Cryptography Stack Exchange is a question and answer site for software developers, mathematicians and others interested in cryptography. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Still trying to design a fully binary cryptography container format for my mobile app, I am here asking if container is ever relevant.

Thanks to Apple, I cannot use GPG directly because I can neither call GPG through GPGME (fork(2) is not allowed) nor embed it (license conflicts)

The container format is simple ar(1) format that store its contents in plaintext, probably then compressed with some algorithm like Gzip. I am wondering if the container format affects the security of the already encrypted data, the MAC and IV that is held as separate entities within.

The plan is to use AES-CTR (or AES-CFB, or AES-OFB, or my homebrew cipher SHA512-HMAC-OFB which turned out to be just as fast as AES thanks to its larger block size) to encrypt the data and put it in the archive file (data.obj in the archive), its initialization vector in plain (iv.bin), asymmetrically encrypted symmetric key (key.bin, if relevant), the SHA512-HMAC of the archive in "check archive" form that is described below (mac.bin) and some metadata (algorithms used, compression of plaintext used, probably a user supplied tag in plain text) in plain (info.json)


ar(1) format stores file name (up to 16 characters), owner UID/GID, modification time, file mode (permissions) in its file header.

MAC (or signature) is calculated on a "check archive": every entity except the MAC (or signature) itself are in the check archive, which is also in ar(1) format, sorted in binary order based on their file names, retaining all their properties (name, owner, modification date and file mode).

share|improve this question
"algorithms used" should be stored with the _key_(s). $\:$ "compression of plaintext used, $\hspace{.8 in}$ probably a user supplied tag" should go into the MAC. $\;\;\;\;$ – Ricky Demer May 6 '14 at 18:00
@RickyDemer That is, algorithm should not appear in the container at all, and everything that is not MAC itself should be included in MAC? – Maxthon Chan May 6 '14 at 18:02
Yes. $\:$ Also note that compression may be problematic. $\;\;\;\;$ – Ricky Demer May 6 '14 at 18:05
@RickyDemer I forgot the asymmetrically encrypted symmetric key may appear in the file as well as key.bin. How to deal with that? – Maxthon Chan May 6 '14 at 18:06
If you're using public keys too, then you should probably use digital signatures, possibly instead of a MAC. $\;$ – Ricky Demer May 6 '14 at 19:15

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.