PDF to Text: extract text from a PDF

Because a PDF stores the order things were DRAWN, not the order they are read. On a two-column page a typesetter usually emits the whole left column and then the whole right column, so a naive extractor gives you one column after the other with every word present and nothing readable. PDFMacro groups items onto lines by their baseline and sorts each line left to right before joining, so columns come back interleaved by row the way you read them.

Keep them if a person is going to read the file: the page's own line breaks are what makes it look like the document. Reflow them if something else is going to read it — a search index, a diff, a language model — because a hard wrap at column width is an artefact of the page layout, and a sentence split across three lines stops matching as a sentence.

Almost certainly nothing went wrong: the document is scanned, so its pages are images with no text layer to extract. PDFMacro tells you which pages produced nothing rather than handing back a silently empty file, and the fix is to run OCR first, which adds a real text layer the extractor can then read.

No. The file is opened, processed and saved back by code running in this browser tab — there is no upload endpoint on our servers to send it to. You can watch the Network panel in your browser's developer tools while you work and confirm nothing carrying the document leaves the page.

PDF to Text is free, with no account required and no page or file-size caps imposed by the plan. There is no page cap and no watermark on the output.

Yes. Once the page has loaded, every tool keeps working with the network disconnected, because the processing never depended on a server round-trip. AI features that use an on-device model need one download the first time; after that they run offline too.