Fixed a bug when padding a zero-dimensional code. Reported by M. Grassl.

Fixed a bug causing unnecessary recomputation of minimum weight words in MinimumWord and MinimumWords. Reported by M. Grassl.

A bug in CosetLeaders over GF(16) has been fixed. Reported by M. Grassl.

A bug in ArithmeticGeometricCode for places over function fields has been fixed. The dimension of the resulting code now matches the dimension of the Riemann-Roch space. (Formerly, the dual of the appropriate code was being returned.) Reported by M. Grassl.

A bug in the Gaussian-integer discrete logarithm algorithm has been fixed.

A bug in the function field generation for curves that caused crashes in some "inseparable" cases has been fixed.

Bugs in HasOnlyOrdinarySingularities and IsNodalCurve have been fixed.

A bug in TestEquations has been fixed. Reported by S. Tengely.

Conversion of abelian groups to permutation groups now handles the trivial group without a runtime error being produced.

Fixed bug when computing stabilizer of subspace in a matrix group. Bug reported by Eamonn O'Brien.

Bug fix in Classes as called by MolienSeries of a matrix group.

Fixed memory leak when using SmallGroup to get certain groups of order 1536. Bug reported by Jack Schmidt.

The check for too many short vectors in lattice enumeration has been fixed so reasonable requests which were banned are now allowed. Problem reported by Ronald van Luijk.

A bug in the code for computing the minimum of an ideal that caused specifically constructed ideals to report wrong (too large) minima, has been fixed.

A bug in the general arithmetic of equation orders and number fields that would cause memory corruption has been fixed.

A bug in class field theory that would cause internal errors in certain computations with trivial class fields over ℚ has been fixed.

A bug in the ext<... constructor for number fields where the RHS is a list of element in the LHS has been fixed. Previously an internal error would be triggered upon creation of such a trival extension.

A bug in the precision management during the p-adic computation of Galois groups over ℚ has been fixed. Previously, an error would have been triggered. Reported by Brumer.

A precision problem in the creation of the Minkowski lattice from an order has been fixed. Due to the too low precision, the creation of the lattice could fail. Reported by N. Bruin.

A bug in factorization over function fields has been fixed (where coefficients were scaled incorrectly).

Fixed a bug in try/catch that could cause surrounding loops to fail in interesting ways.

Fixed a memory leak – errors handled by try/catch were not being deleted properly.

Absolute paths may now be used in package spec files.

A bug in the discriminant computation for p-adic field has been fixed. Previously, certain relative extensions would be reported as haing a discriminant of 0.