[Numpy-discussion] segmentation fault

Shin, Daehyok sdhyok at catchlab.org
Tue Aug 3 08:39:14 EDT 2004


I am just playing with ieeespecial.py  
to see what happens if I tried some intuitive things.
I know the command has an error.
But, I think "segmentation fault" is too harsh as a response to the error.
Am I right?

Daehyok Shin

> -----Original Message-----
> From: Perry Greenfield [mailto:perry at stsci.edu]
> Sent: Tuesday, August 03, 2004 AM 10:22
> To: Shin; numpy
> Subject: Re: [Numpy-discussion] segmentation fault
> 
> 
> On 8/3/04 12:09 AM, "Shin" <sdhyok at email.unc.edu> wrote:
> 
> > The followind code generates "Process Python segmentation fault".
> > Is it expected, or a bug?
> > 
> I'd hazard that it is a bug since I'm not sure when a seg fault is really
> desired, but maybe that's just me ;-)
> 
> > from numarray import *
> > from numarray.ieeespecial import *
> > 
> > x = arange(10, typecode='d')
> > x[5] = nan
> > del x[getnan(x)]
> > 
> But note that this is an error of usage. You shouldn't be trying 
> to delete a
> part of an array. You can delete the array but here you appear to 
> be trying
> to delete a nan value. Arrays don't work like lists in this 
> regard. What are
> you trying to do?
> 
> Perry Greenfield
> 
> 




More information about the NumPy-Discussion mailing list