Merge branch 'main' into 2-25-vinext

Co-authored-by: KinomotoMio <200703522+KinomotoMio@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

.github/workflows/pyrefly-diff-comment.yml

@@ -0,0 +1,88 @@
+name: Comment with Pyrefly Diff
+
+on:
+  workflow_run:
+    workflows:
+      - Pyrefly Diff Check
+    types:
+      - completed
+
+permissions: {}
+
+jobs:
+  comment:
+    name: Comment PR with pyrefly diff
+    runs-on: ubuntu-latest
+    permissions:
+      actions: read
+      contents: read
+      issues: write
+      pull-requests: write
+    if: ${{ github.event.workflow_run.conclusion == 'success' && github.event.workflow_run.pull_requests[0].head.repo.full_name != github.repository }}
+    steps:
+      - name: Download pyrefly diff artifact
+        uses: actions/github-script@v8
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          script: |
+            const fs = require('fs');
+            const artifacts = await github.rest.actions.listWorkflowRunArtifacts({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              run_id: ${{ github.event.workflow_run.id }},
+            });
+            const match = artifacts.data.artifacts.find((artifact) =>
+              artifact.name === 'pyrefly_diff'
+            );
+            if (!match) {
+              throw new Error('pyrefly_diff artifact not found');
+            }
+            const download = await github.rest.actions.downloadArtifact({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              artifact_id: match.id,
+              archive_format: 'zip',
+            });
+            fs.writeFileSync('pyrefly_diff.zip', Buffer.from(download.data));
+
+      - name: Unzip artifact
+        run: unzip -o pyrefly_diff.zip
+
+      - name: Post comment
+        uses: actions/github-script@v8
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          script: |
+            const fs = require('fs');
+            let diff = fs.readFileSync('pyrefly_diff.txt', { encoding: 'utf8' });
+            let prNumber = null;
+            try {
+              prNumber = parseInt(fs.readFileSync('pr_number.txt', { encoding: 'utf8' }), 10);
+            } catch (err) {
+              // Fallback to workflow_run payload if artifact is missing or incomplete.
+              const prs = context.payload.workflow_run.pull_requests || [];
+              if (prs.length > 0 && prs[0].number) {
+                prNumber = prs[0].number;
+              }
+            }
+            if (!prNumber) {
+              throw new Error('PR number not found in artifact or workflow_run payload');
+            }
+
+            const MAX_CHARS = 65000;
+            if (diff.length > MAX_CHARS) {
+              diff = diff.slice(0, MAX_CHARS);
+              diff = diff.slice(0, diff.lastIndexOf('\\n'));
+              diff += '\\n\\n... (truncated) ...';
+            }
+
+            const body = diff.trim()
+              ? `### Pyrefly Diff (base → PR)\\n\\`\\`\\`diff\\n${diff}\\n\\`\\`\\``
+              : '### Pyrefly Diff\\nNo changes detected.';
+
+            await github.rest.issues.createComment({
+              issue_number: prNumber,
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              body,
+            });

.github/workflows/pyrefly-diff.yml

@@ -0,0 +1,85 @@
+name: Pyrefly Diff Check
+
+on:
+  pull_request:
+    paths:
+      - 'api/**/*.py'
+
+permissions:
+  contents: read
+
+jobs:
+  pyrefly-diff:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      issues: write
+      pull-requests: write
+    steps:
+      - name: Checkout PR branch
+        uses: actions/checkout@v6
+        with:
+          fetch-depth: 0
+
+      - name: Setup Python & UV
+        uses: astral-sh/setup-uv@v5
+        with:
+          enable-cache: true
+
+      - name: Install dependencies
+        run: uv sync --project api --dev
+
+      - name: Run pyrefly on PR branch
+        run: |
+          uv run --directory api pyrefly check > /tmp/pyrefly_pr.txt 2>&1 || true
+
+      - name: Checkout base branch
+        run: git checkout ${{ github.base_ref }}
+
+      - name: Run pyrefly on base branch
+        run: |
+          uv run --directory api pyrefly check > /tmp/pyrefly_base.txt 2>&1 || true
+
+      - name: Compute diff
+        run: |
+          diff /tmp/pyrefly_base.txt /tmp/pyrefly_pr.txt > pyrefly_diff.txt || true
+
+      - name: Save PR number
+        run: |
+          echo ${{ github.event.pull_request.number }} > pr_number.txt
+
+      - name: Upload pyrefly diff
+        uses: actions/upload-artifact@v4
+        with:
+          name: pyrefly_diff
+          path: |
+            pyrefly_diff.txt
+            pr_number.txt
+
+      - name: Comment PR with pyrefly diff
+        if: ${{ github.event.pull_request.head.repo.full_name == github.repository }}
+        uses: actions/github-script@v8
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          script: |
+            const fs = require('fs');
+            let diff = fs.readFileSync('pyrefly_diff.txt', { encoding: 'utf8' });
+            const prNumber = context.payload.pull_request.number;
+
+            const MAX_CHARS = 65000;
+            if (diff.length > MAX_CHARS) {
+              diff = diff.slice(0, MAX_CHARS);
+              diff = diff.slice(0, diff.lastIndexOf('\n'));
+              diff += '\n\n... (truncated) ...';
+            }
+
+            const body = diff.trim()
+              ? `### Pyrefly Diff (base → PR)\n\`\`\`diff\n${diff}\n\`\`\``
+              : '### Pyrefly Diff\nNo changes detected.';
+
+            await github.rest.issues.createComment({
+              issue_number: prNumber,
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              body,
+            });

api/extensions/storage/aws_s3_storage.py

@@ -83,5 +83,5 @@ class AwsS3Storage(BaseStorage):
         except:
             return False
 
-    def delete(self, filename):
+    def delete(self, filename: str):
         self.client.delete_object(Bucket=self.bucket_name, Key=filename)

api/extensions/storage/azure_blob_storage.py

@@ -75,7 +75,7 @@ class AzureBlobStorage(BaseStorage):
         blob = client.get_blob_client(container=self.bucket_name, blob=filename)
         return blob.exists()
 
-    def delete(self, filename):
+    def delete(self, filename: str):
         if not self.bucket_name:
             return
 

api/extensions/storage/baidu_obs_storage.py

@@ -53,5 +53,5 @@ class BaiduObsStorage(BaseStorage):
             return False
         return True
 
