fraunhofer-aisec / rabe

rabe is an Attribute Based Encryption library, written in Rust

Rabe

rabe is a rust library implementing several Attribute Based Encryption (ABE) schemes using a modified version of the bn library of zcash (type-3 pairing / Baretto Naering curve). The modification of bn brings in serde or borsh instead of the deprecated rustc_serialize. The standard serialization library is serde. If you want to use borsh, you need to specify it as feature.

For integration in distributed applications contact us.

Implemented Ciphertext Policy Schemes (CP-ABE)

BDABE CP-ABE

Georg Bramm, Mark Gall, Julian Schütte , "Blockchain based Distributed Attribute-based Encryption". In Proceedings of the 15th International Joint Conference on e-Business and Telecommunications (ICETE 2018) - Volume 2: SECRYPT, pages 99-110. Available from https://doi.org/10.5220/0006852602650276

AC17 CP-ABE

Shashank Agrawal, Melissa Chase, "FAME: Fast Attribute-based Message Encryption", (Section 3). In Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security 2017. Available from https://eprint.iacr.org/2017/807.pdf

AW11 CP-ABE

Lewko, Allison, and Brent Waters, "Decentralizing Attribute-Based Encryption.", (Appendix D). In Eurocrypt 2011. Available from http://eprint.iacr.org/2010/351.pdf

BSW CP-ABE

John Bethencourt, Amit Sahai, Brent Waters, "Ciphertext-Policy Attribute-Based Encryption" In IEEE Symposion on Security and Privacy, 2007. Available from https://doi.org/10.1109/SP.2007.11

MKE08 CP-ABE

S Müller, S Katzenbeisser, C Eckert , "Distributed Attribute-based Encryption". Published in International Conference on Information Security and Cryptology, Heidelberg, 2008. Available from http://www2.seceng.informatik.tu-darmstadt.de/assets/mueller/icisc08.pdf

Implemented Key Policy Schemes (KP-ABE)

AC17 KP-ABE

Shashank Agrawal, Melissa Chase, "FAME: Fast Attribute-based Message Encryption". In Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security 2017. Available from https://eprint.iacr.org/2017/807.pdf

LSW KP-ABE

Allison Lewko, Amit Sahai and Brent Waters, "Revocation Systems with Very Small Private Keys". In IEEE Symposium on Security and Privacy, 2010. SP'10. Available from http://eprint.iacr.org/2008/309.pdf

YCT14 KP-ABE

Xuanxia Yao, Zhi Chen, Ye Tian, "A lightweight attribute-based encryption scheme for the Internet of things". In Future Generation Computer Systems. Available from http://www.sciencedirect.com/science/article/pii/S0167739X14002039

WARNING !

The YCT14 scheme was broken in [1] and a 'fixed' version was again broken in [2]. Demonstration how the attack can be implemented in practice in https://www.blackhat.com/eu-21/briefings/schedule/index.html#practical-attacks-against-attribute-based-encryption-25058. [1] https://ieeexplore.ieee.org/document/8651482 [2] https://ieeexplore.ieee.org/document/9291064

Building rabe lib

In order to compile and test:

• install rust nightly
• git clone library
• install build-essential
• and then run cargo build --release && RUST_BACKTRACE=1 cargo test -- --nocapture
• rabe is also available with borsh serialization. just add --no-default-features --features borsh to the build command

Building rabe console app

See README.md

rabe's Issues

How can I use this project in Python?

Hi
I've built the project using cargo build --release.
Now I see librabe.so file. It contains these symbols (using nm command):

0000000000004008 b completed.8061
                 w __cxa_finalize@@GLIBC_2.2.5
0000000000001040 t deregister_tm_clones
00000000000010b0 t __do_global_dtors_aux
0000000000003dd0 d __do_global_dtors_aux_fini_array_entry
0000000000004000 d __dso_handle
0000000000003dd8 d _DYNAMIC
0000000000001114 t _fini
00000000000010f0 t frame_dummy
0000000000003dc8 d __frame_dummy_init_array_entry
0000000000002094 r __FRAME_END__
0000000000003fc8 d _GLOBAL_OFFSET_TABLE_
                 w __gmon_start__
0000000000002000 r __GNU_EH_FRAME_HDR
0000000000001000 t _init
                 w _ITM_deregisterTMCloneTable
                 w _ITM_registerTMCloneTable
0000000000001070 t register_tm_clones
0000000000004008 d __TMC_END__
0000000000001100 t _ZN3std3sys4unix4args3imp15ARGV_INIT_ARRAY12init_wrapper17h9817c52b4f788959E
0000000000003dc0 d _ZN3std3sys4unix4args3imp15ARGV_INIT_ARRAY17hf93f535c4855ea1bE
0000000000004010 b _ZN3std3sys4unix4args3imp4ARGC17h4850cf38500183b7E.0
0000000000004018 b _ZN3std3sys4unix4args3imp4ARGV17hd52966b69df71d64E.0

I don't see any encrypt or decrypt function. Did I have a mistake in building the project? or is there another way to see the exports of this shared object (so) file.

finally I want to use this project in python language.

