Repair a damaged PDF
Recover the pages of a PDF that says it is damaged and cannot be opened
Your files stay with you. Conversion happens inside the browser; no file is ever uploaded to a server.
Did this tool do the job?
Thanks, your feedback came through.
How it works
Tap or click the box to select the PDF that will not open or says it is damaged, then press "Repair". A typical case: an invoice you downloaded from an email, or a thesis copied from a USB stick, makes Acrobat say "There was an error opening this document. The file is damaged and could not be repaired." The tool first tries a structural repair. It ignores the table of contents at the end of the PDF (the cross-reference table), scans the file from start to finish, finds every object where it really is, rebuilds the page list and writes the file with a fresh table. Text stays selectable, and links and fonts are kept. If the structure cannot be saved, the second route renders every readable page as a 150 DPI image and builds a new PDF from them; text is then not selectable. The result line names the method, the number of pages recovered and the problems found, such as "the cross-reference table was broken (12 of 37 entries pointed to the wrong place)". Pages missing from an incomplete download cannot be brought back.
This tool is also known as repair pdf, fix corrupted pdf, pdf file is damaged and could not be repaired, recover pdf file, pdf won't open, repair pdf online.
What is Cross-reference table (xref)?
The cross-reference table is a directory at the end of a PDF that lists the exact byte position of every object in the file: pages, fonts, images. Reader apps look at it first and jump straight to what they need. When a file is copied only partly, a program crashes while saving or an email attachment gets mangled, the table points to the wrong places and the reader reports the file as damaged, even though the pages themselves are often intact. This tool ignores the table, finds each object by scanning, and writes a new, correct table.
What is PDF page tree?
The page tree is the structure that keeps a PDF's pages in order: the document catalogue points to a root node, the root to child nodes, and those to the individual pages. Inherited settings live there too, so a page size or a font list can be written once on a parent node and apply to all pages below it. If a node breaks or the root is lost, the pages are still in the file but the reader cannot reach them, and the document opens blank or incomplete. When the tool detects a broken tree, it collects every page object in the file, copies inherited settings down to each page and rebuilds the tree.
What is the difference between structural repair and recovering pages as images?
Structural repair fixes the file's internal layout and leaves the content alone: text stays selectable and searchable, and fonts, links, form fields and file size stay about the same. Recovering pages as images turns every readable page into a picture inside a new PDF; the page looks identical, but text cannot be selected, the file gets bigger and form fields are lost. The tool always tries the structural repair first and switches to images only if that fails. The result line tells you which method was used.
What can be repaired, and what cannot?
Repairable: a cross-reference table that points to wrong places or is missing, a wrong table address at the end of the file, a missing end-of-file marker, a lost or looping page list, and half-written update sections left by a crash while saving. In these cases the real content is fine but the reader cannot find it, and that is behind most "file is damaged" errors.
Not repairable: parts of an incomplete download that were never written (those pages arrive blank or not at all), corrupted bytes in the middle of a compressed stream (that page is recovered partly or fully through the image route), and a file that is not a PDF at all but was saved with a .pdf extension. A password-protected file must be unlocked first with the Unlock PDF tool.
What to try before repairing
Open the file in another reader first: the PDF viewers in Chrome and Edge are more forgiving than Acrobat, and if the file opens there, printing it to "Save as PDF" also rewrites it. If the file came from an email or a website, download it again; an interrupted download is smaller than expected, and no tool can recreate the missing part.
If the file is on a USB stick or memory card, copy it to your computer first and repair the copy; working on a failing drive can make things worse. After repairing, open the file and check the page count and the last pages. The result line says how many pages, if any, were left blank because their content was missing.
Frequently asked questions
How do I open a PDF that says it is damaged?
Try Chrome or Edge first. If the file still will not open, select it on this page and press "Repair". The tool rebuilds the internal table and page list, which fixes most "file is damaged" errors, and the repaired file opens in any reader.
Why can I not select text in the repaired PDF?
The structural repair failed, so the tool recovered the pages as images; the result line says "method: image". The pages look the same but are pictures. To get the text back you need text extraction with OCR.
Can a partly downloaded PDF be fully recovered?
No. The part that was never downloaded is not in the file. The tool recovers the pages that did arrive and reports how many were left without content. Downloading the file again from its source is the only complete fix.
Is my file uploaded?
No. Scanning, repairing and, if needed, rendering pages as images all happen inside your browser; the damaged file never leaves your device. Once the page has loaded, the tool works even offline.