---
title: "Magma V2.22-9 Patch Notes"
canonical: "https://docs.magma-maths.org/patchnotes/2.22-9.html"
source: "https://docs.magma-maths.org/patchnotes/2.22-9.md"
---

<a id="magma-v2-22-9-patch-notes"></a>
# Magma V2.22-9 Patch Notes

## Aggregates

- `IsEmpty` now also works on tuples.

## Algebraic Fields

- An ideal of an order of an algebraic number or function field can be contructed from an element of the bottom coefficient ring (Z, a polynomial ring or a valuation ring) using `!!`. Reported by C. Quitte. This fixes a bug in [`RingClassGroup`](../GlobalFields/NumberFieldsAndOrders/ideal-class-group.md#function-ringclassgroup-rngord) when the input is an order of a quadratic field (`RngQuad`).

## Analytic Jacobians

- A bug reported by J.-S. Mueller was fixed.

## Class Field Theory

- A bug in `HeckeCharacterGroup` of an abelian number field was fixed. Reported by P. Stucky.

## Coding Theory

- The function `Transversal(C, V)`, where C is a linear code and V is its ambient space, has been added.

## Dirichlet characters

- Non-primitive characters over Q now include their modulus when printed.

## Elliptic Curves

- A problem with `GaloisRepresentation` was fixed. Reported by N. Freitas.
- Magma-level printing of elliptic curves over the rationals no longer prints the universe of the coefficient sequence.

## Extended Reals

- It is now allowed to use `div` to divide an infinity by an integer.

## Finite Fields

- Fixed a bug in `Log` that could cause a crash in rare circumstances.
- New procedure `SetZechLimit(n)`} is added, which allows one to set the limit $L$ so that the Zech representation will be used for any non-prime finite field with size at most $L$ (default is 2^{20} and max is 2^{30}).  The corresponding function `GetZechLimit()` returns the current limit.

## Groebner Bases

- A bug in the `FGLM` parameter handling for the function `GroebnerBasis` has been fixed.  Reported by M. Kudo.

## Groups

- A bug when constructing the centralizer of an element in a general linear group has been fixed. Bug reported by E. O’Brien and H. Dietrich.
- An incorrect type `FinFld` in signatures for `BorelSp` and `ExtendedWeylGroup` has been fixed.  (It is now the correct `FldFin`.)  Reported by G. Williamson.
- A crash to do with finitely presented groups has been fixed. Bug reported by I. M. Isaacs.

## Hyperelliptic Curves

- A bug in [`ShiodaInvariants`](../ArithmeticGeometry/HyperellipticCurves/invariants-isomorphisms.md#function-shiodainvariants-crvhyp) has been fixed.  Problem noted and solution supplied by J. Sijsling.

<a id="i-o"></a>
## I/O

- The intrinsics [`WriteCheck`](../MagmaLanguage/InputOutput/io-operations.md#function-writecheck-io-monstgelt) and [`WriteBytesCheck`](../MagmaLanguage/InputOutput/io-operations.md#function-writebytescheck-io-seqenum) have been added.

## Language

- A break or continue statement is now allowed from within a try-catch statement.
- A problem with `clear` where identifiers had wrong values has been fixed.  Problem reported by C. Meaney.

## Lattices

- A bug with internal precision in [`BKZ`](../LatticesQuadraticForms/Lattices/reduction.md#function-bkz-mtrx-rngintelt) was remedied.

## Newton Polygons

- A bug in the construction of a newton polygon of a polynomial with a zero coefficient over an infinite precision $p$-adic field has been fixed.

## Polynomial Rings

- The previously omitted function `SquarefreeFactorisation` has been added as a synonym for `SquarefreeFactorization`. Reported by C. Meaney.
- A crash in multivariate resultant over the integers has been fixed.  Reported by J. Doyle.

## Series Rings

- Memory handling when taking a fractional power of a series not over the rationals has been fixed. Reported by G. Blanco.

## System

- A slowdown in the memory manager when there were many free blocks of similar size has been fixed.  Problem reported by M. Grassl.

## p-adic Rings

- A bug with coercion sometimes failing for elements between two p-adic quotient rings with the same (large) p was fixed. Reported by J.-S. Mueller.
- It is now possible to apply `HenselLift` to a polynomial and a sequence of 1 polynomial, not just more than 1. Reported by C. Doris.