call_split Split PDF — Extract Pages star

How It Works

Splitting a PDF means extracting specific page objects from the source document's internal structure and writing them into a brand-new PDF file, using the open-source pdf-lib library running entirely in your browser. Each page keeps its own content stream and any resources it references — fonts, images, vector graphics — copied faithfully, so text stays selectable and images keep their original compression exactly as they were in the source.

There is a subtlety worth knowing here: a multi-page PDF commonly shares resources across pages — the clearest example is an embedded font used for body text throughout the whole document, which the original file stores once and every page simply references. When you split that document into several separate output files, page extraction does not deduplicate shared resources across the new, independent files — each output file is a standalone PDF and therefore needs its own complete copy of every font, image or resource its pages use. The practical result: the combined size of all the split files can end up larger than the original single file, because a font that was stored once is now stored once per output file that uses it.

Example: a 50-page report with one 2 MB embedded font, split into 10 separate 5-page files, could turn into ten files that each carry their own copy of that font — roughly 20 MB of font data total instead of the original 2 MB, on top of the actual page content. This is normal PDF behavior, not a bug, and it is the same trade-off any splitting tool makes.

What to Know

• Range syntax: type page numbers and ranges like "1-3, 7, 12-14" — commas separate items, hyphens define ranges, and the output PDF contains exactly those pages in exactly that order.
• Repeats and reversed ranges work: listing a page number twice duplicates it in the output; a range written backward (e.g. "5-3") extracts those pages in reverse order.
• Total size can grow: because shared resources like fonts get copied into every output file independently, do not be surprised if several split files together outweigh the original — this is expected, not a sign of corruption.
• Nothing is re-rendered: pages are copied as structured objects, not repainted as images, so there is no quality loss regardless of how many times you split and re-split a document.