Skip to content

Latest commit

 

History

History
147 lines (112 loc) · 4.24 KB

README.md

File metadata and controls

147 lines (112 loc) · 4.24 KB
Raw

WebP Python bindings

Build status License PyPI GitHub

Installation

pip install webp

On Windows you may encounter the following error during installation:

conans.errors.ConanException: 'settings.compiler' value not defined

This means that you need to install a C compiler and configure Conan so that it knows which compiler to use. See #20 for more details.

Requirements

  • Python 3.8+

Usage

import webp

Simple API

# Save an image
webp.save_image(img, 'image.webp', quality=80)

# Load an image
img = webp.load_image('image.webp', 'RGBA')

# Save an animation
webp.save_images(imgs, 'anim.webp', fps=10, lossless=True)

# Load an animation
imgs = webp.load_images('anim.webp', 'RGB', fps=10)

If you prefer working with numpy arrays, use the functions imwrite, imread, mimwrite, and mimread instead.

Advanced API

# Encode a PIL image to WebP in memory, with encoder hints
pic = webp.WebPPicture.from_pil(img)
config = WebPConfig.new(preset=webp.WebPPreset.PHOTO, quality=70)
buf = pic.encode(config).buffer()

# Read a WebP file and decode to a BGR numpy array
with open('image.webp', 'rb') as f:
  webp_data = webp.WebPData.from_buffer(f.read())
  arr = webp_data.decode(color_mode=WebPColorMode.BGR)

# Save an animation
enc = webp.WebPAnimEncoder.new(width, height)
timestamp_ms = 0
for img in imgs:
  pic = webp.WebPPicture.from_pil(img)
  enc.encode_frame(pic, timestamp_ms)
  timestamp_ms += 250
anim_data = enc.assemble(timestamp_ms)
with open('anim.webp', 'wb') as f:
  f.write(anim_data.buffer())

# Load an animation
with open('anim.webp', 'rb') as f:
  webp_data = webp.WebPData.from_buffer(f.read())
  dec = webp.WebPAnimDecoder.new(webp_data)
  for arr, timestamp_ms in dec.frames():
    # `arr` contains decoded pixels for the frame
    # `timestamp_ms` contains the _end_ time of the frame
    pass

Features

  • Picture encoding/decoding
  • Animation encoding/decoding
  • Automatic memory management
  • Simple API for working with PIL.Image objects

Not implemented

  • Encoding/decoding still images in YUV color mode
  • Advanced muxing/demuxing (color profiles, etc.)
  • Expose all useful fields

Developer notes

Setting up

  1. Install mamba and conda-lock. The easiest way to do this is by installing Mambaforge and then running mamba install conda-lock.
  2. Create and activate the Conda environment: 
    $ conda-lock install -n webp
$ mamba activate webp
  3. Install PyPI dependencies: 
    $ pdm install -G:all

Running tests

$ pytest tests/

Cutting a new release

  1. Ensure that tests are passing and everything is ready for release.
  2. Create and push a Git tag: 
    $ git tag v0.1.6
$ git push --tags
  3. Download the artifacts from GitHub Actions, which will include the source distribution tarball and binary wheels.
  4. Create a new release on GitHub from the tagged commit and upload the packages as attachments to the release.
  5. Also upload the packages to PyPI using Twine: 
    $ twine upload webp-*.tar.gz webp-*.whl
  6. Bump the version number in pyproject.toml and create a commit, signalling the start of development on the next version.

These files should also be added to a GitHub release.

Known issues

  • An animation where all frames are identical will "collapse" in on itself, resulting in a single frame. Unfortunately, WebP seems to discard timestamp information in this case, which breaks webp.load_images when the FPS is specified.
  • There are currently no 32-bit binaries of libwebp uploaded to Conan Center. If you are running 32-bit Python, libwebp will be built from source.