Scanned PDFs on macOS

Translate a scanned PDF
without opening the file.

Press ⌥A, drag a box around the line, and macOS reads it off the screen. The translation comes back next to the original, and any word in it can be looked up on its own and saved with the sentence. Scanned pages, two column papers and locked files all behave the same way, because nothing ever opens the document.

01 · Why it happens

Why the text will not select

A scanned book is a stack of photographs. There is no text layer under those letters, so your cursor slides across the page and selects nothing.

Academic papers are their own kind of hostile. Two columns, and dragging across one line drags in the neighbouring one, plus a header, plus a footnote marker. You end up pasting a sentence that was never in the paper.

Then there are the files that are simply locked, and the ones where selecting works but every space comes out in the wrong place.

None of this is your fault, and none of it is going to be fixed by the person who made the file.

02 · The usual workaround

Retyping it by hand

You squint at the word, type it into a dictionary, get it slightly wrong, try again. Thirty seconds gone. By the time you are back on the page you have lost the thread of the paragraph, which was the thing you were actually doing.

The other option is uploading the whole file to a website that promises free OCR and free translation. For a library book that is fine. For a contract, a medical record or an unpublished paper, it is not.

03 · How to do it

Read the screen, not the file

Leafy never opens your PDF, so the file format stops mattering.

Press ⌥A and drag a box around what you want. macOS text recognition reads that patch of screen, and the translation comes back in a small popup next to the line it came from.

When one word is the problem rather than the whole sentence, drag over it inside that popup. You get it explained in the sense this sentence is using, and it lands in your library with the line still attached.

Scanned page, locked file, two columns, a screenshot somebody sent you, a preview thumbnail. All the same to it. Your reader stays open at the same scroll position and you keep going.

04 · Which readers

Which apps does this work in?

Preview, Skim, Zotero, Adobe Acrobat, a PDF sitting in a browser tab, Books, a Kindle app window. Leafy is not an extension and not a reader, so there is nothing to install into any of them.

The same goes for the things next to your PDF. Ebooks, images, slides, a terminal, anything with letters on it.

05 · The limits

What it will not do

It will not turn a 300 page book into a translated document. Leafy works a line at a time, at the point where you got stuck, because it was built for reading rather than for bulk conversion. If you need the whole thing rendered, use a tool made for that and let it grind.

Handwriting is a coin flip. Very small type in a bad scan is worse. Zoom the page in first and the recognition gets noticeably better, which is annoying advice but it is true.

06 · The honest part

Where the text goes

Your library lives on your Mac. No account, no sync server, no ads. The page image is read on your machine and never leaves it, so nothing gets uploaded to a website.

The text you scan does go out to an AI so it can explain the word as your sentence uses it, with nothing attached that says who you are. The privacy page spells out exactly what is sent and what is kept.