If we adjoin a square root of -1, usually denoted by *i*, to the field of
rationals we obtain a field that is an extension of degree 2. This field
is called the **Gaussian rationals** and its ring of integers is called the
**Gaussian integers**, because C.F. Gauss was the first to study them.

In **GAP** Gaussian rationals are written in the form

,
where `a` + `b`*E(4)`a` and `b` are rationals, because `E(4)`

is **GAP**'s name for
*i*. Because 1 and *i* form an integral base the Gaussian integers are
written in the form

, where `a` + `b`*E(4)`a` and `b` are integers.

The first sections in this chapter describe the operations applicable to Operations for Gaussians).

The next sections describe the functions that test whether an object is a Gaussian rational or integer (see IsGaussRat and IsGaussInt).

The **GAP** object `GaussianRationals`

is the field domain of all Gaussian
rationals, and the object `GaussianIntegers`

is the ring domain of all
Gaussian integers. All set theoretic functions are applicable to those
two domains (see chapter Domains and Set Functions for Gaussians).

The Gaussian rationals form a field so all field functions, e.g., `Norm`

,
are applicable to the domain `GaussianRationals`

and its elements (see
chapter Fields and Field Functions for Gaussian Rationals).

The Gaussian integers form a Euclidean ring so all ring functions, e.g.,
`Factors`

, are applicable to `GaussianIntegers`

and its elements (see
chapter Rings, Ring Functions for Gaussian Integers, and
TwoSquares).

The field of Gaussian rationals is just a special case of cyclotomic fields, so everything that applies to those fields also applies to it (see chapters Cyclotomics and Subfields of Cyclotomic Fields).

All functions are in the library file `LIBNAME/"gaussian.g"`

.

- Comparisons of Gaussians
- Operations for Gaussians
- IsGaussRat
- IsGaussInt
- Set Functions for Gaussians
- Field Functions for Gaussian Rationals
- Ring Functions for Gaussian Integers
- TwoSquares

`x` = `y`

`x` < `y`

The equality operator evaluates to `true`

if the two Gaussians `x` and
`y` are equal, and to `false`

otherwise. The inequality operator `<`

evaluates to `true`

if the two Gaussians `x` and `y` are not equal, and
to `false`

otherwise. It is also possible to compare a Gaussian with an
object of another type, of course they are never equal.

Two Gaussians

and `a` + `b`*E(4)

are considered
equal if `c` + `d`*E(4)

and `a` = `c`

.
`b` = `d`

gap> 1 + E(4) = 2 / (1 - E(4)); true gap> 1 + E(4) = 1 - E(4); false gap> 1 + E(4) = E(6); false

`x` < `y`

`x` <= `y`

`x` `y`

`x` = `y`

The operators `<`

, `<=`

, `, and `

`=`

evaluate to `true`

if the
Gaussian `x` is less than, less than or equal to, greater than, and
greater than or equal to the Gaussian `y`, and to `false`

otherwise.
Gaussians can also be compared to objects of other types, they are
Comparisons of Cyclotomics).

A Gaussian

is considered less than another Gaussian
`a` + `b`*E(4)

if `c` + `d`*E(4)`a` is less than `c`, or if `a` is equal to `c` and
`b` is less than `d`.

gap> 1 + E(4) < 2 + E(4); true gap> 1 + E(4) < 1 - E(4); false gap> 1 + E(4) < 1/2; false

`x` + `y`

`x` - `y`

`x` * `y`

`x` / `y`

The operators `+`

, `-`

, `*`

, and `/`

evaluate to the sum, difference,
product, and quotient of the two Gaussians `x` and `y`. Of course either
operand may also be an ordinary rational (see Rationals), because the
rationals are embedded into the Gaussian rationals. On the other hand
the Gaussian rationals are embedded into other cyclotomic fields, so
either operand may also be a cyclotomic (see Cyclotomics). Division by
0 is as usual an error.

