From help-octave-request at che dot utexas dot edu  Mon Aug 29 16:24:58 1994
Subject: Re: Various questions 
From: John Eaton <jwe at che dot utexas dot edu>
To: wtc at courage dot arlut dot utexas dot edu (Wichaya Changwatchai)
cc: help-octave at che dot utexas dot edu (Octave help mailing list)
Date: Mon, 29 Aug 94 16:24:53 EDT

wtc at courage dot arlut dot utexas dot edu (Wichaya Changwatchai) wrote:

: 1) The section on range constants says that:
: 
: 	Octave does not convert range constants to vectors unless it is
: 	necessary to do so.
: 
: When would it NOT be necessary to convert a range constant to a
: vector?  Would an example be, say, a for loop?

Yes.  Also, assignment of a range to a variable does not cause a
conversion.

Currently, indexing a range will cause the original range to be
converted to a vector, and the result will be a vector, not a range.
This is because it was easier to implement -- the code for indexing
matrices already had to be written, and I have not had time to
implement a new set of functions for indexing ranges as a special
case.  This might be a good project for a volunteer.  :-)

: 2) Is there a way to subscript a constant range, e.g.
: 
: 	[2:2:10](2:4)
: 
: as opposed to having to assign it to a dummy variable:
: 
: 	a = [2:2:10];
: 	a(2:4)

Not yet.  This might be a useful extension, but it is not likely to
make it to the top of my to-do list any time soon.

: 3) The operators .+ and .- don't seem to work, as documented in the
: manual (of course, they ARE redundant with + and -).  Have they
: been disabled? 

Oops.  This will be fixed in the next release.

jwe