A PDF that won't open anymore, shows an error or displays blank pages? PDFRoll automatically tries several repair methods, from the most faithful to the most robust, to recover as many pages as possible.
How it works
- Select the damaged PDF.
- Click “Repair PDF”.
- Download the repaired version.
Frequently asked questions
Can every PDF be repaired?
No: if the data is too damaged, only some pages can be recovered, sometimes as images. If no page is readable, PDFRoll tells you.
Is the original file modified?
No, PDFRoll creates a new, repaired copy.
Is the text still selectable after the repair?
Yes, in most cases. Pages recovered as a last resort become images: PDFRoll keeps the text that's still readable, and the “OCR PDF” tool can fill in the rest if needed.
Why are some pages missing?
Pages that remain unreadable are removed. For a truncated file, the pages at the end of the file can't be recovered.
My PDF asks for a password. Is it damaged?
No, it's protected. Use “Unlock PDF” with the password rather than the repair tool.
Can I repair a PDF on my phone?
Yes, the tool works in your phone's or tablet's browser, for free and with no sign-up. A large, badly damaged file may take a little longer.
Are my files sent to a server?
No. Everything happens in your browser: your documents never leave your device.
Four repair methods, tried one after another
There's nothing to set up: click “Repair PDF” and PDFRoll moves from one method to the next until the file is recovered.
- Rebuilding the file structure: the table that shows where each element is located is recalculated. This method preserves the original document best.
- Object-by-object rereading: the file is reread element by element, without relying on its damaged index table.
- Recovering an incomplete file: for an interrupted download or a truncated copy, the complete pages are reconstructed.
- Page-by-page recovery: as a last resort, each page that's still readable is redrawn as an image.
After the repair, pages that remain unreadable are removed and a message tells you how many pages were recovered.
Typical situations
- An attachment that shows “damaged file” or “unsupported format”.
- An interrupted download, a USB drive transfer or a sync that went wrong.
- A PDF generated by unreliable software, which opens in one reader but not in another.
- A file rejected by another tool, for example when you try to merge or compress it.
The repair happens in your browser: a damaged document stays private, even if it contains personal information.
What to know about the result
When the structure can be rebuilt, the result is identical to the original, selectable text included. If the file is truncated, the missing pages at the end can't be recovered. With the last method, pages are rebuilt as images; when a page's text is still readable, PDFRoll keeps it as an invisible layer so it stays selectable. If not, make it searchable again with “OCR PDF”.
If the file is simply password-protected, it isn't damaged: use “Unlock PDF”. And if no page can be recovered, the data is too damaged: try to find another copy of the document, for example in the original email or folder.