From bug-octave-request at bevo dot che dot wisc dot edu  Fri Jan 15 11:40:15 1999
Subject: Re: [lost mail? ]`list' behaviour different in octave-2.1.10 	 and2.1.12
From: "A. Scottedward Hodel" <scotte at eng dot auburn dot edu>
To: Joao Cardoso <jcardoso at inescn dot pt>
CC: bug-octave at bevo dot che dot wisc dot edu
Date: Fri, 15 Jan 1999 11:39:11 -0600 (CST)

John and I hashed this over a bit when he added the list data structure.  
The problem is a discrepancy in 2.1.10 between what happens when you say

a = list(...)
    x = a(2)
    x = a([2 3 4])
2.1.12 returns a list for both of these, while 2.1.10 returns a the
list value for the first.  That can lead to lots of irritating problems
when doing list operations, since if an index vector drops to length 1 you
get a value instead of a list with a single value.  

The new list function "nth" replaces the singleton case:

    x = nth(a,2)

will do what you want, and allows the above list operations to
consistently return lists, even if there's only one item in the list.
I think the matlab cell structure requires a similar kind of operation,
but I haven't used it in several months so I don't recall the details.

A S Hodel Assoc. Prof. Dept Elect Eng, Auburn Univ,AL  36849-5201
On leave at NASA Marshall Space Flight Center (256) 544-1426
Address:Mail Code ED-13, MSFC, Alabama, 35812
http://www.eng.auburn.edu/~scotte

----------
>From: Joao Cardoso <jcardoso at inescn dot pt>
>To: John Eaton <jwe at bevo dot che dot wisc dot edu>
>Subject: [lost mail? ]`list' behaviour different in octave-2.1.10 and 2.1.12
>Date: Fri, Jan 15, 1999, 10:53 AM
>

>------- start of forwarded message -------
>
>[Some days ago I sent this e-mail that seems to be lost; I resent it
>here again]
>
>[I think the problem was that your address on the list was
>jcardoso at bart dot inescn dot pt, but this message came from
>jcardoso at inescn dot pt dot   I changed your address on the list and am
>reposting the message for you.  The odd thing is that SmartList did
>not notify me that the message had not been posted.  Hmm.  --jwe]
>
>Hi,
>
>In octave-2.1.10 one could retrive an element from a list and use it as
>normal variable. This behaviour has been changed in octave-2.1.12. Is
>this deliberate? If it is, I can't see how to use lists...
>
>In 2.1.10:
>  
>   octave-2.1.10:1> a=list(1,2,3);
>   octave-2.1.10:2> a(2)*3
>   ans = 6
>
>in 2.1.12:
>
>   octave-2.1.12:1> a=list(1,2,3);
>   octave-2.1.12:2> a(2)*3
>   error: binary operator `*' not implemented for `list' by `scalar'
>operations
>   error: evaluating binary operator `*' near line 2, column 5
>   octave-2.1.12:2> t=a(2)
>   t =
>   (
>     [1] = 2
>   )
>   octave-2.1.12:3> t*2
>   error: binary operator `*' not implemented for `list' by `scalar'
>operations
>   error: evaluating binary operator `*' near line 3, column 2
>   octave-2.1.12:3> 
>
>This happens also for matrices.
>
>Thanks,
>Joao
>
>-- 
>Joao Cardoso, INESC  |  e-mail:
>R. Jose Falcao 110   |  tel:    + 351 2 2094345
>4050 Porto, Portugal |  fax:    + 351 2 2008487
>