ColtaneDB Python SDK (coltanedb)

Official Python client and query SDK for ColtaneDB and the Colt (.colt, ColtLang) declarative query language, powered by the Meld Model (Multi-Engine Link Declaration).

Conceived and created by Ayush Bhattacharya.

Installation

Install the package from PyPI:

pip install coltanedb

Features

• The Meld Model: Query, link, and stitch records seamlessly across PostgreSQL, Redis, MongoDB, and SQLite.
• Declarative Query Execution: Execute .colt statements with injection-proof parameter binding.
• Multi-Engine Planning: Inspect native parameterized query plans before executing (client.plan).
• Interactive Language Guide: Export the complete Colt Language Reference via coltanedb.guide(file=True).
• Author & Project Metadata: Inspect author credentials and repository links via coltanedb.author().

Quickstart

1. Initialize the Client

import coltanedb
from coltanedb import ColtaneClient

# Initialize from an app.colt file (resolves connections via environment)
db = ColtaneClient.from_file("app.colt")

# Or connect to a running ColtaneDB bridge (colt serve) or server
client = ColtaneClient(
    base_url="http://localhost:3000",
    api_key="clt_live_demo",
)

2. Execute a Colt Query

# Run with bound parameters
result = client.run("FIND books WHERE views > $min LIMIT 10", min=10)

print("Query success:", result.success)
for row in result.rows:
    print(f"Book: {row['title']}")

# Inspect execution metadata and cross-engine status
print(result.meta)

3. Generate the Complete Colt Language Guide

# Creates a COLT_GUIDE.md file in the current working directory
coltanedb.guide(file=True)

# Or print readable plain text directly to the console
coltanedb.guide(file=False)

4. Author & Attribution

# Prints author info with clickable terminal hyperlinks
coltanedb.author()

5. Inspect Execution Plans

plan = client.plan("""
    FIND reviews FETCH book 
    WHERE book.views > 10 
    LIMIT 20
""")

for step in plan.steps:
    print(f"Backend: {step.backend}, Op: {step.operation}")

License

Apache-2.0 © Ayush Bhattacharya & ColtaneDB Team