Data Warehouse

A small and lightweight FastAPI service to manage SQLite databases through swagger UI. Allows users to import csv files in relational databases, using a small C helper to read through and infer the type affinity for each column.

Coming soon:

1. Support for pointing to Excel files

Quick Start

• Python 3.10+
• GCC/Clang for building the C helper

1. Create a virtual env and install deps:

   Mac/Unix: python -m venv .venv source .venv/bin/activate
   Windows: .venv\Scripts\activate

   pip install -r requirements.txt

2. Build the C helper (outputs bin/csv_type_infer) make

Example Usage

1. Initialize/connect to a database file curl -X POST "http://127.0.0.1:8000/database?db_path=sample.db"

2. Import a CSV into a table (infers types using the C helper) curl -X PUT "http://127.0.0.1:8000/database/importCSV?csv_path=datasets/test_types.csv&table_name=items&create_if_missing=true&replace=true&check_types=true"

3. Query data curl "http://127.0.0.1:8000/database/query?sql=SELECT%20*%20FROM%20items%20LIMIT%2010"

4. Check CSV type affinity worked correctly: "http://127.0.0.1:8000/database/getschemas"