feat: introduce trigger functionality (#27644)

Signed-off-by: lyzno1 <yuanyouhuilyz@gmail.com>
Co-authored-by: Stream <Stream_2@qq.com>
Co-authored-by: lyzno1 <92089059+lyzno1@users.noreply.github.com>
Co-authored-by: zhsama <torvalds@linux.do>
Co-authored-by: Harry <xh001x@hotmail.com>
Co-authored-by: lyzno1 <yuanyouhuilyz@gmail.com>
Co-authored-by: yessenia <yessenia.contact@gmail.com>
Co-authored-by: hjlarry <hjlarry@163.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: WTW0313 <twwu@dify.ai>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

api/repositories/factory.py

@@ -5,7 +5,7 @@ This factory is specifically designed for DifyAPI repositories that handle
 service-layer operations with dependency injection patterns.
 """
 
-from sqlalchemy.orm import sessionmaker
+from sqlalchemy.orm import Session, sessionmaker
 
 from configs import dify_config
 from core.repositories import DifyCoreRepositoryFactory, RepositoryImportError
@@ -25,7 +25,7 @@ class DifyAPIRepositoryFactory(DifyCoreRepositoryFactory):
 
     @classmethod
     def create_api_workflow_node_execution_repository(
-        cls, session_maker: sessionmaker
+        cls, session_maker: sessionmaker[Session]
     ) -> DifyAPIWorkflowNodeExecutionRepository:
         """
         Create a DifyAPIWorkflowNodeExecutionRepository instance based on configuration.
@@ -55,7 +55,7 @@ class DifyAPIRepositoryFactory(DifyCoreRepositoryFactory):
             ) from e
 
     @classmethod
-    def create_api_workflow_run_repository(cls, session_maker: sessionmaker) -> APIWorkflowRunRepository:
+    def create_api_workflow_run_repository(cls, session_maker: sessionmaker[Session]) -> APIWorkflowRunRepository:
         """
         Create an APIWorkflowRunRepository instance based on configuration.
 

api/repositories/sqlalchemy_workflow_trigger_log_repository.py

@@ -0,0 +1,86 @@
+"""
+SQLAlchemy implementation of WorkflowTriggerLogRepository.
+"""
+
+from collections.abc import Sequence
+from datetime import UTC, datetime, timedelta
+
+from sqlalchemy import and_, select
+from sqlalchemy.orm import Session
+
+from models.enums import WorkflowTriggerStatus
+from models.trigger import WorkflowTriggerLog
+from repositories.workflow_trigger_log_repository import WorkflowTriggerLogRepository
+
+
+class SQLAlchemyWorkflowTriggerLogRepository(WorkflowTriggerLogRepository):
+    """
+    SQLAlchemy implementation of WorkflowTriggerLogRepository.
+
+    Optimized for large table operations with proper indexing and batch processing.
+    """
+
+    def __init__(self, session: Session):
+        self.session = session
+
+    def create(self, trigger_log: WorkflowTriggerLog) -> WorkflowTriggerLog:
+        """Create a new trigger log entry."""
+        self.session.add(trigger_log)
+        self.session.flush()
+        return trigger_log
+
+    def update(self, trigger_log: WorkflowTriggerLog) -> WorkflowTriggerLog:
+        """Update an existing trigger log entry."""
+        self.session.merge(trigger_log)
+        self.session.flush()
+        return trigger_log
+
+    def get_by_id(self, trigger_log_id: str, tenant_id: str | None = None) -> WorkflowTriggerLog | None:
+        """Get a trigger log by its ID."""
+        query = select(WorkflowTriggerLog).where(WorkflowTriggerLog.id == trigger_log_id)
+
+        if tenant_id:
+            query = query.where(WorkflowTriggerLog.tenant_id == tenant_id)
+
+        return self.session.scalar(query)
+
+    def get_failed_for_retry(
+        self, tenant_id: str, max_retry_count: int = 3, limit: int = 100
+    ) -> Sequence[WorkflowTriggerLog]:
+        """Get failed trigger logs eligible for retry."""
+        query = (
+            select(WorkflowTriggerLog)
+            .where(
+                and_(
+                    WorkflowTriggerLog.tenant_id == tenant_id,
+                    WorkflowTriggerLog.status.in_([WorkflowTriggerStatus.FAILED, WorkflowTriggerStatus.RATE_LIMITED]),
+                    WorkflowTriggerLog.retry_count < max_retry_count,
+                )
+            )
+            .order_by(WorkflowTriggerLog.created_at.asc())
+            .limit(limit)
+        )
+
+        return list(self.session.scalars(query).all())
+
+    def get_recent_logs(
+        self, tenant_id: str, app_id: str, hours: int = 24, limit: int = 100, offset: int = 0
+    ) -> Sequence[WorkflowTriggerLog]:
+        """Get recent trigger logs within specified hours."""
+        since = datetime.now(UTC) - timedelta(hours=hours)
+
+        query = (
+            select(WorkflowTriggerLog)
+            .where(
+                and_(
+                    WorkflowTriggerLog.tenant_id == tenant_id,
+                    WorkflowTriggerLog.app_id == app_id,
+                    WorkflowTriggerLog.created_at >= since,
+                )
+            )
+            .order_by(WorkflowTriggerLog.created_at.desc())
+            .limit(limit)
+            .offset(offset)
+        )
+
+        return list(self.session.scalars(query).all())

