Merge 0d26f152b0 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

@@ -708,6 +708,7 @@ from .gab import (
     GabTVIE,
 )
 from .gaia import GaiaIE
+from .gamedevtv import GameDevTVDashboardIE
 from .gamejolt import (
     GameJoltCommunityIE,
     GameJoltGameIE,
@@ -2252,6 +2253,10 @@ from .ufctv import (
 from .ukcolumn import UkColumnIE
 from .uktvplay import UKTVPlayIE
 from .umg import UMGDeIE
+from .underline import (
+    UnderlinePosterIE,
+    UnderlinePosterListIE,
+)
 from .unistra import UnistraIE
 from .unity import UnityIE
 from .unsupported import (

yt_dlp/extractor/adobepass.py

@@ -1362,7 +1362,7 @@ class AdobePassIE(InfoExtractor):  # XXX: Conventionally, base classes should en
 
     def _download_webpage_handle(self, *args, **kwargs):
         headers = self.geo_verification_headers()
-        headers.update(kwargs.get('headers', {}))
+        headers.update(kwargs.get('headers') or {})
         kwargs['headers'] = headers
         return super()._download_webpage_handle(
             *args, **kwargs)

yt_dlp/extractor/gamedevtv.py

@@ -0,0 +1,141 @@
+import json
+
+from .common import InfoExtractor
+from ..networking.exceptions import HTTPError
+from ..utils import (
+    ExtractorError,
+    clean_html,
+    int_or_none,
+    join_nonempty,
+    parse_iso8601,
+    str_or_none,
+    url_or_none,
+)
+from ..utils.traversal import traverse_obj
+
+
+class GameDevTVDashboardIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?gamedev\.tv/dashboard/courses/(?P<course_id>\d+)(?:/(?P<lecture_id>\d+))?'
+    _NETRC_MACHINE = 'gamedevtv'
+    _TESTS = [{
+        'url': 'https://www.gamedev.tv/dashboard/courses/25',
+        'info_dict': {
+            'id': '25',
+            'title': 'Complete Blender Creator 3: Learn 3D Modelling for Beginners',
+            'tags': ['blender', 'course', 'all', 'box modelling', 'sculpting'],
+            'categories': ['Blender', '3D Art'],
+            'thumbnail': 'https://gamedev-files.b-cdn.net/courses/qisc9pmu1jdc.jpg',
+            'upload_date': '20220516',
+            'timestamp': 1652694420,
+            'modified_date': '20241027',
+            'modified_timestamp': 1730049658,
+        },
+        'playlist_count': 100,
+    }, {
+        'url': 'https://www.gamedev.tv/dashboard/courses/63/2279',
+        'info_dict': {
+            'id': 'df04f4d8-68a4-4756-a71b-9ca9446c3a01',
+            'ext': 'mp4',
+            'modified_timestamp': 1701695752,
+            'upload_date': '20230504',
+            'episode': 'MagicaVoxel Community Course Introduction',
+            'series_id': '63',
+            'title': 'MagicaVoxel Community Course Introduction',
+            'timestamp': 1683195397,
+            'modified_date': '20231204',
+            'categories': ['3D Art', 'MagicaVoxel'],
+            'season': 'MagicaVoxel Community Course',
+            'tags': ['MagicaVoxel', 'all', 'course'],
+            'series': 'MagicaVoxel 3D Art Mini Course',
+            'duration': 1405,
+            'episode_number': 1,
+            'season_number': 1,
+            'season_id': '219',
+            'description': 'md5:a378738c5bbec1c785d76c067652d650',
+            'display_id': '63-219-2279',
+            'alt_title': '1_CC_MVX MagicaVoxel Community Course Introduction.mp4',
+            'thumbnail': 'https://vz-23691c65-6fa.b-cdn.net/df04f4d8-68a4-4756-a71b-9ca9446c3a01/thumbnail.jpg',
+        },
+    }]
+    _API_HEADERS = {}
+
+    def _perform_login(self, username, password):
+        try:
+            response = self._download_json(
+                'https://api.gamedev.tv/api/students/login', None, 'Logging in',
+                headers={'Content-Type': 'application/json'},
+                data=json.dumps({
