From help-request at octave dot org Tue Dec 14 08:37:25 2004
Subject: Re: writing integer with fwrite
From: "John W. Eaton" <jwe at bevo dot che dot wisc dot edu>
To: Reiner dot Suikat at dlr dot de
cc: help at octave dot org
Date: Tue, 14 Dec 2004 09:38:27 -0500

On 14-Dec-2004, David Bateman <David dot Bateman at motorola dot com> wrote:

| According to Reiner dot Suikat at dlr dot de <Reiner.Suikat@dlr.de> (on 12/14/04):
| > Hello and thanks for the quick answer,
| > 
| > But it doesn't solve the problem.
| > The statement:
| > cnt=fwrite(fid, bytes,'uint32')
| > Returns cnt=2, so it uses two uint32 values to store the one variable,
| > in Matlab it returns cnt=1, converting the value bytes to a
| > 32bit-Integer before writing. I need it to store as a 4byte value when
| > using uint32.
| > Any ideas?
| > I am using the precompiled windows version 2.150 from sourceforge.
| > Is this perhaps a known bug?
| 
| Use the uint32 from the latest versions of octave (2.1.64 for example)
| and cast as uint32 before calling fwrite
| 
| cnt=fwrite(fid, uint32(bytes), 'uint32')

It seems that with 2.1.64 the explicit type conversion is not
necessary:

  octave:1> fid = fopen ("x.out", "w");
  octave:2> cnt = fwrite (fid, pi, "ulong")
  cnt = 1
  octave:3> cnt = fwrite (fid, pi, "uint32")
  cnt = 1
  octave:4> fclose (fid);
  octave:5> fid = fopen ("x.out", "r");
  octave:6> fread (fid, 1, "ulong")
  ans = 3
  octave:7> fread (fid, 1, "uint32")
  ans = 3

This is on an x86 system running Debian.

BTW, the 2.1.50 snapshot was made almost 1-1/2 years ago.  You should
really try a later version.  2.1.64 is the current "testing" or
"recommended" version.

jwe



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------