> < ^ Date: Tue, 15 Aug 2000 11:42:58 -0400 (EDT)
> < ^ From: Alexander Hulpke <hulpke@math.colostate.edu >
< ^ Subject: Re: New functions for Association Scheme and one question

Dear Gap-Forum,

```I encountered one strange behavior of GAP4 4r2 fix4.
That is
gap>W:=[[1,1,1],[1, E(3), E(3)^2],[1,E(3)^2,E(3)]];;
gap>Sum([[1,2]],i-> W{[i]});
[[1,1,1],[1,E(3),E(3)^2]]
```

In this form as you give it, the input produces an error on my machine.

However I believe you might have gotten distracted by the number of
different brackets and ways of indexing:

```gap> Sum([1,2],i-> W[i]);
[ 2, -E(3)^2, -E(3) ]
produces the sum of the first two rows of W: W[1]+W[2]. I suppose this is
what you want.

gap> Sum([1,2],i-> W{[i]});
[ [ 2, -E(3)^2, -E(3) ] ]
```

returns the sum W{[1]}+W{[2]}, the sum of the matrices obtained by cutting out
the first row and cutting out the second row.

```gap> Sum([[1,2]],i-> W{i});
[ [ 1, 1, 1 ], [ 1, E(3), E(3)^2 ] ]
```

(the result you got) is a sum over only one entry (namly the index set
[1,2]). It is the matrix obtained by cutting out the first two rows of W.

As far as I can see, this behaviour is sensible and according to the
description in the manual.

Best regards,

Alexander Hulpke

> < [top]