comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » Re: IDL5.2 and "corrupted descriptors"
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Return to the default flat view Create a new topic Submit Reply
Re: IDL5.2 and "corrupted descriptors" [message #13930] Mon, 28 December 1998 00:00 Go to previous message
rpina is currently offline  rpina
Messages: 6
Registered: December 1997
Junior Member
Mark Rivers wrote:

> In article <3687D882.6CAACF1@astro.ufl.edu>, "Robert K. Pina" <rpina@astro.ufl.edu> writes:
>> Hi,
>>
>> I have written some widget tools which I have been using with no
>> problems under IDL5.1 and 5.0 (and earlier) on a Windows NT 4.0 x86
>> platform. However, under IDL 5.2, I occassionally get the following
>> error message:
>>
>> %Array has a corrupted descriptor: "name"
>>
>> where "name" is the actual name of the array. There is no help
>> pertaining to "corrupted descriptors" in the documentation. I have been
>> unsuccessful in determining the cause of the error mesage. I kept my
>> IDL5.1 installation on the same host machine and the code runs with no
>> problems under 5.1. Any suggestions?
>
> Do your programs use "call_external" at all? That message typically happens
> when a call_external routine messes things with memory.
>
> Mark Rivers

Hi Mark,

No, the function which is producing the error does not use call_external. However, since you
mention something messing with memory, the array which is passed to this function is
manipulated using pointers in the calling program. Basically, the calling program is passed an
input data array for which I create a pointer:

pdata = ptr_new(data)

where "data" is the input data array. After doing various things, I call the function (which
produces the error) as

x = bad_function(*pdata, /misc_keywords)

I have also tried making a copy

y = *pdata

and then calling

x = bad_function(y,/misc_keywords)

but I get the same error. If I do something like

x = bad_function( randomn(seed,100),/misc_keywords)

it does not produce the error. So, you are right that it has something to do with memory
manipulation, in this case through the use of pointers. What I don't understand is why the
function worked with no problems under pre-5.2 versions.

Thanks for your input.

Robert



--
_______________________________________
_______________________________________

Dr. Robert K. Pina

Assistant Professor
Department of Astronomy
University of Florida
_______________________________________

Mailing Address:

211 BSSC / PO Box 112055
University of Florida
Gainesville, FL 32611-2055

Contact Info:

352-392-2054 (office)
352-846-1645 (laboratory)
352-392-5089 (fax)
rpina@astro.ufl.edu
[Message index]
 
Read Message
Read Message
Read Message
Previous Topic: 3D Scatter plot
Next Topic: IDL5.2 and "corrupted descriptors"

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Fri Oct 10 22:27:59 PDT 2025

Total time taken to generate the page: 0.95859 seconds