feat: add a abstract layer for WorkflowNodeExcetion (#18026)

This commit is contained in:
-LAN-
2025-04-17 12:48:52 +09:00
committed by GitHub
parent 77fde04ef7
commit 6d9dd3109e
24 changed files with 807 additions and 251 deletions

View File

@@ -320,10 +320,9 @@ class AdvancedChatAppGenerateTaskPipeline:
session=session, workflow_run_id=self._workflow_run_id
)
workflow_node_execution = self._workflow_cycle_manager._handle_workflow_node_execution_retried(
session=session, workflow_run=workflow_run, event=event
workflow_run=workflow_run, event=event
)
node_retry_resp = self._workflow_cycle_manager._workflow_node_retry_to_stream_response(
session=session,
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
@@ -341,11 +340,10 @@ class AdvancedChatAppGenerateTaskPipeline:
session=session, workflow_run_id=self._workflow_run_id
)
workflow_node_execution = self._workflow_cycle_manager._handle_node_execution_start(
session=session, workflow_run=workflow_run, event=event
workflow_run=workflow_run, event=event
)
node_start_resp = self._workflow_cycle_manager._workflow_node_start_to_stream_response(
session=session,
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
@@ -363,11 +361,10 @@ class AdvancedChatAppGenerateTaskPipeline:
with Session(db.engine, expire_on_commit=False) as session:
workflow_node_execution = self._workflow_cycle_manager._handle_workflow_node_execution_success(
session=session, event=event
event=event
)
node_finish_resp = self._workflow_cycle_manager._workflow_node_finish_to_stream_response(
session=session,
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
@@ -383,18 +380,15 @@ class AdvancedChatAppGenerateTaskPipeline:
| QueueNodeInLoopFailedEvent
| QueueNodeExceptionEvent,
):
with Session(db.engine, expire_on_commit=False) as session:
workflow_node_execution = self._workflow_cycle_manager._handle_workflow_node_execution_failed(
session=session, event=event
)
workflow_node_execution = self._workflow_cycle_manager._handle_workflow_node_execution_failed(
event=event
)
node_finish_resp = self._workflow_cycle_manager._workflow_node_finish_to_stream_response(
session=session,
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
)
session.commit()
node_finish_resp = self._workflow_cycle_manager._workflow_node_finish_to_stream_response(
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
)
if node_finish_resp:
yield node_finish_resp

View File

@@ -279,10 +279,9 @@ class WorkflowAppGenerateTaskPipeline:
session=session, workflow_run_id=self._workflow_run_id
)
workflow_node_execution = self._workflow_cycle_manager._handle_workflow_node_execution_retried(
session=session, workflow_run=workflow_run, event=event
workflow_run=workflow_run, event=event
)
response = self._workflow_cycle_manager._workflow_node_retry_to_stream_response(
session=session,
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
@@ -300,10 +299,9 @@ class WorkflowAppGenerateTaskPipeline:
session=session, workflow_run_id=self._workflow_run_id
)
workflow_node_execution = self._workflow_cycle_manager._handle_node_execution_start(
session=session, workflow_run=workflow_run, event=event
workflow_run=workflow_run, event=event
)
node_start_response = self._workflow_cycle_manager._workflow_node_start_to_stream_response(
session=session,
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
@@ -313,17 +311,14 @@ class WorkflowAppGenerateTaskPipeline:
if node_start_response:
yield node_start_response
elif isinstance(event, QueueNodeSucceededEvent):
with Session(db.engine, expire_on_commit=False) as session:
workflow_node_execution = self._workflow_cycle_manager._handle_workflow_node_execution_success(
session=session, event=event
)
node_success_response = self._workflow_cycle_manager._workflow_node_finish_to_stream_response(
session=session,
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
)
session.commit()
workflow_node_execution = self._workflow_cycle_manager._handle_workflow_node_execution_success(
event=event
)
node_success_response = self._workflow_cycle_manager._workflow_node_finish_to_stream_response(
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
)
if node_success_response:
yield node_success_response
@@ -334,18 +329,14 @@ class WorkflowAppGenerateTaskPipeline:
| QueueNodeInLoopFailedEvent
| QueueNodeExceptionEvent,
):
with Session(db.engine, expire_on_commit=False) as session:
workflow_node_execution = self._workflow_cycle_manager._handle_workflow_node_execution_failed(
session=session,
event=event,
)
node_failed_response = self._workflow_cycle_manager._workflow_node_finish_to_stream_response(
session=session,
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
)
session.commit()
workflow_node_execution = self._workflow_cycle_manager._handle_workflow_node_execution_failed(
event=event,
)
node_failed_response = self._workflow_cycle_manager._workflow_node_finish_to_stream_response(
event=event,
task_id=self._application_generate_entity.task_id,
workflow_node_execution=workflow_node_execution,
)
if node_failed_response:
yield node_failed_response
@@ -627,6 +618,7 @@ class WorkflowAppGenerateTaskPipeline:
workflow_app_log.created_by = self._user_id
session.add(workflow_app_log)
session.commit()
def _text_chunk_to_stream_response(
self, text: str, from_variable_selector: Optional[list[str]] = None