+                    'email': username,
+                    'password': password,
+                    'cart_items': [],
+                }).encode())
+        except ExtractorError as e:
+            if isinstance(e.cause, HTTPError) and e.cause.status == 401:
+                raise ExtractorError('Invalid username/password', expected=True)
+            raise
+
+        self._API_HEADERS['Authorization'] = f'{response["token_type"]} {response["access_token"]}'
+
+    def _real_initialize(self):
+        if not self._API_HEADERS.get('Authorization'):
+            self.raise_login_required(
+                'This content is only available with purchase', method='password')
+
+    def _entries(self, data, course_id, course_info, selected_lecture):
+        for section in traverse_obj(data, ('sections', ..., {dict})):
+            section_info = traverse_obj(section, {
+                'season_id': ('id', {str_or_none}),
+                'season': ('title', {str}),
+                'season_number': ('order', {int_or_none}),
+            })
+            for lecture in traverse_obj(section, ('lectures', lambda _, v: url_or_none(v['video']['playListUrl']))):
+                if selected_lecture and str(lecture.get('id')) != selected_lecture:
+                    continue
+                display_id = join_nonempty(course_id, section_info.get('season_id'), lecture.get('id'))
+                formats, subtitles = self._extract_m3u8_formats_and_subtitles(
+                    lecture['video']['playListUrl'], display_id, 'mp4', m3u8_id='hls')
+                yield {
+                    **course_info,
+                    **section_info,
+                    'id': display_id,  # fallback
+                    'display_id': display_id,
+                    'formats': formats,
+                    'subtitles': subtitles,
+                    'series': course_info.get('title'),
+                    'series_id': course_id,
+                    **traverse_obj(lecture, {
+                        'id': ('video', 'guid', {str}),
+                        'title': ('title', {str}),
+                        'alt_title': ('video', 'title', {str}),
+                        'description': ('description', {clean_html}),
+                        'episode': ('title', {str}),
+                        'episode_number': ('order', {int_or_none}),
+                        'duration': ('video', 'duration_in_sec', {int_or_none}),
+                        'timestamp': ('video', 'created_at', {parse_iso8601}),
+                        'modified_timestamp': ('video', 'updated_at', {parse_iso8601}),
+                        'thumbnail': ('video', 'thumbnailUrl', {url_or_none}),
+                    }),
+                }
+
+    def _real_extract(self, url):
+        course_id, lecture_id = self._match_valid_url(url).group('course_id', 'lecture_id')
+        data = self._download_json(
+            f'https://api.gamedev.tv/api/courses/my/{course_id}', course_id,
+            headers=self._API_HEADERS)['data']
+
+        course_info = traverse_obj(data, {
+            'title': ('title', {str}),
+            'tags': ('tags', ..., 'name', {str}),
+            'categories': ('categories', ..., 'title', {str}),
+            'timestamp': ('created_at', {parse_iso8601}),
+            'modified_timestamp': ('updated_at', {parse_iso8601}),
+            'thumbnail': ('image', {url_or_none}),
+        })
+
+        entries = self._entries(data, course_id, course_info, lecture_id)
+        if lecture_id:
+            lecture = next(entries, None)
+            if not lecture:
+                raise ExtractorError('Lecture not found')
+            return lecture
+        return self.playlist_result(entries, course_id, **course_info)

