Re: [Math] RandomDataGenerator.nextInt vs UniformIntegerDistribution.sample

2013-07-29 Thread Phil Steitz
On 7/25/13 4:50 AM, Gilles wrote:
 Hi.

 The classes
   RandomDataGenerator
 and
  UniformIntegerDistribution
 use different codes for providing random integers in a range.

 Shouldn't the same code be used in both?
 Are both equally valid on the whole integer range?

The version in RandomDataGenerator is more robust (see MATH-936). 
Probably the impl there should be moved to the sample method of
UniformIntegerDistribution and RandomDataGenerator should create and
use a UIDist instance.

Phil


 Regards,
 Gilles

 [1] Or better (IMO) shouldn't RandomDataGenerator be a client
 of UniformIntegerDistribution?


 -
 To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
 For additional commands, e-mail: dev-h...@commons.apache.org




-
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org



[Math] RandomDataGenerator.nextInt vs UniformIntegerDistribution.sample

2013-07-25 Thread Gilles

Hi.

The classes
  RandomDataGenerator
and
 UniformIntegerDistribution
use different codes for providing random integers in a range.

Shouldn't the same code be used in both?
Are both equally valid on the whole integer range?


Regards,
Gilles

[1] Or better (IMO) shouldn't RandomDataGenerator be a client
of UniformIntegerDistribution?


-
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org