View File

@@ -6,7 +6,7 @@ from typing import Any, Optional, Union, cast
from uuid import uuid4
from sqlalchemy import func, select
from sqlalchemy.orm import Session
from sqlalchemy.orm import Session, sessionmaker
from core.app.entities.app_invoke_entities import AdvancedChatAppGenerateEntity, InvokeFrom, WorkflowAppGenerateEntity
from core.app.entities.queue_entities import (
@@ -49,12 +49,14 @@ from core.file import FILE_MODEL_IDENTITY, File
from core.model_runtime.utils.encoders import jsonable_encoder
from core.ops.entities.trace_entity import TraceTaskName
from core.ops.ops_trace_manager import TraceQueueManager, TraceTask
from core.repository import RepositoryFactory
from core.tools.tool_manager import ToolManager
from core.workflow.entities.node_entities import NodeRunMetadataKey
from core.workflow.enums import SystemVariableKey
from core.workflow.nodes import NodeType
from core.workflow.nodes.tool.entities import ToolNodeData
from core.workflow.workflow_entry import WorkflowEntry
from extensions.ext_database import db
from models.account import Account
from models.enums import CreatedByRole, WorkflowRunTriggeredFrom
from models.model import EndUser
@@ -80,6 +82,21 @@ class WorkflowCycleManage:
self._application_generate_entity = application_generate_entity
self._workflow_system_variables = workflow_system_variables
# Initialize the session factory and repository
# We use the global db engine instead of the session passed to methods
# Disable expire_on_commit to avoid the need for merging objects
self._session_factory = sessionmaker(bind=db.engine, expire_on_commit=False)
self._workflow_node_execution_repository = RepositoryFactory.create_workflow_node_execution_repository(
params={
"tenant_id": self._application_generate_entity.app_config.tenant_id,
"app_id": self._application_generate_entity.app_config.app_id,
"session_factory": self._session_factory,
}
)
# We'll still keep the cache for backward compatibility and performance
# but use the repository for database operations
def _handle_workflow_run_start(
self,
*,
@@ -254,19 +271,15 @@ class WorkflowCycleManage:
workflow_run.finished_at = datetime.now(UTC).replace(tzinfo=None)
workflow_run.exceptions_count = exceptions_count
stmt = select(WorkflowNodeExecution.node_execution_id).where(
WorkflowNodeExecution.tenant_id == workflow_run.tenant_id,
WorkflowNodeExecution.app_id == workflow_run.app_id,
WorkflowNodeExecution.workflow_id == workflow_run.workflow_id,
WorkflowNodeExecution.triggered_from == WorkflowNodeExecutionTriggeredFrom.WORKFLOW_RUN.value,
WorkflowNodeExecution.workflow_run_id == workflow_run.id,
WorkflowNodeExecution.status == WorkflowNodeExecutionStatus.RUNNING.value,
# Use the instance repository to find running executions for a workflow run
running_workflow_node_executions = self._workflow_node_execution_repository.get_running_executions(
workflow_run_id=workflow_run.id
)
ids = session.scalars(stmt).all()
# Use self._get_workflow_node_execution here to make sure the cache is updated
running_workflow_node_executions = [
self._get_workflow_node_execution(session=session, node_execution_id=id) for id in ids if id
]
# Update the cache with the retrieved executions
for execution in running_workflow_node_executions:
if execution.node_execution_id:
self._workflow_node_executions[execution.node_execution_id] = execution
for workflow_node_execution in running_workflow_node_executions:
now = datetime.now(UTC).replace(tzinfo=None)
@@ -288,7 +301,7 @@ class WorkflowCycleManage:
return workflow_run
def _handle_node_execution_start(
self, *, session: Session, workflow_run: WorkflowRun, event: QueueNodeStartedEvent
self, *, workflow_run: WorkflowRun, event: QueueNodeStartedEvent
) -> WorkflowNodeExecution:
workflow_node_execution = WorkflowNodeExecution()
workflow_node_execution.id = str(uuid4())
@@ -315,17 +328,14 @@ class WorkflowCycleManage:
)
workflow_node_execution.created_at = datetime.now(UTC).replace(tzinfo=None)
session.add(workflow_node_execution)
# Use the instance repository to save the workflow node execution
self._workflow_node_execution_repository.save(workflow_node_execution)
self._workflow_node_executions[event.node_execution_id] = workflow_node_execution
return workflow_node_execution
def _handle_workflow_node_execution_success(
