If we want to be consistent, we could deprecate [lang]'s min/max and point
to [math]. This would parallel nicely with c.lang for java.lang and c.math
for java.math. It does not seem right to add all primitive types to
c.lang.NumberUtils if min/max routines are in c.math.

Gary

-----Original Message-----
From: Mark R. Diggory [mailto:[EMAIL PROTECTED] 
Sent: Saturday, June 21, 2003 15:18
To: Jakarta Commons Developers List
Subject: Re: [lang] NumberUtils minimum, maximum, and xor

Just to note: we have moved somwhat along these lines in the the commons 
[math] sandbox component. Currently we have o.a.c.m.stat.StatUtils:

double min(double[] doubleArr)
double max(double[] doubleArr)

available there.

-Mark Diggory

_matthewHawthorne wrote:
> I have 2 observations:
> 
> (1) Currently, the following methods are in o.a.c.l.NumberUtils
> 
> int maximum(int a, int b, int c)
> long maximum(long a, long b, long c)
> int minimum(int a, int b, int c)
> long minimum(long a, long b, long c)
> 
> I think it be more flexible to replace them with the following:
> 
> int minimum(int[] intArr)
> int maximum(int[] intArr)
> long minimum(long[] longArr)
> long maximum(long[] longArr)
> 
> It also may be a good time to add any missing methods such as:
> 
> short minimum(short[] shortArr)
> short maximum(short[] shortArr)
> float minimum(float[] floatArr)
> float maximum(float[] floatArr)
> double minimum(double[] doubleArr)
> double maximum(double[] doubleArr)
> 
> Any thoughts?
> 
> 
> (2) After searching for an easy way to xor booleans, and not finding
> anything, I created a method:
> 
> boolean xor(boolean[] boolArr)
> 
> Would this be a good addition to NumberUtils?
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
> 


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to