imgwrench 0.17.0

A highly opinionated image processor for the commandline.

Image Wrench

A highly opinionated image processor for the commandline. Multiple subcommands can be executed sequentially to form a processing pipeline.

imgwrench is free software available under the MIT license. Detailed documentation can be found at https://imgwrench.readthedocs.io.

Features

• Subcommands can be executed sequentially to form a pipeline

• Command blackwhite for converting images to black and white

• Command collage creates a collage from multiple images

• Command colorfix for fixing the colors of aged photographs

• Command crop for cropping images to give aspect ratio

• Command dither for converting images to black and white and dithering

• Command filmstrip to stack images horizontally forming a filmstrip

• Command flip to flip/mirror images left-right

• Command frame to put a monocolor frame around images

• Command framecrop top frame and crop an image to a target aspect ratio

• Command quad collects four images to a quad

• Command resize for resizing images

• Command save for no processing, but saving images with the given parameters

• Command stack for vertically stacking images

Installation

Make sure you have Python and pip installed and available in your $PATH. Then imgwrench can be installed with

pip install imgwrench

In order to install for the current user only, you may want to execute

pip install --user imgwrench

instead. In this case you will have to ensure that the local bin directory (typically ~/.local/bin on Linux systems) is contained in your $PATH.

Note that legacy Python2 is not supported. If your system still ships Python2 as the default Python interpreter, you may have to execute

pip3 install imgwrench

or

python3 -m pip install imgwrench

Usage

imgwrench is used on the command line by piping file paths into it, e.g.

ls /path/to/my/images/*.jpg | imgwrench blackwhite

Full command line help is

Usage: imgwrench [OPTIONS] COMMAND1 [ARGS]... [COMMAND2 [ARGS]...]...

A highly opinionated image processor for the commandline. Multiple
subcommands can be executed sequentially to form a processing pipeline.

Options:
-i, --image-list FILENAME  File containing paths to images for processing,
                        defaults to stdin

-r, --repeat INTEGER       repeat every image in input sequence  [default:
                             1]

-p, --prefix TEXT          prefix for all output filenames before numbering
                        [default: img_]

-d, --digits INTEGER       number of digits for file numbering  [default: 4]
-c, --increment INTEGER    increment for file numbering  [default: 1]
-k, --keep-names           keep original file names instead of numbering
                        [default: False]

-f, --force-overwrite      force overwriting output image file if it exists
                        [default: False]

-o, --outdir DIRECTORY     output directory  [default: .]
-q, --quality INTEGER      quality of the output images, integer 0 - 100
                        [default: 88]

-e, --preserve-exif        preserve image exif and xmp metadata if available
                        [default: False]

-j, --jpg / --png          save output images in JPEG format (otherwise PNG)
                        [default: True]

--help                     Show this message and exit.

Commands:
blackwhite  Convert color images to black and white.
collage     Create a collage from multiple images.
colorfix    Fix colors by stretching channel histograms to full range.
crop        Crop images to the given aspect ratio.
dither      Apply black-white dithering to images.
filmstrip   Stack all images horizontally, creating a filmstrip.
flip        Flip/mirror images left-right.
frame       Put a monocolor frame around images.
framecrop   Crop and frame an image to a target aspect ratio.
quad        Collects four images to a quad.
resize      Resize images to a maximum side length preserving aspect...
save        No-op to enable saving of images without any processing.
stack       Stacks pairs of images vertically, empty space in the middle.

Pipelines

imgwrench subcommands can be combined into pipelines. This saves you from generating intermediate files cluttering your filesystem or reducing the quality of the final results. For example, if you want to convert all images in the current directory to black and white, put a white frame around them and have them cut to an aspect ratio of 3:2 (for standard format printing), you would execute the following command:

ls *.JPG | \
imgwrench -o out -q 95 -p oldschool_img_ \
        blackwhite \
        framecrop -a 3:2 -w 0.03 -c white

Please refer to the detailed subcommand documentation for the individual parameters.

Developer Notes

Should you run into the following exception while running imgwrench from an editable install

importlib_metadata.PackageNotFoundError: No package metadata was found for imgwrench

try executing make dist to regenerate the egg files required bei importlib which have likely been removed by a call to make clean.

