zignal-processing 0.5.0

Zero-dependency image processing library

Zignal Python Bindings

Zero‑dependency image processing primitives in Zig, exposed to Python.

Install

pip install zignal-processing

Docs: https://bfactory-ai.github.io/zignal/python/zignal.html

Quickstart

import zignal

# Load / create
img = zignal.Image.load("photo.jpg")                # PNG/JPEG
img = zignal.Image(480, 640, color=(30, 144, 255))  # RGB fill

# Process
img = img.gaussian_blur(1.5)
img = img.resize(0.5, zignal.InterpolationMethod.BILINEAR)

# Pixels and NumPy interop with shared memory
img[10, 20] = zignal.Hsv(60, 100, 100)
arr = img.to_numpy()                 # (rows, cols, 3) uint8 view
img2 = zignal.Image.from_numpy(arr)  # zero‑copy with shared memory

# Draw
canvas = img.canvas()
canvas.draw_line((10, 10), (100, 60), zignal.Rgb(255, 0, 0))

# Terminal preview (auto: kitty → sixel → sgr)
print(f"{img:auto}")
print(f"{img2:auto}")  # also modified

# Save
img.save("out.png")

Highlights

• Image ops: load/save, resize/warp, crop/letterbox, extract/insert, blur/sharpen, flips
• Pixels: direct get/set; slice assignment; NumPy zero‑copy to/from
• Colors: 12 spaces (Rgb/Rgba, Hsl/Hsv, Lab/Lch, Xyz/Xyb, Oklab/Oklch, Lms, Ycbcr)
• Canvas: lines, circles, polygons, text via bitmap BDF/PCF fonts
• Geometry: Rectangle, ConvexHull, SimilarityTransform, AffineTranform, ProjectiveTransform
• Matrix: float64 matrices with NumPy bridge
• Terminal: SGR, Braille, Sixel, Kitty
• Advanced: FeatureDistributionMatching
• Core: Zig, no external dependencies

Development

• Build native extension: zig build python-bindings
• Generate stubs (.pyi): zig build python-stubs
• Editable install: cd bindings/python && uv venv && uv pip install -e .
• Tests: uv run pytest -q

If Python headers/libs aren’t auto‑detected during build, set: PYTHON_INCLUDE_DIR, PYTHON_LIBS_DIR, PYTHON_LIB_NAME.

License

MIT