Python SVG to PPTX Compiler

A high-fidelity, two-pass compiler that translates structured SVG metadata into native PowerPoint (.pptx) presentations. It uses a strictly decoupled Intermediate Representation (IR) layer to bridge SVG semantics with PowerPoint's DrawingML.

Features

• Decoupled Architecture: Separate Frontend (Parser), IR Layer, and Backend (Generator).
• Geometric Mapping: Translates SVG shapes (<rect>, <circle>, <ellipse>, <polygon>) into native PowerPoint shapes.
• Smart Text Handling: Maps nested <text> and <tspan> elements to native text frames with custom styling (font-family, size, weight, color).
• Intelligent Connectors: Support for straight, elbow, and curve routing with automatic shape anchoring (begin_connect / end_connect).
• Arrowhead Support: Recognizes standard SVG markers and maps them to PowerPoint's arrow, diamond, and stealth line-end types.
• Complex Icons: Supports embedding nested <svg> fragments as high-quality pictures.
• Hierarchical Grouping: Reconstructs SVG group structures (<g>) as native PowerPoint GroupShapes, supporting nested hierarchies.
• Styling Preservation: Handles hex colors, CSS color names (e.g., orange, blue), stroke widths, corner radii, and translucency (alpha).
• Logging Implementation: Uses standard Python logging module for production-ready monitoring.

SVG Metadata Schema & Requirements

The compiler expects SVG files to conform to specific structural and metadata rules to guide the transformation properly:

Canvas & Styling

• Dimensions: SVG must have viewBox and full widht and height, e.g. viewBox="0 0 960 540" width="100%" height="100%".
• Styling: Use presentation attributes (fill="#FF5733", stroke="#e0e0e0") ONLY. Do not use inline CSS (style="...") or <style> blocks.
• Colors & Transparency: Use 3 or 6-digit hex colors. For opacity, use explicit fill-opacity="..." or stroke-opacity="..." attributes (0.0 to 1.0) rather than 8-digit hex codes.
• Typography: Use standard fonts only (e.g., Arial, Calibri, Segoe UI). Use <tspan> for rich text formatting.

Structural Rules

• InfoBox (data-element-type="infoBox"): Wrap logical components in <g id="[unique_id]" data-element-type="infoBox">.
• Connectors (data-element-type="connector"): Must be kept isolated at the root level (never nested inside other <g> groups) as <line> or <polyline>.
  • Required attributes: data-start="[source_id]", data-end="[target_id]", data-connector-type="straight|elbow|curve".
  • Arrowheads: Supported via marker-start and marker-end attributes referencing standard marker defs (none, arrow, diamond, stealth).
• Icons (data-element-type="icon"): Isolate inside a group and nest a child <svg> with explicit x, y, width, height, and viewBox attributes.

Usage

Installation

pip install surquest-utils-svg2pptx

Simple Conversion (Single Slide)

from surquest.utils.svg2pptx import SVG2Pptx

# Initialize converter
converter = SVG2Pptx()

# Convert SVG file/string to PPTX
converter.convert(
    svg_input="input.svg",
    output_path="output.pptx"
)

Multi-Slide Conversion

You can generate a multi-slide presentation by providing a list of SVG paths. The slides will be generated in the order provided.

from surquest.utils.svg2pptx import SVG2Pptx

converter = SVG2Pptx()
converter.convert(
    svg_input=["slide1.svg", "slide2.svg", "slide3.svg"],
    output_path="multi_slide.pptx"
)

Project Structure

.
├── data/               # Sample SVG and JSON IR files
├── prompts/            # System instructions for LLM-based SVG generation
├── scripts/            # Utility scripts (e.g., run.py)
├── src/
│   └── surquest/
│       └── utils/
│           └── svg2pptx/
│               ├── generator/  # PPTX Generator (Backend)
│               ├── models/     # Intermediate Representation (IR) Models
│               ├── parser/     # SVG Parser (Frontend)
│               └── svg2pptx.py # Orchestrator & Main API
├── tests/              # Test suite (unit and integration tests)
├── pyproject.toml      # Build and dependency configuration
└── README.md

Advanced Usage

Exporting and Importing Intermediate Representation (JSON)

The compiler allows you to export the Intermediate Representation (IR) to JSON, and recreate presentations directly from the JSON. This is particularly useful for debugging or programmatic manipulation before PPTX generation.

from surquest.utils.svg2pptx import SVG2Pptx

converter = SVG2Pptx()

# 1. Export SVG directly to JSON IR
json_string = converter.to_json("input.svg")

# 2. Convert to PPTX and save the JSON IR simultaneously
converter.convert(
    svg_input="input.svg",
    output_path="output.pptx",
    export_as_json=True # Will also generate output.json
)

# 3. Create a presentation from a saved JSON IR
converter.from_json(
    json_input="output.json",
    output_path="restored_presentation.pptx"
)

Low level APIs

from surquest.utils.svg2pptx.parser import SVGParser
from surquest.utils.svg2pptx.generator import PPTXBackend

# 1. Parse SVG into Intermediate Representation
parser = SVGParser(svg_source, slide_width=12192000, slide_height=6858000, svg_ns="{http://www.w3.org/2000/svg}")
ir_slide = parser.parse()

# 2. Render IR to PPTX
prs = PPTXBackend([ir_slide]).render()
prs.save("output.pptx")

API Deployment (FastAPI)

This project includes a FastAPI application in the app/ folder that can be deployed to Vercel.

Endpoint: POST /sources/SVG/target/PPTX:convert

Converts raw SVG metadata to a downloadable .pptx file.

Query Parameters:

• filename: (Required) The name of the output PPTX file (e.g., presentation.pptx).

Request Body:

• The request body must be the raw SVG text strings (Content-Type should be text/plain or image/svg+xml).

Example Request:

curl -X POST "https://your-app.vercel.app/sources/SVG/target/PPTX:convert?filename=my_slide.pptx" \
     -H "Content-Type: text/plain" \
     --data-binary "@input.svg"

Local Development

1. Install requirements: pip install -r requirements.txt
2. Run the server: uvicorn app.main:app --reload

Vercel Deployment

The project is configured for Vercel via vercel.json. Simply connect your repository to Vercel and it will automatically detect the Python configuration.

Development & Testing

Development Environment (VS Code Dev Container)

This repository includes a VS Code Dev Container configuration to provide a consistent development environment. It uses Python 3.11 on Alpine Linux and comes pre-installed with all necessary C dependencies and Python packages (python-pptx, lxml, pytest, etc.).

1. Ensure you have Docker and the Dev Containers extension installed in VS Code.
2. Open the project in VS Code and click Reopen in Container when prompted (or use the Command Palette: Dev Containers: Reopen in Container).

Running Tests

We use pytest for testing. You can run the full test suite with coverage reporting:

pytest tests/

Running the Example

You can use the provided run script to process SVG files in the data/ directory:

python scripts/run.py

License

This project is licensed under the MIT License.