self, *, session: Session, event: QueueNodeSucceededEvent
) -> WorkflowNodeExecution:
workflow_node_execution = self._get_workflow_node_execution(
session=session, node_execution_id=event.node_execution_id
)
def _handle_workflow_node_execution_success(self, *, event: QueueNodeSucceededEvent) -> WorkflowNodeExecution:
workflow_node_execution = self._get_workflow_node_execution(node_execution_id=event.node_execution_id)
inputs = WorkflowEntry.handle_special_values(event.inputs)
process_data = WorkflowEntry.handle_special_values(event.process_data)
outputs = WorkflowEntry.handle_special_values(event.outputs)
@@ -344,13 +354,13 @@ class WorkflowCycleManage:
workflow_node_execution.finished_at = finished_at
workflow_node_execution.elapsed_time = elapsed_time
workflow_node_execution = session.merge(workflow_node_execution)
# Use the instance repository to update the workflow node execution
self._workflow_node_execution_repository.update(workflow_node_execution)
return workflow_node_execution
def _handle_workflow_node_execution_failed(
self,
*,
session: Session,
event: QueueNodeFailedEvent
| QueueNodeInIterationFailedEvent
| QueueNodeInLoopFailedEvent
@@ -361,9 +371,7 @@ class WorkflowCycleManage:
:param event: queue node failed event
:return:
"""
workflow_node_execution = self._get_workflow_node_execution(
session=session, node_execution_id=event.node_execution_id
)
workflow_node_execution = self._get_workflow_node_execution(node_execution_id=event.node_execution_id)
inputs = WorkflowEntry.handle_special_values(event.inputs)
process_data = WorkflowEntry.handle_special_values(event.process_data)
@@ -387,14 +395,14 @@ class WorkflowCycleManage:
workflow_node_execution.elapsed_time = elapsed_time
workflow_node_execution.execution_metadata = execution_metadata
workflow_node_execution = session.merge(workflow_node_execution)
return workflow_node_execution
def _handle_workflow_node_execution_retried(
self, *, session: Session, workflow_run: WorkflowRun, event: QueueNodeRetryEvent
self, *, workflow_run: WorkflowRun, event: QueueNodeRetryEvent
) -> WorkflowNodeExecution:
"""
Workflow node execution failed
:param workflow_run: workflow run
:param event: queue node failed event
:return:
"""
@@ -439,15 +447,12 @@ class WorkflowCycleManage:
workflow_node_execution.execution_metadata = execution_metadata
workflow_node_execution.index = event.node_run_index
session.add(workflow_node_execution)
# Use the instance repository to save the workflow node execution
self._workflow_node_execution_repository.save(workflow_node_execution)
self._workflow_node_executions[event.node_execution_id] = workflow_node_execution
return workflow_node_execution
#################################################
# to stream responses #
#################################################
def _workflow_start_to_stream_response(
self,
*,
@@ -455,7 +460,6 @@ class WorkflowCycleManage:
task_id: str,
workflow_run: WorkflowRun,
) -> WorkflowStartStreamResponse:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
return WorkflowStartStreamResponse(
task_id=task_id,
@@ -521,14 +525,10 @@ class WorkflowCycleManage:
def _workflow_node_start_to_stream_response(
self,
*,
session: Session,
event: QueueNodeStartedEvent,
task_id: str,
workflow_node_execution: WorkflowNodeExecution,
) -> Optional[NodeStartStreamResponse]:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
if workflow_node_execution.node_type in {NodeType.ITERATION.value, NodeType.LOOP.value}:
return None
if not workflow_node_execution.workflow_run_id:
@@ -571,7 +571,6 @@ class WorkflowCycleManage:
def _workflow_node_finish_to_stream_response(
self,
*,
session: Session,
event: QueueNodeSucceededEvent
| QueueNodeFailedEvent
| QueueNodeInIterationFailedEvent
@@ -580,8 +579,6 @@ class WorkflowCycleManage:
task_id: str,
workflow_node_execution: WorkflowNodeExecution,
) -> Optional[NodeFinishStreamResponse]:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
if workflow_node_execution.node_type in {NodeType.ITERATION.value, NodeType.LOOP.value}:
return None
if not workflow_node_execution.workflow_run_id:
@@ -621,13 +618,10 @@ class WorkflowCycleManage:
def _workflow_node_retry_to_stream_response(
self,
*,
session: Session,
event: QueueNodeRetryEvent,
task_id: str,
