120 lines
5.3 KiB
Python
120 lines
5.3 KiB
Python
# Copyright 2025 Google LLC
|
|
#
|
|
# Licensed 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.
|
|
|
|
"""
|
|
Initializes the BigQuery and Vertex AI Search environment for the application.
|
|
|
|
This script performs the following actions:
|
|
1. Retrieves configuration from environment variables or uses defaults.
|
|
2. Creates the specified BigQuery dataset if it doesn't already exist.
|
|
3. Creates the 'search_applications' table within that dataset.
|
|
4. Creates a Vertex AI Search Datastore if it doesn't already exist.
|
|
5. Imports documents from a specified GCS bucket into the Datastore.
|
|
6. Creates a Vertex AI Search Engine (App) linked to the Datastore.
|
|
|
|
Usage:
|
|
Run this script directly (e.g., `python setup.py`).
|
|
Set environment variables to override defaults:
|
|
- 'BIG_QUERY_DATASET'
|
|
- 'GOOGLE_CLOUD_PROJECT'
|
|
- 'VERTEX_AI_SEARCH_LOCATION'
|
|
- 'VERTEX_AI_DATASTORE_ID'
|
|
- 'VERTEX_AI_ENGINE_ID'
|
|
"""
|
|
|
|
from os import getenv
|
|
from scripts.big_query_setup import create_dataset, create_table
|
|
from src.service.search_application import SEARCH_APPLICATION_TABLE
|
|
from src.model.search import SearchApplication
|
|
from scripts.vertexai_search_setup import create_vertex_ai_datastore, create_vertex_ai_engine, import_documents_to_datastore
|
|
|
|
|
|
def main():
|
|
# 1. BigQuery Setup
|
|
print("--- Setting up BigQuery ---")
|
|
BIG_QUERY_DATASET = getenv("BIG_QUERY_DATASET", "quickbot_default_bq_dataset")
|
|
GCLOUD_PROJECT = getenv("GCLOUD_PROJECT", "my-gcloud-project")
|
|
create_dataset(BIG_QUERY_DATASET)
|
|
create_table(
|
|
BIG_QUERY_DATASET, SEARCH_APPLICATION_TABLE, SearchApplication.__schema__()
|
|
)
|
|
|
|
# 2. Vertex AI Search Setup
|
|
print("--- Setting up Vertex AI Search ---")
|
|
VERTEX_AI_LOCATION = getenv("VERTEX_AI_LOCATION", "global")
|
|
VERTEX_AI_DATASTORE_ID = getenv("VERTEX_AI_DATASTORE_ID", "quickbot_alphabet_pdfs_ds")
|
|
VERTEX_AI_ENGINE_ID = getenv("VERTEX_AI_ENGINE_ID", "quickbot_alphabet_search_engine")
|
|
GCS_SOURCE_URI = "gs://cloud-samples-data/gen-app-builder/search/alphabet-investor-pdfs/*.pdf"
|
|
DATASTORE_DISPLAY_NAME_PREFIX = "Alphabet Investor Docs DS"
|
|
ENGINE_DISPLAY_NAME_PREFIX = "Alphabet Investor Engine"
|
|
|
|
datastore_display_name = f"{DATASTORE_DISPLAY_NAME_PREFIX} ({VERTEX_AI_DATASTORE_ID})"
|
|
engine_display_name = f"{ENGINE_DISPLAY_NAME_PREFIX} ({VERTEX_AI_ENGINE_ID})"
|
|
try:
|
|
# Create/Get Datastore
|
|
print(f"Attempting to create/get Datastore '{VERTEX_AI_DATASTORE_ID}' in project '{GCLOUD_PROJECT}' location '{VERTEX_AI_LOCATION}'...")
|
|
datastore = create_vertex_ai_datastore(
|
|
GCLOUD_PROJECT, VERTEX_AI_LOCATION, VERTEX_AI_DATASTORE_ID, datastore_display_name
|
|
)
|
|
if not datastore:
|
|
print("Datastore creation/retrieval failed. Aborting further Vertex AI Search setup.")
|
|
print("--- Application setup finished (with errors) ---")
|
|
raise
|
|
print(f"Successfully ensured Datastore exists: {datastore.name}")
|
|
|
|
# Import documents into Datastore
|
|
# Note: This will attempt to import documents every time the script runs.
|
|
# For production, you might want to add a check to skip this if documents
|
|
# are already present or if a previous import was successful.
|
|
print(f"\nAttempting to import documents from '{GCS_SOURCE_URI}' into datastore: {datastore.name}")
|
|
import_documents_to_datastore(
|
|
GCLOUD_PROJECT, VERTEX_AI_LOCATION, VERTEX_AI_DATASTORE_ID, GCS_SOURCE_URI
|
|
)
|
|
# Note: Document import can take a long time. The script waits.
|
|
print("Document import process initiated/completed.\n")
|
|
|
|
# Create/Get Engine
|
|
print(f"Attempting to create/get Engine '{VERTEX_AI_ENGINE_ID}' in project '{GCLOUD_PROJECT}' location '{VERTEX_AI_LOCATION}'...")
|
|
# The create_vertex_ai_engine function expects a list of datastore IDs (not full resource names).
|
|
engine = create_vertex_ai_engine(
|
|
GCLOUD_PROJECT,
|
|
VERTEX_AI_LOCATION,
|
|
VERTEX_AI_ENGINE_ID,
|
|
engine_display_name,
|
|
[VERTEX_AI_DATASTORE_ID] # Pass the Datastore ID string
|
|
)
|
|
|
|
if not engine:
|
|
print("Engine creation/retrieval failed.")
|
|
print("--- Application setup finished (with errors) ---")
|
|
raise
|
|
print(f"Successfully ensured Engine exists: {engine.name}")
|
|
|
|
print("\nVertex AI Search setup completed successfully.")
|
|
|
|
except Exception as e:
|
|
print(f"A critical error occurred during the setup process: {e}")
|
|
import traceback
|
|
print("Detailed traceback:")
|
|
print(traceback.format_exc())
|
|
# If running in Docker build, exiting with non-zero will fail the build
|
|
import sys
|
|
sys.exit(1)
|
|
|
|
print("\n--- Application setup finished ---")
|
|
print("\nSuccess! All resources should now be configured.\n")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main() |