Release it on crates.io

See https://doc.rust-lang.org/cargo/commands/cargo-publish.html

Duplicate attributes not supported in BSW scheme

The following policy with duplicated attribute "B" would caused decryption in BSW scheme to fail:

    let policy = String::from(r#"{"name": "or", "children": [{"name": "and", "children":  [{"name": "A"}, {"name": "B"}]}, {"name": "and", "children":  [{"name": "C"}, {"name": "B"}]}]}"#);

I've tested with the Python implementation by the original BSW authors and duplicated attributes are supported. Their paper also mentioned duplicated attributes are supported.

Please help to resolve this.

Library fails to process policies with more than 60 attributes.

More or less.
You need to create a policy with ~80 attributes to produces the bug and call encrypt/keygen, respectively.
Unclear if this is a parser issue or scheme issue.
Happens with AC17/BSW schemes (all).

YCT14 KP-ABE scheme is broken

Hello,

The YCT14 scheme was broken in [1] and a 'fixed' version was again broken in [2].
We have demonstrated how the attack can be implemented in practice in https://www.blackhat.com/eu-21/briefings/schedule/index.html#practical-attacks-against-attribute-based-encryption-25058.

[1] https://ieeexplore.ieee.org/document/8651482
[2] https://ieeexplore.ieee.org/document/9291064

cffi

Almost got cffi working, just missing 1 little thing, I guess...

from cffi import FFI
ffi = FFI()
ffi.cdef("""
    struct CpAbeCiphertext;
    struct CpAbeContext;
    struct CpAbeSecretKey;

    struct CpAbeContext* rabe_bsw_context_create();
    void rabe_bsw_context_destroy(struct CpAbeContext* ctx);
    struct CpAbeSecretKey* rabe_bsw_keygen(const struct CpAbeContext* ctx, const char* attributes);
    void rabe_bsw_keygen_destroy(void* sk);
    int32_t rabe_bsw_encrypt(const void* pk, char* policy, char* pt, int32_t pt_len, char** ct, int32_t *ct_len);
    int32_t rabe_bsw_decrypt(const struct CpAbeSecretKey* sk, const char* ct, uint32_t ct_len, char** pt_buf, uint32_t *pt_len);
""")

C = ffi.dlopen("librabe.so")

ctx = C.rabe_bsw_context_create()
print(ctx)
sk = C.rabe_bsw_keygen(ctx, b"[ \"test1\", \"test2\", \"test3\" ]")
print(sk)
pt = b"testing123"
ct = ffi.new("char **")
ct_len = ???

output = C.rabe_bsw_encrypt(ctx, b"{\"OR\": [{\"ATT\": \"A\"}, {\"ATT\": \"B\"}]}", pt, len(pt), ct, ct_len)
print(output)

How should I define ct_len?

Change some function signature

Hi, I'm writing a C-FFI binding for this project, but i think some function signature is not reasonable enough.

Example

pub fn keygen(
    _pk: &KpAbePublicKey,
    _msk: &KpAbeMasterKey,
    _policy: &String,
    _language: PolicyLanguage,
) -> Result<KpAbeSecretKey, RabeError> 

the parameter _policy in this function is &String, but this function only need a immutable reference, i recommend use &str.
Similarly, in this function:

pub fn encrypt(
    _gk: &Aw11GlobalKey,
    _pks: &Vec<Aw11PublicKey>,
    _policy: &String,
    _language: PolicyLanguage,
    _plaintext: &[u8],
) -> Result<Aw11Ciphertext, RabeError> 

the type of _pks should be &[Aw11PublicKey] instead of &Vec<Aw11PublicKey>

Reason

• I can construct a String from a *const c_char by let s = String::from_raw_parts(policy as *mut u8,len, len) during FFI calling.But, because this String not really own its memory, so we need a std::mem::forget(s) to make sure not free the memory passed by FFI calling.
• If use &str, i can construct a slice from pointer directly and no need to care about memory.
• Also, When i pass a array of pointer for &Vec<Aw11PublicKey>, i need to use Vec::from_raw_partsand not free the memory.
• Not only in FFI, if i want to pass a let policy = "xxxxx", i also need to construct a String and pass it to function. But if the parameter is &str, i can pass policy directly.
• And because String had impl trait Deref<str>, pass a &String to a &str is ok.
• Similarly, use &Vec<T> instead of &[T] has the same problem. Such as &Vec<String>, all of these are FFI unfriendly and may cause unnecessary overhead under certain circumstances.

in order to create example, how to calculate time based on number of Attributes?