`x` ^ `n`

The operator `^`

evaluates to the `n`-th power of the Gaussian rational
`x`. If `n` is positive, the power is defined as the `n`-fold product

; if `x`*`x`*...`x``n` is negative, the power is defined as
`(1/`

; and if `x`)^(-`n`)`n` is zero, the power is 1, even if `x` is 0.

gap> (1 + E(4)) * (E(4) - 1); -2

`IsGaussRat( `

`obj` )

`IsGaussRat`

returns `true`

if `obj`, which may be an object of arbitrary
type, is a Gaussian rational and `false`

otherwise. Will signal an error
if `obj` is an unbound variable.

gap> IsGaussRat( 1/2 ); true gap> IsGaussRat( E(4) ); true gap> IsGaussRat( E(6) ); false gap> IsGaussRat( true ); false

`IsGaussInt`

can be used to test whether an object is a Gaussian integer
(see IsGaussInt).

`IsGaussInt( `

`obj` )

`IsGaussInt`

returns `true`

if `obj`, which may be an object of arbitrary
type, is a Gaussian integer, and false otherwise. Will signal an error
if `obj` is an unbound variable.

gap> IsGaussInt( 1 ); true gap> IsGaussInt( E(4) ); true gap> IsGaussInt( 1/2 + 1/2*E(4) ); false gap> IsGaussInt( E(6) ); false

`IsGaussRat`

can be used to test whether an object is a Gaussian rational
(see IsGaussRat).

As already mentioned in the introduction of this chapter the objects
`GaussianRationals`

and `GaussianIntegers`

are the domains of Gaussian
rationals and integers respectively. All set theoretic functions, i.e.,
`Size`

and `Intersection`

, are applicable to these domains and their
elements (see chapter Domains). There does not seem to be an important
use of this however. All functions not mentioned here are not treated
specially, i.e., they are implemented by the default function mentioned
in the respective section.

`in`

The membership test for Gaussian rationals is implemented via
`IsGaussRat`

(IsGaussRat). The membership test for Gaussian integers
is implemented via `IsGaussInt`

(see IsGaussInt).

`Random`

A random Gaussian rational

is computed by combining two
random rationals `a` + `b`*E(4)`a` and `b` (see Set Functions for Rationals).
Likewise a random Gaussian integer

is computed by
Set Functions for Integers).
`a` + `b`*E(4)

gap> Size( GaussianRationals ); "infinity" gap> Intersection( GaussianIntegers, [1,1/2,E(4),-E(6),E(4)/3] ); [ 1, E(4) ]

As already mentioned in the introduction of this chapter, the domain of Gaussian rationals is a field. Therefore all field functions are applicable to this domain and its elements (see chapter Fields). This section gives further comments on the definitions and implementations of those functions for the the Gaussian rationals. All functions not mentioned here are not treated specially, i.e., they are implemented by the default function mentioned in the respective section.

`Conjugates`

The field of Gaussian rationals is an extension of degree 2 of the
rationals, its prime field. Therefore there is one further conjugate of
every element

, namely `a` + `b`*E(4)

.
`a` - `b`*E(4)

`Norm`

, `Trace`

According to the definition of conjugates above, the norm of a Gaussian
rational

is `a` + `b`*E(4)

and the trace is
`a`^2 + `b`^2`2*`

.
`a`

As already mentioned in the introduction to this chapter, the ring of Gaussian integers is a Euclidean ring. Therefore all ring functions are applicable to this ring and its elements (see chapter Rings). This section gives further comments on the definitions and implementations of those functions for the Gaussian integers. All functions not mentioned here are not treated specially, i.e., they are implemented by the default function mentioned in the respective section.

`IsUnit`

, `Units`

, `IsAssociated`

, `Associates`

The units of `GaussianIntegers`

are `[ 1, E(4), -1, -E(4) ]`

.

