Metadata-Version: 2.4
Name: lightningclean
Version: 1.2.1
Summary: Blazing fast hardware-accelerated tabular firewall and regulated sanitization engine.
Author: AI Research Lab
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: C++
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: POSIX :: Linux
Classifier: Operating System :: MacOS
Classifier: Operating System :: Microsoft :: Windows
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: numpy>=1.22.0
Requires-Dist: pyarrow>=8.0.0
Provides-Extra: web
Requires-Dist: fastapi>=0.100.0; extra == "web"
Requires-Dist: uvicorn>=0.20.0; extra == "web"
Requires-Dist: pydantic>=2.0.0; extra == "web"
Dynamic: requires-python

# LightningClean

Hardware-Accelerated Tabular Firewall and Regulated Low-Latency Data Sanitization Engine.

LightningClean is an enterprise-grade high-performance Python package built with a native C++ backend designed to sanitize massive tabular datasets at bare-metal speeds. By utilizing hardware-level AVX-512/AVX2 SIMD vectorization and breaking Python's execution limits via OpenMP multi-core multithreading, it isolates and rectifies structural data anomalies seamlessly without memory copies.

---

## Key Architectural Capabilities

1. **CPUID Dynamic Dispatcher**: Automatically detects host microprocessors at runtime to deploy tailored hardware vectors seamlessly.
2. **True Zero-Copy Memory Linkage**: Shares direct RAM memory data pointer tracks across Python/C++ loops to preserve system memory space.
3. **Shield Mode Page Isolation**: Wraps executions inside sandboxed memory barriers to capture segmentation faults safely without process termination.
4. **Deterministic Audit Control**: Locks parallel reduction variances to deliver strict bit-for-bit mathematical reproducibility across compliance audits.
5. **PII Masking Engine**: Performs regex scans during string array extraction to obscure protected data structures natively.

---

## Installation

### Standard Production Core
```bash
pip install lightningclean
```

### Full Enterprise Web Extra
```bash
pip install "lightningclean[web]"
```

---

## Operational Code Example

```python
import pandas as pd
import lightningclean as lc

# Load a massive contaminated analytical database
df = pd.read_csv("unstable_enterprise_dataset.csv")

# 1-Line Execution Pass with Compliance Constraints Enabled
clean_df = lc.clean(
    df, 
    shield=True, 
    deterministic=True,      # Bit-exact audit reproducibility
    pii_mode=True,           # Automatic masking of protected rows
    numa_aware=True          # Pins tasks directly across physical hardware CPU sockets
)

# Extract structured summaries instantly
report = clean_df.attrs["shield_report"]
print(f"Sanitized: {report['cleaned_count']} | Quarantined: {report['corrupted_count']}")
```
