| Re: Fatal Bug in CONTOUR [message #30088] |
Tue, 09 April 2002 03:49 |
Wolfgang.Dobler
Messages: 4
Registered: April 2002
|
Junior Member |
|
|
Yes, 5.5 really solves the problem.
Sebastian Moeller <sebastian.moeller@lur.rwth-aachen.de> writes:
> it seems that V5.5 solves your problem. It gives a nice error message
> and does not hang, at least if the commands are given in the "command
> line" of the IDE. Here is the error message:
> % Program caused arithmetic error: Floating illegal operand
This one is just a warning, but then, indeed, it says
% CONTOUR: Invalid value specified for keyword LEVELS: No finite elements.
and exits gracefully. Exactly what one would expect from a reliable piece
of software.
W o l f g a n g
--
------------------------------------------------------------ -----------
| Wolfgang Dobler, Kiepenheuer Institute for Solar Physics, Freiburg |
| http://www.kis.uni-freiburg.de/~dobler/ |
------------------------------------------------------------ -----------
|
|
|
|
| Re: Fatal Bug in CONTOUR [message #30110 is a reply to message #30088] |
Mon, 08 April 2002 10:45  |
Sebastian Moeller
Messages: 7
Registered: April 2002
|
Junior Member |
|
|
Wolfgang.Dobler@kis.uni-freiburg.de wrote:
> Did anybody encounter this before:
>
> levs = alog(-2)*[1,2,3] & print, levs
> contour, randomu(!pi,10,10),LEVELS=levs
>
> hangs terminally (i.e. I cannot interrupt it and need to kill the IDL
> session). Happens with 5.4 and 5.2 (at least).
>
> Of course there is no point in contouring with all levels equal to
> NaNs. But I consider it a bad bug if my programs will hang terminally,
> just because I've fed them some bad data.
>
>
> W o l f g a n g
>
Hi there,
it seems that V5.5 solves your problem. It gives a nice error message
and oes not hang, at least if the commands are given in the "command
line" of the IDE. Here is the error message:
IDL Version 5.5, Microsoft Windows (Win32 x86). (c) 2001, Research
Systems, Inc.
IDL> levs = alog(-2)*[1,2,3] & print, levs
-NaN -NaN -NaN
% Program caused arithmetic error: Floating illegal operand
the contour command produces just an empty polot window (with axis), the
IDE was not blocked.
Ahoi
Sebastian Moeller
|
|
|
|
| Re: Fatal Bug in CONTOUR [message #30123 is a reply to message #30110] |
Mon, 08 April 2002 06:59 |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Wolfgang.Dobler (Wolfgang.Dobler@kis.uni-freiburg.de) writes:
> Of course there is no point in contouring with all levels equal to
> NaNs. But I consider it a bad bug if my programs will hang terminally,
> just because I've fed them some bad data.
I had another encounter with this unjustified faith in
technology this weekend. My middle son put the baked-on,
cheese-encrusted bowls of my famous Broccoli Ziti Bake
in the dishwasher without any pre-soaking, etc. I wonder
if this is another sign the apocalypse is upon us. :-(
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
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
