comp.lang.idl-pvwave archive: archive

Home » Public Forums » archive » Re: array handling

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: array handling [message #52113]

Mon, 08 January 2007 09:43

news.qwest.net
Messages: 137
Registered: September 2005

Senior Member

"Brian Larsen" <balarsen@gmail.com> wrote in message
news:1168100109.828745.131590@11g2000cwr.googlegroups.com...
> Something I have always wondered is the difference between using
> finite() and just using ne?
>
> As an example:
> IDL> array = [ 1.0, 2.0, !Values.F_NAN, 4.0, !Values.F_NAN ]
> IDL> Print, Where( NOT Float( Finite(array) ) )
> 2 4
> IDL> print, where(array ne array)
> 2 4
>
> What is the fundamental difference here? Speed? Style? or something
> else entirely?
>
> Cheers,
>
> Brian

while where(array ne array) is clever, short, and possibly fast
one should note (from IDL help):
"On Windows, using relational operators such as EQ and NE
with the values infinity or NaN (Not a Number) causes an "illegal operand"
error"

Programmers will generally avoid depending on a result that flags an error,
and is not strictly defined in the language. The results may change in
different versions
of IDL, and perahps on different operating systems. Who knows, in version 7,
IDL may
decide to make nan eq nan return 1.

Cheers,
bob

Report message to a moderator

Re: array handling [message #52121 is a reply to message #52113]

Sat, 06 January 2007 08:51

David Fanning
Messages: 11724
Registered: August 2001

Senior Member

Michael Galloy writes:

> By the way, it took me a second to figure out why you were converting to
> float in the FINITE example. Maybe
>
> IDL> print, where(~finite(array))
>
> would be better since FINITE returns only 0's and 1's (as long as you
> have IDL 6.0+)?

That confused me, too. I had to go read the damn article myself. :-)

But I did update the page to this more compact formulation
while I was doing it.

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.")

Report message to a moderator

Re: array handling [message #52122 is a reply to message #52121]

Sat, 06 January 2007 08:47

David Fanning
Messages: 11724
Registered: August 2001

Senior Member

Brian Larsen writes:

> Something I have always wondered is the difference between using
> finite() and just using ne?
>
> As an example:
> IDL> array = [ 1.0, 2.0, !Values.F_NAN, 4.0, !Values.F_NAN ]
> IDL> Print, Where( NOT Float( Finite(array) ) )
> 2 4
> IDL> print, where(array ne array)
> 2 4
>
> What is the fundamental difference here? Speed? Style? or something
> else entirely?

I don't know the answer, but I would guess the latter formulation
would lead some programmers to a general feeling of unease
regarding treating things that aren't numbers as if they were.
There are no hard and fast rules for such things, I don't think,
so you are at the mercy of programmers you don't know and
implementations that could, I guess, change in the future.

Plus, you get that damn "Arithmetic error: Floating illegal operation"
message when you try it. That generally causes a loss of confidence
among customers. :-(

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.")

Report message to a moderator

Re: array handling [message #52123 is a reply to message #52122]

Sat, 06 January 2007 08:39

Michael Galloy
Messages: 1114
Registered: April 2006

Senior Member

Brian Larsen wrote:
> Something I have always wondered is the difference between using
> finite() and just using ne?
>
> As an example:
> IDL> array = [ 1.0, 2.0, !Values.F_NAN, 4.0, !Values.F_NAN ]
> IDL> Print, Where( NOT Float( Finite(array) ) )
> 2 4
> IDL> print, where(array ne array)
> 2 4
>
> What is the fundamental difference here? Speed? Style? or something
> else entirely?

One thing is that FINITE can handle NaN's and infinity all in one step.

By the way, it took me a second to figure out why you were converting to
float in the FINITE example. Maybe

IDL> print, where(~finite(array))

would be better since FINITE returns only 0's and 1's (as long as you
have IDL 6.0+)?

Mike
--
www.michaelgalloy.com

Report message to a moderator

Re: array handling [message #52124 is a reply to message #52123]

Sat, 06 January 2007 08:15

Brian Larsen
Messages: 270
Registered: June 2006

Senior Member

Something I have always wondered is the difference between using
finite() and just using ne?

As an example:
IDL> array = [ 1.0, 2.0, !Values.F_NAN, 4.0, !Values.F_NAN ]
IDL> Print, Where( NOT Float( Finite(array) ) )
2 4
IDL> print, where(array ne array)
2 4

What is the fundamental difference here? Speed? Style? or something
else entirely?

Cheers,

Brian

On Jan 6, 8:01 am, David Fanning <n...@dfanning.com> wrote:
> jochem.vere...@gmail.com writes:
>> I am a newbie in IDL. Although I cannot handle the following problem, I
>> am sure for the 'Masters' this is peanuts.
>
>> The problem:
>> I wish to filter an array of redundant data in one line. This is a part
>> of the line to be filtered:
>
>> -Inf -NaN -Inf 0.542373 2.39394 2.28125
>> 0.000000 0.000000 0.000000
>> 3.21053 1.21839 0.000000 -NaN 0.967213
>> 2.03448 0.724138 0.000000 0.000000 0.000000
>> -NaN
>
>> With WHERE I can get rid of the zeros, but I dont know how to exclude
>> the '-NaN' and '-Inf'.
>
>> Any help would be most welcome,Here is an article that will help:
>
> http://www.dfanning.com/tips/check_nan.html
>
> 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.")

Report message to a moderator

Re: array handling [message #52125 is a reply to message #52124]

Sat, 06 January 2007 07:01

David Fanning
Messages: 11724
Registered: August 2001

Senior Member

jochem.verelst@gmail.com writes:

> I am a newbie in IDL. Although I cannot handle the following problem, I
> am sure for the 'Masters' this is peanuts.
>
> The problem:
> I wish to filter an array of redundant data in one line. This is a part
> of the line to be filtered:
>
> -Inf -NaN -Inf 0.542373 2.39394 2.28125
> 0.000000 0.000000 0.000000
> 3.21053 1.21839 0.000000 -NaN 0.967213
> 2.03448 0.724138 0.000000 0.000000 0.000000
> -NaN
>
> With WHERE I can get rid of the zeros, but I dont know how to exclude
> the '-NaN' and '-Inf'.
>
> Any help would be most welcome,

Here is an article that will help:

http://www.dfanning.com/tips/check_nan.html

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.")

Report message to a moderator

Re: array handling [message #52218 is a reply to message #52113]

Thu, 11 January 2007 07:27

Brian Larsen
Messages: 270
Registered: June 2006

Senior Member

>> Brian while where(array ne array) is clever, short, and possibly fast
> one should note (from IDL help):
> "On Windows, using relational operators such as EQ and NE
> with the values infinity or NaN (Not a Number) causes an "illegal operand"
> error"

This would explain why I had never seen the error that David mentioned
before. In Linux it just goes on its merry way w/o errors. But in the
future I think I will move to finite.

Brian

Report message to a moderator

Previous Topic:	Recorrido a Espa�a
Next Topic:	Re: Recorrido a España

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 14:00:12 PDT 2025

Total time taken to generate the page: 0.00619 seconds