| Re: !null values in arrays [message #74040] |
Thu, 16 December 2010 07:39 |
penteado
Messages: 866
Registered: February 2018
|
Senior Member
Administrator |
|
|
On Dec 16, 1:14 pm, Paul van Delst <paul.vande...@noaa.gov> wrote:
> Purely for perspective into the issue, ruby handles operations with null values thusly:
>
> irb(main):001:0> 3 + nil
> TypeError: nil can't be coerced into Fixnum
> from (irb):1:in `+'
> from (irb):1
>
> and python similarly:
>
>>>> 3 + None
>
> Traceback (most recent call last):
> File "<stdin>", line 1, in ?
> TypeError: unsupported operand type(s) for +: 'int' and 'NoneType'
Both appropriate. !null and NaN are not the same, and are not the same
as 0. In fact, their lack in more primitive languages is an important
hassle.
0 is not, intrinsically, the same as missing data. In particular, most
mathematical operations on NaNs (say, adding a value to a NaN) should
always return NaN to make sense. If an array has missing data, data
cannot be magically created on the missing elements because, say, a
scalar was added to the array.
NaN is special in the sense that it is a floating point value, part of
the IEEE 754 standard, and properly recognized by hardware and other
software. !null is a language construct, which necessarily varies
among languages, as they vary in what constitutes a variable. It is an
undefined variable, so one can assign to it, but not operate on it,
the same way that one cannot operate on an undefined variable, but can
create it by assignment.
|
|
|
|
| Re: !null values in arrays [message #74042 is a reply to message #74040] |
Thu, 16 December 2010 07:14  |
Paul Van Delst[1]
Messages: 1157
Registered: April 2002
|
Senior Member |
|
|
Hello,
Axel M wrote:
> Hi Gray,
>
> I see your point and agree to some extent. But I would also see the
> opposite arguments: what should then be the result of 3 * !NULL or
> 3 / !NULL ? Do you propose that !NULL is handled as a 0? This might
> often lead to more confusion, and for indexing purposes it is
> definitely not a 0.
In IDL, storing a null in an otherwise homogeneous array makes said array a list. Why not use a list? The OP could still
call it an array.... :o)
Purely for perspective into the issue, ruby handles operations with null values thusly:
irb(main):001:0> 3 + nil
TypeError: nil can't be coerced into Fixnum
from (irb):1:in `+'
from (irb):1
and python similarly:
>>> 3 + None
Traceback (most recent call last):
File "<stdin>", line 1, in ?
TypeError: unsupported operand type(s) for +: 'int' and 'NoneType'
cheers,
paulv
|
|
|
|
| Re: !null values in arrays [message #74043 is a reply to message #74042] |
Thu, 16 December 2010 03:30 |
Axel Martínez
Messages: 22
Registered: June 2010
|
Junior Member |
|
|
Hi Gray,
I see your point and agree to some extent. But I would also see the
opposite arguments: what should then be the result of 3 * !NULL or
3 / !NULL ? Do you propose that !NULL is handled as a 0? This might
often lead to more confusion, and for indexing purposes it is
definitely not a 0.
The eventual advantage is probably not worth it...
|
|
|
|
| Re: !null values in arrays [message #74049 is a reply to message #74043] |
Wed, 15 December 2010 10:20 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Gray writes:
> However, writing this post is worth it just to include that
> tautological error message.
Now I have to go looking for my glasses *and* my dictionary. :-(
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