yt_dlp/extractor/underline.py

@@ -0,0 +1,123 @@
+import functools
+
+from .common import InfoExtractor
+from ..utils import (
+    OnDemandPagedList,
+    filter_dict,
+    update_url,
+    url_or_none,
+)
+from ..utils.traversal import traverse_obj
+
+
+class UnderlinePosterIE(InfoExtractor):
+    _VALID_URL = r'https://(?:www\.)?underline\.io/events/\d+/posters/\d+/poster/(?P<id>\d+)-[\w-]+/?(?:[#?]|$)'
+
+    _TESTS = [{
+        # Video and PPT
+        'url': 'https://underline.io/events/342/posters/12863/poster/66466-towards-a-general-purpose-machine-translation-system-for-sranantongo',
+        'info_dict': {
+            'id': '66466',
+            'ext': 'mp4',
+            'title': 'Towards a general purpose machine translation system for Sranantongo',
+            'thumbnail': 'https://assets.underline.io/lecture/66466/poster_document_thumbnail_extract/2a8d3abd5a5edbf0117d8d9bdf018e6d.jpg',
+            'live_status': 'not_live',
+        },
+    }, {
+        # PPT only
+        'url': 'https://underline.io/events/342/posters/12863/poster/66459-low-resourced-multilingual-neural-machine-translation-for-ometo-english',
+        'info_dict': {
+            'id': '66459',
+            'title': 'Low Resourced Multilingual Neural Machine Translation for Ometo-English',
+            'thumbnail': 'https://assets.underline.io/lecture/66459/poster_document_thumbnail_extract/69a3fd48e3bb137ddcf3099db637d184.jpg',
+            'live_status': 'not_live',
+        },
+        'params': {
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+        'expected_warnings': ['No video formats found!', 'Requested format is not available'],
+    }, {
+        'url': 'https://www.underline.io/events/342/posters/12863/poster/66463-mbti-personality-prediction-approach-on-persian-twitter/?tab=video',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        data = self._search_nextjs_data(webpage, video_id)['props']['pageProps']['fallback']
+
+        m3u8_url = traverse_obj(data, (..., 'data', 'attributes', 'playlist', {url_or_none}), get_all=False)
+        subtitle_urls = filter_dict(dict(traverse_obj(data, (
+            ..., 'data', lambda _, v: v['type'] == 'transcripts', {lambda x: (
+                x['relationships']['language']['data']['id'],
+                url_or_none(x['attributes']['subtitleUrl']),
+            )}))))
+
+        return {
+            'id': video_id,
+            'title': traverse_obj(data, (..., 'data', 'attributes', 'title', {str}), get_all=False),
+            'thumbnail': traverse_obj(data, (
+                ..., 'data', 'attributes', 'originalPosterDocumentThumbnailExtractUrl', {url_or_none}), get_all=False),
+            'formats': self._extract_m3u8_formats(m3u8_url, video_id) if m3u8_url else [],
+            'subtitles': filter_dict(dict(traverse_obj(data, (
+                ..., 'included', lambda _, v: v['type'] == 'transcript_languages', {lambda x: (
+                    x['attributes']['locale'],
+                    [{'url': subtitle_urls[x['id']]}],
+                )})))),
+            'live_status': 'not_live',
+        }
+
+
+class UnderlinePosterListIE(InfoExtractor):
+    _VALID_URL = r'https://(?:www\.)?underline\.io/events/(?P<event_id>\d+)/posters/?\?(?:[^&]+&)*?eventSessionId=(?P<event_session_id>\d+)(?:&|#)?'
+    _HEADERS = {'Accept': 'application/vnd.api+json'}
+    _PAGE_SIZE = 16
+
+    _TESTS = [{
+        'url': 'https://underline.io/events/342/posters?eventSessionId=12863',
+        'info_dict': {
+            'id': '12863',
+            'title': 'W13 (WiNLP) The Sixth Widening NLP Workshop',
+            'description': r're:.+Isidora Tourni and Surangika Ranathunga.+',
+        },
+        'playlist_mincount': 10,
+    }, {
+        'url': 'https://www.underline.io/events/342/posters?searchGroup=lecture&eventSessionId=12860',
+        'only_matching': True,
+    }]
+
+    def _fetch_paged_channel_video_list(self, list_url, event_id, event_session_id, page):
+        page_num = page + 1
+        video_url_tmpl = f'https://underline.io/events/{event_id}/posters/{event_session_id}/poster/{{id}}-{{slug}}'
+
+        page_info = self._download_json(update_url(
+            list_url, query_update={'page[number]': page_num}), event_session_id, headers=self._HEADERS,
+            note=f'Downloading list info (page {page_num})', errnote=f'Failed to download list info (page {page_num})')
+
+        yield from traverse_obj(page_info, (
+            'data', ..., {lambda x: video_url_tmpl.format(id=x['id'], slug=x['attributes']['slug'])}, {self.url_result}))
+
+    def _real_extract(self, url):
+        event_id, event_session_id = self._match_valid_url(url).groups()
+
+        event_info = self._download_json(
+            f'https://app.underline.io/api/v1/thin_event_sessions/{event_session_id}', event_session_id, headers=self._HEADERS,
+            note='Downloading event session info', errnote='Failed to download event session info')
+
+        return self.playlist_result(OnDemandPagedList(functools.partial(
+            self._fetch_paged_channel_video_list,
+            update_url('https://app.underline.io/api/v1/thin_lectures', query_update={
+                'filter[poster_lecture]': 'true',
+                'filter[published]': 'all',
+                'sort': 'held_at,id',
+                'include': 'sorted_profiles,event_session,tag,event',
+                'filter[scope]': 'all',
+                'filter[event_id]': event_id,
+                'filter[event_session_id]': event_session_id,
+                'page[size]': self._PAGE_SIZE,
+            }), event_id, event_session_id), self._PAGE_SIZE),
+            event_session_id,
+            traverse_obj(event_info, ('data', 'attributes', 'name'), get_all=False),
+            traverse_obj(event_info, ('data', 'attributes', 'description'), get_all=False))