From bug-octave-request at bevo dot che dot wisc dot edu  Fri Sep 15 22:00:53 1995
Subject: Is that a bug in filter documentation?
From: John Eaton <jwe at bevo dot che dot wisc dot edu>
To: Michael Smolsky <fnsiguc at weizmann dot weizmann dot ac dot il>
Cc: bug-octave at bevo dot che dot wisc dot edu
Date: Fri, 15 Sep 1995 22:00:53 -0500

Michael Smolsky <fnsiguc at weizmann dot weizmann dot ac dot il> wrote:

: Well, compatibility with Matlab is really more important than with 
: sanity (I'm quite serious here). I found octave more compatible after I 
: added 
: 
: do_fortran_indexing =           'true';
: treat_neg_dim_as_zero =         'true';
: empty_list_elements_ok =        'true';
: implicit_str_to_num_ok =        'true';
: whitespace_in_literal_matrix =  'traditional';
: prefer_zero_one_indexing =      'true';
: prefer_column_vectors =         'false';
: 
: to ~/.octaverc. The last line wasn't recommended in your docs, as far as 
: I remember.

Version 1.2 will have a --traditional option that will set the
following variables:

  PS1				= ">> "
  PS2				= ""
  default_save_format		= "mat-binary"
  define_all_return_values	= "true"
  do_fortran_indexing		= "true"
  empty_list_elements_ok	= "true"
  implicit_str_to_num_ok	= "true"
  ok_to_lose_imaginary_part	= "true"
  page_screen_output		= "false"
  prefer_column_vectors		= "false"
  prefer_zero_one_indexing	= "true"
  print_empty_dimensions	= "false"
  treat_neg_dim_as_zero		= "true"
  warn_function_name_clash	= "false"
  whitespace_in_literal_matrix	= "traditional"


Thanks,

jwe