| IDL5 and large structures: Program code area full [message #10511] |
Tue, 09 December 1997 00:00  |
Alex Schuster
Messages: 124
Registered: February 1997
|
Senior Member |
|
|
Hi there!
There is this program, I didn't use it for a long time, and today I
started it for the first time under IDL 5.02. I have heard about some
strange bugs with structures containing many tags, but as I didn't have
problems with IDL 4, I am surprised I got a 'Program code full error'.
Any ideas? Hey, didn't RSI tell us all limitations are gone now? And why
do I have more limitations than before?
This is the code:
-------------------------snip------------------------------
some_variable = {_refima, $
disp : intarr (256, 256, 3), $
disp_3d : intarr (256, 256, 3), $
ident : 0, $
image_type : 0, $
filename : '', $
pix : fltarr(3), $
dim : intarr(3), $
dim_old : intarr(3), $
num_planes : 0, $
start_plane : 0, $
frame : 0, $
frame_old : 0, $
num_frames : 0, $
gate : 0, $
gate_old : 0, $
num_gates : 0, $
bed_pos : 0, $
bed_pos_old : 0, $
num_bed_pos : 0, $
header_size : 0L, $
offset : fltarr(3), $
angles : fltarr(3), $
center : fltarr(3), $
interpolation : 0, $
inv_flag : 0, $
scale_factor : 0.0, $
min : 0, $
max : 0, $
scal : intarr(2), $
diff : intarr(2), $
surf : intarr(2), $
grad : intarr(2), $
cont : intarr(2), $
quant : fltarr(10), $
num_contours : 0, $
contour_mode : 0, $
sigma : 0.0, $
average : intarr(3), $
filter_type : 0, $
kernel_size : 0, $
color_table : 0, $
gamma : 0.0, $
vox_flag : 0, $
render_control : 0, $
viewport : intarr (3), $
rotation : fltarr (3,3), $
surf_levels : intarr(2), $
mask : intarr(2), $
int_off : 0, $
int_len : 0, $
annotation : '' }
-------------------------snap------------------------------
I just pasted into the shell, and IDL says:
[...]
IDL> mask : intarr(2), $
IDL> int_off : 0, $
int_off : 0, $
^
% Program code area full.
IDL> int_len : 0, $
int_len : 0, $
^
% Syntax error.
IDL> annotation : !STR80}
IDL> help, !version, /structure
** Structure !VERSION, 5 tags, length=40:
ARCH STRING 'sparc'
OS STRING 'sunos'
OS_FAMILY STRING 'unix'
RELEASE STRING '5.0'
BUILD_DATE STRING 'Apr 28 1997'
Can anyone using IDL 5.02 be so kind check if this works? And, no, I
don't even think about modifying the program in any way.
Many thanks in advance,
Alex
--
Alex Schuster Wonko@weird.cologne.de PGP Key available
alex@pet.mpin-koeln.mpg.de
|
|
|
|
| Re: IDL5 and large structures: Program code area full [message #10538 is a reply to message #10511] |
Wed, 17 December 1997 00:00  |
R. Bauer
Messages: 137
Registered: November 1996
|
Senior Member |
|
|
Alex Schuster wrote:
> Wow, I get fast responses today!
>
> David Fanning helped me a lot when he wrote:
>
>> [reasonably long structure snipped]
>>
>>> Can anyone using IDL 5.02 be so kind check if this works?
>>
>> Well, it certainly works as you describe when it is "pasted"
>> onto the IDL command line, but why in the world would you
>> be doing this?
>
> This code is part of a script, which I invoke via @.
> The code I posted isn't the original code, I tried to make it more
> readable. In the original code, the variables (some_variable, str80
> etc.) are defined as system variables via DEFSYSV. I didn't like common
> blocks when I wrote the program, so I chose to use global system
> variables.
>
>>> And, no, I
>>> don't even think about modifying the program in any way.
>>
>> Well, I think I would modify it by adding an END statement
>> and at least running it as a main-level program. At least
>> then it will work. :-)
>>
>> IDL> .Run structure_def
>
> Good idea! I really should have tried this myself.
> After some changes (I define the structure in the program, and then,
> outside, the system variable) the code now compiles completely, and
> works.
>
An other way is to make a procedure from it e.g.
pro some,some_variable
some_variable = {_refima, $
disp : intarr (256, 256, 3), $
disp_3d : intarr (256, 256, 3), $
ident : 0, $
image_type : 0 }
end
The main program can compile it by resolve_routine and use execute in this
way.
pro main
resolve_routine,'some'
a=execute('some,some_variable')
help,some_variable,/str
end
I would prefer this solution because I don't like storing unused
variables or common blocks
--
R.Bauer
Institut fuer Stratosphaerische Chemie (ICG-1)
Forschungszentrum Juelich
email: R.Bauer@fz-juelich.de
|
|
|
|
| Re: IDL5 and large structures: Program code area full [message #10539 is a reply to message #10511] |
Wed, 17 December 1997 00:00 |
R. Bauer
Messages: 137
Registered: November 1996
|
Senior Member |
|
|
David Fanning wrote:
> Alex Schuster (alex@rosa.mpin-koeln.mpg.de) writes:
>
>> There is this program, I didn't use it for a long time, and today I
>> started it for the first time under IDL 5.02. I have heard about some
>> strange bugs with structures containing many tags, but as I didn't have
>> problems with IDL 4, I am surprised I got a 'Program code full error'.
>>
>> Any ideas? Hey, didn't RSI tell us all limitations are gone now? And why
>> do I have more limitations than before?
>>
>> This is the code:
>
> [reasonably long structure snipped]
>
>> Can anyone using IDL 5.02 be so kind check if this works?
>
> Well, it certainly works as you describe when it is "pasted"
> onto the IDL command line, but why in the world would you
> be doing this?
>
I did the same whith idl5.03
IDL> help,!version,/str
** Structure !VERSION, 5 tags, length=40:
ARCH STRING 'ibmr2'
OS STRING 'AIX'
OS_FAMILY STRING 'unix'
RELEASE STRING '5.0.3'
BUILD_DATE STRING 'Nov 3 1997'
and there NO Problems.
R.Bauer
--
R.Bauer
Institut fuer Stratosphaerische Chemie (ICG-1)
Forschungszentrum Juelich
email: R.Bauer@fz-juelich.de
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
