Skip to content

v1.0

Latest
Latest

Choose a tag to compare

View all tags
@hjk-maker hjk-maker released this 28 Mar 10:26
· 2 commits to main since this release
1.0
5f49714
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

⚡ ClathSQL Pro — Intelligent Data Routing Engine

HJK-INC Enterprise Core | Virtual SQL Organisation
Automate Your Storage Decisions with AI-Powered Intelligence

License
Python
Status

📖 Overview

ClathSQL Pro is an autonomous data ingestion and routing platform designed for high-volume, multi-format data pipelines. By leveraging TinyLlama-based neural inference, it classifies incoming unstructured content into three execution paths:

Route Technology Use Case
SQL SQLite (Relational) Structured tabular data, queryable records
NOSQL JSON Document Store Unstructured documents, nested schemas
PLOT Matplotlib Visualization Time-series, numerical sequences, metrics

Built by hjk-inc, this system embodies our philosophy: "Build once, automate everywhere." No manual classification needed—the brain decides, the vault stores, the GUI visualizes.

🎯 Key Features

🔹 Hybrid AI Routing Logic

  • Neural Inference: TinyLlama-1.1B Chat model for semantic understanding
  • Regex Bypass: Zero-latency pattern matching for obvious cases (JSON → NOSQL, SELECT → SQL)
  • Warmup Sequence: Pre-loaded model weights eliminate first-inference lag

🔹 Dual-Storage Vault

  • SQLite Backend: Indexed temporal database with auto-timestamps
  • Append-Only JSON: Lock-safe NoSQL document storage with PID concurrency control

🔹 Enterprise Dashboard

  • CustomTkinter UI: Dark-mode optimized for extended operations
  • Real-time Monitoring: CPU/RAM telemetry via psutil
  • Embedded Visualization: Matplotlib plots rendered directly in GUI
  • Export Automation: One-click CSV export of vault contents

🔹 Thread-Safe Architecture

  • All heavy operations run on background threads
  • Non-blocking UI during AI inference (5-15s models)
  • Graceful shutdown with resource cleanup

🚀 Installation & Setup

Prerequisites

# Python 3.8+ required
python --version

# Install dependencies
pip install customtkinter torch transformers pandas matplotlib psutil

Quick Start

git clone https://github.com/virtualsql/clathsql-pro.git
cd clathsql-pro
python clathsql_pro.py

First Run Checklist

  1. ✅ Model downloads automatically (~1GB TinyLlama cache)
  2. ✅ Watch status indicator turn green ("Brain Online")
  3. ✅ Paste test data → Click "CLUTCH DATA"
  4. ✅ Verify vault files created (hjk_clath_vault.db, hjk_clath_docs.json)

🧪 Usage Examples

1. Routing to SQL

Paste: "SELECT * FROM users WHERE id > 100"
→ AI detects SQL keywords → Stored in relational table

2. Routing to NoSQL

Paste: {"user_id": 42, "action": "purchase", "amount": 99.99}
→ Regex bypasses AI → Appended as JSON document

3. Routing to Plot

Paste:
10.5
20.3
15.8
25.1
30.4
→ AI identifies numerical sequence → Generated trend plot

🛠️ Configuration

Environment Variables (Optional)

# .env file at project root
MODEL_ID=TinyLlama/TinyLlama-1.1B-Chat-v1.0   # Switch to Phi-3 or Llama-3
DEVICE=0                                       # 0 for GPU, -1 for CPU
MAX_TOKENS=5                                   # Response length limit

Customizing AI Model

Edit ClathSQL_Brain.__init__:

model_id = "mistralai/Mistral-7B-Instruct-v0.1"  # Higher accuracy
torch_dtype=torch.float16                        # Reduce VRAM usage

📊 Vault Statistics

Access real-time metrics via the sidebar panel:

Metric Description
SQL Entries Count of records in clath_stream table
NoSQL Docs Lines written to document store
CPU Power Real-time processor utilization
RAM Usage System memory consumption percentage

Refresh manually or watch auto-updates every 30 seconds.

🤝 Integration Guide

Connect to Maker's Newspaper Platform

Expose ClathSQL routing logic via FastAPI for cross-service automation:

from fastapi import FastAPI
app = FastAPI()

@app.post("/route-data")
def route_data(data: str):
    decision = brain.clutch_logic(data)
    return {"route": decision}

Then consume from your other HJK-INC microservices without duplicating infrastructure.

Embed in Existing Applications

Import core classes directly:

from clathsql_pro import ClathSQL_Brain, ClathSQL_Vault

brain = ClathSQL_Brain(lambda x, y: None)  # Disable UI callback
vault = ClathSQL_Vault()
decision = brain.clutch_logic(user_input)
vault.store_sql(user_input) if decision == "SQL" else vault.store_nosql(user_input)

🔒 Security Considerations

Risk Mitigation
Path Injection Input sanitized before file I/O
Concurrent Writes PID lock file prevents race conditions
Memory Leaks Explicit thread cleanup on app close
Model Exploits Temperature=0.1 limits adversarial outputs

⚠️ Warning: For production use, add input size limits (max 1MB per payload) and authentication middleware.

🐛 Known Issues

Issue Status Workaround
First inference takes ~15s Expected Warmup runs automatically on launch
Plot fails with <2 numbers Intentional Display warning instead of fake data
Large JSON slows append Ongoing Batch writes planned v2.1

📅 Roadmap

  • V2.0 — Stable release with hybrid routing
  • V2.1 — Multi-threaded batch processing
  • V2.2 — REST API layer for remote routing
  • V2.3 — PostgreSQL/Redis adapters
  • V2.4 — Cloud deployment templates (Docker/Kubernetes)

👥 Contributing

Pull requests welcome! Please follow these guidelines:

  1. Fork the repository
  2. Create feature branch (git checkout -b feature/amazing-feature)
  3. Commit with conventional messages (feat:, fix:, docs:)
  4. Push to branch and open Pull Request
  5. Ensure all tests pass before merge

📄 License

MIT License © 2024 HJK-INC. See LICENSE for full details.

📞 Support

For enterprise support, integration consulting, or bulk deployment:

Built with ❤️ by the HJK-INC Engineering Team

"Lazy by design, powerful by necessity"

Assets 2
Loading