AuraDoc
Sign inSign up

What is a cross-reference table?

The cross-reference table, often called the xref, is an index stored near the end of a PDF file. It lists the position of every object, such as pages, fonts, images and annotations, so that a reader can jump straight to what it needs instead of reading the file from the start. Without it, a program has to scan the whole file to find its parts. Most of the time you never see it, but it decides how quickly and how reliably a document opens.

How it works

A PDF is built from numbered objects. The page tree points to pages, and pages point to fonts and images. The cross-reference table maps each number to a byte offset, and a trailer at the very end tells the reader where the table starts. When a file is edited, new objects are added and the table is updated to match. A single wrong offset is enough to make a reader report that the file is damaged or show a blank page.

Damaged tables / Rebuilding

A table is usually damaged when a download stops early, when a transfer corrupts bytes or when a program writes the file incompletely. Many readers try to rebuild it silently by scanning the file, which is why some broken PDFs still open on one device and fail on another. A rebuilt index can be saved into a clean file, but anything that was lost from the content itself cannot be recovered, so keep the original in case the damage goes deeper than the index.

Using AuraDoc for it

Repair PDF reads a damaged file and writes a new one with a rebuilt structure, Linearize PDF reorganizes the file so the first page can show before the whole file has loaded and PDF quality report checks the structure and tells you whether problems were found. Each result is a new file, so your original stays as it was.

Tools that use this

Guides that put it into practice

Common questions

Why does one program open my PDF while another reports an error?
Readers differ in how strictly they treat a bad table. Some rebuild it quietly and others stop. The file may still be incomplete in either case.
Does a larger file have a larger table?
Yes, roughly in proportion to the number of objects. A PDF with thousands of small objects has a longer table than one with a few large images.
Can a repair bring back missing content?
Only if the content is still in the file. A repair fixes the index, not data that was never fully saved or downloaded.

Related terms

Need more than a quick fix?

AuraDoc is a document platform with OCR, AI search and summaries, and more than 60 PDF tools for teams. Try it free for 7 days.

Tools marked “Secure server” send your file once, over an encrypted connection, to an isolated AuraDoc server. It is processed in a throw-away process and deleted as soon as the result is sent back. Fair use and anti-abuse policy