"Compressing a PDF to a Target Size in the Browser", a developer write-up with code from the site's own compressor

In-browser PDF compression involves using JavaScript libraries like PDF.js and pdf-lib to shrink PDF files directly in a web browser. The process breaks down PDF pages into HTML5 canvas elements, which are then rendered at various image quality settings to achieve a target file size.

A binary search algorithm dynamically adjusts image quality and scale to compress the PDF to the desired size. By rendering each page as a JPEG image on the canvas and reassembling the pages into a new PDF, the technique allows precise control over the final file size. The algorithm iteratively refines the image quality and scale until the compressed PDF reaches or falls below the specified target size, such as 2 MB.

Performance optimizations include offloading CPU-intensive tasks to Web Workers and cleaning up DOM objects to prevent memory leaks. Handling hybrid PDFs with both vector text and raster images requires separate strategies for rasterizing images versus rendering vector content. This client-side solution provides privacy, cost savings, and instant processing speed, making it an efficient alternative to server-side compression methods.