/lcp-decrypt

lcp-decrypt takes an ePUB file protected with a Readium LCP protection (sometimes also called CARE) and decrypts it into a regular ePUB file.

Primary LanguageGoMIT LicenseMIT

lcp-decrypt - a quick&dirty tool to remove LCP/CARE DRMs

lcp-decrypt takes an ePUB file protected with a Readium LCP protection (sometimes also called CARE) and decrypts it into a regular ePUB file. The decryption requires the LCP user key. The retrieval of the key is not handled by this program, in other words, lcp-decrypt does not crack any DRM. It only makes a DRMed ePUB you already have legitimate access to usable on any ePUB compatible reader.

You can use lcp-decrypt online or build it as a CLI application that you can then use locally.

Building the lcp-decrypt CLI

Until binaries are provided, you need to compile the tool yourself by running

go build ./cmd/lcp-decrypt

Running lcp-decrypt

Once you have your user key (as a hex encoded string), getting a decoded ePUB is as simple as running

# decrypts ebook_with_drm.epub into ebook_without_drm.epub
lcp-decrypt -userKey 012345 ebook_with_drm.epub ebook_without_drm.epub

Retrieving the LCP user key

The process to retrieve the user key depends on how you officially access the ebook you purchased.

Vivlio Reader

I must give credits to Vivlio for providing a Linux version of their reader. The reader is an Electron application, which means it's easy to tell it to forward all requests through mitmproxy. Assuming mitmproxy is running on port 8080, run ./Vivlio-3.3.0.AppImage --proxy-server=127.0.0.1:8080. Log into your ebook reseller through the app and open the book. There should be one request in the mitmproxy console that looks like this:

GET https://api.your-book-store.com/v1/lcp/keys/user?device_id=XXX

and the response should look like

[{"user_key": "0123456789abcdef0123456789abcdef0123456789abcdef0123456789abcdef"}]

The 0123... string is the value you should pass to the -userKey command line flag.

Limitations

As mentioned above, this is a quick&dirty tool. The ePUB parsing was tested on the one file I have access to, and I only checked that the resulting ePUB worked in Calibre and on a Kindle. Bug reports and contributions are welcome.