PDF Metadata Viewer

Every PDF can carry six text fields — title, author, subject, keywords, creator and producer — plus a creation and a modification date. Most files fill in Producer and nothing else, because that is what the writing software stamps. The ones worth looking at are Author, which is usually a real name pulled from the account that made the file, and Creator and Producer, which name the exact application and build. All of it travels with the file when it is emailed or published, which is why those three are flagged here. Dates are stored in PDF’s own D:YYYYMMDDHHmmSS format with an optional offset; a missing offset is read as UTC, so a stamp can be out by a few hours.

A PDF page is measured in points at 72 to the inch, and the page tree can hold a different box for every page. The sizes are grouped, matched against the ISO 216 A and B series and the common North American sizes within a tolerance you can see, and anything else is reported as Custom rather than rounded to the nearest familiar name. Mixed sizes are the useful signal: they almost always mean pages were merged from different sources, and they print badly unless the printer is told to scale each page. Orientation is read from the box itself, so a landscape page in a portrait document shows up immediately.

The document information dictionary only. XMP metadata, embedded fonts and their licences, attachments, JavaScript, form fields, digital signatures, annotations and the revision history a PDF accumulates from incremental saves are all outside what this page reads. An encrypted file that opens without a password is the common “owner password only” case, and the restrictions it declares are honoured by well-behaved viewers and enforced by nothing. The page reads and never rewrites: to strip metadata you need a tool that writes a new file. Nothing leaves the browser; the same four anonymous usage counts as the rest of the site apply.