LatticeSVG¶

Declarative Vector Layout Engine Powered by CSS Grid

Describe layouts with Python dicts → Engine handles measuring, typesetting & rendering → Pixel-perfect SVG / PNG output

Example output — LatticeSVG rendering preview

CSS Grid Layout¶

Full CSS Grid Level 1 implementation — fixed tracks, fr flexible units, minmax(), repeat(), named areas, auto-placement algorithm, all at your fingertips.

Precise Text Typesetting¶

FreeType-based glyph-level measurement with automatic line breaking, CJK typesetting, justification, rich text markup, vertical writing, and auto-hyphenation.

Multiple Node Types¶

TextNode, ImageNode, SVGNode, MplNode (Matplotlib), MathNode (LaTeX formulas) — freely combine and nest them.

Complete Visual Styling¶

63 CSS properties: box model, border-radius, border styles, gradient backgrounds, shadows, transforms, filters, clip-path, opacity.

Declarative API¶

All configuration via Python dict with CSS-compatible property names. No new DSL to learn — your CSS knowledge transfers directly.

SVG & PNG Output¶

Vector SVG by default, optional high-resolution PNG via CairoSVG. Supports WOFF2 font subsetting and embedding.

Minimal Example¶

from latticesvg import GridContainer, TextNode, Renderer

# 1. Create a Grid container
page = GridContainer(style={
    "width": "600px",
    "padding": "24px",
    "grid-template-columns": ["1fr", "1fr"],
    "gap": "16px",
    "background-color": "#ffffff",
})

# 2. Add child nodes
page.add(TextNode("Hello", style={"font-size": "24px", "color": "#2c3e50"}))
page.add(TextNode("World", style={"font-size": "24px", "color": "#e74c3c"}))

# 3. Render output
Renderer().render(page, "hello.svg")

Installation¶

pip install latticesvg

# For PNG output
pip install latticesvg[png]

# For auto-hyphenation
pip install latticesvg[hyphens]

System Dependencies

LatticeSVG requires the FreeType library installed on your system. Most Linux distributions and macOS come with it pre-installed. Windows users please see the Installation Guide.

Project Status¶

Metric	Data
Version	v0.1.0
License	MIT
Python	≥ 3.8
Core Code	~8,900 lines
Tests	352 test functions
Demos	50 example scripts