Repair a damaged PDF
Try a bounded structural repair on a damaged PDF.
Working…
Questions people ask
- What kinds of damage can the repair fix?
- The classic builder mistakes: broken or missing cross-reference tables, missing trailer dictionaries, truncated endings, and object streams that need re-reading. The repair rebuilds the file's structure around the readable pages and then checks its own work. Files missing their %PDF header entirely, or whose object bodies are unreadable, are refused with the reason quoted.
- Can a repair lose my pages?
- It can, and the tool refuses to lie about it: the report names the page count before and after, and a page-count change is reported as exactly what it is. When recovery lands fewer pages than the reader could see, the report says which number survived and asks you to keep the original.
- What happens to signatures and form fields?
- Any structural rewrite is a new document to a digital signature, so signature validity does not survive. Form fields and annotations that the rebuild finds intact travel with the file, and the report warns when you uploaded a signed or form-carrying document so nothing changes silently.
- What if the file cannot be repaired?
- You get a clear refusal with the checker's complaint quoted, not half-repaired bytes. A repair tool that shipped anything it could not verify would be worse than no tool at all.
- My PDF opens fine. What do I get?
- A normalised copy: the same document re-serialised through the repair machinery with no damage found, and the report says "none found". Useful when a vendor swears the problem is your file and you want evidence either way.
- What happens to my files?
- The upload is removed as soon as processing finishes; the repaired file and report stay downloadable for about 10 minutes, then they are deleted automatically.