 edited description
Weird behaviour of SemigroupIdeal (?)
I just tried creating some ideals of matrix semigroups like so:
gap> S := RandomMatrixSemigroup(GF(4), 2, 4, [1,2]); <monoid of 4x4 smatrices over GF(2^2) with 2 generators> gap> Size(S); 21 gap> I := SemigroupIdeal(S, Idempotents(S)); <ideal of semigroup of 4x4 smatrices over GF(2^2) with 8 generators> gap> Size(I); 57
This strikes me as rather odd, so I thought it's a problem with the matrix code, but the transformations code behaves the same:
gap> T := AsTransformationSemigroup(S); <transformation monoid on 256 pts with 2 generators> gap> Size(T); 21 gap> I := SemigroupIdeal(T, Idempotents(T)); <regular transformation semigroup ideal on 256 pts with 8 generators> gap> Size(I); 57
The generators for the matrix semigroup are the following:
[ NewSMatrix(IsPlistSMatrixRep, GF(2^2), 4, [ [ Z(2)^0, 0*Z(2), 0*Z(2), 0*Z(2) ], [ 0*Z(2), Z(2)^0, 0*Z(2), 0*Z(2) ], [ 0*Z(2), 0*Z(2), Z(2)^0, 0*Z(2) ], [ 0*Z(2), 0*Z(2), 0*Z(2), Z(2)^0 ] ]), NewSMatrix(IsPlistSMatrixRep, GF(2^2), 4, [ [ Z(2^2), Z(2)^0, Z(2^2), Z(2)^0 ], [ Z(2^2)^2, Z(2^2), Z(2^2)^2, Z(2^2) ], [ Z(2)^0, Z(2^2)^2, Z(2)^0, Z(2^2)^2 ], [ Z(2)^0, Z(2^2)^2, Z(2)^0, Z(2^2)^2 ] ]), NewSMatrix(IsPlistSMatrixRep, GF(2^2), 4, [ [ 0*Z(2), Z(2^2)^2, 0*Z(2), Z(2)^0 ], [ Z(2^2)^2, Z(2)^0, 0*Z(2), Z(2)^0 ], [ Z(2^2), Z(2)^0, 0*Z(2), 0*Z(2) ], [ Z(2^2)^2, Z(2)^0, 0*Z(2), Z(2)^0 ] ]) ]
Sorry if this is a known issue on the main branch, I didn't dare merging yet.
Comments (8)


 edited description

This exists in other branches too, I wasn't aware of it. Something is wrong in the main algorithm for regular acting semigroup ideals. I will investigate and update.

 changed status to resolved
resolved Issue
#127→ <<cset 82761a0a7213>>

resolved Issue
#127→ <<cset 36b688bee7a8>>

resolved Issue
#127→ <<cset 00c3b081dec2>>

I fixed it, and I grafted the changes onto the tip of matrixsemigroups (so you don't have to merge to get it). Please let me know if something still doesn't work.

 removed milestone
Removing milestone: 2.5 (automated comment)
 Log in to comment