Skip to main content

Python bindings for Wuffs the Library

Project description

pywuffs: Python bindings for Wuffs the Library

Linux CI Windows CI macOS CI Downloads

This project is intended to enable using Wuffs the Library from Python code. For now, it only provides bindings for image and JSON decoding parts of the Auxiliary C++ API as being of the most interest since it provides for "ridiculously fast" decoding of images of some types.

Current version of Wuffs library used in this project is unsupported snapshot taken from this tag. The primary rationale for using the snapshot version instead of a stable release is that it provides JPEG decoder.

Installation

Using pip

python3 -m pip install pywuffs

Using CMake

CMake build support is mostly intended for development purposes, so the process might be not so smooth.

Building the Python module using CMake requires pybind11 library to be installed in the system, for example in Ubuntu it can be installed like this:

sudo apt install pybind11-dev

Linux

mkdir _build && cd _build
cmake ..
make

Windows

mkdir _build && cd _build
cmake -A x64 ..
cmake --build .

Usage example

The example below demonstrates how to decode a PNG image and its EXIF metadata:

from pywuffs import ImageDecoderType, PixelFormat
from pywuffs.aux import (
    ImageDecoder,
    ImageDecoderConfig,
    ImageDecoderFlags
)

config = ImageDecoderConfig()

# All decoders are enabled by default
config.enabled_decoders = [ImageDecoderType.PNG]

# No metadata is reported by default
config.flags = [ImageDecoderFlags.REPORT_METADATA_EXIF]

# Pixel format is PixelFormat.BGRA_PREMUL by default
config.pixel_format = PixelFormat.BGR

decoder = ImageDecoder(config)

decoding_result = decoder.decode("lena.png")

# Decoded image data in BGR format
image_data = decoding_result.pixbuf

# Shape of the decoded image
image_shape = decoding_result.pixbuf.shape

# Parsed EXIF metadata
meta_minfo = decoding_result.reported_metadata[0].minfo
meta_bytes = decoding_result.reported_metadata[0].data.tobytes()

API reference

API documentation is available at https://pywuffs.readthedocs.io.

Implementation goals

  1. Bindings are supposed to be as close as possible to the original C and C++ Wuffs API. The differences are only justified when it's hardly possible to transfer the API entries to Python as is.
  2. Bindings are not supposed to add much overhead. Because of that some parts of the API are not as convenient as they expected to be.

Roadmap

  1. Bindings for other parts of wuffs_aux API (CBOR decoding).
  2. Bindings for the C API of Wuffs the Library.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pywuffs-2.0.0.tar.gz (422.4 kB view details)

Uploaded Source

File details

Details for the file pywuffs-2.0.0.tar.gz.

File metadata

  • Download URL: pywuffs-2.0.0.tar.gz
  • Upload date:
  • Size: 422.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.10.12

File hashes

Hashes for pywuffs-2.0.0.tar.gz
Algorithm Hash digest
SHA256 948b5649ef79c5d3d50231ff01e483d7791c543493c4bf76afeeb0229a0540a7
MD5 6438ba6a05745165e919fa4613e0e33e
BLAKE2b-256 16bfe2af56244ec05fb3711c0927d2dd4359581844642b0325d31a87ed5f92fd

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page