Download the PHP package codemasher/image-optimizer without Composer

Image Optimizer

This library is handy and very easy to use optimizer for image files. It uses optipng, pngquant, jpegoptim, svgo and few more libraries, so before use it you should install proper libraries on your server. Project contains Vagrantfile that defines testing virtual machine with all libraries installed, so you can check Vagrantfile how to install all those stuff.

Thanks to ImageOptimizer and libraries that it uses, your image files can be 10%-70% smaller.

Installation

Using composer:

composer require ps/image-optimizer

Basic usage

Configuration

By default optimizer does not throw any exception, if file can not be optimized or optimizing library for given file is not installed, optimizer will not touch original file. This behaviour is ok when you want to eventually optimize files uploaded by user. When in your use case optimization fault should cause exception, ignore_errors option was created especially for you.

This library is very smart, you do not have to configure paths to all binaries of libraries that are used by ImageOptimizer, library will be looking for those binaries in few places, so if binaries are placed in standard places, it will be found automatically.

Supported options:

• ignore_errors (default: true)
• single_optimizer_timeout_in_seconds (default: 60) - useful when you want to have control how long optimizing lasts. For example in some cases optimizing may not be worth when it takes big amount of time. Pass null in order to turn off timeout.
• output_filepath_pattern (default: %basename%/%filename%%ext%) - destination where optimized file will be stored. By default it overrides original file. There are 3 placehoders: %basename%, %filename% (without extension and dot) and %ext% (extension with dot) which will be replaced by values from original file.
• execute_only_first_png_optimizer (default: true) - execute the first successful or all png optimizers
• execute_only_first_jpeg_optimizer (default: true) - execute the first successful or all jpeg optimizers
• optipng_options (default: array('-i0', '-o2', '-quiet')) - an array of arguments to pass to the library
• pngquant_options (default: array('--force'))
• pngcrush_options (default: array('-reduce', '-q', '-ow'))
• pngout_options (default: array('-s3', '-q', '-y'))
• advpng_options (default: array('-z', '-4', '-q'))
• gifsicle_options (default: array('-b', '-O5'))
• jpegoptim_options (default: array('--strip-all', '--all-progressive'))
• jpegtran_options (default: array('-optimize', '-progressive'))
• svgo_options (default: array('--disable=cleanupIDs'))
• custom_optimizers (default array())
• optipng_bin (default: will be guessed) - you can enforce paths to binaries, but by default it will be guessed
• pngquant_bin
• pngcrush_bin
• pngout_bin
• advpng_bin
• gifsicle_bin
• jpegoptim_bin
• jpegtran_bin
• svgo_bin

You can pass array of options as first argument of ImageOptimizer\OptimizerFactory constructor. Second argument is optionally Psr\LoggerInterface.

Supported optimizers

• default (smart) - it guess file type and choose optimizer for this file type
• png - chain of optimizers for png files, by default it uses pngquant and optipng. pngquant is lossy optimization
• jpg - first of two optimizations will be executed: jpegtran or jpegoptim
• gif - alias to gifsicle
• pngquant - homepage
• optipng - homepage
• pngcrush - homepage
• pngout - homepage
• advpng - homepage
• jpegtran - homepage
• jpegoptim - homepage
• gifsicle - homepage
• svgo - homepage

You can obtain concrete optimizer by passing his name to ImageOptimizer\OptimizerFactory::get method:

Custom optimizers

You can easily define custom optimizers:

And then usage:

I got "All optimizers failed to optimize the file"

Probably you don't have required optimazers installed. Let's have a look at Vagrantfile file in order to see an example how to install those commands.

In order to see all intermediate errors, you can use logger (be default NullLogger is used, so logs are not available):

License

MIT