-    def delete(self, filename):
+    def delete(self, filename: str):
         self.client.delete_object(bucket_name=self.bucket_name, key=filename)

api/extensions/storage/base_storage.py

@@ -28,7 +28,7 @@ class BaseStorage(ABC):
         raise NotImplementedError
 
     @abstractmethod
-    def delete(self, filename):
+    def delete(self, filename: str):
         raise NotImplementedError
 
     def scan(self, path, files=True, directories=False) -> list[str]:

api/extensions/storage/google_cloud_storage.py

@@ -61,6 +61,6 @@ class GoogleCloudStorage(BaseStorage):
         blob = bucket.blob(filename)
         return blob.exists()
 
-    def delete(self, filename):
+    def delete(self, filename: str):
         bucket = self.client.get_bucket(self.bucket_name)
         bucket.delete_blob(filename)

api/extensions/storage/huawei_obs_storage.py

@@ -41,7 +41,7 @@ class HuaweiObsStorage(BaseStorage):
             return False
         return True
 
-    def delete(self, filename):
+    def delete(self, filename: str):
         self.client.deleteObject(bucketName=self.bucket_name, objectKey=filename)
 
     def _get_meta(self, filename):

api/extensions/storage/oracle_oci_storage.py

@@ -55,5 +55,5 @@ class OracleOCIStorage(BaseStorage):
         except:
             return False
 
-    def delete(self, filename):
+    def delete(self, filename: str):
         self.client.delete_object(Bucket=self.bucket_name, Key=filename)

api/extensions/storage/supabase_storage.py

@@ -51,7 +51,7 @@ class SupabaseStorage(BaseStorage):
             return True
         return False
 
-    def delete(self, filename):
+    def delete(self, filename: str):
         self.client.storage.from_(self.bucket_name).remove([filename])
 
     def bucket_exists(self):

api/extensions/storage/tencent_cos_storage.py

@@ -47,5 +47,5 @@ class TencentCosStorage(BaseStorage):
     def exists(self, filename):
         return self.client.object_exists(Bucket=self.bucket_name, Key=filename)
 
-    def delete(self, filename):
+    def delete(self, filename: str):
         self.client.delete_object(Bucket=self.bucket_name, Key=filename)

api/extensions/storage/volcengine_tos_storage.py

@@ -60,7 +60,7 @@ class VolcengineTosStorage(BaseStorage):
             return False
         return True
 
-    def delete(self, filename):
+    def delete(self, filename: str):
         if not self.bucket_name:
             return
         self.client.delete_object(bucket=self.bucket_name, key=filename)

api/migrations/versions/2026_02_26_1336-e288952f2994_add_partial_indexes_on_conversations_.py

@@ -1,37 +0,0 @@
-"""add partial indexes on conversations for app_id with created_at and updated_at
-
-Revision ID: e288952f2994
-Revises: fce013ca180e
-Create Date: 2026-02-26 13:36:45.928922
-
-"""
-from alembic import op
-import sqlalchemy as sa
-
-# revision identifiers, used by Alembic.
-revision = 'e288952f2994'
-down_revision = 'fce013ca180e'
-branch_labels = None
-depends_on = None
-
-
-def upgrade():
-    with op.batch_alter_table('conversations', schema=None) as batch_op:
-        batch_op.create_index(
-            'conversation_app_created_at_idx',
-            ['app_id', sa.literal_column('created_at DESC')],
-            unique=False,
-            postgresql_where=sa.text('is_deleted = false'),
-        )
-        batch_op.create_index(
-            'conversation_app_updated_at_idx',
-            ['app_id', sa.literal_column('updated_at DESC')],
-            unique=False,
-            postgresql_where=sa.text('is_deleted = false'),
-        )
-
-
-def downgrade():
-    with op.batch_alter_table('conversations', schema=None) as batch_op:
-        batch_op.drop_index('conversation_app_updated_at_idx', postgresql_where=sa.text('is_deleted = false'))
-        batch_op.drop_index('conversation_app_created_at_idx', postgresql_where=sa.text('is_deleted = false'))

api/models/model.py

@@ -711,18 +711,6 @@ class Conversation(Base):
     __table_args__ = (
         sa.PrimaryKeyConstraint("id", name="conversation_pkey"),
         sa.Index("conversation_app_from_user_idx", "app_id", "from_source", "from_end_user_id"),
-        sa.Index(
-            "conversation_app_created_at_idx",
-            "app_id",
-            sa.text("created_at DESC"),
-            postgresql_where=sa.text("is_deleted = false"),
-        ),
-        sa.Index(
-            "conversation_app_updated_at_idx",
-            "app_id",
-            sa.text("updated_at DESC"),
-            postgresql_where=sa.text("is_deleted = false"),
-        ),
     )
 
     id: Mapped[str] = mapped_column(StringUUID, default=lambda: str(uuid4()))

api/pyproject.toml

@@ -176,6 +176,7 @@ dev = [
     "sseclient-py>=1.8.0",
     "pytest-timeout>=2.4.0",
     "pytest-xdist>=3.8.0",
+    "pyrefly>=0.54.0",
 ]
 
 ############################################################

api/tests/test_containers_integration_tests/tasks/test_deal_dataset_vector_index_task.py

@@ -50,8 +50,26 @@ class TestDealDatasetVectorIndexTask:
             mock_factory.return_value = mock_instance
             yield mock_factory
 
+    @pytest.fixture
+    def account_and_tenant(self, db_session_with_containers, mock_external_service_dependencies):
+        """Create an account with an owner tenant for testing.
+
+        Returns a tuple of (account, tenant) where tenant is guaranteed to be non-None.
+        """
+        fake = Faker()
+        account = AccountService.create_account(
+            email=fake.email(),
+            name=fake.name(),
+            interface_language="en-US",
+            password=fake.password(length=12),
+        )
+        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
+        tenant = account.current_tenant
+        assert tenant is not None
+        return account, tenant
+
     def test_deal_dataset_vector_index_task_remove_action_success(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test successful removal of dataset vector index.
@@ -63,16 +81,7 @@ class TestDealDatasetVectorIndexTask:
         4. Completes without errors
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset
         dataset = Dataset(
@@ -118,7 +127,7 @@ class TestDealDatasetVectorIndexTask:
         assert mock_processor.clean.call_count >= 0  # For now, just check it doesn't fail
 
     def test_deal_dataset_vector_index_task_add_action_success(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test successful addition of dataset vector index.
@@ -132,16 +141,7 @@ class TestDealDatasetVectorIndexTask:
         6. Updates document status to completed
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset
         dataset = Dataset(
@@ -227,7 +227,7 @@ class TestDealDatasetVectorIndexTask:
         mock_processor.load.assert_called_once()
 