workflow_node_execution: WorkflowNodeExecution,
) -> Optional[Union[NodeRetryStreamResponse, NodeFinishStreamResponse]]:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
if workflow_node_execution.node_type in {NodeType.ITERATION.value, NodeType.LOOP.value}:
return None
if not workflow_node_execution.workflow_run_id:
@@ -668,7 +662,6 @@ class WorkflowCycleManage:
def _workflow_parallel_branch_start_to_stream_response(
self, *, session: Session, task_id: str, workflow_run: WorkflowRun, event: QueueParallelBranchRunStartedEvent
) -> ParallelBranchStartStreamResponse:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
return ParallelBranchStartStreamResponse(
task_id=task_id,
@@ -692,7 +685,6 @@ class WorkflowCycleManage:
workflow_run: WorkflowRun,
event: QueueParallelBranchRunSucceededEvent | QueueParallelBranchRunFailedEvent,
) -> ParallelBranchFinishedStreamResponse:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
return ParallelBranchFinishedStreamResponse(
task_id=task_id,
@@ -713,7 +705,6 @@ class WorkflowCycleManage:
def _workflow_iteration_start_to_stream_response(
self, *, session: Session, task_id: str, workflow_run: WorkflowRun, event: QueueIterationStartEvent
) -> IterationNodeStartStreamResponse:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
return IterationNodeStartStreamResponse(
task_id=task_id,
@@ -735,7 +726,6 @@ class WorkflowCycleManage:
def _workflow_iteration_next_to_stream_response(
self, *, session: Session, task_id: str, workflow_run: WorkflowRun, event: QueueIterationNextEvent
) -> IterationNodeNextStreamResponse:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
return IterationNodeNextStreamResponse(
task_id=task_id,
@@ -759,7 +749,6 @@ class WorkflowCycleManage:
def _workflow_iteration_completed_to_stream_response(
self, *, session: Session, task_id: str, workflow_run: WorkflowRun, event: QueueIterationCompletedEvent
) -> IterationNodeCompletedStreamResponse:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
return IterationNodeCompletedStreamResponse(
task_id=task_id,
@@ -790,7 +779,6 @@ class WorkflowCycleManage:
def _workflow_loop_start_to_stream_response(
self, *, session: Session, task_id: str, workflow_run: WorkflowRun, event: QueueLoopStartEvent
) -> LoopNodeStartStreamResponse:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
return LoopNodeStartStreamResponse(
task_id=task_id,
@@ -812,7 +800,6 @@ class WorkflowCycleManage:
def _workflow_loop_next_to_stream_response(
self, *, session: Session, task_id: str, workflow_run: WorkflowRun, event: QueueLoopNextEvent
) -> LoopNodeNextStreamResponse:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
return LoopNodeNextStreamResponse(
task_id=task_id,
@@ -836,7 +823,6 @@ class WorkflowCycleManage:
def _workflow_loop_completed_to_stream_response(
self, *, session: Session, task_id: str, workflow_run: WorkflowRun, event: QueueLoopCompletedEvent
) -> LoopNodeCompletedStreamResponse:
# receive session to make sure the workflow_run won't be expired, need a more elegant way to handle this
_ = session
return LoopNodeCompletedStreamResponse(
task_id=task_id,
@@ -934,11 +920,22 @@ class WorkflowCycleManage:
return workflow_run
def _get_workflow_node_execution(self, session: Session, node_execution_id: str) -> WorkflowNodeExecution:
if node_execution_id not in self._workflow_node_executions:
def _get_workflow_node_execution(self, node_execution_id: str) -> WorkflowNodeExecution:
# First check the cache for performance
if node_execution_id in self._workflow_node_executions:
cached_execution = self._workflow_node_executions[node_execution_id]
# No need to merge with session since expire_on_commit=False
return cached_execution
# If not in cache, use the instance repository to get by node_execution_id
execution = self._workflow_node_execution_repository.get_by_node_execution_id(node_execution_id)
if not execution:
raise ValueError(f"Workflow node execution not found: {node_execution_id}")
cached_workflow_node_execution = self._workflow_node_executions[node_execution_id]
return session.merge(cached_workflow_node_execution)
# Update cache
self._workflow_node_executions[node_execution_id] = execution
return execution
def _handle_agent_log(self, task_id: str, event: QueueAgentLogEvent) -> AgentLogStreamResponse:
"""