`StandardAssociate`

The standard associate of a Gaussian integer `x` is the associated
element `y` of `x` that lies in the first quadrant of the complex plane.
That is `y` is that element from

that has
positive real part and nonnegative imaginary part.
`x` * [1,-1,E(4),-E(4)]

`EuclideanDegree`

The Euclidean degree of a Gaussian integer `x` is the product of `x` and
its complex conjugate.

`EuclideanRemainder`

Define the integer part `i` of the quotient of `x` and `y` as the point
of the lattice spanned by 1 and `E(4)`

that lies next to the rational
quotient of `x` and `y`, rounding towards the origin if there are several
such points. Then `EuclideanRemainder( `

is defined as `x`, `y` )

. With this definition the ordinary Euclidean algorithm for
the greatest common divisor works, whereas it does not work if you always
round towards the origin.
`x` -
`i` * `y`

`EuclideanQuotient`

The Euclidean quotient of two Gaussian integers `x` and `y` is the
quotient of *w* and `y`, where *w* is the difference between `x` and the
Euclidean remainder of `x` and `y`.

`QuotientRemainder`

`QuotientRemainder`

uses `EuclideanRemainder`

and `EuclideanQuotient`

.

`IsPrime`

, `IsIrreducible`

Since the Gaussian integers are a Euclidean ring, primes and irreducibles
are equivalent. The primes are the elements `1 + E(4)`

and `1 - E(4)`

of
norm 2, the elements

and `a` + `b`*E(4)

of norm
`a` - `b`*E(4)

with `p` = `a`^2 + `b`^2`p` a rational prime congruent to 1 mod 4,
and the elements `p` of norm

with `p`^2`p` a rational prime congruent
to 3 mod 4.

`Factors`

The list returned by `Factors`

is sorted according to the norms of the
primes, and among those of equal norm with respect to `<`

. All elements
in the list are standard associates, except the first, which is
multiplied by a unit as necessary.

The above characterization already shows how one can factor a Gaussian
integer. First compute the norm of the element, factor this norm over
the rational integers and then split 2 and the primes congruent to 1 mod
4 with `TwoSquares`

(see TwoSquares).

gap> Factors( GaussianIntegers, 30 ); [ -1-E(4), 1+E(4), 3, 1+2*E(4), 2+E(4) ]

`TwoSquares( `

`n` )

`TwoSquares`

returns a list of two integers *x<=y* such that the sum of
the squares of *x* and *y* is equal to the nonnegative integer `n`, i.e.,
*n = x^2+y^2*. If no such representation exists `TwoSquares`

will return
`false`

. `TwoSquares`

will return a representation for which the gcd of
*x* and *y* is as small as possible. If there are several such
representations, it is not specified which one `TwoSquares`

returns.

Let *a* be the product of all maximal powers of primes of the form *4k+3*
dividing *n*. A representation of *n* as a sum of two squares exists if
and only if *a* is a perfect square. Let *b* be the maximal power of *2*
dividing *n*, or its half, whichever is a perfect square. Then the
minimal possible gcd of *x* and *y* is the square root *c* of *a b*. The
number of different minimal representations with *x<=y* is *2^{l-1}*,
where *l* is the number of different prime factors of the form *4k+1* of
*n*.

gap> TwoSquares( 5 ); [ 1, 2 ] gap> TwoSquares( 11 ); false # no representation exists gap> TwoSquares( 16 ); [ 0, 4 ] gap> TwoSquares( 45 ); [ 3, 6 ] # 3 is the minimal possible gcd because 9 divides 45 gap> TwoSquares( 125 ); [ 2, 11 ] # not [ 5, 10 ] because this has not minimal gcd gap> TwoSquares( 13*17 ); [ 5, 14 ] # [10,11] would be the other possible representation gap> TwoSquares( 848654483879497562821 ); [ 6305894639, 28440994650 ] # 848654483879497562821 is prime

GAP 3.4.4

April 1997