Check EPUB for DRM before converting it

Checked 3 September 2026

A book bought from a store, or borrowed from a library, usually will not convert to anything. The reason is a small file inside the EPUB, and which file it is tells you which store locked the book. One of the files people mistake for DRM is nothing of the kind, and a book carrying it converts normally. You can tell the cases apart in about a minute with tools already on your machine.

An EPUB is a zip

Rename one to .zip and it opens. Inside is a folder called META-INF, and that folder is where every protection scheme leaves its mark. The book's text and images may be encrypted, but the marker that says so is a plain file with a fixed name, and the name is enough. Our EPUB to PDF converter does exactly this check when you drop a book on it, and the table below is the same list it uses, in the same order.

The five cases

None of these files present, and no encryption.xml at all, means the book is open and converts. The three named schemes are checked first, in the order listed, so a book carrying both rights.xml and an encryption.xml is reported as Adobe ADEPT.

How to look

On a Mac or Linux, open a terminal in the folder holding the book and list the archive. The one line you care about is the filter on META-INF.

unzip -l book.epub | grep META-INF

For a book with no DRM and no obfuscated fonts, the output is one directory and one file. This is what it printed for a test EPUB built for this article, which carries a font-obfuscation entry and nothing else.

        0  09-03-2026 22:47   META-INF/
       12  09-03-2026 22:47   META-INF/container.xml
      236  09-03-2026 22:47   META-INF/encryption.xml

container.xml is in every EPUB and means nothing. If rights.xml, sinf.xml or license.lcpl appears alongside it, read the row for that file in the table and stop. If only encryption.xml appears, print it and read the algorithm it names.

unzip -p book.epub META-INF/encryption.xml

On Windows, copy the book, rename the copy to end in .zip, and open it in Explorer. META-INF is a folder at the top level, and encryption.xml opens in Notepad.

The one that looks like DRM and is not

An encryption.xml on its own frightens people, and it frightens a lot of converters, which refuse the book on sight. Most of the time it is there for the fonts. A publisher who embeds a licensed typeface is not allowed to ship it as a file anyone can lift out of the zip, so the EPUB standard added a way to scramble the font bytes. Two algorithms do this, and they are the only two that count as fonts rather than content.

• http://www.idpf.org/2008/embedding
• http://ns.adobe.com/pdf/enc#RC

If every EncryptionMethod in the file names one of those two, the text and images are untouched and the book converts. Our converter treats it that way, with a test that exists so nobody later "fixes" it into refusing ordinary books. The scrambled font itself is not restored, so a book that used one gets a plain serif in its place, which is a cosmetic loss on a book that otherwise comes through whole.

If the file names anything else, for instance an AES algorithm from the XML Encryption namespace, the content is encrypted with something we do not have a name for. The converter reports it as an encryption scheme it does not recognise and refuses, the same as the named three.

Kindle books are a different question

A book from Amazon is not an EPUB. It is AZW3 or the older MOBI, and none of those are zips with a META-INF folder. There is nothing to check because the check does not apply. Renaming one to .epub does not change what is inside, and our converter will tell you it is not a zip. This page reads EPUB only, and a browser tool claiming to read all three is worth checking before you trust it.

What happens on our page

Drop a locked book on the converter and it names the scheme and the store it came from, then declines. The book has not left your device, because the check runs in your browser as the zip is opened, before anything else happens. We also count the refusal, per scheme, as a failed operation. That count is why this page exists, and it carries the scheme name and nothing from the book.

filetity does not remove DRM and is not going to. There is no setting for it, no plan for it, and no hint anywhere on this site about how someone else might. A book you own without a lock converts. A book with one belongs to the reader app that sold or lent it to you.

If the book is open and the PDF is still wrong

Then DRM was never the problem, and the likely cause is the book's own markup. Project Gutenberg books are the clearest example. Tested against their Frankenstein on 3 September 2026, the file put class="chapter" on 60 elements for a book with 31 chapters, so a page-break rule keyed on that class scattered breaks through the text. The converter now marks chapters with its own attribute instead of trusting the book's class names, and page breaks come from the spine, which is the book's own reading order. A Gutenberg or Standard Ebooks title has no DRM and should come through with one chapter per page.

The marker list on this page is copied from the converter's own detection code as of the date at the top. If a store changes its scheme, the code changes first and this page follows. Corrections to support@filetity.com.

Related: how local processing works here, including how to watch the network while a book is checked.