Merge 3302588c77 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/facebook.py

@@ -443,7 +443,9 @@ class FacebookIE(InfoExtractor):
         try:
             login_results = self._download_webpage(request, None,
                                                    note='Logging in', errnote='unable to fetch login page')
-            if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
+            if 'Your Request Couldn' in login_results:
+                self.raise_login_required('Failed to login with credentials', method='cookies')
+            elif re.search(r'<form[^>]*name="login"[^<]*</form>', login_results):
                 error = self._html_search_regex(
                     r'(?s)<div[^>]+class=(["\']).*?login_error_box.*?\1[^>]*><div[^>]*>.*?</div><div[^>]*>(?P<error>.+?)</div>',
                     login_results, 'login error', default=None, group='error')
@@ -476,13 +478,41 @@ class FacebookIE(InfoExtractor):
             return
 
     def _extract_from_url(self, url, video_id):
+        cookies = self._get_cookies(url)
+        # user passed logged-in cookies or attempted to login
+        login_data = cookies.get('c_user') and cookies.get('xs')
+        logged_in = False
+
         webpage = self._download_webpage(
             url.replace('://m.facebook.com/', '://www.facebook.com/'), video_id)
 
+        sjs_data = [self._parse_json(j, video_id, fatal=False) for j in re.findall(
+            r'data-sjs>({.*?ScheduledServerJS.*?})</script>', webpage)]
+
+        if login_data:
+            logged_in = get_first(sjs_data, (
+                'require', ..., ..., ..., '__bbox', 'define',
+                lambda _, v: 'CurrentUserInitialData' in v, ..., 'ACCOUNT_ID'), default='0') != '0'
+            if logged_in and (info := get_first(sjs_data, ('require', ..., ..., ..., '__bbox', 'require', ..., ..., ...,
+                                                           '__bbox', 'result', 'data', (('ufac_client', 'state',
+                                                                                         (('set_contact_point_state_renderer', 'title'),
+                                                                                          ('intro_state_renderer', 'header_title'))),
+                                                                                        ('epsilon_checkpoint', 'screen', 'title'))))):
+                if any(content in info for content in ['days left to appeal', 'suspended your account']):
+                    raise ExtractorError('Your account is suspended', expected=True)
+                if 'Enter mobile number' == info:
+                    raise ExtractorError('Facebook is requiring mobile number confirmation', expected=True)
+                if 'your account has been locked' in info:
+                    raise ExtractorError('Your account has been locked', expected=True)
+
+        if props := get_first(sjs_data, (
+                'require', ..., ..., ..., '__bbox', 'require', ..., ..., ..., 'rootView',
+                lambda _, v: v['title'].startswith('This content isn\'t available'))):
+            raise ExtractorError(
+                f'Content unavailable. Facebook said: {props.get("body") or props["title"]}', expected=True)
+
         def extract_metadata(webpage):
-            post_data = [self._parse_json(j, video_id, fatal=False) for j in re.findall(
+            post = traverse_obj(sjs_data, (
-                r'data-sjs>({.*?ScheduledServerJS.*?})</script>', webpage)]
-            post = traverse_obj(post_data, (
                 ..., 'require', ..., ..., ..., '__bbox', 'require', ..., ..., ..., '__bbox', 'result', 'data'), expected_type=dict) or []
             media = traverse_obj(post, (..., 'attachments', ..., lambda k, v: (
                 k == 'media' and str(v['id']) == video_id and v['__typename'] == 'Video')), expected_type=dict)
@@ -810,6 +840,10 @@ class FacebookIE(InfoExtractor):
             video_data = extract_from_jsmods_instances(tahoe_js_data)
 
         if not video_data:
+            if not login_data:
+                raise ExtractorError('Cannot parse data. Try logging in.', expected=True)
+            if not logged_in:
+                raise ExtractorError('Failed to login with provided data.', expected=True)
             raise ExtractorError('Cannot parse data')
 
         if len(video_data) > 1: