---
title: "Magma V2.19-4 Patch Notes"
canonical: "https://docs.magma-maths.org/patchnotes/2.19-4.html"
source: "https://docs.magma-maths.org/patchnotes/2.19-4.md"
---

<a id="magma-v2-19-4-patch-notes"></a>
# Magma V2.19-4 Patch Notes

## Algebraic Number Fields

- The check on the `Al` parameter to `Subfields` has been improved.
- A bug has been fixed in `RelativeField` when the first argument is a degree 1 extension of $\mathbb{Q}$.
- A bug has been fixed in the `KluenersvanHoeij` `Subfields` algorithm.
- `Subfields` are now returned in the same order each time they are requested for a field.

## Class Field Theory for Global Function Fields

- The computation of maximal orders of class fields of function fields has been improved.

## Commutative Algebra

- A bug when computing Gröbner bases of the integers has been fixed. Reported by E. Rains.

## General Local Fields

- A bug has been fixed when an attempt is made to construct a local field from a reducible polynomial. Reported by Lassina Dembele.

## Groups

- Two bugs in [`LowIndexNormalSubgroups`](../FinitelyPresentedGroups/FPGroups/subgroup-fin-index.md#function-lowindexnormalsubgroups-grpfp-rngintelt) have been fixed. Bugs reported by Primoz Potocnik.
- A bug in `pQuotient` has been fixed. The bug may have caused crashes or incorrect answers to be returned.
- The Magma level printing of pc-groups has been changed to use the [`CompactPresentation`](../FiniteGroups/FiniteSolubleGroups/more-presentations.md#function-compactpresentation-grppc) format, and the Magma level printing of elements of a pc-group within sets and sequences has been changed to use the `Eltseq` format. This has the advantage that Magma will read such sets and sequences correctly, and the disadvantage that the printing is not easily human readable.
- A crash in [`RepresentativeCocycles`](../FiniteGroups/FiniteSolubleGroups/central-extensions.md#function-representativecocycles-grppc-grppc-algmatelt-algmatelt) for a pc-group has been fixed. Bug reported by Stephen Humphries.

## Number Fields

- Additional functionality has been provided for characters over number fields with [`TotallyUnitTrivialSubgroup`](../GlobalFields/DirichletAndHeckeCharacters/introduction.md#function-totallyunittrivialsubgroup-grpdrchnf) for Dirichlet characters, [`NormInduction`](../GlobalFields/DirichletAndHeckeCharacters/introduction.md#function-norminduction-fldnum-grpdrchelt) for Hecke characters, and also [`GrossenTwist`](../GlobalFields/DirichletAndHeckeCharacters/introduction.md#function-grossentwist-grossenchar-list) for Grossencharacters.

## Polynomials

- A fix to [`GaussianBinomial`](../LieTheory/QuantumGroups/gauss.md#function-gaussianbinomial-rngintelt-rngintelt-rngelt) was made to automatically coerce an element from a polynomial ring into a function field. The same is also now done for [`GaussianFactorial`](../LieTheory/QuantumGroups/gauss.md#function-gaussianfactorial-rngintelt-rngelt) and [`GaussNumber`](../LieTheory/QuantumGroups/gauss.md#function-gaussnumber-rngintelt-rngelt).

## Polytopes

- A particular sequence of transformations on the `Zero` element of a toric lattice could result in `IsPrimitive` giving an incorrect answer; this has now been corrected.
- Under very special conditions, isomorphism testing of rational polytopes could fail with an assertion error. This has now been fixed. Reported by Antonio Laface and Damiano Testa.
- A bug has been fixed in `IsEquivalent` when analysing rational polytopes of non-zero codimension.
- The computation of the automorphism group of a polytope has been improved.

## Representation Theory

- A crash (runtime error) while computing [`SchurIndices`](../RepresentationTheory/CharactersOfFiniteGroups/schur-index.md#function-schurindices-algchtrelt) has been fixed. Bug reported by Alex Bartel.

## Toric Geometry

- The new fan constructor [`FanWithWeights`](../AlgebraicGeometry/ToricVarieties/toric-fans.md#function-fanwithweights-seqenum) has been added. This allows the creation of a fan based on the given weights and optional ample divisor.
- The choice of generators selected by `QuotientGradings` has been improved. Suggested by Antonio Laface and Damiano Testa.
- Attempts to construct a fan from contradictory cone data now gives the explanation of what went wrong as part of the error message, rather than as a separate piece of output.
- The computation time required by `IsFano` has been significantly reduced. A new intrinsic [`IsWeakFano`](../AlgebraicGeometry/ToricVarieties/toric-varieties.md#function-isweakfano-torvar) has been added.