     def test_deal_dataset_vector_index_task_update_action_success(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test successful update of dataset vector index.
@@ -242,16 +242,7 @@ class TestDealDatasetVectorIndexTask:
         7. Updates document status to completed
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset with parent-child index
         dataset = Dataset(
@@ -338,7 +329,7 @@ class TestDealDatasetVectorIndexTask:
         mock_processor.load.assert_called_once()
 
     def test_deal_dataset_vector_index_task_dataset_not_found_error(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test task behavior when dataset is not found.
@@ -358,7 +349,7 @@ class TestDealDatasetVectorIndexTask:
         mock_processor.load.assert_not_called()
 
     def test_deal_dataset_vector_index_task_add_action_no_documents(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test add action when no documents exist for the dataset.
@@ -367,16 +358,7 @@ class TestDealDatasetVectorIndexTask:
         a dataset exists but has no documents to process.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset without documents
         dataset = Dataset(
@@ -399,7 +381,7 @@ class TestDealDatasetVectorIndexTask:
         mock_processor.load.assert_not_called()
 
     def test_deal_dataset_vector_index_task_add_action_no_segments(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test add action when documents exist but have no segments.
@@ -408,16 +390,7 @@ class TestDealDatasetVectorIndexTask:
         documents exist but contain no segments to process.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset
         dataset = Dataset(
@@ -464,7 +437,7 @@ class TestDealDatasetVectorIndexTask:
         mock_processor.load.assert_not_called()
 
     def test_deal_dataset_vector_index_task_update_action_no_documents(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test update action when no documents exist for the dataset.
@@ -473,16 +446,7 @@ class TestDealDatasetVectorIndexTask:
         a dataset exists but has no documents to process during update.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset without documents
         dataset = Dataset(
@@ -506,7 +470,7 @@ class TestDealDatasetVectorIndexTask:
         mock_processor.load.assert_not_called()
 
     def test_deal_dataset_vector_index_task_add_action_with_exception_handling(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test add action with exception handling during processing.
@@ -515,16 +479,7 @@ class TestDealDatasetVectorIndexTask:
         during document processing and updates document status to error.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset
         dataset = Dataset(
@@ -611,7 +566,7 @@ class TestDealDatasetVectorIndexTask:
         assert "Test exception during indexing" in updated_document.error
 
     def test_deal_dataset_vector_index_task_with_custom_index_type(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test task behavior with custom index type (QA_INDEX).
@@ -620,16 +575,7 @@ class TestDealDatasetVectorIndexTask:
         and initializes the appropriate index processor.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset with custom index type
         dataset = Dataset(
@@ -696,7 +642,7 @@ class TestDealDatasetVectorIndexTask:
         mock_processor.load.assert_called_once()
 
     def test_deal_dataset_vector_index_task_with_default_index_type(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test task behavior with default index type (PARAGRAPH_INDEX).
@@ -705,16 +651,7 @@ class TestDealDatasetVectorIndexTask:
         when dataset.doc_form is None.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset without doc_form (should use default)
         dataset = Dataset(
@@ -781,7 +718,7 @@ class TestDealDatasetVectorIndexTask:
         mock_processor.load.assert_called_once()
 
     def test_deal_dataset_vector_index_task_multiple_documents_processing(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test task processing with multiple documents and segments.
@@ -790,16 +727,7 @@ class TestDealDatasetVectorIndexTask:
         and their segments in sequence.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset
         dataset = Dataset(
@@ -893,7 +821,7 @@ class TestDealDatasetVectorIndexTask:
         assert mock_processor.load.call_count == 3
 
     def test_deal_dataset_vector_index_task_document_status_transitions(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test document status transitions during task execution.
@@ -902,16 +830,7 @@ class TestDealDatasetVectorIndexTask:
         'completed' to 'indexing' and back to 'completed' during processing.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset
         dataset = Dataset(
@@ -999,7 +918,7 @@ class TestDealDatasetVectorIndexTask:
         assert updated_document.indexing_status == "completed"
 
     def test_deal_dataset_vector_index_task_with_disabled_documents(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test task behavior with disabled documents.
@@ -1008,16 +927,7 @@ class TestDealDatasetVectorIndexTask:
         during processing.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset
         dataset = Dataset(
@@ -1129,7 +1039,7 @@ class TestDealDatasetVectorIndexTask:
         mock_processor.load.assert_called_once()
 
     def test_deal_dataset_vector_index_task_with_archived_documents(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test task behavior with archived documents.
@@ -1138,16 +1048,7 @@ class TestDealDatasetVectorIndexTask:
         during processing.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset
         dataset = Dataset(
@@ -1259,7 +1160,7 @@ class TestDealDatasetVectorIndexTask:
         mock_processor.load.assert_called_once()
 
     def test_deal_dataset_vector_index_task_with_incomplete_documents(
-        self, db_session_with_containers, mock_index_processor_factory, mock_external_service_dependencies
+        self, db_session_with_containers, mock_index_processor_factory, account_and_tenant
     ):
         """
         Test task behavior with documents that have incomplete indexing status.
@@ -1268,16 +1169,7 @@ class TestDealDatasetVectorIndexTask:
         incomplete indexing status during processing.
         """
         fake = Faker()
-
-        # Create test data
-        account = AccountService.create_account(
-            email=fake.email(),
-            name=fake.name(),
-            interface_language="en-US",
-            password=fake.password(length=12),
-        )
-        TenantService.create_owner_tenant_if_not_exist(account, name=fake.company())
-        tenant = account.current_tenant
+        account, tenant = account_and_tenant
 
