sqlalchemy

SQLAlchemy Recipes (Standalone Scripts)

Six copy-paste friendly SQLAlchemy 2.0 scripts for CUBRID. Each recipe is a single .py file you can run directly with python3.

Connection used by all recipes:

cubrid+pycubrid://dba@localhost:33000/testdb

Prerequisites

• Python 3.10+
• CUBRID running on localhost:33000 (testdb, user dba)

From repository root:

make up

Install

pip install -r requirements.txt

Recipes

1. 01_connect_and_session.py
   • Engine creation, Session usage, SELECT 1, connection metadata, clean shutdown
2. 02_orm_crud.py
   • ORM model with DeclarativeBase and mapped_column, full CRUD lifecycle
3. 03_relationships.py
   • Department/Employee relationship, lazy/eager loading, child-first deletion
4. 04_bulk_insert.py
   • Bulk insert via session.execute(insert(), rows) vs add_all, timing comparison
5. 05_cubrid_upsert.py
   • CUBRID ON DUPLICATE KEY UPDATE and REPLACE INTO via sqlalchemy_cubrid.dml
6. 06_reflection.py
   • Runtime schema discovery with inspect() and Inspector API

Run

python3 01_connect_and_session.py
python3 02_orm_crud.py
python3 03_relationships.py
python3 04_bulk_insert.py
python3 05_cubrid_upsert.py
python3 06_reflection.py

Notes

• Every script creates its own cookbook_ tables, seeds data, demonstrates the feature, and drops tables in finally.
• Boolean-like values are stored as Integer (0/1).
• Money values are stored as integer cents.
• No DB-level cascade is used; related deletes are handled in application code.