View File

@@ -5,6 +5,7 @@ from datetime import datetime, timedelta
from typing import Optional
from langfuse import Langfuse # type: ignore
from sqlalchemy.orm import sessionmaker
from core.ops.base_trace_instance import BaseTraceInstance
from core.ops.entities.config_entity import LangfuseConfig
@@ -28,9 +29,9 @@ from core.ops.langfuse_trace.entities.langfuse_trace_entity import (
UnitEnum,
)
from core.ops.utils import filter_none_values
from core.repository.repository_factory import RepositoryFactory
from extensions.ext_database import db
from models.model import EndUser
from models.workflow import WorkflowNodeExecution
logger = logging.getLogger(__name__)
@@ -110,36 +111,18 @@ class LangFuseDataTrace(BaseTraceInstance):
)
self.add_trace(langfuse_trace_data=trace_data)
# through workflow_run_id get all_nodes_execution
workflow_nodes_execution_id_records = (
db.session.query(WorkflowNodeExecution.id)
.filter(WorkflowNodeExecution.workflow_run_id == trace_info.workflow_run_id)
.all()
# through workflow_run_id get all_nodes_execution using repository
session_factory = sessionmaker(bind=db.engine)
workflow_node_execution_repository = RepositoryFactory.create_workflow_node_execution_repository(
params={"tenant_id": trace_info.tenant_id, "session_factory": session_factory},
)
for node_execution_id_record in workflow_nodes_execution_id_records:
node_execution = (
db.session.query(
WorkflowNodeExecution.id,
WorkflowNodeExecution.tenant_id,
WorkflowNodeExecution.app_id,
WorkflowNodeExecution.title,
WorkflowNodeExecution.node_type,
WorkflowNodeExecution.status,
WorkflowNodeExecution.inputs,
WorkflowNodeExecution.outputs,
WorkflowNodeExecution.created_at,
WorkflowNodeExecution.elapsed_time,
WorkflowNodeExecution.process_data,
WorkflowNodeExecution.execution_metadata,
)
.filter(WorkflowNodeExecution.id == node_execution_id_record.id)
.first()
)
if not node_execution:
continue
# Get all executions for this workflow run
workflow_node_executions = workflow_node_execution_repository.get_by_workflow_run(
workflow_run_id=trace_info.workflow_run_id
)
for node_execution in workflow_node_executions:
node_execution_id = node_execution.id
tenant_id = node_execution.tenant_id
app_id = node_execution.app_id