         # Create dataset
         dataset = Dataset(

api/tests/unit_tests/tasks/test_clean_dataset_task.py

@@ -143,234 +143,8 @@ def mock_upload_file():
 # ============================================================================
 # Test Basic Cleanup
 # ============================================================================
-
-
-class TestBasicCleanup:
-    """Test cases for basic dataset cleanup functionality."""
-
-    def test_clean_dataset_task_empty_dataset(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test cleanup of an empty dataset with no documents or segments.
-
-        Scenario:
-        - Dataset has no documents or segments
-        - Should still clean vector database and delete related records
-
-        Expected behavior:
-        - IndexProcessorFactory is called to clean vector database
-        - No storage deletions occur
-        - Related records (DatasetProcessRule, etc.) are deleted
-        - Session is committed and closed
-        """
-        # Arrange
-        mock_db_session.session.scalars.return_value.all.return_value = []
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert
-        mock_index_processor_factory["factory"].assert_called_once_with("paragraph_index")
-        mock_index_processor_factory["processor"].clean.assert_called_once()
-        mock_storage.delete.assert_not_called()
-        mock_db_session.session.commit.assert_called_once()
-        mock_db_session.session.close.assert_called_once()
-
-    def test_clean_dataset_task_with_documents_and_segments(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-        mock_document,
-        mock_segment,
-    ):
-        """
-        Test cleanup of dataset with documents and segments.
-
-        Scenario:
-        - Dataset has one document and one segment
-        - No image files in segment content
-
-        Expected behavior:
-        - Documents and segments are deleted
-        - Vector database is cleaned
-        - Session is committed
-        """
-        # Arrange
-        mock_db_session.session.scalars.return_value.all.side_effect = [
-            [mock_document],  # documents
-            [mock_segment],  # segments
-        ]
-        mock_get_image_upload_file_ids.return_value = []
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert
-        mock_db_session.session.delete.assert_any_call(mock_document)
-        # Segments are deleted in batch; verify a DELETE on document_segments was issued
-        execute_sqls = [" ".join(str(c[0][0]).split()) for c in mock_db_session.session.execute.call_args_list]
-        assert any("DELETE FROM document_segments" in sql for sql in execute_sqls)
-        mock_db_session.session.commit.assert_called_once()
-
-    def test_clean_dataset_task_deletes_related_records(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test that all related records are deleted.
-
-        Expected behavior:
-        - DatasetProcessRule records are deleted
-        - DatasetQuery records are deleted
-        - AppDatasetJoin records are deleted
-        - DatasetMetadata records are deleted
-        - DatasetMetadataBinding records are deleted
-        """
-        # Arrange
-        mock_query = mock_db_session.session.query.return_value
-        mock_query.where.return_value = mock_query
-        mock_query.delete.return_value = 1
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert - verify query.where.delete was called multiple times
-        # for different models (DatasetProcessRule, DatasetQuery, etc.)
-        assert mock_query.delete.call_count >= 5
-
-
-# ============================================================================
-# Test Doc Form Validation
-# ============================================================================
-
-
-class TestDocFormValidation:
-    """Test cases for doc_form validation and default fallback."""
-
-    @pytest.mark.parametrize(
-        "invalid_doc_form",
-        [
-            None,
-            "",
-            "   ",
-            "\t",
-            "\n",
-            "  \t\n  ",
-        ],
-    )
-    def test_clean_dataset_task_invalid_doc_form_uses_default(
-        self,
-        invalid_doc_form,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test that invalid doc_form values use default paragraph index type.
-
-        Scenario:
-        - doc_form is None, empty, or whitespace-only
-        - Should use default IndexStructureType.PARAGRAPH_INDEX
-
-        Expected behavior:
-        - Default index type is used for cleanup
-        - No errors are raised
-        - Cleanup proceeds normally
-        """
-        # Arrange - import to verify the default value
-        from core.rag.index_processor.constant.index_type import IndexStructureType
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form=invalid_doc_form,
-        )
-
-        # Assert - IndexProcessorFactory should be called with default type
-        mock_index_processor_factory["factory"].assert_called_once_with(IndexStructureType.PARAGRAPH_INDEX)
-        mock_index_processor_factory["processor"].clean.assert_called_once()
-
-    def test_clean_dataset_task_valid_doc_form_used_directly(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test that valid doc_form values are used directly.
-
-        Expected behavior:
-        - Provided doc_form is passed to IndexProcessorFactory
-        """
-        # Arrange
-        valid_doc_form = "qa_index"
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form=valid_doc_form,
-        )
-
-        # Assert
-        mock_index_processor_factory["factory"].assert_called_once_with(valid_doc_form)
-
-
+# Note: Basic cleanup behavior is now covered by testcontainers-based
+# integration tests; no unit tests remain in this section.
 # ============================================================================
 # Test Error Handling
 # ============================================================================
@@ -379,156 +153,6 @@ class TestDocFormValidation:
 class TestErrorHandling:
     """Test cases for error handling and recovery."""
 
-    def test_clean_dataset_task_vector_cleanup_failure_continues(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-        mock_document,
-        mock_segment,
-    ):
-        """
-        Test that document cleanup continues even if vector cleanup fails.
-
-        Scenario:
-        - IndexProcessor.clean() raises an exception
-        - Document and segment deletion should still proceed
-
-        Expected behavior:
-        - Exception is caught and logged
-        - Documents and segments are still deleted
-        - Session is committed
-        """
-        # Arrange
-        mock_db_session.session.scalars.return_value.all.side_effect = [
-            [mock_document],  # documents
-            [mock_segment],  # segments
-        ]
-        mock_index_processor_factory["processor"].clean.side_effect = Exception("Vector database error")
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert - documents and segments should still be deleted
-        mock_db_session.session.delete.assert_any_call(mock_document)
-        # Segments are deleted in batch; verify a DELETE on document_segments was issued
-        execute_sqls = [" ".join(str(c[0][0]).split()) for c in mock_db_session.session.execute.call_args_list]
-        assert any("DELETE FROM document_segments" in sql for sql in execute_sqls)
-        mock_db_session.session.commit.assert_called_once()
-
-    def test_clean_dataset_task_storage_delete_failure_continues(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test that cleanup continues even if storage deletion fails.
-
-        Scenario:
-        - Segment contains image file references
-        - Storage.delete() raises an exception
-        - Cleanup should continue
-
-        Expected behavior:
-        - Exception is caught and logged
-        - Image file record is still deleted from database
-        - Other cleanup operations proceed
-        """
-        # Arrange
-        # Need at least one document for segment processing to occur (code is in else block)
-        mock_document = MagicMock()
-        mock_document.id = str(uuid.uuid4())
-        mock_document.tenant_id = tenant_id
-        mock_document.data_source_type = "website"  # Non-upload type to avoid file deletion
-
-        mock_segment = MagicMock()
-        mock_segment.id = str(uuid.uuid4())
-        mock_segment.content = "Test content with image"
-
-        mock_upload_file = MagicMock()
-        mock_upload_file.id = str(uuid.uuid4())
-        mock_upload_file.key = "images/test-image.jpg"
-
-        image_file_id = mock_upload_file.id
-
-        mock_db_session.session.scalars.return_value.all.side_effect = [
-            [mock_document],  # documents - need at least one for segment processing
-            [mock_segment],  # segments
-        ]
-        mock_get_image_upload_file_ids.return_value = [image_file_id]
-        mock_db_session.session.query.return_value.where.return_value.all.return_value = [mock_upload_file]
-        mock_storage.delete.side_effect = Exception("Storage service unavailable")
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert - storage delete was attempted for image file
-        mock_storage.delete.assert_called_with(mock_upload_file.key)
-        # Upload files are deleted in batch; verify a DELETE on upload_files was issued
-        execute_sqls = [" ".join(str(c[0][0]).split()) for c in mock_db_session.session.execute.call_args_list]
-        assert any("DELETE FROM upload_files" in sql for sql in execute_sqls)
-
-    def test_clean_dataset_task_database_error_rollback(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test that database session is rolled back on error.
-
-        Scenario:
-        - Database operation raises an exception
-        - Session should be rolled back to prevent dirty state
-
-        Expected behavior:
-        - Session.rollback() is called
-        - Session.close() is called in finally block
-        """
-        # Arrange
-        mock_db_session.session.commit.side_effect = Exception("Database commit failed")
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert
-        mock_db_session.session.rollback.assert_called_once()
-        mock_db_session.session.close.assert_called_once()
-
     def test_clean_dataset_task_rollback_failure_still_closes_session(
         self,
         dataset_id,
@@ -754,296 +378,6 @@ class TestSegmentAttachmentCleanup:
         assert any("DELETE FROM segment_attachment_bindings" in sql for sql in execute_sqls)
 
