# self_encryption
**Self encrypting files (convergent encryption plus obfuscation)**
|Crate|Travis|Appveyor|Coverage|
|:------:|:-------:|:-------:|:-------:|
|[](https://crates.io/crates/self_encryption)|[](https://travis-ci.org/dirvine/self_encryption)|[](https://ci.appveyor.com/project/dirvine/self-encryption) | [](https://coveralls.io/r/dirvine/self_encryption?branch=master)|
| [API Documentation](http://dirvine.github.io/self_encryption/self_encryption/) | [MaidSafe System Documention](http://systemdocs.maidsafe.net/) | [MaidSafe web site](http://www.maidsafe.net) | [Safe Community site](https://forum.safenetwork.io) |
#Overview
A version of [convergent encryption](http://en.wikipedia.org/wiki/Convergent_encryption) with an additional obfuscation step. This pattern allows secured data that can also be [de-duplicated](http://en.wikipedia.org/wiki/Data_deduplication). This library presents an API that can be utilised in any application that provides POSIX like a filesystem interface, where it will deal very effectively with the content part of any data (in tests the parallelised approach here can actually be faster than reading/writing data as a single stream). It is important to realise two important aspects of this library:
1. This library deals with file content **only**
2. This library provides very secure data, but does return a data structure (DataMap) that in turn requires to be secured.
![image of self encryption] (https://github.com/dirvine/self_encryption/blob/master/img/self-encryption.png?raw=true)
## ToDo list
- [x] Initial structure
- [x] Test set-up
- [x] Travis integration
- [x] Docs creation
- [x] Docs hosting (github.io)
- [x] Windows CI set-up (ci.AppVeyor.com)
- [x] Read/Write file in memory based buffer
- [x] API version 0.0.8
- [x] Implement disk based interface as example
- [x] Full unit tests in lib.rs
- [x] Integrations tests in tests module
- [x] Benchmark tests for varying file sizes from 1 byte to 10 M/b
- [x] API stable version 0.1.0
- [ ] Add compression pre encrypt and post encrypt in encrypt and decrypt methods
- [ ] Add task passing to allow cores to be lit up when handling chunks
- [ ] Allow any size file
- [ ] Replace sequencer with new struct and use BufferedStream to offload to disk (MemoryMapped file)
- [ ] Clean up any cache chunks when disk space is low (start + now < 90%)
- [ ] Store intermediate chunks when disk space is low (start + now < 90%)
- [ ] Add another functor to constructor to allow storage query for chunks (not get)
- [ ] Check for first last middle chunks on net and presume file stored
- [x] Coverage analysis (coveralls ?)
- [ ] Uncomment benchmark tests read methods (require bench in beta channel or stabilised first)
#Video of the process
[self_encryption process and use case video] (https://www.youtube.com/watch?v=Jnvwv4z17b4)