Hi,
in order to create msk and pk of an AC17 KP-ABE scheme, I used all as you say but it is not working. it says "--scheme: command not found".
where should I use this command? in my rabe path "ubuntu:~/rabe" ?
I think the problem is path, I did what you said in "compile and test" and it is work and I have no problem with it.
also I have another issue, how can I calculate time of run based on number of attributes? where can I change the number?
I hope you help me in this, I used so many ways but none works and I can't change number of attributes.
thanks

Error in encrypt files

Hi. My name is Humberto.
I am studying about ABE, and I get rabe-console to test it.
But i don't take success in file encrypt e decryption. For example:

Start with:

$ rabe --s AC17CP setup
$ rabe --s AC17CP keygen --a 'A B'

In the encryption example 1:

rabe --s AC17CP encrypt pk.key 'B' teste.doc

It finish at error:

error: The following required arguments were not provided:

USAGE:
rabe encrypt [ARGS]

And other example, 2:

rabe encrypt pk.key "B" ./teste.doc

error: The following required arguments were not provided:

USAGE:
rabe --s encrypt [ARGS]

How to encrypt the teste.doc, and after decrypt it?

Would you add a full use scenario in the Rabe Console README file?

Thank you for attention.

CLI utilities do not work

The CLI tools do not work anymore for BSW and AC17 C/KP-ABE.
Apparently, the added schemes cause this.

The problem: https://github.com/Fraunhofer-AISEC/rabe/blob/master/src/main.rs#L1228
This line fails with:

  0: std::sys::unix::backtrace::tracing::imp::unwind_backtrace
             at src/libstd/sys/unix/backtrace/tracing/gcc_s.rs:39
   1: std::sys_common::backtrace::_print
             at src/libstd/sys_common/backtrace.rs:71
   2: std::panicking::default_hook::{{closure}}
             at src/libstd/sys_common/backtrace.rs:59
             at src/libstd/panicking.rs:197
   3: std::panicking::default_hook
             at src/libstd/panicking.rs:211
   4: std::panicking::rust_panic_with_hook
             at src/libstd/panicking.rs:474
   5: std::panicking::continue_panic_fmt
             at src/libstd/panicking.rs:381
   6: std::panicking::begin_panic_fmt
             at src/libstd/panicking.rs:336
   7: rabe::main::read_file
             at src/main.rs:1574
   8: rabe::main::run_encrypt
             at src/main.rs:1228
   9: rabe::main::run
             at src/main.rs:406
  10: rabe::main
             at src/main.rs:393
  11: std::rt::lang_start::{{closure}}
             at /rustc/a53f9df32fbb0b5f4382caaad8f1a46f36ea887c/src/libstd/rt.rs:64
  12: std::panicking::try::do_call
             at src/libstd/rt.rs:49
             at src/libstd/panicking.rs:293
  13: __rust_maybe_catch_panic
             at src/libpanic_unwind/lib.rs:85
  14: std::rt::lang_start_internal
             at src/libstd/panicking.rs:272
             at src/libstd/panic.rs:394
             at src/libstd/rt.rs:48
  15: std::rt::lang_start
             at /rustc/a53f9df32fbb0b5f4382caaad8f1a46f36ea887c/src/libstd/rt.rs:64
  16: main
  17: __libc_start_main
  18: _start

Reason:

https://github.com/Fraunhofer-AISEC/rabe/blob/master/src/main.rs#L329

The wrong variable is assigned.

Getting "The following required arguments were not provided" when compiling rabe-console

Hi all,

I compiled and built the lib based on the readme in the parent directory, but for some reason running rabe commands led to a 'zsh: command not found: rabe' error, suggesting something went wrong.

WHen I went to compile rabe-console, everything but the last step worked. Running ../target/release/rabe returned the following error:

    --s <s>

USAGE:
    rabe [OPTIONS] --s <s> [SUBCOMMAND]

For more information try --help```

which is cyclical because `rabe --help` doesn't work either.

How should I troubleshoot this?

About the AES dependency

I've been using this library for part of my Ph.D. research. it has proven very useful so far, but I was wondering about the AES cipher used underneath.

I've seen that rabe calls aes from aes-0.7.0. This crate only offers AES or AES-CTR. Seemed weird to me, and I have checked the newest version of aes crate and developers comment how:

This crate implements only the low-level block cipher function, and is intended for use for implementing higher-level constructions only. It is NOT intended for direct use in applications.

So now I'm wondering what mode of AES does Rabe use. Or whether at some point in the code you build AES-CBC or AES-GCM.

In case Rabe is using pure AES, since this is not recommended, are you planning on implementing AEADs?

BTW thanks for all the hard work 😄

HumanPolicy failed

The following will cause panic:

    let human = String::from("(B and C)");
    let _human_val: PolicyValue = parse(&human, PolicyLanguage::HumanPolicy).expect("unsuccessful parse");

code is inactive due to #[cfg] directives: feature = "borsh" is disabled

use crate::error::RabeError;
#[cfg(feature = "borsh")]
use borsh::{BorshSerialize, BorshDeserialize};
#[cfg(not(feature = "borsh"))]
use serde::{Serialize, Deserialize};