| Re: Defining Integer Field in a Structure [message #27695] |
Fri, 02 November 2001 12:06  |
Logan Lindquist
Messages: 50
Registered: October 2001
|
Member |
|
|
"David Fanning" <david@dfanning.com> wrote in message
news:MPG.164b817bd6ec291d989743@news.frii.com...
> Pavel A. Romashkin (pavel.romashkin@noaa.gov) writes:
> I'm going to have a class on How to Write Questions
> that Normal People Can Answer.
I think the real problem is that people are just too different. In the
phrasing of a question, terminology and the logic of the problem can be
confused too easly.
> But questions, many times, are symptoms of confusion,
> not actual entities that should be answered specifically.
> Any good technical support person knows that, but fewer
> have time to do it right.
Everyone is busy these days, our lives continually find ways to move our
interaction with the world on to new things. If the person genuinely needs
help and the 'expert' genuinely wants to give assistance than there can be
no bad questions and no bad answers. The communication received on both ends
is the result of each persons intentions.
Regards,
Logan
|
|
|
|
| Re: Defining Integer Field in a Structure [message #27726 is a reply to message #27695] |
Thu, 01 November 2001 15:36  |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
JD Smith <jdsmith@astro.cornell.edu> writes:
> Brian Bell wrote:
>>
>> Is there a way to define a field within a structure whose type is
>> integer? In my case, I want the field to be any integer from 0 to 4.
>> I know you can assign an actual integer to the field, but I don't want
>> to do this because I want the field value to change within certain
>> procedures and to return that value to the procedures calling it. I
>> would appreciate any help you might be able to offer.
>>
>
> This is illegal:
>
> IDL> a={0:"this", 1:"is",2:"a",3:"funny",4:"test"}
>
> so the short answer is no. However, if you just want 4 numbered
> "slots", try an array:
>
> IDL> a={vals:['this','is','a','funny','test']}
> IDL> print,a.vals[1]
> is
>
> if you'd like the types of each slot to be different, try at a pointer
> array. If you'd like the number of "slots" to be variable, try a
> pointer to a pointer array.
Following up on this diversion, it was in fact possible in IDL 5.2 and
earlier to make structures whose tag names were not valid IDL variable
names. For example, this statement was perfectly valid and executable:
x = create_struct('1',"this",'2',"is",'3',"funny",'4',"test ")
and it made exactly the structure you were expecting. However, I'm
not sure it would be useful, and more importantly these types of
antics are now forbidden by the IDL runtime.
Craig
--
------------------------------------------------------------ --------------
Craig B. Markwardt, Ph.D. EMAIL: craigmnet@cow.physics.wisc.edu
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
------------------------------------------------------------ --------------
|
|
|
|
|
|
| Re: Defining Integer Field in a Structure [message #27730 is a reply to message #27728] |
Thu, 01 November 2001 14:27 |
Pavel A. Romashkin
Messages: 531
Registered: November 2000
|
Senior Member |
|
|
I am getting confused by David and JD. Isn't all you want is this:
my_str = {field_1: 0s, field_2: 'Something_else'}
The only problem I see is that FIELD_1 is not going to be limited to
1...4 range, any valid integer (or any numeric type, for that matter)
can be used to assigne the value of FIELD_1. If you want it to be in
1...4 range, a check for that will have to be implemented.
When you pass the structure from one pro to another, they can read and
change FIELD_1 as required, as long as the entire MY_STR was passed.
Cheers,
Pavel
Brian Bell wrote:
>
> Is there a way to define a field within a structure whose type is
> integer? In my case, I want the field to be any integer from 0 to 4.
> I know you can assign an actual integer to the field, but I don't want
> to do this because I want the field value to change within certain
> procedures and to return that value to the procedures calling it. I
> would appreciate any help you might be able to offer.
>
> Thanks,
>
> Brian
|
|
|
|
| Re: Defining Integer Field in a Structure [message #27732 is a reply to message #27730] |
Thu, 01 November 2001 14:05 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
JD Smith (jdsmith@astro.cornell.edu) writes:
> Brian Bell wrote:
>>
>> Is there a way to define a field within a structure whose type is
>> integer? In my case, I want the field to be any integer from 0 to 4.
