playlist_id, video_id) and unsmuggled_data.get('extract_playlist') is not False:
+                return self.playlist_result(
+                    (self.url_result(
+                        # it may be better to use self.cache for storing device_id
+                        smuggle_url(f'https://ani.gamer.com.tw/animeVideo.php?sn={ep["videoSn"]}', {
+                            'extract_playlist': False,
+                            'device_id': device_id,
+                        }), ie=BahamutIE,
+                        video_id=ep['videoSn'], thumbnail=ep.get('cover')) for ep in traverse_obj(
+                            api_result,
+                            # This (the first ellipsis) extracts episodes of all languages,
+                            # maybe just extract episodes of the current language?
+                            ('anime', 'episodes', ..., ...))),
+                    playlist_id=playlist_id, **metadata)
+
+            # video-specific metadata, extract after returning the playlist result
+            metadata.update(traverse_obj(api_result, ('video', {
+                'thumbnail': 'cover',
+                'title': 'title',
+                'timestamp': ('upTime', {unified_timestamp}),
+                'duration': ('duration', {float_or_none(scale=60)}),
+                'age_limit': ('rating', {lambda x: self.RATING_TO_AGE_LIMIT.get(x)}),
+            })))
+
+        m3u8_info, urlh = self._download_json_handle(
+            'https://ani.gamer.com.tw/ajax/m3u8.php', video_id,
+            note='Downloading m3u8 URL', errnote='Failed to download m3u8 URL', query={
+                'sn': video_id,
+                'device': device_id,
+            }, impersonate=True, headers=self.geo_verification_headers(), expected_status=400)
+
+        formats_fatal = True
+        if urlh.status == 400:
+            # TODO: handle more error codes, search for /case \d+{4}:/g in anime_player.js
+            error_code = traverse_obj(m3u8_info, ('error', 'code'))
+            if error_code == 1011:
+                self.raise_geo_restricted(metadata_available=True)
+                formats_fatal = False
+            elif error_code == 1007:
+                if unsmuggled_data.pop('device_id', None) is not None:
+                    return self.url_result(
+                        smuggle_url(f'https://ani.gamer.com.tw/animeVideo.php?sn={video_id}',
+                                    unsmuggled_data), ie=BahamutIE, video_id=video_id)
+                raise ExtractorError('Invalid device id!')
+            elif error_code == 1017:
+                self.raise_login_required(metadata_available=True)
+                formats_fatal = False
+            else:
+                raise ExtractorError(
+                    traverse_obj(m3u8_info, ('error', 'message')) or 'Failed to download m3u8 URL')
+
+        return {
+            **metadata,
+            'id': video_id,
+            'formats': self._extract_m3u8_formats(
+                m3u8_info.get('src'), video_id, ext='mp4', fatal=formats_fatal, headers={
+                    'Origin': 'https://ani.gamer.com.tw',
+                    **self.geo_verification_headers(),
+                }),
+            'http_headers': {'Origin': 'https://ani.gamer.com.tw'},
+        }