 
-# ============================================================================
-# Test Upload File Cleanup
-# ============================================================================
-
-
-class TestUploadFileCleanup:
-    """Test cases for upload file cleanup."""
-
-    def test_clean_dataset_task_deletes_document_upload_files(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test that document upload files are deleted.
-
-        Scenario:
-        - Document has data_source_type = "upload_file"
-        - data_source_info contains upload_file_id
-
-        Expected behavior:
-        - Upload file is deleted from storage
-        - Upload file record is deleted from database
-        """
-        # Arrange
-        mock_document = MagicMock()
-        mock_document.id = str(uuid.uuid4())
-        mock_document.tenant_id = tenant_id
-        mock_document.data_source_type = "upload_file"
-        mock_document.data_source_info = '{"upload_file_id": "test-file-id"}'
-        mock_document.data_source_info_dict = {"upload_file_id": "test-file-id"}
-
-        mock_upload_file = MagicMock()
-        mock_upload_file.id = "test-file-id"
-        mock_upload_file.key = "uploads/test-file.txt"
-
-        mock_db_session.session.scalars.return_value.all.side_effect = [
-            [mock_document],  # documents
-            [],  # segments
-        ]
-        mock_db_session.session.query.return_value.where.return_value.all.return_value = [mock_upload_file]
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert
-        mock_storage.delete.assert_called_with(mock_upload_file.key)
-        # Upload files are deleted in batch; verify a DELETE on upload_files was issued
-        execute_sqls = [" ".join(str(c[0][0]).split()) for c in mock_db_session.session.execute.call_args_list]
-        assert any("DELETE FROM upload_files" in sql for sql in execute_sqls)
-
-    def test_clean_dataset_task_handles_missing_upload_file(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test that missing upload files are handled gracefully.
-
-        Scenario:
-        - Document references an upload_file_id that doesn't exist
-
-        Expected behavior:
-        - No error is raised
-        - Cleanup continues normally
-        """
-        # Arrange
-        mock_document = MagicMock()
-        mock_document.id = str(uuid.uuid4())
-        mock_document.tenant_id = tenant_id
-        mock_document.data_source_type = "upload_file"
-        mock_document.data_source_info = '{"upload_file_id": "nonexistent-file"}'
-        mock_document.data_source_info_dict = {"upload_file_id": "nonexistent-file"}
-
-        mock_db_session.session.scalars.return_value.all.side_effect = [
-            [mock_document],  # documents
-            [],  # segments
-        ]
-        mock_db_session.session.query.return_value.where.return_value.all.return_value = []
-
-        # Act - should not raise exception
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert
-        mock_storage.delete.assert_not_called()
-        mock_db_session.session.commit.assert_called_once()
-
-    def test_clean_dataset_task_handles_non_upload_file_data_source(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test that non-upload_file data sources are skipped.
-
-        Scenario:
-        - Document has data_source_type = "website"
-
-        Expected behavior:
-        - No file deletion is attempted
-        """
-        # Arrange
-        mock_document = MagicMock()
-        mock_document.id = str(uuid.uuid4())
-        mock_document.tenant_id = tenant_id
-        mock_document.data_source_type = "website"
-        mock_document.data_source_info = None
-
-        mock_db_session.session.scalars.return_value.all.side_effect = [
-            [mock_document],  # documents
-            [],  # segments
-        ]
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert - storage delete should not be called for document files
-        # (only for image files in segments, which are empty here)
-        mock_storage.delete.assert_not_called()
-
-
-# ============================================================================
-# Test Image File Cleanup
-# ============================================================================
-
-
-class TestImageFileCleanup:
-    """Test cases for image file cleanup in segments."""
-
-    def test_clean_dataset_task_deletes_image_files_in_segments(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test that image files referenced in segment content are deleted.
-
-        Scenario:
-        - Segment content contains image file references
-        - get_image_upload_file_ids returns file IDs
-
-        Expected behavior:
-        - Each image file is deleted from storage
-        - Each image file record is deleted from database
-        """
-        # Arrange
-        # Need at least one document for segment processing to occur (code is in else block)
-        mock_document = MagicMock()
-        mock_document.id = str(uuid.uuid4())
-        mock_document.tenant_id = tenant_id
-        mock_document.data_source_type = "website"  # Non-upload type
-
-        mock_segment = MagicMock()
-        mock_segment.id = str(uuid.uuid4())
-        mock_segment.content = '<img src="file://image-1"> <img src="file://image-2">'
-
-        image_file_ids = ["image-1", "image-2"]
-        mock_get_image_upload_file_ids.return_value = image_file_ids
-
-        mock_image_files = []
-        for file_id in image_file_ids:
-            mock_file = MagicMock()
-            mock_file.id = file_id
-            mock_file.key = f"images/{file_id}.jpg"
-            mock_image_files.append(mock_file)
-
-        mock_db_session.session.scalars.return_value.all.side_effect = [
-            [mock_document],  # documents - need at least one for segment processing
-            [mock_segment],  # segments
-        ]
-
-        # Setup a mock query chain that returns files in batch (align with .in_().all())
-        mock_query = MagicMock()
-        mock_where = MagicMock()
-        mock_query.where.return_value = mock_where
-        mock_where.all.return_value = mock_image_files
-        mock_db_session.session.query.return_value = mock_query
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert - each expected image key was deleted at least once
-        calls = [c.args[0] for c in mock_storage.delete.call_args_list]
-        assert "images/image-1.jpg" in calls
-        assert "images/image-2.jpg" in calls
-
-    def test_clean_dataset_task_handles_missing_image_file(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test that missing image files are handled gracefully.
-
-        Scenario:
-        - Segment references image file ID that doesn't exist in database
-
-        Expected behavior:
-        - No error is raised
-        - Cleanup continues
-        """
-        # Arrange
-        # Need at least one document for segment processing to occur (code is in else block)
-        mock_document = MagicMock()
-        mock_document.id = str(uuid.uuid4())
-        mock_document.tenant_id = tenant_id
-        mock_document.data_source_type = "website"  # Non-upload type
-
-        mock_segment = MagicMock()
-        mock_segment.id = str(uuid.uuid4())
-        mock_segment.content = '<img src="file://nonexistent-image">'
-
-        mock_get_image_upload_file_ids.return_value = ["nonexistent-image"]
-
-        mock_db_session.session.scalars.return_value.all.side_effect = [
-            [mock_document],  # documents - need at least one for segment processing
-            [mock_segment],  # segments
-        ]
-
-        # Image file not found
-        mock_db_session.session.query.return_value.where.return_value.all.return_value = []
-
-        # Act - should not raise exception
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert
-        mock_storage.delete.assert_not_called()
-        mock_db_session.session.commit.assert_called_once()
-
-
 # ============================================================================
 # Test Edge Cases
 # ============================================================================
@@ -1052,114 +386,6 @@ class TestImageFileCleanup:
 class TestEdgeCases:
     """Test edge cases and boundary conditions."""
 
