#657 addrow creates strange matrix

[Anonymous]

The following commands show problems with addrow:

(%i260) m : matrix([0, 0]);

(%o260) [ 0 0 ]
(%i261) m : addrow(m, m);

                             \[ 0  0 \]

(%o261) [ ]
[ 0 0 ]
(%i262) m[1, 1] : 10;

(%o262) 10
(%i263) m;
[ 10 0 ]
(%o263) [ ]
[ 10 0 ]

i.e. two elements, instead of one are modified!

Maxima information:

Maxima version: 5.9.1
Maxima build date: 7:34 9/24/2004
host type: i686-pc-mingw32
lisp-implementation-type: Kyoto Common Lisp
lisp-implementation-version: GCL 2.6.5

----

Ola Dahl
ola.dahl@ts.mah.se

[Robert Dodier]

Logged In: YES
user_id=501686

It appears that addrow copies a reference to the new rows,
instead of copying the rows. I would suggest that copying
the rows is more generally useful, but at present addrow
appears to function as intended.

Another example which shows that a change to a row changes
the matrix returned by addrow:

(%i8) m1:[0,0];
(%o8) [0, 0]
(%i9) m2:[0,0];
(%o9) [0, 0]
(%i10) m: matrix([0,0]);
(%o10) [ 0 0 ]
(%i11) n: addrow(m,m1,m2);
[ 0 0 ]
[ ]
(%o11) [ 0 0 ]
[ ]
[ 0 0 ]
(%i12) m1[2]:12;
(%o12) 12
(%i13) m2[1]:21;
(%o13) 21
(%i14) m;
(%o14) [ 0 0 ]
(%i15) n;
[ 0 0 ]
[ ]
(%o15) [ 0 12 ]
[ ]
[ 21 0 ]
(%i16) m[1,1]:11;
(%o16) 11
(%i17) n;
[ 11 0 ]
[ ]
(%o17) [ 0 12 ]
[ ]
[ 21 0 ]

[Robert Dodier]

Logged In: YES
user_id=501686

The stuff I wrote before, "at present addrow appears to
function as intended" is just wishful thinking. This is a
bug. Should be easy to fix.

[Dieter Kaiser]

Fixed in comm2.lisp revision 1.35.
Closing this bug report as fixed.
Dieter Kaiser