View File

@@ -7,6 +7,7 @@ from typing import Optional, cast
from langsmith import Client
from langsmith.schemas import RunBase
from sqlalchemy.orm import sessionmaker
from core.ops.base_trace_instance import BaseTraceInstance
from core.ops.entities.config_entity import LangSmithConfig
@@ -27,9 +28,9 @@ from core.ops.langsmith_trace.entities.langsmith_trace_entity import (
LangSmithRunUpdateModel,
)
from core.ops.utils import filter_none_values, generate_dotted_order
from core.repository.repository_factory import RepositoryFactory
from extensions.ext_database import db
from models.model import EndUser, MessageFile
from models.workflow import WorkflowNodeExecution
logger = logging.getLogger(__name__)
@@ -134,36 +135,22 @@ class LangSmithDataTrace(BaseTraceInstance):
self.add_run(langsmith_run)
# through workflow_run_id get all_nodes_execution
workflow_nodes_execution_id_records = (
db.session.query(WorkflowNodeExecution.id)
.filter(WorkflowNodeExecution.workflow_run_id == trace_info.workflow_run_id)
.all()
# through workflow_run_id get all_nodes_execution using repository
session_factory = sessionmaker(bind=db.engine)
workflow_node_execution_repository = RepositoryFactory.create_workflow_node_execution_repository(
params={
"tenant_id": trace_info.tenant_id,
"app_id": trace_info.metadata.get("app_id"),
"session_factory": session_factory,
},
)
for node_execution_id_record in workflow_nodes_execution_id_records:
node_execution = (
db.session.query(
WorkflowNodeExecution.id,
WorkflowNodeExecution.tenant_id,
WorkflowNodeExecution.app_id,
WorkflowNodeExecution.title,
WorkflowNodeExecution.node_type,
WorkflowNodeExecution.status,
WorkflowNodeExecution.inputs,
WorkflowNodeExecution.outputs,
WorkflowNodeExecution.created_at,
WorkflowNodeExecution.elapsed_time,
WorkflowNodeExecution.process_data,
WorkflowNodeExecution.execution_metadata,
)
.filter(WorkflowNodeExecution.id == node_execution_id_record.id)
.first()
)
if not node_execution:
continue
# Get all executions for this workflow run
workflow_node_executions = workflow_node_execution_repository.get_by_workflow_run(
workflow_run_id=trace_info.workflow_run_id
)
for node_execution in workflow_node_executions:
node_execution_id = node_execution.id
tenant_id = node_execution.tenant_id
app_id = node_execution.app_id

