This repository has been archived by the owner on Mar 3, 2020. It is now read-only.
Releases: facebookarchive/conceal
Releases · facebookarchive/conceal
v2.0.1 - All C++ port
v1.1.3 - Smaller native libraries
Compiler configuration was changed to include only used code from OpenSSL library. Especially important if containing application is not splitted by architecture, therefore containing binaries for all archs at once.
WARNING: This release is from Sep'16.
Look for latest version!
v1.1.2 - 256-bit encryption
AndroidConceal
factory class (and Conceal base class for other configurations)- 256-bits encryption available
Entity.create(...)
for platform-independent entities: see #138
See README.md
for new ways of creating a Crypto
and Entity
and how old code still works.
v1.0.7 - OpenSSL 1.0.2g + works in Android M
This is v1.0.6 fixed for Android M
- native code is compiled with Position Independent Code (PIC) flag
- an extra instrumentation test is added that test the fix when run on an Android 6 emulator (x86)
v1.0.6 - OpenSSL 1.0.2g
IMPORTANT: don't use this version. Use 1.0.7 instead which includes a fix in native code compilation.
- Integrates OpenSSL 1.0.2g
- 1.0.2e was tagged as buggy by Google
- not updated apps will not be able to be in Google Play from July
- Fixes a 4 bytes memory access in native when copying the IV (on encryption and decryption)
v1.0.5 - PBKDF2, InputStream.skip
New features:
- PBKDF2 (class PasswordBasedKeyDerivation)
- Decrypting InputStream implements skip
Fixes:
- Decrypting InputStream: Initial IV read could been truncated if underlying input stream didn't have enough available bytes
- Documentation: troubleshooting section added with proguard typical issue
v1.0.4 - 64bit archs, openssl 1.0.2e, perf improvements
This release includes:
- upgrade to openssl 1.0.2e encryption
- support for 64 bits platforms
- minor code and documentation fixes
- Performance improvement for streams:
- reading uses the same client buffer, no need for extra allocation
- no artificially chunked decryption reduces JNI calls (N div 256 for a reading buffer of size N)
- writing can receive an auxiliar buffer allocated by the client reducing JNI calls by the same proportion and enabling buffer reuse