comp.lang.idl-pvwave archive: archive

Home » Public Forums » archive » Re: Keyword checking

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

Re: Keyword checking [message #31009 is a reply to message #31007]

Wed, 29 May 2002 07:55

thompson
Messages: 584
Registered: August 1991

Senior Member

Paul van Delst <paul.vandelst@noaa.gov> writes:

> Randall Skelton wrote:
>>
>> I have run into a great deal of trouble checking keywords in IDL and I
>> thought I would relay my thoughts and frustrations. I am trying to
>> prevent a user from passing an undefined variable in a keyword...
>>

(stuff deleted)

> Why the arg_present? Do you specifically want to tell the user that what they
> passed was undefined? I thought arg_present was for using when you wanted to
> return something in the variable? If the data is for input only, won't
> n_elements() suffice?

> me confused.

I agree. I can't think of any legitimate reason why one would want to generate
an error message if an undefined keyword was passed.

In the case of an input keyword, and undefined keyword should be treated
exactly as if the keyword was not passed at all. The reason for this is quite
simple. If one has embedded subroutines with keyword inheritance, there has to
be some way to pass the keywords along. For example

pro test1, key1=key1
test2, key1=key1
return
end

pro test2, key1=key1
if n_elements(key1) ne 0 then help, key1 else print, 'KEY1 not passed'
return
end

This obviously very trivial example is enough to illustrate my point. If the
user calls the procedure TEST1, the keyword KEY1 will be passed along to TEST2
whether or not the user called TEST1 with that keyword. If the user simply
calls TEST1 without the keyword, it will still appear in TEST2 as an undefined
value. There's nothing wrong with that.

In the output case, all one cares about is whether or not there's a variable to
receive the output. It doesn't matter if the variable was defined previously
or not.

I can only conclude that trapping an undefined keyword as an error is bad IDL
programming practice.

William Thompson

Report message to a moderator

[Message index]

		Re: Keyword checking By: R.Bauer on Wed, 29 May 2002 09:02
		Re: Keyword checking By: Liam E. Gumley on Wed, 29 May 2002 08:22
		Re: Keyword checking By: thompson on Wed, 29 May 2002 07:55
		Re: Keyword checking By: Paul Van Delst[1] on Wed, 29 May 2002 07:15
		Re: Keyword checking By: Jaco van Gorkom on Wed, 29 May 2002 07:04
		Re: Keyword checking By: Jaco van Gorkom on Wed, 29 May 2002 06:52
		Re: Keyword checking By: David Fanning on Wed, 29 May 2002 06:43

Previous Topic:	Re: accessing reverse_indices
Next Topic:	Re: color in widgets

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Tue Jun 09 18:28:33 PDT 2026

Total time taken to generate the page: 0.02472 seconds