-    def test_clean_dataset_task_multiple_documents_and_segments(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test cleanup of multiple documents and segments.
-
-        Scenario:
-        - Dataset has 5 documents and 10 segments
-
-        Expected behavior:
-        - All documents and segments are deleted
-        """
-        # Arrange
-        mock_documents = []
-        for i in range(5):
-            doc = MagicMock()
-            doc.id = str(uuid.uuid4())
-            doc.tenant_id = tenant_id
-            doc.data_source_type = "website"  # Non-upload type
-            mock_documents.append(doc)
-
-        mock_segments = []
-        for i in range(10):
-            seg = MagicMock()
-            seg.id = str(uuid.uuid4())
-            seg.content = f"Segment content {i}"
-            mock_segments.append(seg)
-
-        mock_db_session.session.scalars.return_value.all.side_effect = [
-            mock_documents,
-            mock_segments,
-        ]
-        mock_get_image_upload_file_ids.return_value = []
-
-        # Act
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert - all documents and segments should be deleted (documents per-entity, segments in batch)
-        delete_calls = mock_db_session.session.delete.call_args_list
-        deleted_items = [call[0][0] for call in delete_calls]
-
-        for doc in mock_documents:
-            assert doc in deleted_items
-        # Verify a batch DELETE on document_segments occurred
-        execute_sqls = [" ".join(str(c[0][0]).split()) for c in mock_db_session.session.execute.call_args_list]
-        assert any("DELETE FROM document_segments" in sql for sql in execute_sqls)
-
-    def test_clean_dataset_task_document_with_empty_data_source_info(
-        self,
-        dataset_id,
-        tenant_id,
-        collection_binding_id,
-        mock_db_session,
-        mock_storage,
-        mock_index_processor_factory,
-        mock_get_image_upload_file_ids,
-    ):
-        """
-        Test handling of document with empty data_source_info.
-
-        Scenario:
-        - Document has data_source_type = "upload_file"
-        - data_source_info is None or empty
-
-        Expected behavior:
-        - No error is raised
-        - File deletion is skipped
-        """
-        # Arrange
-        mock_document = MagicMock()
-        mock_document.id = str(uuid.uuid4())
-        mock_document.tenant_id = tenant_id
-        mock_document.data_source_type = "upload_file"
-        mock_document.data_source_info = None
-
-        mock_db_session.session.scalars.return_value.all.side_effect = [
-            [mock_document],  # documents
-            [],  # segments
-        ]
-
-        # Act - should not raise exception
-        clean_dataset_task(
-            dataset_id=dataset_id,
-            tenant_id=tenant_id,
-            indexing_technique="high_quality",
-            index_struct='{"type": "paragraph"}',
-            collection_binding_id=collection_binding_id,
-            doc_form="paragraph_index",
-        )
-
-        # Assert
-        mock_storage.delete.assert_not_called()
-        mock_db_session.session.commit.assert_called_once()
-
     def test_clean_dataset_task_session_always_closed(
         self,
         dataset_id,

api/uv.lock

@@ -1471,6 +1471,7 @@ dev = [
     { name = "lxml-stubs" },
     { name = "mypy" },
     { name = "pandas-stubs" },
+    { name = "pyrefly" },
     { name = "pytest" },
     { name = "pytest-benchmark" },
     { name = "pytest-cov" },
@@ -1671,6 +1672,7 @@ dev = [
     { name = "lxml-stubs", specifier = "~=0.5.1" },
     { name = "mypy", specifier = "~=1.17.1" },
     { name = "pandas-stubs", specifier = "~=2.2.3" },
+    { name = "pyrefly", specifier = ">=0.54.0" },
     { name = "pytest", specifier = "~=8.3.2" },
     { name = "pytest-benchmark", specifier = "~=4.0.0" },
     { name = "pytest-cov", specifier = "~=4.1.0" },
@@ -5107,6 +5109,22 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/5a/dc/491b7661614ab97483abf2056be1deee4dc2490ecbf7bff9ab5cdbac86e1/pyreadline3-3.5.4-py3-none-any.whl", hash = "sha256:eaf8e6cc3c49bcccf145fc6067ba8643d1df34d604a1ec0eccbf7a18e6d3fae6", size = 83178, upload-time = "2024-09-19T02:40:08.598Z" },
 ]
 