>> I know you can assign an actual integer to the field, but I don't want
>> to do this because I want the field value to change within certain
>> procedures and to return that value to the procedures calling it. I
>> would appreciate any help you might be able to offer.
>>
>
> This is illegal:
>
> IDL> a={0:"this", 1:"is",2:"a",3:"funny",4:"test"}
>
> so the short answer is no. However, if you just want 4 numbered
> "slots", try an array:
>
> IDL> a={vals:['this','is','a','funny','test']}
> IDL> print,a.vals[1]
> is
>
> if you'd like the types of each slot to be different, try at a pointer
> array. If you'd like the number of "slots" to be variable, try a
> pointer to a pointer array.
I have a feeling Brian was looking for something a LOT
simpler than this. :-)
I think he is confused between a specific *instance* of
a structure and the *definition* of a structure. Especially
because he has probably seen most structured both defined
and populated with the same statement. For example, like
this:
IDL> struct = {MYSTRUCT, a:3, b:5.4}
Here the variable struct is a specific *instance* of the
structure whose definition is defined in IDL as "MYSTUCT".
For example, you can create another structure like this:
IDL> struct2 = {MYSTRUCT}
IDL> Help, struct2, /Structure
** Structure MYSTRUCT, 2 tags, length=8, data length=6:
A INT 0
B FLOAT 0.000000
The fields in the second structure are defined *by* the
fields that appeared in the original structure definition
statement. These are the "null" field values, if you like.
You can assign any value you like to them:
IDL struct2.a = 4
If you wish to create a stucture definition without
actually creating an instance of a structure, you
put the structure definition in a file, like this:
PRO newstruct__define.pro
struct = {NEWSTRUCT, a:0, b:0.0, c:fltarr(10)}
END
Then you can use the structure definition whenever you
like:
IDL> var = {NEWSTRUCT}
IDL> var.a = 5
IDL> var.c = Findgen(10)*5
Cheers,
David
Cheers,
David
--
David W. Fanning, Ph.D.
Fanning Software Consulting
Phone: 970-221-0438, E-mail: david@dfanning.com
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Toll-Free IDL Book Orders: 1-888-461-0155
|
|
|
|
| Re: Defining Integer Field in a Structure [message #27735 is a reply to message #27732] |
Thu, 01 November 2001 13:24 |
John-David T. Smith
Messages: 384
Registered: January 2000
|
Senior Member |
|
|
Brian Bell wrote:
>
> Is there a way to define a field within a structure whose type is
> integer? In my case, I want the field to be any integer from 0 to 4.
> I know you can assign an actual integer to the field, but I don't want
> to do this because I want the field value to change within certain
> procedures and to return that value to the procedures calling it. I
> would appreciate any help you might be able to offer.
>
This is illegal:
IDL> a={0:"this", 1:"is",2:"a",3:"funny",4:"test"}
so the short answer is no. However, if you just want 4 numbered
"slots", try an array:
IDL> a={vals:['this','is','a','funny','test']}
IDL> print,a.vals[1]
is
if you'd like the types of each slot to be different, try at a pointer
array. If you'd like the number of "slots" to be variable, try a
pointer to a pointer array.
JD
|
|
|
|
| Re: Defining Integer Field in a Structure [message #27824 is a reply to message #27732] |
Tue, 06 November 2001 17:02 |
Harvey Rarback
Messages: 24
Registered: September 1998
|
Junior Member |
|
|
"David Fanning" <david@dfanning.com> wrote
> If you wish to create a stucture definition without
> actually creating an instance of a structure, you
> put the structure definition in a file, like this:
>
> PRO newstruct__define.pro
> struct = {NEWSTRUCT, a:0, b:0.0, c:fltarr(10)}
> END
This is something that has always confused me. It seems like the variable
"struct" is created (at least help will tell you that it exists). So, does
IDL
"actually create an instance of a structure" with this mechanism or not?
Thanks (in advance) for helping clear up my confusion.
--Harvey
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
