Repair PDF
Recover the readable pages of a damaged PDF into a new, clean file that opens again.
Your damaged PDF is repaired inside this page and never uploaded.
What this tool does
To repair a damaged PDF, choose the file and let the tool recover it. It first tries to copy the readable pages into a clean new document, keeping their text. If the structure is too broken for that, it redraws each page as an image so the file opens again. It cannot bring back data that is truly gone.
Parses the damaged file leniently and copies each readable page into a fresh PDF, one page at a time, so a single bad page does not lose the rest.
If that fails, renders every page with pdf.js, which is more forgiving, and builds a new PDF from the images.
How to use it
- Choose the PDF that will not open properly.
- Wait while the tool tries the first repair method.
- If needed, it automatically tries rebuilding the pages as images.
- Read the message to see which method worked and how many pages were recovered.
- Press Download repaired PDF and open it to check.
Why use this PDF repair tool
Tries a clean re-copy first, which keeps your text and graphics, and only falls back to redrawing pages as images when the file structure is too damaged to read. PDFs get damaged by interrupted downloads, failing disks, email gateways and software bugs, and often only the index at the end of the file is broken while every page is still intact. Rebuilding that structure recovers the document.
Why we built it
We built the repair PDF tool to try a clean re-copy first, which keeps text and graphics, and fall back to redrawing pages as images only when needed. Check the result with the PDF metadata viewer and shrink it with compress PDF.
If your task goes further, try the Unlock PDF.
Who this tool is for
- People whose PDF will not open
- Office staff recovering damaged downloads
- Students with corrupted submissions
When to use it
- A PDF will not open or shows an error about the file being damaged.
- A download was interrupted partway.
- A file opens but pages are blank or missing.
- A PDF works in one reader but not in another.
How it works
A PDF ends with an index (the cross-reference table) that says where each object lives. When it is damaged, strict readers give up even though the page data is intact. The first method ignores the broken index, reads what it can, and writes a new file with a fresh index.
If the page data itself is too broken to copy, the second method asks pdf.js - which can reconstruct a file by scanning it - to draw each page, then stores those pictures in a new PDF.
No repair can recreate content that is missing from the file. If the download was cut short, the pages at the end may simply not exist.
What you give it, and what you get back
| Item | Details |
|---|---|
| Input | A PDF that will not open or displays errors. |
| Method used | Re-copy of pages (keeps text), or page images (text lost). |
| Output | A new PDF made from the pages that could be recovered. |
| Report | Pages recovered, original size and repaired size. |
Issues you may run into
| What you see | Why it happens | What to do |
|---|---|---|
| No pages could be recovered | The page data is missing or the file is not really a PDF. | Try to get a fresh copy of the file. |
| Does not look like a PDF | No PDF header was found. | It may be another file type renamed to .pdf. |
| Text not selectable afterwards | The image fallback was used. | That is expected - the structure was too damaged to keep text. |
| Fewer pages than before | Some pages were unreadable. | Those pages are skipped; the rest are kept. |
Assumptions and limitations
- Cannot recover content that is not in the file.
- The image fallback loses selectable text, links and bookmarks.
- Password-protected PDFs must be unlocked first.
- Results depend on how and where the file is damaged.
Frequently asked questions
How do I repair a corrupted PDF?
Choose the file. The tool recovers the readable pages into a new, clean PDF and tells you how it did it. Download and open the result.
Why will my PDF not open?
Its internal index is often damaged by an interrupted download, a disk fault or an email gateway. The page data may still be intact and recoverable.
Will I lose any content?
Pages that are unreadable are skipped, and the image fallback loses selectable text. Pages whose data is intact are kept.
Is my file uploaded to repair it?
No. The repair runs in your browser and the file never leaves your device.
What if it says no pages could be recovered?
The page data itself is missing, so no repair can bring it back. Try to obtain the file again from its source.