bug in finite fields of degree 1

[tthsqe12]

LibNormal.normal can still return the wrong base_ring. Two steps to reproduce:

1. Change

   Singular.jl/src/number/NumberTypes.jl

   Line 244 in 4ee35af

   if n == 1

   to
   if false && n == 1

2. include this file multiple times in a row from the repl. It should eventually give an assertion error:

using Singular

   FiniteField(7, 1, "x", cached=false)

   F, (Fa,) = FunctionField(Fp(7), ["a"])
   K, a = AlgebraicExtensionField(F, Fa^2 + 1)
@show K
@show K.ptr.cpp_object
   R, (x, y, z) = PolynomialRing(K, ["x", "y", "z"])
   I = Ideal(R, a*z-x^4, z-y^6)
   l = Singular.LibNormal.normal(I)
   S = l[1][1][1]
@show K
@show K.ptr.cpp_object
@show base_ring(S)
   @assert base_ring(S) == K

[fingolfin]

Is this issue still around? The linked code comments suggests this is due to a bug in Singular, so perhaps @hannes14 needs to take a look ?

[tthsqe12]

Well, with

1. the above change to NumberTypes.jl
2. libSingular_julia at oscar-system/libsingular-julia@85d0dec
3. Singular at Singular/Singular@2ea97f9

things have improved. The above code just segfaults right away instead of making me include it multiple times!