Delete Blank Pages

Detecting a blank page in a PDF is surprisingly difficult. Unlike a raster image where you can just check if every pixel is white, a PDF page is a complex stream of rendering instructions.

This tool uses a multi-layered heuristic approach to scan the PDF object stream:

• Operator Counting: We parse the page's contents and count the rendering operators. If the only operators present are structural (like q and Q for saving/restoring graphics states) but no text drawing (Tj) or path painting (S, f) operators exist, the page is likely blank.
• Invisible Text: Sometimes software generates "blank" pages containing invisible spaces or white text on a white background. We extract text streams and strip out whitespace to calculate meaningful content length.
• Annotation Checks: A page might have no text or paths, but contain a visible Sticky Note annotation or a Form Field. The algorithm scans the /Annots dictionary to ensure no interactive elements exist before declaring it blank.

Parsing the raw instruction stream of every page in a 1,000-page document requires massive CPU cycles. We offload this heavy parsing logic to a Web Worker running compiled WebAssembly, keeping the UI perfectly smooth.