View File

@@ -7,6 +7,7 @@ from typing import Optional, cast
from opik import Opik, Trace
from opik.id_helpers import uuid4_to_uuid7
from sqlalchemy.orm import sessionmaker
from core.ops.base_trace_instance import BaseTraceInstance
from core.ops.entities.config_entity import OpikConfig
@@ -21,9 +22,9 @@ from core.ops.entities.trace_entity import (
TraceTaskName,
WorkflowTraceInfo,
)
from core.repository.repository_factory import RepositoryFactory
from extensions.ext_database import db
from models.model import EndUser, MessageFile
from models.workflow import WorkflowNodeExecution
logger = logging.getLogger(__name__)
@@ -147,36 +148,22 @@ class OpikDataTrace(BaseTraceInstance):
}
self.add_trace(trace_data)
# through workflow_run_id get all_nodes_execution
workflow_nodes_execution_id_records = (
db.session.query(WorkflowNodeExecution.id)
.filter(WorkflowNodeExecution.workflow_run_id == trace_info.workflow_run_id)
.all()
# through workflow_run_id get all_nodes_execution using repository
session_factory = sessionmaker(bind=db.engine)
workflow_node_execution_repository = RepositoryFactory.create_workflow_node_execution_repository(
params={
"tenant_id": trace_info.tenant_id,
"app_id": trace_info.metadata.get("app_id"),
"session_factory": session_factory,
},
)
for node_execution_id_record in workflow_nodes_execution_id_records:
node_execution = (
db.session.query(
WorkflowNodeExecution.id,
WorkflowNodeExecution.tenant_id,
WorkflowNodeExecution.app_id,
WorkflowNodeExecution.title,
WorkflowNodeExecution.node_type,
WorkflowNodeExecution.status,
WorkflowNodeExecution.inputs,
WorkflowNodeExecution.outputs,
WorkflowNodeExecution.created_at,
WorkflowNodeExecution.elapsed_time,
WorkflowNodeExecution.process_data,
WorkflowNodeExecution.execution_metadata,
)
.filter(WorkflowNodeExecution.id == node_execution_id_record.id)
.first()
)
if not node_execution:
continue
# Get all executions for this workflow run
workflow_node_executions = workflow_node_execution_repository.get_by_workflow_run(
workflow_run_id=trace_info.workflow_run_id
)
for node_execution in workflow_node_executions:
node_execution_id = node_execution.id
tenant_id = node_execution.tenant_id
app_id = node_execution.app_id

View File

@@ -0,0 +1,15 @@
"""
Repository interfaces for data access.
This package contains repository interfaces that define the contract
for accessing and manipulating data, regardless of the underlying
storage mechanism.
"""
from core.repository.repository_factory import RepositoryFactory
from core.repository.workflow_node_execution_repository import WorkflowNodeExecutionRepository
__all__ = [
"RepositoryFactory",
"WorkflowNodeExecutionRepository",
]

