🔥 DraftForge — 3D→2D CAD Drafting Engine

A modular, automated Python pipeline to convert 3D CAD files (IGS/STEP) and 2D engineering drawings (PDF/images) into high-fidelity, standardized 2D SVG vector graphics and PDF files.

---

Architecture Overview

The pipeline has two specialized input paths feeding into a unified geometry representation and rendering stage:

                  ┌───────────────────────┐
                  │      Input File       │
                  └──────────┬────────────┘
                             │
              Is it a CAD (IGS/STEP) or Drawing (PDF/Img)?
                             │
              ┌──────────────┴──────────────┐
              ▼                             ▼
       [ IGS / STEP ]                [ PDF / Image ]
     Path A (CAD Parser)          Path B (Drawing Parser)
    FreeCAD Headless & TechDraw    pdfplumber & OpenCV Hough
              │                             │
              └──────────────┬──────────────┘
                             ▼
              ┌─────────────────────────────┐
              │   Unified Geometry Data     │
              │  (Lines, Texts, Dimensions) │
              └──────────────┬───────────── ┘
                             ▼
              ┌─────────────────────────────┐
              │  Unified Canvas & Renderer  │
              │  (svgwrite & reportlab PDF) │
              └──────────────┬──────────────┘
                             ▼
                    ┌──────────────────┐
                    │  SVG & PDF Docs  │
                    └──────────────────┘

• Path A (cad_parser.py): Utilizes headless FreeCAD to load 3D parts and uses the TechDraw module to dynamically project Front, Side, and Top orthographic views.
• Path B (pdf_parser.py): Parses PDF drawings using pdfplumber to extract native vector geometry (lines, rects) and OCR/text, and uses computer vision (pdf2image + opencv-python) to apply Hough Line detection to raster images.
• Unified Canvas (renderer.py): Draws clean technical drawings with standardised colors, scale scales, sheet margins, and dimensions.

---

Setup & Dependencies

1. System Dependencies

PDF Rendering (Poppler)

pdf2image requires poppler to convert PDF pages into images for computer vision line detection.

• macOS: brew install poppler
• Linux (Ubuntu/Debian): sudo apt-get install poppler-utils
• Windows: Download poppler and add the bin/ folder to your System PATH.

FreeCAD (For 3D CAD Path A)

To run Path A natively, FreeCAD must be installed on your system.

• macOS: brew install --cask freecad (installed in /Applications/FreeCAD.app)
• Linux: sudo apt-get install freecad
• Python Setup: The pipeline dynamically attempts to detect standard FreeCAD library paths. If FreeCAD is installed in a non-standard location, you can set the FREECAD_LIB_PATH environment variable:

  export FREECAD_LIB_PATH="/Applications/FreeCAD.app/Contents/Resources/lib"

  Note: If FreeCAD is not installed, the pipeline runs in a "Mock CAD Mode" for demonstration, producing drawing placeholders so testing does not break.

2. Python Dependencies

Install the required packages:

pip install -r requirements.txt

---

CLI Usage

Run the pipeline using the CLI entrypoint pipeline.py:

# Parse a 3D CAD file (IGS/STEP) and project views to SVG/PDF
python pipeline.py --input path/to/model.igs --output output_drawing --path cad

# Parse a 2D PDF drawing (vector or raster) and output clean vector paths
python pipeline.py --input path/to/drawing.pdf --output output_drawing --path pdf

# Parse an image drawing (PNG/JPG) using Hough Line detection
python pipeline.py --input path/to/blueprint.png --output output_drawing --path pdf

Options:

• --input, -i: Path to the input file (IGS, STEP, PDF, PNG, JPG).
• --output, -o: Base name for the outputs (will generate .svg and .pdf).
• --path, -p: Explicitly force the pipeline path (cad or pdf). If not provided, it is auto-detected from the file extension.
• --view, -v: (For CAD path) Space-separated list of views to project. Options: top, front, side, iso. Defaults to all views.