修复 PDF

It rebuilds the file's internal structure, which recovers broken or missing cross-reference tables, wrong object numbers and stream lengths, endings cut off by an incomplete download, and quirks left by buggy PDF generators. If the actual content data is still in the file, it usually comes back.

Files where most of the content bytes are gone, files that aren't really PDFs despite the extension, and encrypted files whose password you don't have. It also can't regenerate image data: a corrupted picture inside an otherwise fixed PDF stays corrupted.

Each file is passed through QPDF, which reads whatever structure it can find, reconstructs the object table, and writes a clean new PDF. The rewritten file downloads as repaired-<original name>.pdf.

Yes, if you know the password. You're asked for it before the repair starts, and the repaired copy is written without encryption, so run it through Protect PDF again if it still needs a password.

Yes. One repaired file downloads directly, and several download together as repaired_pdfs.zip. Files that couldn't be repaired are listed by name in a message, and the rest still download.

Open the file in a web browser such as Chrome or Firefox, which have their own recovery logic, and use the browser's print or save function to write a new PDF. If the file is encrypted, remove the password first with the password removal tool and then repair the result.

No. The page content, fonts, and images are carried over as they are; only the surrounding file structure is rewritten.

No. QPDF runs as WebAssembly inside your browser, so the broken file is read and rewritten on your own device. A file that fails to repair here was never copied anywhere either.