Format problems

[John W. Eaton]

On  4-Aug-1998, address@hidden <address@hidden> wrote:

| I'm using octave-2.0.11 on an HPUX 10.20 platform. Being unable to
| compile octave due to the lack of a fortran compiler or f2c on my
| machine, I was forced to download the binary. I am now facing an
| annoying bug that is driving me out of my mind.
| 
| Specifically, the bug is the following: the size of a vector is always 
| given in exponential form that makes it hard to read. For example, if
| A is  a 20x20 matrix, the command size(A) gives the following :
| 
| >>  size(A)
| ans =
| 
|   1.0e+01  *
| 
|    2   2
| 
| 
| It gets worse. If A is a 20x200 matrix, here is what size(A) yeilds:
| 
| >>  size(A)
|  1.0e+02  *
| 
|     0    2
| 
| Note that I have the format set to short, and in fact, all other data
| is displayed properly. Is there any way around it?

This is a bug in the way formatting is handled when fixed_point_format
is nonzero.  As a quick fix, try adding

  fixed_point_format = 0;

to your ~/.octaverc file.

This problem is fixed in 2.0.13 so that it works correctly even if you
specify fixed_point_format = 1.

jwe