Documentation for PG Library

Introduction

The PG library offers a suite of utilities to manage and interact with PostgreSQL databases using SQLAlchemy. It provides functionality such as initializing the database engine, creating tables, and managing sessions.

Installation

pip3 install pg-alchemy-kit or if you prefer poetry add pg-alchemy-kit

Usage

1. Initialization

To initialize the PostgreSQL utility class, use:

from pg_alchemy_kit import PG

db = PG()
db.initialize(url="postgresql://username:password@localhost:5432/mydatabase")

• url (Optional): The connection string for the PostgreSQL database. If not provided, the default is derived from get_engine_url(), which uses the following environment variables:
  • PG_USER: The username for the database.
  • PG_PASSWORD: The password for the database.
  • PG_HOST: The host for the database.
  • PG_PORT: The port for the database.
  • PG_DB: The name of the database.

2. Creating Tables

To create tables in your PostgreSQL database:

from your_orm_module import BaseModel1, BaseModel2

# Create tables for the models in the provided list
db.create_tables([BaseModel1, BaseModel2])

• Bases: A list of SQLAlchemy base models.
• schemas: A list of schema names. Default is ["public"].

3. Managing Sessions

Context Manager

Use the get_session_ctx() to manage your session using a context manager:

with db.get_session_ctx() as session:
    # Use session for database operations here
    ...

Generator

You can also use get_session() to get a session:

session = next(db.get_session())

4. Closing the Connection

After all operations, ensure you close the database connection:

db.close()

Logging

The PG class sets up a logger to capture messages. If you want to use your own logger, pass it during initialization:

import logging

logger = logging.getLogger('my_custom_logger')
db.initialize(url="postgresql://username:password@localhost:5432/mydatabase", logger=logger)

Documentation for PGUtils Class

Introduction

PGUtils is a utility class that provides various database-related methods for performing CRUD operations, transforming SQL statements, and managing connections.

Initialization

Before using PGUtils, it should be initialized:

from your_module_path import PGUtils

logger = logging.getLogger('my_custom_logger')
db_utils = PGUtils(logger)

• logger: A logging instance to capture any log messages.

Setting up a session:

After initializing, you should set up a session for further operations:

session = db.get_session_ctx()  # Get this from the PG class
db_utils.initialize(session)

Methods

1. SQL Execution

Select Query

To select records from the database:

results = db_utils.select(session, "SELECT * FROM your_table WHERE condition=:condition", {'condition': value})

Insert Query

To insert records:

status = db_utils.insert(session, "INSERT INTO your_table(column) VALUES (:value)", {'value': value})

Delete Query

To delete records:

status = db_utils.delete(session, "DELETE FROM your_table WHERE condition=:condition", {'condition': value})

Execute Query

To execute any SQL:

status = db_utils.execute(session, "YOUR SQL QUERY HERE")

Update Query

To update records:

status = db_utils.update(session, ModelClass, {'key': key_value}, {'column_to_update': new_value})

2. ORM Operations

Insert ORM Record

To insert a single ORM record:

record = db_utils.insert_orm(session, ModelClass, {'column': value})

Bulk Insert ORM Records

To insert multiple ORM records:

ids, records = db_utils.bulk_insert_orm(session, ModelClass, [{'column1': value1}, {'column2': value2}])

Insert ORM on Conflict

To insert ORM records with conflict handling:

db_utils.insert_orm_on_conflict(session, ModelClass, [{'column': value}])

Delete ORM Records

To delete ORM records:

status = db_utils.delete_orm(session, ModelClass, [uuid1, uuid2])

Get UUID

To get the UUID of a record:

record_uuid = db_utils.get_uuid(session, ModelClass, {'column': value})

3. Utilities

Wrap SQL to JSON

To wrap a SQL statement such that its result is returned as a JSON array:

json_sql = PGUtils.wrap_to_json("YOUR SQL QUERY HERE")

4. Engine Management

Get Engine

To get an engine:

engine = get_engine("postgresql://username:password@localhost:5432/mydatabase")

Get Engine URL

To get an engine URL:

url = get_engine_url(connection_type="postgresql", pg_username="username", pg_password="password", pg_host="localhost", pg_port="5432", pg_db="mydatabase")

Conclusion

The combination of PG and PGUtils offers a robust solution for database operations with PostgreSQL using SQLAlchemy. Whether you're using raw SQL or ORM models, these classes simplify your database tasks. Always ensure to handle exceptions and roll back sessions where necessary to maintain data integrity.