Fixing Page Order When Cropping Two-Column Persian PDFs

Some e-book readers render two-column scanned PDFs poorly. Splitting each spread into individual pages fixes readability — but for Persian (right-to-left) books, a naive split leaves the pages in the wrong order.

Credit to Jadi's original post for the convert-based approach — this uses mutool instead, which handles the splitting in one step.

1. Split Each Two-Column Page Into Two Single Pages

sudo apt install mupdf-tools

mutool poster -x 2 haj_sayah.pdf output.pdf

What this does: Splits each two-column spread into two separate single-column pages.

2. Fix the Resulting Page Order

Why it breaks: Persian is written right-to-left, so pages in a Persian book are ordered in reverse relative to a left-to-right book. mutool poster splits pages assuming left-to-right order, which works fine for English books but scrambles the order for Persian ones.

The fix: Swap each even page with the odd page next to it.

sudo apt install pdftk

pdftk output.pdf shuffle even odd output haj_sayyah_.pdf

3. Restore the Original Cover

The shuffle step also reorders the cover. Swap it back from the original file:

pdftk A=haj_sayyah_.pdf B=haj_sayah.pdf cat B1 A3-end output out1.pdf

Result

out1.pdf is the final, correctly-ordered, single-column file — readable page by page on any e-reader.