pdfpix

Repair a PDF

or drop one here
Choose a PDF file first.

A PDF that will not open is often a file with a broken cross-reference table rather than lost pages. This tool rebuilds that table in your browser and opens the result again to prove it. If the pages come back, you download them. If they do not, the tool says so instead of guessing.

How to repair a PDF

  1. Choose the damaged PDF

    Press Choose a PDF, or drop one on the page. The file stays in this browser tab.

  2. Repair the file structure

    Repair rebuilds the cross-reference data that tells a PDF reader where each object and page lives.

  3. Wait for the output check

    The repaired file is opened again and its pages, page sizes and selectable text are compared with the readable source.

  4. Download the repaired PDF

    Download appears only after the rebuilt file passes its checks. The original file is never changed.

What a rebuild reaches, and what it cannot

A PDF keeps its pages in numbered objects and keeps a table at the end of the file that records the byte offset of every one of them. A reader jumps to that table first. When the table is wrong, the reader has no map, so it reports a damaged file even though every page is still sitting in the bytes. Rebuilding the table means finding the objects that are really there and writing a fresh, correct index for them.

That is the whole job, and it recovers a large share of files that stopped opening after a bad copy, an interrupted download or an editor that closed part way through a save. It does not recover everything. If the file was cut short and the last bytes never arrived, those pages are gone and no tool can invent them. If a page object survives but the drawing instructions inside it are corrupt, that page can come back blank. If the file is encrypted, the repair has nothing to read until you unlock it with a password you know.

Because those cases exist, the rebuilt file is opened again before any download is offered. The check counts the pages, reads each page size and reads the selectable characters, then compares them with what the source still gives up. When the numbers disagree the page says so and keeps the download closed, rather than handing you a file that looks repaired and is not.

Questions

Does the damaged PDF leave my browser?
No. The PDF, repair and verification stay in this browser tab.
What damage can Repair fix?
It rebuilds broken cross-reference data when the PDF objects are still present.
What can Repair not restore?
It cannot recreate bytes missing from a truncated file, recover a password or infer content that is no longer present.
Can it repair a protected PDF?
Unlock the PDF first using a password you know, then repair the unlocked copy.
What does the repair check cover?
It opens the rebuilt output again, compares its pages, page sizes and selectable text with the readable source, and refuses a failed output.
Why does a PDF reader refuse to open a file?
Most often the cross-reference table at the end of the file points at the wrong byte offsets. That table is the index a reader uses to find every object and every page. Damage it and the objects are still in the file, but nothing can locate them.
Will the repaired file look the same?
The page content is copied without being rewritten, so text, fonts and images stay as they were. The check compares page count, page sizes and selectable characters against the readable source, and blocks the download when they do not match.

The other pdfpix tools

All 30 of them run in a browser tab like this one. Each reads its own output back and checks it before offering the download, and none of them needs an account.

Where to go next