api/repositories/workflow_trigger_log_repository.py

@@ -0,0 +1,111 @@
+"""
+Repository protocol for WorkflowTriggerLog operations.
+
+This module provides a protocol interface for operations on WorkflowTriggerLog,
+designed to efficiently handle a potentially large volume of trigger logs with
+proper indexing and batch operations.
+"""
+
+from collections.abc import Sequence
+from enum import StrEnum
+from typing import Protocol
+
+from models.trigger import WorkflowTriggerLog
+
+
+class TriggerLogOrderBy(StrEnum):
+    """Fields available for ordering trigger logs"""
+
+    CREATED_AT = "created_at"
+    TRIGGERED_AT = "triggered_at"
+    FINISHED_AT = "finished_at"
+    STATUS = "status"
+
+
+class WorkflowTriggerLogRepository(Protocol):
+    """
+    Protocol for operations on WorkflowTriggerLog.
+
+    This repository provides efficient access patterns for the trigger log table,
+    which is expected to grow large over time. It includes:
+    - Batch operations for cleanup
+    - Efficient queries with proper indexing
+    - Pagination support
+    - Status-based filtering
+
+    Implementation notes:
+    - Leverage database indexes on (tenant_id, app_id), status, and created_at
+    - Use batch operations for deletions to avoid locking
+    - Support pagination for large result sets
+    """
+
+    def create(self, trigger_log: WorkflowTriggerLog) -> WorkflowTriggerLog:
+        """
+        Create a new trigger log entry.
+
+        Args:
+            trigger_log: The WorkflowTriggerLog instance to create
+
+        Returns:
+            The created WorkflowTriggerLog with generated ID
+        """
+        ...
+
+    def update(self, trigger_log: WorkflowTriggerLog) -> WorkflowTriggerLog:
+        """
+        Update an existing trigger log entry.
+
+        Args:
+            trigger_log: The WorkflowTriggerLog instance to update
+
+        Returns:
+            The updated WorkflowTriggerLog
+        """
+        ...
+
+    def get_by_id(self, trigger_log_id: str, tenant_id: str | None = None) -> WorkflowTriggerLog | None:
+        """
+        Get a trigger log by its ID.
+
+        Args:
+            trigger_log_id: The trigger log identifier
+            tenant_id: Optional tenant identifier for additional security
+
+        Returns:
+            The WorkflowTriggerLog if found, None otherwise
+        """
+        ...
+
+    def get_failed_for_retry(
+        self, tenant_id: str, max_retry_count: int = 3, limit: int = 100
+    ) -> Sequence[WorkflowTriggerLog]:
+        """
+        Get failed trigger logs that are eligible for retry.
+
+        Args:
+            tenant_id: The tenant identifier
+            max_retry_count: Maximum retry count to consider
+            limit: Maximum number of results
+
+        Returns:
+            A sequence of WorkflowTriggerLog instances eligible for retry
+        """
+        ...
+
+    def get_recent_logs(
+        self, tenant_id: str, app_id: str, hours: int = 24, limit: int = 100, offset: int = 0
+    ) -> Sequence[WorkflowTriggerLog]:
+        """
+        Get recent trigger logs within specified hours.
+
+        Args:
+            tenant_id: The tenant identifier
+            app_id: The application identifier
+            hours: Number of hours to look back
+            limit: Maximum number of results
+            offset: Number of results to skip
+
+        Returns:
+            A sequence of recent WorkflowTriggerLog instances
+        """
+        ...