View File

@@ -0,0 +1,97 @@
"""
Repository factory for creating repository instances.
This module provides a simple factory interface for creating repository instances.
It does not contain any implementation details or dependencies on specific repositories.
"""
from collections.abc import Callable, Mapping
from typing import Any, Literal, Optional, cast
from core.repository.workflow_node_execution_repository import WorkflowNodeExecutionRepository
# Type for factory functions - takes a dict of parameters and returns any repository type
RepositoryFactoryFunc = Callable[[Mapping[str, Any]], Any]
# Type for workflow node execution factory function
WorkflowNodeExecutionFactoryFunc = Callable[[Mapping[str, Any]], WorkflowNodeExecutionRepository]
# Repository type literals
RepositoryType = Literal["workflow_node_execution"]
class RepositoryFactory:
"""
Factory class for creating repository instances.
This factory delegates the actual repository creation to implementation-specific
factory functions that are registered with the factory at runtime.
"""
# Dictionary to store factory functions
_factory_functions: dict[str, RepositoryFactoryFunc] = {}
@classmethod
def _register_factory(cls, repository_type: RepositoryType, factory_func: RepositoryFactoryFunc) -> None:
"""
Register a factory function for a specific repository type.
This is a private method and should not be called directly.
Args:
repository_type: The type of repository (e.g., 'workflow_node_execution')
factory_func: A function that takes parameters and returns a repository instance
"""
cls._factory_functions[repository_type] = factory_func
@classmethod
def _create_repository(cls, repository_type: RepositoryType, params: Optional[Mapping[str, Any]] = None) -> Any:
"""
Create a new repository instance with the provided parameters.
This is a private method and should not be called directly.
Args:
repository_type: The type of repository to create
params: A dictionary of parameters to pass to the factory function
Returns:
A new instance of the requested repository
Raises:
ValueError: If no factory function is registered for the repository type
"""
if repository_type not in cls._factory_functions:
raise ValueError(f"No factory function registered for repository type '{repository_type}'")
# Use empty dict if params is None
params = params or {}
return cls._factory_functions[repository_type](params)
@classmethod
def register_workflow_node_execution_factory(cls, factory_func: WorkflowNodeExecutionFactoryFunc) -> None:
"""
Register a factory function for the workflow node execution repository.
Args:
factory_func: A function that takes parameters and returns a WorkflowNodeExecutionRepository instance
"""
cls._register_factory("workflow_node_execution", factory_func)
@classmethod
def create_workflow_node_execution_repository(
cls, params: Optional[Mapping[str, Any]] = None
) -> WorkflowNodeExecutionRepository:
"""
Create a new WorkflowNodeExecutionRepository instance with the provided parameters.
Args:
params: A dictionary of parameters to pass to the factory function
Returns:
A new instance of the WorkflowNodeExecutionRepository
Raises:
ValueError: If no factory function is registered for the workflow_node_execution repository type
"""
# We can safely cast here because we've registered a WorkflowNodeExecutionFactoryFunc
return cast(WorkflowNodeExecutionRepository, cls._create_repository("workflow_node_execution", params))

View File

@@ -0,0 +1,88 @@
from collections.abc import Sequence
from dataclasses import dataclass
from typing import Literal, Optional, Protocol
from models.workflow import WorkflowNodeExecution
@dataclass
class OrderConfig:
"""Configuration for ordering WorkflowNodeExecution instances."""
order_by: list[str]
order_direction: Optional[Literal["asc", "desc"]] = None
class WorkflowNodeExecutionRepository(Protocol):
"""
Repository interface for WorkflowNodeExecution.
This interface defines the contract for accessing and manipulating
WorkflowNodeExecution data, regardless of the underlying storage mechanism.
Note: Domain-specific concepts like multi-tenancy (tenant_id), application context (app_id),
and trigger sources (triggered_from) should be handled at the implementation level, not in
the core interface. This keeps the core domain model clean and independent of specific
application domains or deployment scenarios.
"""
def save(self, execution: WorkflowNodeExecution) -> None:
"""
Save a WorkflowNodeExecution instance.
Args:
execution: The WorkflowNodeExecution instance to save
"""
...
def get_by_node_execution_id(self, node_execution_id: str) -> Optional[WorkflowNodeExecution]:
"""
Retrieve a WorkflowNodeExecution by its node_execution_id.
Args:
node_execution_id: The node execution ID
Returns:
The WorkflowNodeExecution instance if found, None otherwise
"""
...
def get_by_workflow_run(
self,
workflow_run_id: str,
order_config: Optional[OrderConfig] = None,
) -> Sequence[WorkflowNodeExecution]:
"""
Retrieve all WorkflowNodeExecution instances for a specific workflow run.
Args:
workflow_run_id: The workflow run ID
order_config: Optional configuration for ordering results
order_config.order_by: List of fields to order by (e.g., ["index", "created_at"])
order_config.order_direction: Direction to order ("asc" or "desc")
Returns:
A list of WorkflowNodeExecution instances
"""
...
def get_running_executions(self, workflow_run_id: str) -> Sequence[WorkflowNodeExecution]:
"""
Retrieve all running WorkflowNodeExecution instances for a specific workflow run.
Args:
workflow_run_id: The workflow run ID
Returns:
A list of running WorkflowNodeExecution instances
"""
...
def update(self, execution: WorkflowNodeExecution) -> None:
"""
Update an existing WorkflowNodeExecution instance.
Args:
execution: The WorkflowNodeExecution instance to update
"""
...