+[[package]]
+name = "pyrefly"
+version = "0.54.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/81/44/c10b16a302fda90d0af1328f880b232761b510eab546616a7be2fdf35a57/pyrefly-0.54.0.tar.gz", hash = "sha256:c6663be64d492f0d2f2a411ada9f28a6792163d34133639378b7f3dd9a8dca94", size = 5098893, upload-time = "2026-02-23T15:44:35.111Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/5f/99/8fdcdb4e55f0227fdd9f6abce36b619bab1ecb0662b83b66adc8cba3c788/pyrefly-0.54.0-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:58a3f092b6dc25ef79b2dc6c69a40f36784ca157c312bfc0baea463926a9db6d", size = 12223973, upload-time = "2026-02-23T15:44:14.278Z" },
+    { url = "https://files.pythonhosted.org/packages/90/35/c2aaf87a76003ad27b286594d2e5178f811eaa15bfe3d98dba2b47d56dd1/pyrefly-0.54.0-py3-none-macosx_11_0_arm64.whl", hash = "sha256:615081414106dd95873bc39c3a4bed68754c6cc24a8177ac51d22f88f88d3eb3", size = 11785585, upload-time = "2026-02-23T15:44:17.468Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/4a/ced02691ed67e5a897714979196f08ad279ec7ec7f63c45e00a75a7f3c0e/pyrefly-0.54.0-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0cbcaf20f5fe585079079a95205c1f3cd4542d17228cdf1df560288880623b70", size = 33381977, upload-time = "2026-02-23T15:44:19.736Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/ce/72a117ed437c8f6950862181014b41e36f3c3997580e29b772b71e78d587/pyrefly-0.54.0-py3-none-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:66d5da116c0d34acfbd66663addd3ca8aa78a636f6692a66e078126d3620a883", size = 35962821, upload-time = "2026-02-23T15:44:22.357Z" },
+    { url = "https://files.pythonhosted.org/packages/85/de/89013f5ae0a35d2b6b01274a92a35ee91431ea001050edf0a16748d39875/pyrefly-0.54.0-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6ef3ac27f1a4baaf67aead64287d3163350844794aca6315ad1a9650b16ec26a", size = 38496689, upload-time = "2026-02-23T15:44:25.236Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/9a/33b097c7bf498b924742dca32dd5d9c6a3fa6c2b52b63a58eb9e1980ca89/pyrefly-0.54.0-py3-none-win32.whl", hash = "sha256:7d607d72200a8afbd2db10bfefb40160a7a5d709d207161c21649cedd5cfc09a", size = 11295268, upload-time = "2026-02-23T15:44:27.551Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/21/9263fd1144d2a3d7342b474f183f7785b3358a1565c864089b780110b933/pyrefly-0.54.0-py3-none-win_amd64.whl", hash = "sha256:fd416f04f89309385696f685bd5c9141011f18c8072f84d31ca20c748546e791", size = 12081810, upload-time = "2026-02-23T15:44:29.461Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/5b/fad062a196c064cbc8564de5b2f4d3cb6315f852e3b31e8a1ce74c69a1ea/pyrefly-0.54.0-py3-none-win_arm64.whl", hash = "sha256:f06ab371356c7b1925e0bffe193b738797e71e5dbbff7fb5a13f90ee7521211d", size = 11564930, upload-time = "2026-02-23T15:44:33.053Z" },
+]
+
 [[package]]
 name = "pytest"
 version = "8.3.5"

web/app/components/develop/template/template.en.mdx

@@ -1,5 +1,5 @@
 import { CodeGroup } from '../code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '../md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
 
 # Completion App API
 

web/app/components/develop/template/template.ja.mdx

@@ -1,5 +1,5 @@
 import { CodeGroup } from '../code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '../md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
 
 # Completion アプリ API
 

web/app/components/develop/template/template_advanced_chat.en.mdx

@@ -1,5 +1,5 @@
 import { CodeGroup, Embed } from '../code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '../md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
 
 # Advanced Chat App API
 

web/app/components/develop/template/template_advanced_chat.ja.mdx

@@ -1,5 +1,5 @@
 import { CodeGroup } from '../code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '../md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
 
 # 高度なチャットアプリ API
 

web/app/components/develop/template/template_chat.en.mdx

@@ -1,5 +1,5 @@
 import { CodeGroup } from '../code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '../md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
 
 # Chat App API
 

web/app/components/develop/template/template_chat.ja.mdx

@@ -1,5 +1,5 @@
 import { CodeGroup } from '../code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '../md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
 
 # チャットアプリ API
 

web/app/components/develop/template/template_workflow.en.mdx

@@ -1,5 +1,5 @@
 import { CodeGroup } from '../code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '../md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
 
 # Workflow App API
 

web/app/components/develop/template/template_workflow.ja.mdx

@@ -1,5 +1,5 @@
 import { CodeGroup } from '../code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '../md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
 
 # ワークフローアプリ API
 

web/app/components/develop/template/template_workflow.zh.mdx

@@ -1,5 +1,5 @@
 import { CodeGroup } from '../code.tsx'
-import { Row, Col, Properties, Property, Heading, SubProperty, Paragraph } from '../md.tsx'
+import { Row, Col, Properties, Property, Heading, SubProperty } from '../md.tsx'
 
 # Workflow 应用 API
 

web/app/components/header/account-dropdown/index.spec.tsx

@@ -1,10 +1,9 @@
-import type { AppRouterInstance } from 'next/dist/shared/lib/app-router-context.shared-runtime'
 import type { AppContextValue } from '@/context/app-context'
 import type { ModalContextState } from '@/context/modal-context'
 import type { ProviderContextState } from '@/context/provider-context'
 import { QueryClient, QueryClientProvider } from '@tanstack/react-query'
 import { fireEvent, render, screen, waitFor } from '@testing-library/react'
-import { AppRouterContext } from 'next/dist/shared/lib/app-router-context.shared-runtime'
+import { useRouter } from 'next/navigation'
 import { Plan } from '@/app/components/billing/type'
 import { useAppContext } from '@/context/app-context'
 import { useGlobalPublicStore } from '@/context/global-public-context'
@@ -50,6 +49,14 @@ vi.mock('@/service/use-common', () => ({
   useLogout: vi.fn(),
 }))
 
