Softonic

Repair PDF

Step 1 / 3
Next

A PDF that will not open often has only structural damage: a broken cross-reference table, a mismatched or leftover trailer, or object offsets that no longer line up after a bad save. This tool opens the file with a tolerant parser, keeps every page it can read, and writes a brand-new PDF from scratch, so the cross-reference table and trailer are regenerated cleanly. It cannot invent data that is genuinely gone, but it fixes the surprising number of PDFs whose pages are intact and whose only problem is a broken structure. Everything runs in your browser, so the file never leaves your device.

How the repair works

  1. 1

    Upload the PDF

    Drag and drop the file that will not open. It stays on your device; nothing is sent to a server.

  2. 2

    The tool re-reads it

    A tolerant PDF engine opens the file, ignoring the broken structure, and reads every page it can recover.

  3. 3

    A clean file is rebuilt

    The pages are copied into a brand-new document, so the cross-reference table and trailer are written from scratch and are valid by construction.

  4. 4

    Download the result

    You get a fresh PDF that opens cleanly. Open it in your reader to confirm the pages are all there.

What this repair can and cannot fix

The tool works by re-reading the PDF and writing a completely new file. That regenerates the structure around your pages, which fixes structural faults but cannot bring back data that is truly missing.

Problem Fixed? Why
Broken or mismatched cross-reference table Yes The new file is written with a fresh, correct xref
Stale or leftover trailer after a bad save Yes A clean trailer is generated for the rebuilt file
Cruft from repeated incremental saves Yes Only the current pages are kept; old revisions are dropped
Objects a reader rejects but the parser can read Often Re-serialising them in a clean file usually satisfies strict readers
File truncated so the parser cannot open it No If the engine cannot parse the file, there is nothing to rebuild
Object data genuinely missing or unreadable No The tool keeps what it can read; it cannot invent the rest
Encrypted with an unknown password No The content stays locked without the password

Why PDF structures break

  • Partial downloads or uploads. A file that lost its last bytes in transit loses the cross-reference table that sits at the tail.
  • Interrupted writes. A program that crashed while saving may never have flushed a valid trailer.
  • Bad binary transfers. Moving a PDF in a text mode that rewrites line endings corrupts bytes inside compressed streams.
  • Messy incremental edits. Repeated save-on-top edits can leave a structure that some readers refuse even though the pages are fine.

What it does not do

  • It does not recover a PDF the parser cannot open at all; a badly truncated file may be beyond this tool.
  • It does not repair a damaged image or font stored inside the PDF; a broken JPEG stays broken even in a clean container.
  • It does not remove a password or decrypt protected content.

For those cases, a desktop tool such as Ghostscript or qpdf, run locally, can sometimes go further.

Avoiding broken PDFs

When you generate PDFs in code, close the file properly so the trailer is written. When you serve them over HTTP, send a correct Content-Length so the client knows when the file is complete. Keep an original copy before editing, so a bad save is never your only version.

Frequently Asked Questions

It keeps every page the parser can read, which for structurally damaged files is usually all of them. If some pages are genuinely gone (their data is missing from the file), the tool cannot bring them back; it rebuilds cleanly around what remains.

Yes, when they are structurally intact: they are copied into the new file as they are. If a font or image stream is itself corrupted, that specific glyph or picture may still render blank, but the rest of the page is fine.

Only when the content is not truly locked. The tool tries to read past basic encryption, but it has no password field and cannot break real protection. A file locked with a password you do not have cannot be recovered.

No. The file is read and rebuilt entirely in your browser with an in-page PDF engine. Nothing is sent anywhere, so you can repair confidential documents safely.

Nothing. The tool never changes your source; it writes a separate, repaired copy. Download the new file, confirm it opens, and keep the original until you are sure.

Related Tools

Compress PDF

Compress PDF files in your browser by re-rendering pages as smaller JPEG-based PDFs for easier sharing and uploading.

Merge PDF and Images

Combine PDFs plus JPG and PNG images into one PDF, reorder the uploaded files, and download the merged document from your browser.

JPG to PDF Converter

Combine JPG or PNG images into one PDF with A4, Letter, A3 or A5 pages in portrait or landscape.

PNG to PDF Converter

Combine PNG screenshots, scans, or design exports into one PDF. Choose A4 or Letter pages, orientation, and margins before downloading.

HTML to PDF Converter

Convert basic HTML and inline CSS into an A4, Letter, or Legal raster PDF in your browser, without uploading the source or finished file.

Remove Pages from PDF

Delete selected pages or page ranges from a PDF. Enter numbers such as 2, 5-7, keep at least one page, and download a trimmed PDF.

Tool available in other languages