variations

A simple interface that allows processing of images.

Compatibility

python >= 3.7

Installation

Run pip install variations
(optional) If you want to use StackBlur

pip install pillow-stackblur
(optional) If you want to use Face Detection

pip install face_recognition

Usage

from PIL import Image
from variations import processors
from variations.variation import Variation
from variations.utils import prepare_image

variation = Variation(
    size=(400, 0),
    max_height=800,
    clip=False,
    upscale=False,
    anchor=processors.Anchor.TOP_LEFT,
    jpeg=dict(
        quality=92,
    ),
    webp=dict(
        lossless=True,
        quality=90,
    ),
    postprocessors=[
        processors.ColorOverlay('#FF0000', overlay_opacity=0.25),
    ],
)

img = Image.open('source.jpg')
img = prepare_image(img, draft_size=variation.get_output_size(img.size))
new_img = variation.process(img)
dest_path = variation.replace_extension('dest.jpg')
variation.save(new_img, dest_path)

Options

	Type	Examples	Description
size	tuple list	`(640, 480)` `(640, 0)`	The canvas size of image. If you set the width or height to zero, the corresponding value will be automatically adjusted based on the aspect ratio
max_width	int	`640`	It specifies the maximum width in pixels.This option have meaning only when corresponding value in `size` is zero
max_height	int	`480`	It specifies the maximum height in pixels.This option have meaning only when corresponding value in `size` is zero
clip	bool		When set to `True`, the image can be cropped when filling the canvas.
upscale	bool		When set to `True`, the image can be upscaled when filling the canvas.
anchor	str tuple list	`'tr'` (top right) `'c'` (center) `(1, 1)` (bottom right)	Defines the anchor point.
face_detection	bool		Use a face detection system to find anchor point. You must install facial recognition api to use this.
format	str	`'JPEG'` `'png'` `'WebP'`	Enforce output image format. Defaults to `'AUTO'`, which means keep input format.
preprocessors	list	`[processors.Crop(width=200, height=120, x=50, y=50)]`	PilKit processors are invoked before the main processing stage
postprocessors	list	`[processors.ColorOverlay('#0000FF', 0.10)]`	PilKit processors are invoked after the main processing stage

Additional options for specific formats

It is possible to pass additional options to Image.save() method.

Variation(
    # ...
    jpeg=dict(
        quality=80,
        progressive=True,
    ),
    webp=dict(
        autoconvert=False,
        quality=80,
    ),
    tiff=dict(
        compression='tiff_jpeg',
    )
)

Development and Testing

After cloning the Git repository, you should install this in a virtualenv and set up for development:

virtualenv .venv
source .venv/bin/activate
pip install -r ./requirements.txt
pre-commit install

Metadata

Release files for variations 0.3.3

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for variations 0.3.3
File	Size	Uploaded
variations-0.3.3.tar.gz	19.4 kB	Apr 25, 2023	Details

Built distribution (wheel)

Table of built distributions (wheels) for variations 0.3.3
File	Interpreter	ABI	Platform	Reset
variations-0.3.3-py2.py3-none-any.whl 15.7 kB Apr 25, 2023	Python 2, Python 3	none	any	Details

Total release size: 35.1 kB

Release files / variations-0.3.3.tar.gz

Download URL	variations-0.3.3.tar.gz
Size	19.4 kB
Tags	Source
SHA-256 checksum How to use checksums	`0d5e0abf74b1ffca27afdbdf3595f9cb0b6bc0c70616c83d4a3235830d05af80`
BLAKE2b-256 checksum How to use checksums	`6c10746e6cfccff1fa2a29140bb071f34241321d4800e733e909a0c4285f9400`
Upload date	Apr 25, 2023
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/4.0.2 CPython/3.10.6`

Release files / variations-0.3.3-py2.py3-none-any.whl

Download URL	variations-0.3.3-py2.py3-none-any.whl
Size	15.7 kB
Tags	Python 2 Python 3
SHA-256 checksum How to use checksums	`b30b10e979bc232060953cf17efd8f9af11f7efae4523b81bfb1b77a7bc52697`
BLAKE2b-256 checksum How to use checksums	`f99743d7d5ce691dd8194285f973529eeddd4659a8e581f6a432068b481910f0`
Upload date	Apr 25, 2023
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/4.0.2 CPython/3.10.6`