From help-octave-request at che dot utexas dot edu  Wed Mar 29 15:35:14 1995
Subject: Re: Octave binary format 
From: John Eaton <jwe at che dot utexas dot edu>
To: Ted dot Harding at nessie dot mcc dot ac dot uk (Ted Harding)
cc: help-octave at che dot utexas dot edu
Date: Wed, 29 Mar 95 09:34:56 CST

Ted dot Harding at nessie dot mcc dot ac dot uk (Ted Harding) wrote:

: Hello All:
: 
: A question in an earlier posting may have got lost in the discussion
: about speed of fprintf.
: 
: The question was: can anyone point me to a specification of the
: octave binary-save format (or perhaps the MatLab one)?

The Octave binary format is briefly described in the comments before
the function read_binary_data() in load-save.cc.  The value of the
`magic number' mentioned there can be either Octave-1-L or Octave-1-B,
which serves to identify that the file is an Octave binary data file,
version 1, and that it was written on either a little-endian or
big-endian machine.

The Matlab data file format is described in some Matlab manual.  I
think it is the `External Reference Guide'.

Code for reading and writing both of these formats is in load-save.cc.
(Yes, it would be useful to have these functions in a separate library
that could be called from C, C++, etc.).

jwe