comp.lang.idl-pvwave archive: archive » Re: odd behaviour from array

Home » Public Forums » archive » Re: odd behaviour from array_equal() with NaN, Inf values

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

Re: odd behaviour from array_equal() with NaN, Inf values [message #79030]

Mon, 23 January 2012 15:27

wallabadah
Messages: 28
Registered: November 2005

Junior Member

Here's a quick wrapper that checks for NaNs in the input, and returns true for the above test case. I haven't tested it extensively, but it seems to solve my current problem.

function wjp_array_equal, op1, op2, no_typeconv = no_typeconv
compile_opt idl2

op1NaN = where(finite(op1, /NaN), op1NaNCount, complement = op1Finite)
op2NaN = where(finite(op2, /NaN), op2NaNCount, complement = op2Finite)

; if no NaN values are present, use the normal array_equal() function
if op1NaNCount eq 0 && op2NaNCount eq 0 then $
return, array_equal(op1, op2, no_typeconv = no_typeconv)

; if counts of NaN values are equal, and results of the two where()
; functions are the same, compare the non-NaN values using array_equal()
if op1NaNCount eq op2NaNCount && array_equal(op1Finite, op2Finite) then $
return, array_equal(op1[op1Finite], op2[op2Finite], no_typeconv = no_typeconv)

; otherwise return 0
return, 0
end

Report message to a moderator

[Message index]

		Re: odd behaviour from array_equal() with NaN, Inf values By: wallabadah on Mon, 23 January 2012 15:27
		Re: odd behaviour from array_equal() with NaN, Inf values By: Carsten Lechte on Mon, 23 January 2012 04:37
		Re: odd behaviour from array_equal() with NaN, Inf values By: Craig Markwardt on Sun, 22 January 2012 21:46

Previous Topic:	Cramér–von Mises criterion
Next Topic:	to read excel file in IDL

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 17:26:06 PDT 2025

Total time taken to generate the page: 0.00440 seconds