| Need Help Resolving ENDIF Errors [message #65538] |
Tue, 10 March 2009 13:38  |
einszweilieb
Messages: 4
Registered: March 2009
|
Junior Member |
|
|
I'm trying to run a procedure in IDL. The procedure has a series of
IF LOOPS, each of minimal complexity - they are fairly simple.
For example:
XX=readfits('/home/kphil/img/img213.fits')
YY=readfits('/home/kphil/img/img213.fits', EXTEN_NO=1)
ZZ =YY
G = where(ZZ ne 0)
invarmap= ZZ * 0.0
invarmap[G] = (1./(ZZ[G])^2)
if ((where(XX ne XX))(0) ne -1) then begin
ZZ[where(XX ne XX)] = 0
XX[where(XX ne XX)] = 0
endif
The problem is that everything runs just great, until the ENDIF is
reached. I CONSTANTLY get ENDIF errors.
Why is this happening? I feel that my code is solid, but all I get
are ENDIF errors.
The .pro I'm running has about 7 IF LOOPS, and the ONLY errors I get
are on the lines where ENDIF is.
Any idea what the problem is? Help!
Thanks in advance.
|
|
|
|
| Re: Need Help Resolving ENDIF Errors [message #65581 is a reply to message #65538] |
Thu, 12 March 2009 07:43  |
Carsten Lechte
Messages: 124
Registered: August 2006
|
Senior Member |
|
|
Carsten Lechte wrote:
> David Fanning wrote:
>> it. Better to use something that is actually defined
>> in the operating system, I think.
>
> Or, more precisely, the language environment. The C library for
> GNU cc correctly implements the 'NaN ne NaN' behaviour on all
> operating systems (at least according to the manual). IDL
> inherits its environment from the compiler that is used.
And I forgot to add:
That is why David and I suggest one should prefer documented features
of a system (like FINITE()). On the other hand, who would have
thought to test for basic IEEE 754 compliance before using an
IEEE 754 feature?
chl
|
|
|
|
| Re: Need Help Resolving ENDIF Errors [message #65582 is a reply to message #65538] |
Thu, 12 March 2009 07:35 |
Carsten Lechte
Messages: 124
Registered: August 2006
|
Senior Member |
|
|
David Fanning wrote:
> it. Better to use something that is actually defined
> in the operating system, I think.
Or, more precisely, the language environment. The C library for
GNU cc correctly implements the 'NaN ne NaN' behaviour on all
operating systems (at least according to the manual). IDL
inherits its environment from the compiler that is used.
chl
|
|
|
|
| Re: Need Help Resolving ENDIF Errors [message #65583 is a reply to message #65538] |
Thu, 12 March 2009 07:28 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Jeremy Bailin writes:
> Ye gods! I must admit to never having used IDL in windows, so I've
> never found that issue. Or used windows at all in a decade, for that
> matter. Hmmm.
As Warren Buffet said the other day, it's not who
YOU are sleeping with, it's who THEY are sleeping with.
Hard to say where those programs we write will show up
next. :-)
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.")
|
|
|
|
| Re: Need Help Resolving ENDIF Errors [message #65585 is a reply to message #65538] |
Thu, 12 March 2009 07:14 |
Jeremy Bailin
Messages: 618
Registered: April 2008
|
Senior Member |
|
|
On Mar 11, 9:11 pm, David Fanning <n...@dfanning.com> wrote:
> Jeremy Bailin writes:
>> 1. Yes, there is a use of asking where(XX ne XX) - it tells you where
>> there are NaNs. I use it all the time.
>
> Oh, dear! That's sorta like using the rhythm method of
> birth control, isn't it? I mean, maybe combined with
> hope and prayer, but...I certainly wouldn't rely on
> it. Better to use something that is actually defined
> in the operating system, I think. Here is an article:
>
> http://www.dfanning.com/math_tips/nans.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.")
Ye gods! I must admit to never having used IDL in windows, so I've
never found that issue. Or used windows at all in a decade, for that
matter. Hmmm.
Any idea whether this statement: "If this turns out to be true there a
good chance that in the future the behavior of the Windows and Unix
versions of IDL will be consistent. [That is, versions after IDL
5.6.]" is indeed true for recent versions?
-Jeremy.
|
|
|
|
| Re: Need Help Resolving ENDIF Errors [message #65599 is a reply to message #65538] |
Wed, 11 March 2009 18:11 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Jeremy Bailin writes:
> 1. Yes, there is a use of asking where(XX ne XX) - it tells you where
> there are NaNs. I use it all the time.
Oh, dear! That's sorta like using the rhythm method of
birth control, isn't it? I mean, maybe combined with
hope and prayer, but...I certainly wouldn't rely on
it. Better to use something that is actually defined
in the operating system, I think. Here is an article:
http://www.dfanning.com/math_tips/nans.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.")
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
