FaB Boost for Images is a lightweight, efficient image processing microservice designed to simplify remote image hosting for client sites. It provides a straightforward, URL-based API for hassle-free image transformations.

• On-the-fly image processing
• Secure domain whitelisting
• Image caching
• Support for multiple image formats (JPG, PNG, GIF, BMP, WebP, AVIF)
• Configurable image manipulations
• Cache-Control headers for optimal performance
• File size and dimension limits for security

• PHP 7.4
• Composer
• Required PHP extensions:
  • GD or ImageMagick
  • curl
  • fileinfo
• Composer packages:
  • league/glide
  • nyholm/psr7
  • laminas/laminas-httphandlerrunner

1. Clone the repository:

git clone https://github.com/sleggat/fab-boost-for-images.git

2. Install dependencies:

composer install

3. Create required directories:

mkdir cache_local cache_remote
chmod 777 cache_local cache_remote

4. Configure your web server (example Nginx configuration included)

1. Update the $whitelistedDomains array in image.php to include your allowed domains:

$whitelistedDomains = [
    'your-domain.com',
    'otherdomain.com'
];

2. Adjust size limits if needed:

$maxFileSize = 10 * 1024 * 1024; // 10MB default
// Maximum dimensions
if ((isset($glideParams['w']) && $glideParams['w'] > 5000)...

The service accepts URLs in the following format:

https://your-domain.com/image/example.com/path/to/image.jpg

Add image manipulation parameters as query strings:

• w: Width
• h: Height
• q: Quality (0-100)
• blur: Blur effect
• sharp: Sharpening
• fm: Format conversion
• crop: Crop mode
• bri: Brightness
• con: Contrast
• gam: Gamma
• flip: Flip image
• or: Orientation
• bg: Background colour

Example:

https://your-domain.com/image/example.com/image.jpg?w=800&h=600&q=80

The service currently includes several security measures:

• Domain whitelisting
• File size limits (10MB default)
• Dimension limits (5000px default)
• File type verification
• Input sanitization
• Secure file handling

Images are cached in two layers:

1. Original downloaded images in the cache_remote/ directory
2. Processed images in the cache_local/ directory

1. This is a fairly new project and I wouldn't recommend using this in production, yet.
2. You might want to implement more rigorous checks or disable CPU intensive processes like blur.

Contact me at [email protected] https://frontandback.co.nz