What Is PDF Corruption?

A PDF is a structured binary format: header, body (objects), cross-reference table (xref), trailer. Corruption means one of these is broken:

• Truncation: File cut off mid-stream (incomplete download, disk full). Missing trailer/xref.
• Bad xref: Offsets point to wrong locations. Viewer can't find objects.
• Invalid streams: Compressed data doesn't decompress (checksum mismatch, wrong filter).
• Missing objects: Referenced object doesn't exist (font, image, page).
• Header damage: Doesn't start with %PDF-.

Symptoms: "File damaged," "Cannot open," blank pages, missing content, viewer crashes.

Repair Levels: From Gentle to Nuclear

{`// Repair Strategy Escalation

| Level | Method                    | Privacy    | Success Rate | Data Loss Risk |
|-------|---------------------------|------------|--------------|----------------|
| 1     | Open in different viewer  | Local      | 30%          | None           |
| 2     | Linearize (qpdf/pikepdf)  | Local      | 60%          | None           |
| 3     | Save As (Acrobat/Editor)  | Local      | 70%          | Low            |
| 4     | Compress (Low)            | Local*     | 50%          | Low            |
| 5     | Server repair             | Upload     | 80%          | Medium         |
| 6     | Rasterize → Rebuild       | Local      | 95%          | High (text)    |`}

*Our Compress PDF Low level is client-side.

Client-Side Repair (Our Tools)

Compress PDF — Low Level

Our gentlest repair. Rebuilds cross-reference table, linearizes (Fast Web View), strips metadata, recompresses streams. Often fixes:

• Minor xref errors
• Incremental save debris
• Non-linearized files that load slowly

How to: Open Compress PDF → drag file → choose Low → download. Free, private, no upload.

PDF to JPG → JPG to PDF (Nuclear)

Rasterizes every page to pixels (300 DPI recommended), then rebuilds a clean PDF from those images. Bypasses ALL PDF structure issues. Always works if pages render at all.

Trade-offs: loses text layer (no search, no select), larger file, quality depends on DPI. But you get a working PDF.

Desktop Repair Tools

Adobe Acrobat Pro

File → Save As Other → Optimized PDF. Rebuilds structure, linearizes, can fix many corruptions. Expensive ($239/yr).

PDF-XChange Editor (Windows, Freemium)

File → Save As. Rebuilds xref. Free tier works for repair. Watermarks on some outputs.

Other Viewers (Save As)

Sometimes just opening in a different viewer and doing "Save As" rewrites the file cleanly. Try: SumatraPDF, PDF-XChange, Preview (macOS), Chrome Print → Save as PDF.

CLI Repair (Advanced)

For developers/power users. Local, scriptable, powerful.

qpdf

{`# Linearize (rebuilds xref, enables fast web view)
qpdf --linearize damaged.pdf repaired.pdf
Force repair (attempts to recover from truncation)
qpdf –force-version=1.7 –linearize damaged.pdf repaired.pdf
Check structure without repairing
qpdf –check damaged.pdf`}

pikepdf (Python)

{`import pikepdf
Open and save (rebuilds structure)
pdf = pikepdf.open(“damaged.pdf”)
pdf.save(“repaired.pdf”)
With linearization
pdf.save(“repaired.pdf”, linearize=True)
Inspect objects
for obj in pdf.objects:
print(obj)`}

Ghostscript

{`# Rebuild via PostScript interpretation (heavy but thorough)
gs -o repaired.pdf -sDEVICE=pdfwrite -dPDFSETTINGS=/prepress damaged.pdf`}

Server-Based Repair (Privacy Note)

Services: iLovePDF Repair, PDF2Go Repair, Sejda Repair, PDF24 Repair.

• Upload → they run repair engines → download
• Often higher success rate (commercial engines, more heuristics)
• Your corrupted file uploads to their servers
• For sensitive docs: use local tools only

Last Resort: Rasterize & Rebuild

When structure is too broken for any repair:

1. PDF to JPG at 300 DPI (or PNG for lossless)
2. If that fails: open in any viewer that renders it → screenshot each page → JPG to PDF
3. Result: visually identical PDF, no text layer, clean structure

Then: Compress PDF (Moderate) to reduce size. Protect PDF if needed.

Prevention

• Verify downloads: Check file size matches source. Redownload if mismatch.
• Don't edit PDFs in text editors: Never open .pdf in Notepad/VS Code and save.
• Use "Save As" not "Save" in viewers: "Save" often does incremental update (appends). "Save As" rewrites clean.
• Backup originals: Before any repair attempt, copy the damaged file.
• Linearize for web: Use Compress PDF Low on PDFs you host — prevents partial-load corruption.

FAQ

How do I fix a corrupted PDF for free?

Try our Compress PDF (Low level) — client-side, rebuilds structure. Or qpdf (CLI, free). Or PDF-XChange Editor (Windows, free tier).

Can a corrupted PDF be repaired?

Often yes — minor structure issues (xref, linearization) are fixable. Severe truncation (missing end of file) = data loss, may need rasterize-rebuild.

How do I repair a PDF without uploading?

Use our Compress PDF (Low) — client-side. Or qpdf/pikepdf (CLI, local). Or PDF-XChange Editor (local).

What does "linearize a PDF" mean?

Reorganize file so page 1 loads first, then page 2, etc. Enables streaming in browser. Also rebuilds cross-reference table — fixes many corruptions.

Can I fix a PDF that says "file damaged"?

Yes — try Compress PDF Low, or qpdf --linearize, or PDF-XChange Save As. If those fail: rasterize (PDF to JPG → JPG to PDF).

How do I fix a truncated PDF?

Truncated = missing bytes at end. qpdf --force-version with --linearize sometimes recovers. Otherwise: rasterize-rebuild. Or redownload.

Is there a free PDF repair tool online?

Server-based: iLovePDF, PDF2Go, Sejda (free tiers, upload required). Client-side: our Compress PDF Low (free, private). CLI: qpdf, pikepdf (free, local).

Got a damaged PDF? Start with Compress PDF (Low) — free, private, often fixes it. If not, try PDF to JPG → JPG to PDF (nuclear option).