This document describes a package for the GAP3 group theory langauge which enables computations with the equivalent notions of finite, permutation crossed modules and cat1-groups.
The package divides into six parts, each of which has its own introduction:
• for constructing crossed modules and their morphisms in section \refAbout crossed modules: About crossed modules;
• for cat1-groups, their morphisms, and for converting between crossed modules and cat1-groups, in section \refAbout cat1-groups: About cat1-groups;
• for derivations and sections and the monoids which they form under the Whitehead multiplication, in section \refAbout derivations and sections: About derivations and sections;
• for actor crossed modules, actor cat1-groups and the actors squares which they form, in section \refAbout actors: About actors;
• for the construction of induced crossed modules and induced cat1-groups, in section \refAbout induced constructions: About induced constructions;
• for a collection of utility functions in section \refAbout utilities: About utilities.
These seven About...
sections are collected together in a separate
LaTeX file, xmabout.tex
, which forms a short introduction to the package.
The package may be obtained as a compressed file by ftp from one of
the sites with a GAP3 archive. After decompression, instructions
for installing the package may be found in the README
file.
The following constructions are planned for the next version of the package. Firstly, although sub-crossed module functions have been included, the equivalent set of sub-cat1-groups functions is not complete. Secondly, functions for pre-crossed modules, the Peiffer subgroup of a pre-crossed module and the associated crossed modules, will be added. Group-graphs provide examples of pre-crossed modules and their implementation will require interaction with graph-theoretic functions in GAP3. Crossed squares and the equivalent cat2-groups are the structures which arise as "three-dimensional groups". Examples of these are implicitly included already, namely inclusions of normal sub-crossed modules, and the inner morphism from a crossed module to its actor (section \refInnerMorphism for crossed modules).
The term crossed module was introduced by J. H. C. Whitehead in xmodW2, xmodW1. In xmodL1 Loday reformulated the notion of a crossed module as a cat1-group. Norrie xmodN1, xmodN2 and Gilbert xmodG1 have studied derivations, automorphisms of crossed modules and the actor of a crossed module, while Ellis xmodE1 has investigated higher dimensional analogues. Properties of induced crossed modules have been determined by Brown, Higgins and Wensley in xmodBH1, xmodBW1 and xmodBW2. For further references see xmodAW1 where we discuss some of the data structures and algorithms used in this package, and also tabulate isomorphism classes of cat1-groups up to size 30.
We first recall the descriptions of three equivalent categories: \textbfXMod, the category of crossed modules and their morphisms; \textbfCat1, the category of cat1-groups and their morphisms; and \textbfGpGpd, the subcategory of group objects in the category \textbfGpd of groupoids. We also include functors between these categories which exhibit the equivalences. Most papers on crossed modules use left actions, but we give the alternative right action axioms here, which are more suitable for use in computational group theory programs.
A crossed module X = (∂ : S → R ) consists of a group homomorphism ∂ , called the boundary of \mathcalX, with source S and range R, together with an action α : R → \mathrmAut(S) satisfying, for all s,s1,s2 ∈ S and r ∈ R,
• A conjugation crossed module is an inclusion of a normal subgroup S \unlhd R, where R acts on S by conjugation.
• A central extension crossed module has as boundary a surjection ∂ : S → R with central kernel, where r ∈ R acts on S by conjugation with ∂-1r.
• An automorphism crossed module has as range a subgroup R of the automorphism group Aut(S) of S which contains the inner automorphism group of S. The boundary maps s ∈ S to the inner automorphism of S by s.
• A trivial action crossed module ∂ : S → R has sr = s for all s ∈ S, r ∈ R, the source is abelian and the image lies in the centre of the range.
• An R-Module crossed module has an R-module as source and the zero map as boundary.
• The direct product X1 × X2 of two crossed modules has source S1 × S2, range R1 × R2 and boundary ∂1 × ∂2, with R1, R2 acting trivially on S2, S1 respectively.