+vi.mock('next/navigation', async (importOriginal) => {
+  const actual = await importOriginal<typeof import('next/navigation')>()
+  return {
+    ...actual,
+    useRouter: vi.fn(),
+  }
+})
+
 vi.mock('@/context/i18n', () => ({
   useDocLink: () => (path: string) => `https://docs.dify.ai${path}`,
 }))
@@ -119,15 +126,6 @@ describe('AccountDropdown', () => {
   const mockSetShowAccountSettingModal = vi.fn()
 
   const renderWithRouter = (ui: React.ReactElement) => {
-    const mockRouter = {
-      push: mockPush,
-      replace: vi.fn(),
-      prefetch: vi.fn(),
-      back: vi.fn(),
-      forward: vi.fn(),
-      refresh: vi.fn(),
-    } as unknown as AppRouterInstance
-
     const queryClient = new QueryClient({
       defaultOptions: {
         queries: {
@@ -138,9 +136,7 @@ describe('AccountDropdown', () => {
 
     return render(
       <QueryClientProvider client={queryClient}>
-        <AppRouterContext.Provider value={mockRouter}>
-          {ui}
-        </AppRouterContext.Provider>
+        {ui}
       </QueryClientProvider>,
     )
   }
@@ -166,6 +162,14 @@ describe('AccountDropdown', () => {
     vi.mocked(useLogout).mockReturnValue({
       mutateAsync: mockLogout,
     } as unknown as ReturnType<typeof useLogout>)
+    vi.mocked(useRouter).mockReturnValue({
+      push: mockPush,
+      replace: vi.fn(),
+      prefetch: vi.fn(),
+      back: vi.fn(),
+      forward: vi.fn(),
+      refresh: vi.fn(),
+    })
   })
 
   afterEach(() => {

web/app/components/workflow/utils/workflow.ts

@@ -184,5 +184,5 @@ export const changeNodesAndEdgesId = (nodes: Node[], edges: Edge[]) => {
 }
 
 export const hasErrorHandleNode = (nodeType?: BlockEnum) => {
-  return nodeType === BlockEnum.LLM || nodeType === BlockEnum.Tool || nodeType === BlockEnum.HttpRequest || nodeType === BlockEnum.Code
+  return nodeType === BlockEnum.LLM || nodeType === BlockEnum.Tool || nodeType === BlockEnum.HttpRequest || nodeType === BlockEnum.Code || nodeType === BlockEnum.Agent
 }

web/package.json

@@ -28,9 +28,12 @@
   "scripts": {
     "dev": "next dev",
     "dev:inspect": "next dev --inspect",
+    "dev:vinext": "vinext dev",
     "build": "next build",
     "build:docker": "next build && node scripts/optimize-standalone.js",
+    "build:vinext": "vinext build",
     "start": "node ./scripts/copy-and-start.mjs",
+    "start:vinext": "vinext start",
     "lint": "eslint --cache --concurrency=auto",
     "lint:ci": "eslint --cache --concurrency 2",
     "lint:fix": "pnpm lint --fix",
@@ -173,6 +176,7 @@
     "@iconify-json/ri": "1.2.9",
     "@mdx-js/loader": "3.1.1",
     "@mdx-js/react": "3.1.1",
+    "@mdx-js/rollup": "3.1.1",
     "@next/eslint-plugin-next": "16.1.6",
     "@next/mdx": "16.1.5",
     "@rgrove/parse-xml": "4.2.0",
@@ -211,6 +215,7 @@
     "@typescript-eslint/parser": "8.54.0",
     "@typescript/native-preview": "7.0.0-dev.20251209.1",
     "@vitejs/plugin-react": "5.1.2",
+    "@vitejs/plugin-rsc": "0.5.20",
     "@vitest/coverage-v8": "4.0.17",
     "autoprefixer": "10.4.21",
     "code-inspector-plugin": "1.3.6",
@@ -233,6 +238,7 @@
     "postcss": "8.5.6",
     "postcss-js": "5.0.3",
     "react-scan": "0.4.3",
+    "react-server-dom-webpack": "19.2.4",
     "sass": "1.93.2",
     "serwist": "9.5.4",
     "storybook": "10.2.0",
@@ -240,6 +246,7 @@
     "tsx": "4.21.0",
     "typescript": "5.9.3",
     "uglify-js": "3.19.3",
+    "vinext": "https://pkg.pr.new/hyoban/vinext@5be0ce0",
     "vite": "7.3.1",
     "vite-tsconfig-paths": "6.0.4",
     "vitest": "4.0.17",

web/vite.config.ts

@@ -1,16 +1,60 @@
+import type { Plugin } from 'vite'
 import path from 'node:path'
 import { fileURLToPath } from 'node:url'
+import mdx from '@mdx-js/rollup'
 import react from '@vitejs/plugin-react'
+import vinext from 'vinext'
 import { defineConfig } from 'vite'
 import tsconfigPaths from 'vite-tsconfig-paths'
 
 const __dirname = path.dirname(fileURLToPath(import.meta.url))
+const isCI = !!process.env.CI
 
-export default defineConfig({
-  plugins: [tsconfigPaths(), react()],
-  resolve: {
-    alias: {
-      '~@': __dirname,
+export default defineConfig(({ mode }) => {
+  return {
+    plugins: [
+      ...(mode === 'test'
+        ? [
+            react(),
+            {
+              // Stub .mdx files so components importing them can be unit-tested
+              name: 'mdx-stub',
+              enforce: 'pre',
+              transform(_, id) {
+                if (id.endsWith('.mdx'))
+                  return { code: 'export default () => null', map: null }
+              },
+            } as Plugin,
+          ]
+        : [
+            mdx(),
+            vinext(),
+          ]),
+      tsconfigPaths(),
+    ],
+    resolve: {
+      alias: {
+        '~@': __dirname,
+      },
     },
-  },
+
+    // vinext related config
+    optimizeDeps: {
+      exclude: ['nuqs'],
+    },
+    server: {
+      port: 3000,
+    },
+
+    // Vitest config
+    test: {
+      environment: 'jsdom',
+      globals: true,
+      setupFiles: ['./vitest.setup.ts'],
+      coverage: {
+        provider: 'v8',
+        reporter: isCI ? ['json', 'json-summary'] : ['text', 'json', 'json-summary'],
+      },
+    },
+  }
 })

web/vitest.config.ts

@@ -1,27 +0,0 @@
-import { defineConfig, mergeConfig } from 'vitest/config'
-import viteConfig from './vite.config'
-
-const isCI = !!process.env.CI
-
-export default mergeConfig(viteConfig, defineConfig({
-  plugins: [
-    {
-      // Stub .mdx files so components importing them can be unit-tested
-      name: 'mdx-stub',
-      enforce: 'pre',
-      transform(_, id) {
-        if (id.endsWith('.mdx'))
-          return { code: 'export default () => null', map: null }
-      },
-    },
-  ],
-  test: {
-    environment: 'jsdom',
-    globals: true,
-    setupFiles: ['./vitest.setup.ts'],
-    coverage: {
-      provider: 'v8',
-      reporter: isCI ? ['json', 'json-summary'] : ['text', 'json', 'json-summary'],
-    },
-  },
-}))