AlejandroMorgante commented on code in PR #68479: URL: https://github.com/apache/airflow/pull/68479#discussion_r3483869762
########## providers/google/src/airflow/providers/google/cloud/operators/vertex_ai/agent_engine.py: ########## @@ -0,0 +1,494 @@ +# +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. +"""This module contains Google Vertex AI Agent Engine operators.""" + +from __future__ import annotations + +from collections.abc import Sequence +from functools import cached_property +from typing import TYPE_CHECKING, Any + +from airflow.providers.common.compat.sdk import conf +from airflow.providers.google.cloud.hooks.vertex_ai.agent_engine import AgentEngineHook, _serialize_value +from airflow.providers.google.cloud.operators.cloud_base import GoogleCloudBaseOperator +from airflow.providers.google.cloud.triggers.vertex_ai import ( + AgentEngineDeleteTrigger, + AgentEngineQueryJobTrigger, +) + +if TYPE_CHECKING: + from vertexai._genai import types + + from airflow.providers.common.compat.sdk import Context + + +def _serialize_agent_engine(agent_engine: types.AgentEngine) -> dict[str, Any]: + api_resource = getattr(agent_engine, "api_resource", None) + if api_resource is not None: + return _serialize_value(api_resource) + return _serialize_value(agent_engine) + + +class CreateAgentEngineOperator(GoogleCloudBaseOperator): + """ + Create a Vertex AI Agent Engine. + + :param project_id: Required. The ID of the Google Cloud project that the service belongs to. + :param location: Required. The ID of the Google Cloud location that the service belongs to. + :param agent: Optional. The agent object to deploy. + :param config: Optional. Configuration for the Agent Engine. + :param gcp_conn_id: The connection ID to use connecting to Google Cloud. + :param impersonation_chain: Optional service account to impersonate using short-term credentials. + """ + + template_fields = ( + "project_id", + "location", + "agent", + "config", + "gcp_conn_id", + "impersonation_chain", + ) + + def __init__( + self, + *, + project_id: str, + location: str, + agent: Any | None = None, + config: types.AgentEngineConfigOrDict | None = None, + gcp_conn_id: str = "google_cloud_default", + impersonation_chain: str | Sequence[str] | None = None, + **kwargs, + ) -> None: + super().__init__(**kwargs) + self.project_id = project_id + self.location = location + self.agent = agent + self.config = config + self.gcp_conn_id = gcp_conn_id + self.impersonation_chain = impersonation_chain + + @cached_property + def hook(self) -> AgentEngineHook: + return AgentEngineHook( + gcp_conn_id=self.gcp_conn_id, + impersonation_chain=self.impersonation_chain, + ) + + def execute(self, context: Context) -> dict[str, Any]: + self.log.info("Creating Agent Engine.") + agent_engine = self.hook.create_agent_engine( + project_id=self.project_id, + location=self.location, + agent=self.agent, + config=self.config, + ) + result = _serialize_agent_engine(agent_engine) + self.log.info("Agent Engine was created.") + return result + + +class GetAgentEngineOperator(GoogleCloudBaseOperator): + """ + Get a Vertex AI Agent Engine. + + :param project_id: Required. The ID of the Google Cloud project that the service belongs to. + :param location: Required. The ID of the Google Cloud location that the service belongs to. + :param agent_engine_id: Required. The Agent Engine ID. + :param gcp_conn_id: The connection ID to use connecting to Google Cloud. + :param impersonation_chain: Optional service account to impersonate using short-term credentials. + """ + + template_fields = ("project_id", "location", "agent_engine_id", "gcp_conn_id", "impersonation_chain") + + def __init__( + self, + *, + project_id: str, + location: str, + agent_engine_id: str, + gcp_conn_id: str = "google_cloud_default", + impersonation_chain: str | Sequence[str] | None = None, + **kwargs, + ) -> None: + super().__init__(**kwargs) + self.project_id = project_id + self.location = location + self.agent_engine_id = agent_engine_id + self.gcp_conn_id = gcp_conn_id + self.impersonation_chain = impersonation_chain + + @cached_property + def hook(self) -> AgentEngineHook: + return AgentEngineHook( + gcp_conn_id=self.gcp_conn_id, + impersonation_chain=self.impersonation_chain, + ) + + def execute(self, context: Context) -> dict[str, Any]: + self.log.info("Getting Agent Engine %s.", self.agent_engine_id) + agent_engine = self.hook.get_agent_engine( + project_id=self.project_id, + location=self.location, + agent_engine_id=self.agent_engine_id, + ) + result = _serialize_agent_engine(agent_engine) + self.log.info("Agent Engine %s was retrieved.", self.agent_engine_id) + return result + + +class QueryAgentEngineOperator(GoogleCloudBaseOperator): + """ + Run a query job on a Vertex AI Agent Engine. + + :param project_id: Required. The ID of the Google Cloud project that the service belongs to. + :param location: Required. The ID of the Google Cloud location that the service belongs to. + :param agent_engine_id: Required. The Agent Engine ID. + :param config: Required. Configuration for the query job (``query``, ``output_gcs_uri``). + :param gcp_conn_id: The connection ID to use connecting to Google Cloud. + :param impersonation_chain: Optional service account to impersonate using short-term credentials. + """ + + template_fields = ( + "project_id", + "location", + "agent_engine_id", + "config", + "gcp_conn_id", + "impersonation_chain", + ) + + def __init__( + self, + *, + project_id: str, + location: str, + agent_engine_id: str, + config: types.RunQueryJobAgentEngineConfigOrDict, + gcp_conn_id: str = "google_cloud_default", + impersonation_chain: str | Sequence[str] | None = None, + **kwargs, + ) -> None: + super().__init__(**kwargs) + self.project_id = project_id + self.location = location + self.agent_engine_id = agent_engine_id + self.config = config + self.gcp_conn_id = gcp_conn_id + self.impersonation_chain = impersonation_chain + + @cached_property + def hook(self) -> AgentEngineHook: + return AgentEngineHook( + gcp_conn_id=self.gcp_conn_id, + impersonation_chain=self.impersonation_chain, + ) + + def execute(self, context: Context) -> dict[str, Any]: + self.log.info("Running query job on Agent Engine %s.", self.agent_engine_id) + query_job = self.hook.query_agent_engine( + project_id=self.project_id, + location=self.location, + agent_engine_id=self.agent_engine_id, + config=self.config, + ) + self.log.info("Query job was started on Agent Engine %s.", self.agent_engine_id) + return _serialize_value(query_job) Review Comment: Done. I updated the query operator to handle the full user workflow: it runs the query job, extracts the operation id from the returned job name, waits for completion, and returns the final query job result to XCom. I also removed the separate check operator from the public workflow -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
