Skip to content
Your files stay on your device

File processing runs in your browser. Documents are not uploaded to our servers. The site may download processing engines and send usage or feedback events without file contents.

How we handle your data

Repair PDF

Rebuild a PDF record with a corrupted cross-reference table or objects into a consistent copy, keeping the original untouched. Files stay in your browser.

Small converter download on first use.

One PDF, up to 20 MiB. The original file stays unchanged.

Try an example

Download sample

The sample has a broken cross-reference table. The repaired copy should open normally with both pages and Total: 148.50.

What to expect

QPDF reads the file, rebuilds a broken or missing cross-reference table, skips unreadable objects where it can and writes a new, consistent PDF. Your original file stays unchanged. Content that is missing or overwritten in the damaged file cannot be restored, so check every page of the repaired copy.

One PDF up to 20 MiB; output up to 30 MiB. Processing stops after 30 seconds. Password-protected PDFs must be unlocked first. Rewriting a PDF can invalidate digital signatures.

Encrypted file?Unlock PDF with a known password first.

The first run may download converter files and take a moment. If it fails, check your connection or try a smaller file. Cancel stops active processing; you can choose another file and retry.

Engine licenses and source code

A practical guide

Repair corrupted PDF records

  1. Select the corrupted PDF record.
  2. Repair it to reconstruct the cross-reference table and objects.
  3. Verify page count and key values against a known source.

Try this example

Repair a 30-page test log PDF whose xref table was truncated; confirm the repaired copy reports 30 pages and the final results table.

Before you start: Rewriting invalidates digital signatures and overwritten data cannot be recovered; unlock encrypted records first and keep the original.