Thank you for finding and reporting these errors.


----- Original Message -----
From: "Mark D. Niemiec" <[EMAIL PROTECTED]>
Date: Tuesday, March 18, 2008 0:19
Subject: [Jprogramming] Some more NaN anomalies
To: [email protected]

> While not useful, base-infinity numbers should have predictable 
> semantics:   
> _b1           NB. Correct 
> in both versions
> 1
>    
> _b10          NB. Should 
> be _ (correct in J6.01)
> _j_.
>    _b100      NB. Should be _ (6.01 also incorrect, gives 
> _. instead of _)
> _.j_.
>    _ #. 1 0   NB. Correct in both versions
> _
>    _ #. 1 0 0 NB. Correct in both versions
> _
>    
> 4b_           NB. _ is 
> recognized as a digit, but value is 0 (same in 4b_. and J6.01)
> 0
>    
> 4b__          NB. after 
> b, _ can be sign, or infinity, but not both
> |ill-formed number
> 
> (I also notice that J6.02 finally correctly displays things like _j4
> instead of silently ignoring the imaginary part)
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to