A morphism between two crossed modules X1 = (∂1 : S1 → R1) and X2 = (∂2 : S2 → R2) is a pair (σ, ρ), where σ : S1 → S2 and ρ : R1 → R2 are homomorphisms satisfying
∂2 σ = ρ ∂1, σ(sr) = (σ s)ρ r. |
When X1 = X2 and σ, ρ are automorphisms then (σ, ρ) is an automorphism of X1. The group of automorphisms is denoted by \mathrmAut( X1 ).
X
is a record with fields:
beg-tabularll
X.source
, & the source S of ∂,
X.boundary
, & the homomorphsim ∂,
X.range
, & the range R of ∂,
X.aut
, & a group of automorphisms of S,
X.action
, & a homomorphism from R to X.aut
,
X.isXMod
, & a boolean flag, normally true
,
X.isDomain
, & always true,
X.operations
, & special set of operations XModOps
(see \refOperations for crossed modules),
X.name
, & a concatenation of the names of the source and range.
end-tabular
Here is a simple example of an automorphism crossed module, the holomorph of the cyclic group of size five.
gap> c5 := CyclicGroup( 5 );; c5.name := "c5";; gap> X1 := AutomorphismXMod( c5 ); Crossed module [c5->PermAut(c5)] gap> XModPrint( X1 ); Crossed module [c5->PermAut(c5)] :- : Source group c5 has generators: [ (1,2,3,4,5) ] : Range group = PermAut(c5) has generators: [ (1,2,4,3) ] : Boundary homomorphism maps source generators to: [ () ] : Action homomorphism maps range generators to automorphisms: (1,2,4,3) --> { source gens --> [ (1,3,5,2,4) ] } This automorphism generates the group of automorphisms.
Implementation of the standard constructions is described in sections
ConjugationXMod
, CentralExtensionXMod
, AutomorphismXMod
,
TrivialActionXMod
and RModuleXMod
. With these building blocks,
sub-crossed modules SubXMod
, quotients of normal sub-crossed modules
FactorXMod
and direct products XModOps.DirectProduct
may be
constructed. An extra function XModSelect
is used to call these
constructions using groups of order up to 47 and data from file in
Cat1List
.
A morphism from a crossed module X1 to a crossed module
X2 is a pair of homomorphisms (σ, ρ), where
σ, ρ are respectively homomorphisms between the sources and
ranges of X1 and X2, which commute with the two
boundary maps and which are morphisms for the two actions. In the
following code we construct a simple automorphism of X1
.
gap> sigma1 := GroupHomomorphismByImages( c5, c5, [ (1,2,3,4,5) ], > [ (1,5,4,3,2) ] );; gap> rho1 := InclusionMorphism( X1.range, X1.range );; gap> mor1 := XModMorphism( X1, X1, [ sigma1, rho1 ] ); Morphism of crossed modules <[c5->PermAut(c5)] >-> [c5->PermAut(c5)]> gap> IsXModMorphism( mor1 ); true gap> XModMorphismPrint( mor1 ); Morphism of crossed modules :- : Source = Crossed module [c5->PermAut(c5)] with generating sets: [ (1,2,3,4,5) ] [ (1,2,4,3) ] : Range = Source : Source Homomorphism maps source generators to: [ (1,5,4,3,2) ] : Range Homomorphism maps range generators to: [ (1,2,4,3) ] : isXModMorphism? true gap> IsAutomorphism( mor1 ); true
The functors between \textbfXMod and \textbfCat1, are implemented
as functions XModCat1
and Cat1XMod
.
An integer variable XModPrintLevel
is set initially equal to 1
.
If it is increased, additional information is printed out during the
execution of many of the functions.
