Bug found on version 0.94: error when test performance on ORACLE-10G algorithm

Yesterday I discovered this error. When I tried to test performance of the ORACLE-10G algorithm on my laptop, I get this error:

$ ./jbrute.sh --test --algorithm=A

Number of cores detected: 8
Number of threads to use: 1

Testing performance for ORACLE-10G() ...
Exception in thread "Thread-0" java.lang.IllegalArgumentException: hexBinary needs to be even-length: SYSTEM/2F4C3799A242887767
at javax.xml.bind.DatatypeConverterImpl.parseHexBinary(Unknown Source)
at javax.xml.bind.DatatypeConverter.parseHexBinary(Unknown Source)
at specialAlgorithm.Oracle10g.toByteArray(Oracle10g.java:237)
at specialAlgorithm.MyMessageDigest.hashToByteArray(MyMessageDigest.java:118)
at entities.MyBruteDecryptThread.testBruteForceSpecial(MyBruteDecryptThread.java:255)
at entities.MyBruteDecryptThread.run(MyBruteDecryptThread.java:128)
You could compute 0 ORACLE-10G() hashes/sec approx.

The bug was not on the algorithm itself (decrypt works fine), it was an error on the "testBruteForceSpecial" method on MyBruteDecryptThread class, and it was corrected. The change will be included in the next release, but, I commited it. So, if you need a binary with this error fixed now, you can check out the source and build your own "JBrute.jar" package.

I´m sorry about it, but it is very difficult to me to test all scenarios after a change, and, until now, I did not receive any comments or reports of JBrute from users, so I must discover this kind of bugs by myself, and it is a really slow process...

Regards,
Gonzalo