[Biojava-dev] 1.5ism in OracleDBHelper

mark.schreiber at novartis.com mark.schreiber at novartis.com
Tue May 31 06:08:02 EDT 2005


Hello -

This is fixed now. I have added javac -source 1.4 and javac -target  1.4 
information to the ant build script as well. While this would not have 
prevented this slip up it will prevent someone putting in code with 
generics, auto unboxing etc. Also means that JAR files built with ant on a 
JDK1.5 environment should work with JRE1.4. Interestingly code that makes 
use of classes only found in Java 1.5 will compile but not run (in 1.4) 
while classes that have new methods will probably compile and run unless 
they call a new class (it's only bytecode after all).

- Mark





Bradford Powell <bcpowell at email.unc.edu>
Sent by: biojava-dev-bounces at portal.open-bio.org
05/07/2005 11:21 PM

 
        To:     biojava-dev at biojava.org
        cc:     (bcc: Mark Schreiber/GP/Novartis)
        Subject:        [Biojava-dev] 1.5ism in OracleDBHelper


This is probably just an oversight, but I notice (through compilation
errors using a 1.4.2 VM), that the recent changes to OracleDBHelper
introduced the line:

putString.invoke(rs.getClob(column), new 
Object[]{Long.valueOf(1L),value});

The problem with this is that Long.valueOf is an addition to the Java
class libraries in 1.5. Using potential object caching in this case seems
to be not that useful relative to maintaining 1.4 compatability. (If
biojava should go over to 1.5, it should be for better reasons than this,
and I haven't seen a discussion of 1.5 here.

So, for the mean time, could someone replace 'Long.valueOf(1L)' with
'new Long(1L)'?

Thanks,

-- Bradford Powell
-- bradford_powell at unc.edu


_______________________________________________
biojava-dev mailing list
biojava-dev at biojava.org
http://biojava.org/mailman/listinfo/biojava-dev





More information about the biojava-dev mailing list