XMod( f, a )
A crossed module is determined by its boundary and action homomorphisms, f and a. All the standard constructions described below call this function after constructing the two homomorphisms. In the following example we construct a central extension crossed module s3 × c4 → s3 directly by defining the projection on to the first factor to be the boundary map, and constructing the automorphism group by taking two inner automorphisms as generators.
gap> s3c4 := Group( (1,2),(2,3),(4,5,6,7));; gap> s3c4.name := "s3c4";; gap> s3 := Subgroup( s3c4, [ (1,2), (2,3) ] );; gap> s3.name := "s3";; gap> # construct the boundary gap> gen := s3c4.generators;; gap> imb := [ (1,2), (2,3), () ];; gap> bX := GroupHomomorphismByImages( s3c4, s3, gen, imb );; gap> # construct the inner automorphisms by (1,2) and (2,3) gap> im1 := List( gen, g -> g^(1,2) );; gap> a1 := GroupHomomorphismByImages( s3c4, s3c4, gen, im1 );; gap> im2 := List( gen, g -> g^(2,3) );; gap> a2 := GroupHomomorphismByImages( s3c4, s3c4, gen, im2 );; gap> A := Group( a1, a2 );; gap> # construct the action map from s3 to A gap> aX := GroupHomomorphismByImages( s3, A, [(1,2),(2,3)], [a1,a2] );; gap> X := XMod( bX, aX ); Crossed module [s3c4->s3]
IsXMod( X )
This Boolean function checks that the five main fields of X exist and that the crossed module axioms are satisfied.
gap> IsXMod( X ); true
XModPrint( X )
This function is used to display the main fields of a crossed module.
gap> XModPrint( X ); Crossed module [s3c4->s3] :- : Source group s3c4 has generators: [ (1,2), (2,3), (4,5,6,7) ] : Range group has parent ( s3c4 ) and has generators: [ (1,2), (2,3) ] : Boundary homomorphism maps source generators to: [ (1,2), (2,3), () ] : Action homomorphism maps range generators to automorphisms: (1,2) --> { source gens --> [ (1,2), (1,3), (4,5,6,7) ] } (2,3) --> { source gens --> [ (1,3), (2,3), (4,5,6,7) ] } These 2 automorphisms generate the group of automorphisms.
ConjugationXMod( R [,S] )
This construction returns the crossed module whose source S is a normal subgroup of the range R, the boundary is the inclusion map, the group of automorphisms is the inner automorphism group of S, and the action maps an element of r ∈ R to conjugation of S by r. The default value for S is R.
gap> s4 := Group( (1,2,3,4), (1,2) );; gap> a4 := Subgroup( s4, [ (1,2,3), (2,3,4) ] );; gap> k4 := Subgroup( a4, [ (1,2)(3,4), (1,3)(2,4) ] );; gap> s4.name := "s4";; a4.name := "a4";; k4.name := "k4";; gap> CX := ConjugationXMod( a4, k4 ); Crossed module [k4->a4]
XModName( X )
Whenever the names of the source or range of X
are changed, this
function may be used to produce the new standard form
[X.source.name->X.range.name]
for the name of X
. This function is
called automatically by XModPrint
.
gap> k4.name := "v4";; gap> XModName( CX ); "[v4->a4]"
CentralExtensionXMod( f )
This construction returns the crossed module whose boundary f is a surjection from S to R having as kernel a subgroup of the centre of S. The action maps an element of r ∈ R to conjugation of S by f-1r.
gap> d8 := Subgroup( s4, [ (1,2,3,4), (1,3) ] );; d8.name := "d8";; gap> gend8 := d8.generators;; genk4 := k4.generators;; gap> f := GroupHomomorphismByImages( d8, k4, gend8, genk4 );; gap> EX := CentralExtensionXMod( f ); Crossed module [d8->v4] gap> XModPrint( EX ); Crossed module [d8->v4] :- : Source group d8 has parent s4 and generators: [ (1,2,3,4), (1,3) ] : Range group k4 has parent s4 and generators: [ (1,2)(3,4), (1,3)(2,4) ] : Boundary homomorphism maps source generators to: [ (1,2)(3,4), (1,3)(2,4) ] : Action homomorphism maps range generators to automorphisms: (1,2)(3,4) --> { source gens --> [ (1,2,3,4), (2,4) ] } (1,3)(2,4) --> { source gens --> [ (1,4,3,2), (1,3) ] } These 2 automorphisms generate the group of automorphisms.