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

History

0.17.0 (2022-11-12)

• -r/--repeat option for repeating input files

• Constrain click version to <8.1 due to breaking API change; this will be relaxed in a future version

• Constrain Pillow version to <9.0 due to breaking tests; this will be relaxed in a future version

• Drop support for Python 3.6 and 3.7

• Add support for Python 3.9 and 3.10

• Upgrade dependencies

• Migrate from travis-ci.com to GitHub actions

0.16.1 (2021-06-19)

• Fix support for click>=8.0 which has changed its behaviour regarding custom parameter types

0.16.0 (2021-01-23)

• quad subcommand supports doubling inner frame using the -d/--double-inner-frame flag

0.15.0 (2021-01-22)

• collage subcommand selects best layout based on score function

• collage subcommand supports -n/--number-tries parameter to specify number of layout tries

0.14.0 (2021-01-21)

• BREAKING CHANGE: replace golden collage approach with BRIC algorithm in collage subcommand

• BREAKING CHANGE: drop support for Python 3.5

• format code with black

0.13.0 (2020-10-26)

• quad subcommand to to collect four images into a quad

• improve documentation

0.12.0 (2020-07-24)

• flip subcommand to flip/mirror images left-right

• Monkey patch IFDRational.__eq__ method of Pillow in tests to avoid regression with Pillow 7.2.0

0.11.1 (2020-04-05)

• -x/--seed option for collage to control initialization of random number generator

0.11.0 (2020-03-21)

• collage subcommand for creating a framed collage from images

• BREAKING CHANGE: default method for colorfix is now quantiles-fixed-cutoff

• preserve xmp metadata when -e/--preserve-exif is used (in addition to exif metadata)

0.10.0 (2020-03-04)

• -m/--method option to colorfix (default: quantiles)

• add fixed-cutoff as new method to colorfix accepting fixed colors as color cutoff boundaries

• add quantiles-fixed-cutoff as new method to colorfix combining quantiles and fixed-cutoff

• deprecate running colorfix without specifying method (as default will change in next version)

0.9.0 (2020-02-19)

• add numpy as dependency

• change colorfix algorithm to vectorized numpy code for performance

• support Python 3.8

0.8.1 (2020-01-12)

• fix crash when orientation is missing in exif headers

0.8.0 (2019-07-10)

• dither subcommand for dithering

• filmstrip subcommand to stack images horizontally

• images can be saved in PNG format using --png CLI flag

0.7.1 (2019-05-16)

• fix development status

0.7.0 (2019-05-16)

• option for preserving exif image metadata

• fix error when running with -k/--keep-names

• status progress to Alpha

0.6.0 (2019-03-14)

• framecrop subcommand to crop and frame an image to a target aspect ratio incl. tests and docs

• breaking change: moved command modules to commands package

• introduced ImageInfo as a container for additional meta information in the pipeline

• increased test coverage

• more documentation

0.5.2 (2019-03-10)

• use a custom parameter type for colors

0.5.1 (2019-03-09)

• changed default frame width to 0.025

• usage doc for frame subcommand

• consistent alphabetic sorting of subcommands

• use a custom parameter type for ratios

0.5.0 (2019-03-07)

• blackwhite subcommand to convert color images to black and white; incl. doc

• frame subcommand to put a monocolor frame around images; incl. tests

0.4.0 (2019-02-26)

• convert RGBA mode PNG images to RGB (to enable saving as JPG)

• crop subcommand to crop images to a specified aspect ratio

• documentation for colorfix and crop

0.3.0 (2019-02-17)

• -d/--digits option to specify number of digits in file names

• -c/--increment option to define increment for file numbering

• create non-existing output folder instead of complaining

0.2.0 (2019-01-30)

• no-op save command for only saving images

• raise exception if output image already exists

• -f/--force-overwrite flag to enable overwriting output

• tests for cli, pipeline and resize

0.1.1 (2019-01-29)

• Fix __main__ module

0.1.0 (2019-01-29)

• First release on PyPI.