How to use the Repair PDF tool
- Add the damaged PDFDrop the file onto the box or click to choose it.
- Automatic diagnosisThe header, end-of-file marker and readability of every page are checked.
- RepairThe file structure is rebuilt. If that’s not enough, the readable pages are recovered one by one.
- DownloadDownload the repaired PDF.
What it fixes
- Broken cross-reference tableThe index that tells readers where each object is gets rebuilt from scratch.
- Cut-off downloadsFiles with a missing end and trailer, from interrupted downloads or copies, are completed.
- Extra dataJunk before the PDF header (for example from badly saved e-mail attachments) is removed.
- Unreadable objectsDamaged objects are skipped so the rest of the document can still be used.
- Page recoveryAs a last resort, every page that can be displayed is saved into a new PDF, with searchable text.
- PrivateRepairing happens in your browser. Nothing is uploaded.
Why PDFs get damaged
A PDF is a collection of numbered objects (pages, fonts, images) plus a table at the end of the file that lists where each object starts. If a download is interrupted, a USB stick is removed too early, an e-mail system re-encodes an attachment, or a program crashes while saving, that table or the end of the file is often missing or wrong. Readers then show messages such as “There was an error opening this document”, “The file is damaged and could not be repaired” or simply a blank page.
How the repair works
- Diagnosis. The tool checks for the %PDF header (and removes any junk before it), the %%EOF end marker, and tries to open the file and read every page.
- Rebuilding the structure. The file is read object by object, without relying on the broken table, and written out again with a new, correct cross-reference table and trailer. Text, fonts, images and links are kept exactly as they were. The repaired file is then opened again to make sure every page can be read.
- Page recovery. If the structure can’t be fixed, a second engine with its own recovery logic opens the file and renders every page it can read. These pages are saved into a new PDF as high-quality images with an invisible text layer, so you can still search and copy the text. Pages that can’t be read are listed.
If the repair doesn’t work
- A file that is mostly empty, or only a few kilobytes when you expect megabytes, was probably cut off. The missing part can’t be recovered; download or copy the file again from the original source.
- If the file is password-protected, unlock it first with Unlock PDF (you need the password).
- Files that are not really PDFs (for example a Word document renamed to .pdf) can’t be repaired. Try opening them with the right program, or convert them with Word to PDF.