Split PDF

How it works

The document is opened in the browser with pdf-lib and its pages are counted. A range like 1-3, 5 means pages one to three and page five, in that order, with duplicates dropped; the numbers you type count from one while the library counts from zero, so the last page you can name is the page count shown on screen. To extract, the chosen pages are copied into a fresh document; to split fully, that copy is made once per page and each is downloaded in turn, which is why the whole-document mode is capped. A page number the document does not have, a reversed range such as 8-3 or an empty range are each refused with their own message rather than producing an empty file.

Worked example

On a ten-page PDF the range 1-3, 5 gives four pages: one, two, three and five, in that order, because the parser reads 1-3 as three pages and then 5, and has already seen nothing twice. The same range on a five-page document also works, but 1-12 is refused — page 12 does not exist — and 8-3 is refused because the end comes before the start.

Questions and answers