harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Tim Ellison <t.p.elli...@gmail.com>
Subject Re: ITC: Contribution of java.math and javax.crypto
Date Tue, 14 Mar 2006 16:30:36 GMT
Excellent!  -- thanks Miguel and team.

Regards,
Tim

Miguel Montes wrote:
>  As it was announced by Iris Gastañaga we are contributing the packages
> javax.crypto and java.math on behalf of ITC (Córdoba Technolgy Institute).
> We have been developing this code for several months and we believe it is a
> valid contribution. Our code not only implements the full 5.0 API but also
> uses 5.0 features and syntax.  As 5.0 is a stated goal of Harmony we hope
> that a 5.0 VM will be available soon.
> We are also contributing a set of test cases for both packages.
> 
> Below there is a short description of the contribution.  We will soon
> contribute also  an implementation of java.rmi.
> 
> 
> 
>  *Package name*
> **javax.crypto
> *
> Package Description*
> This is a clean room implementation of the package javax.crypto as specified
> in the JDK 5.0. It requires the existence of other java packages, in
> particular java.security and java.util. It can be used to replace jce.jar in
> the jdk.
> 
> *Current Status*
> The package is almost fully implemented. It does not implement the
> ExemptionMechanism logic.
> 
> *Testing
> * Unit and integration tests (and their documentation) are provided with the
> code.*
> *
> 
> *Implementation Notes
> The code uses heavily J2SE 5.0 features, such as generics, so it requires
> 5.0 VM and libraries (for instance java.security).
> It has been tested against Sun SDK, removing the original jce.jar and
> replacing it by ours.
> 
> *
> 
> 
> 
> 
>  *Package name*
> **java.math
> 
>  *Package Description*
> This is a clean room implementation of the package java.math as specified in
> JDK 5.0.
> 
> *Current Status*
> The package is fully implemented.  Some methods are fairly optimized (for
> example multiplication combines paper-and-pencil and Karatsuba algorithms).
> 
> *Testing
> * Unit and integration tests (and their documentation) are provided with the
> code.*
> *
> 
> *Implementation Notes
> The internal representation for BigInteger is two-complement (this is
> different from the sign-magnitude implementation already donated to
> Harmony). BigDecimal uses only BigInteger's public interface and implements
> the full 5.0 specification (which has about twice as many methods as in 1.4).
> The code uses J2SE 5.0 features, so it requires 5.0 VM. It has been tested
> against Sun SDK.
> 
> *
> **
> 
> 
> 
> 
> --
> Miguel Montes
> 

-- 

Tim Ellison (t.p.ellison@gmail.com)
IBM Java technology centre, UK.

Mime
View raw message