PDF Validator

How to use the pdf validator

1. Choose a PDF; every page is read.
2. Read the findings, which explain what each observation means.
3. Check the font and page tables for the detail behind them.

About this tool

This reads a PDF and reports what is actually in it: how many pages, what size each one is, which carry rotation flags, whether the fonts are embedded, how much extractable text each page holds, and what the metadata claims.

It deliberately does not return a pass or a fail, which makes it an odd validator by the usual standard. That is the point: "valid" is not a property a PDF simply has — readers open files that break the specification every day, and a file that satisfies it can still be useless. A scan with no text layer is perfectly valid and impossible to search, which is exactly the finding someone needs before sending it to a lawyer or a search index.

The font check is the one that explains a common mystery. When a font is not embedded, the reader substitutes whatever it has, and the substitute has different metrics — so line breaks move, spacing changes and a document that looked right on the machine that made it looks wrong everywhere else. Seeing an unembedded font named is usually the whole answer.

Mixed page sizes and rotation flags are reported for the same reason. Both are legal, both are often intentional, and both explain otherwise puzzling behaviour: a document that prints with alternate pages scaled, or one whose extracted text comes out in an order nobody expected.

Common uses

• Finding out whether a PDF is searchable or just a scan.
• Diagnosing why a document looks different on another machine.
• Checking a file before sending it to a printer.

Frequently asked questions

Why does it not say whether my PDF is valid?

Because that is not a single property. Readers accept files that break the specification, and a fully compliant scan with no text layer is still unsearchable. Findings are more useful than a verdict.

What does an unembedded font mean?

The reader substitutes another font with different metrics, so line breaks and spacing shift. It is the usual reason a document looks wrong on someone else's machine.

How can I tell if a PDF is a scan?

By whether its pages contain extractable text. This reports the character count per page, and zero across the document means images rather than text.