| IDL 8.0.1 crashes [message #74229] |
Wed, 05 January 2011 16:08  |
b_gom
Messages: 105
Registered: April 2003
|
Senior Member |
|
|
Is any one else experiencing random crashes with IDL 8 \ eclipse? IDL dies and disappears about 3 time an hour for me. There is no warning, and I can't find a pattern. Sometimes it happens when the help is open, sometimes not. Sometimes it is when I'm running an object graphics program, sometimes its just when I'm typing into the editor. I get the same behavior on two new windows 7 machines, 64 and 32 bits. I think I was getting similar crashes on XP. All the other software on these systems runs fine. I'm guessing it is eclipse related, but I could be wrong.
If anyone else is seeing this, is there any hope for a solution? I wish ITT had some sort of crash logging so that users could report how their buggy software is dying..
|
|
|
|
| Re: IDL 8.0.1 crashes [message #74304 is a reply to message #74229] |
Thu, 06 January 2011 11:01  |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Kenneth P. Bowman writes:
> So is the problem the Workbench? I run IDL 8 on Macs and never
> see crashes*, but then I don't use the Workbench.
As I understand the "theory of the problem" (and my
information is several months out of date), there
are problems in communicating between the IDL and
JAVA processes. These may well be JAVA problems,
rather than IDL problems, but in any case, there are
interaction problems.
In IDL 7 this was not that big a deal because the IDL
process and the JAVA (Workbench) process were two
separate processes, so that if the Workbench went down
the IDL process didn't. (Remember, you had to go spend
a couple of hours figuring out how to kill the IDL
process after one of these episodes so you could get things
going again.)
In IDL 8, the two processes (I'm waving my hands here)
were "merged" into a single process. The effect of this
was to make the Workbench significantly more "fragile"
on some machines, and on Windows machines in particular.
That's all I know about it, except that my way of working
in IDL (apparently!) is almost guaranteed to tickle
whatever problem it is that causes things to go South.
I basically can't get any work done in IDL 8 because it
won't stay alive long enough for me to finish the job. :-(
I will say this, though. I can't remember the last time
I had a problem with IDL 7.1 on my Windows machine. Whatever
IDL 8 put into the registry solved a heck of a lot of problems
that I used to have in IDL 7.1. I'm really happy with it,
although I would like to be participating more fully in
the discussion of the new language features. :-)
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.dfanning.com/
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|
|
|
|
| Re: IDL 8.0.1 crashes [message #80866 is a reply to message #74229] |
Thu, 19 July 2012 12:02 |
Leonardo R. Sbaraini
Messages: 26
Registered: October 2007
|
Junior Member |
|
|
The new Garbage collector detonated my old programs.
Example ?
Inside routine r2 getting an object-attribute from class, and then returning it to r1. The class cleanup will be called after r2 scope.
;----------------------------------------
pro class_define
struct = {class, obj : obj_new() }
end
;----------------------------------------
pro Class::Init
self.obj = obj_new('GenericClass')
end
;----------------------------------------
pro r2, obj
c = obj_new('class')
c->Get, obj
help, obj
end
;----------------------------------------
pro r1
r2, obj
help, obj
end
;----------------------------------------
OLD IDL:
OBJ OBJREF = <ObjHeapVar2(CLASS_OBJECT)>
OBJ OBJREF = <ObjHeapVar2(CLASS_OBJECT)>
SAME CODE, RUNNED IN THE NEW IDL
OBJ OBJREF = <ObjHeapVar2(CLASS_OBJECT)>
OBJ OBJREF = <ObjHeapVar2>
-----> FAIL <-----
Question: There is a compilation option ?
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
