[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: adding an op
From: |
John W. Eaton |
Subject: |
Re: adding an op |
Date: |
Wed, 1 Jun 2005 14:29:00 -0400 |
On 1-Jun-2005, Tom Holroyd wrote:
[Some background: Tom asked about extending Octave to handle CELL ==
CELL operations. Apparently, this is not allowed by Matlab, but it
might be useful. --jwe]
| When I saw that error, though, I just assumed that matlab could do
| it. It seems obvious. The current implementation of setdiff would
| "just work" if == could do cells, though.
I think we might also want to support other logical tests. What would
the return value be for CELL OP CELL where OP is one of <, <=, ==, >=,
>, or != (I'm assuming that it would work element-by-element in all
cases)?
It seems to me that it should be a cell array of values, and that the
values would be the result of the element-by-element operations. Then
an expression like
{[1,2,3], 1, "xyz"} == {[3,2,1], 2, "zyx"}
would produce
{[0,1,0], 0, [0,1,0]}
(with all elements having a logical type).
Does that seem reasonable? Attempting to compare structures would
fail until someone implements the corresponding operators for
structures, which I suppose could be handled in a similar
element-by-element fashion. Would this really be useful?
jwe
- Re: adding an op,
John W. Eaton <=