Extract attachments
What Extract attachments actually does
An enclosed file inside a PDF is easy to miss and, in some readers, awkward to get back out. This page reads the document's list of embedded files, reports what is in there and hands the contents back as ordinary downloads. The part worth knowing about is the name. A name stored inside a document is data supplied by whoever built that document, and nothing in the format stops it being a path with directory separators and parent references in it rather than a plain name. Software that writes such a name straight to disk can be made to write outside the folder the user chose, which is a well understood class of problem and the reason archive tools have hardened against it for years. Every name recovered here is reduced to a single segment before anything is offered for download, so an enclosure stored as a traversal path arrives as a plain file name and lands where you told your browser to put it. That is the practical reason to unpack a document from an untrusted source here rather than in whatever reader happens to open it.
How to use it
- Choose the PDF.
- The document's list of embedded files is read and what it holds is reported.
- Check the names and sizes shown, which are what will be written.
- Download. Each recovered file arrives under a plain name with no directory part.
Useful for
- Get the source data back out of a report that was distributed as a single file.
- Recover the annex from a signed agreement without asking the sender for it again.
- Check whether a document from an unknown source is carrying anything before it is opened elsewhere.
- Unpack an archived document whose original companion files are long gone.
Limits worth knowing
- Only enclosures held in the document's embedded file list are found. Content drawn onto a page, such as a chart image, is not an enclosure and is not recovered here.
- The recovered bytes are exactly as stored. Nothing is scanned, validated or opened, and an enclosure from an untrusted document deserves the same caution as any other unknown file.
- Names are reduced to a single path-free segment, so an enclosure stored as a path arrives under the last part of that path.
- An encrypted document that will not open cannot be unpacked here.
Questions people ask
It says there is nothing to extract. Why?
The document has no embedded files. Images and charts that are part of a page are page content rather than enclosures, and the image extractor is the tool for those.
What happens to a strange file name?
It is reduced to one plain segment before download. A name stored as a path with parent references comes back as just the final name.
Is the recovered file checked for safety?
No. The bytes come back exactly as they were stored. Treat an enclosure from an unfamiliar document as you would any unknown download.
Does the document leave my browser?
No. It is opened and unpacked in this tab.