X #= -Y fails in clp(FD)

Any ideas why? (Running Sicstus Prolog)

| ?- X in -1..1, Y in -1..1, X#= -Y.
! Domain error in user: #= /2
! expected constraint, found _1126#= -(_1124)
! goal:  _1126#= -(_1124)
! goal:  user:(_451#= -(_506))
! Domain error in user: #= /2
! expected constraint, found _140#= -(_105)
! goal:  _140#= -(_105)


Bes regards,

Andrey

after a quick look at the manual:

unary negation symbol is missing in clp(FD) language.
a workaround is to multiply by -1:

| ?- X in -1..1, Y in -1..1, X#= -1*(Y).
X in-1..1,
Y in-1..1 ?
yes

> Any ideas why? (Running Sicstus Prolog)
>
> | ?- X in -1..1, Y in -1..1, X#= -Y.
> ! Domain error in user: #= /2
> ! expected constraint, found _1126#= -(_1124)
> ! goal:  _1126#= -(_1124)
> ! goal:  user:(_451#= -(_506))
> ! Domain error in user: #= /2
> ! expected constraint, found _140#= -(_105)
> ! goal:  _140#= -(_105)
>
>
> Bes regards,
>
> Andrey

I think this is something that should be standardized for CLP(FD). In
B-Prolog, I got what are expected.

| ?- X in -1..1, Y in -1..1, X#= -Y,labeling([X,Y])
X = -1
Y = 1 ?;
X = 0
Y = 0 ?;
X = 1
Y = -1 ?;
no
| ?- X in -1..1, T = (X#=0), #\ T, labeling([], [X]).
X = -1
T = -1#=0 ?;
X = 1
T = 1#=0 ?;
no

Cheers,
Neng-Fa

"Andrey Rybalchenko" <first_dot_last_at_epfl_dot_ch> wrote in message
news:1187210291_654@sicinfo3.epfl.ch...
> after a quick look at the manual:
>
> unary negation symbol is missing in clp(FD) language.
> a workaround is to multiply by -1:
>
> | ?- X in -1..1, Y in -1..1, X#= -1*(Y).
> X in-1..1,
> Y in-1..1 ?
> yes
>