---
license: apache-2.0
tags:
  - deckergui
  - source-code
  - cleaning
  - gemini
  - text-processing
  - en
  - license:apache-2.0
library_name: deckergui
pipeline_tag: text2text-generation
---

## DGUI SourceCleanR

**Source Code Cleaning & Transformation Utility** for DeckerGUI.

### Overview
Notepad++-styled web utility for transforming raw/unstructured source code (HTML), text, and PDFs into clean, human-readable formats using Gemini AI.

### Modes
- **Normal Mode**: Process a single source file with AI-powered cleaning
- **AutoClean Mode**: Batch process up to 100 files simultaneously

### Output Formats
- **TXT** — Plain text output
- **JSON** — Structured data output
- **XML** — Marked-up output format

### Features
- HTML source code cleanup and formatting
- PDF text extraction and normalization
- Batch processing with progress tracking
- AI-powered understanding of code structure
- Preserves semantic meaning while removing noise

### Integration
SourceCleanR connects to the DeckerGUI ecosystem through the DSYNC module for dataset generation and the KPI Tokenizer for usage tracking.

### Citation
```bibtex
@software{deckergui-sourcecleanr,
  title={DGUI SourceCleanR},
  author={wan mohd azizi},
  year={2025},
  publisher={Hugging Face},
  url={https://huggingface.co/ctaxnagomi/DGUI-SourceCleanR}
}
```