AutomorphismXMod( S [, A] )
This construction returns the crossed module whose range R is a permutation representation of a group A which is a group of automorphisms of the source S and which contains the inner automorphism group of S as a subgroup. When A is not specified the full automorphism group is used. The boundary morphism maps s ∈ S to the representation of the inner automorphism of S by s. The action is the isomorphism R → A.
In the following example, recall that the automorphism group of the
quaternion group is isomorphic to the symmetric group of degree 4 and
that the inner automorphism group is isomorphic to k4
. The group
A
is a subgroup of Aut(q8)
isomorphic to d8
.
gap> q8 := Group( (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8) );; gap> q8.name := "q8";; genq8 := q8.generators;; gap> iaq8 := InnerAutomorphismGroup( q8 );; gap> a := GroupHomomorphismByImages( q8, q8, genq8, > [(1,5,3,7)(2,6,4,8),(1,4,3,2)(5,6,7,8)]);; gap> genA := Concatenation( iaq8.generators, [a] ); [ InnerAutomorphism( q8, (1,2,3,4)(5,8,7,6) ), InnerAutomorphism( q8, (1,5,3,7)(2,6,4,8) ), GroupHomomorphismByImages( q8, q8, [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8) ], [ (1,5,3,7)(2,6,4,8), (1,4,3,2)(5,6,7,8) ] ) ] gap> id := IdentityMapping( q8 );; gap> A := Group( genA, id );; gap> AX := AutomorphismXMod( q8, A ); Crossed module [q8->PermSubAut(q8)] gap> RecFields( AX ); [ "isDomain", "isParent", "source", "range", "boundary", "action", "aut", "isXMod", "operations", "name", "isAutomorphismXMod" ]
InnerAutomorphismXMod( S )
This function is equivalent to AutomorphismXMod(S,A)
in the case
when A
is the inner automorphism group of S
.
gap> IX := InnerAutomorphismXMod( q8 ); Crossed module [q8->PermInn(q8)]
TrivialActionXMod( f )
For a crossed module to have trivial action, the axioms require the
source to be abelian and the image of the boundary to lie in the
centre of the range. A homomorphism f
can act as the boundary map
when these conditions are satisfied.
gap> imf := [ (1,3)(2,4), (1,3)(2,4) ];; gap> f := GroupHomomorphismByImages( k4, d8, genk4, imf );; gap> TX := TrivialActionXMod( f ); Crossed module [v4->d8] gap> XModPrint( TX ); Crossed module [v4->d8] :- : Source group has parent ( s4 ) and has generators: [ (1,2)(3,4), (1,3)(2,4) ] : Range group has parent ( s4 ) and has generators: [ (1,2,3,4), (1,3) ] : Boundary homomorphism maps source generators to: [ (1,3)(2,4), (1,3)(2,4) ] The automorphism group is trivial
IsRModule( Rmod )
IsRModuleRecord( Rmod )
An R-module consists of a permutation group R with an action
α : R → A where A is a group of automorphisms of an
abelian group M. When R
is not specified, the function
AutomorphismPair
is automatically called to construct it.
Rmod
with fields:
beg-tabularll
Rmod.module
, & the abelian group M,
Rmod.perm
, & the group R,
Rmod.auto
, & the action group A,
Rmod.isRModule
, & set true.
end-tabular
The IsRModule
distributor calls this function when the parameter is
a record but not a crossed module.
gap> k4gen := k4.generators;; gap> k4im := [ (1,3)(2,4), (1,4)(2,3) ];; gap> a := GroupHomomorphismByImages( k4, k4, k4gen, k4im );; gap> Ak4 := Group( a );; gap> R := rec( );; gap> R.module := k4;; gap> R.auto := Ak4;; gap> IsRModule( R ); true gap> RecFields( R ); [ "module", "auto", "perm", "isRModule" ] gap> R.perm; PermSubAut(v4)
RModuleXMod( Rmod )
The crossed module RX
obtained from an R-module has the abelian
group M as source, the zero map as boundary, the group R which
acts on M as range, the group A of automorphisms of M as
RX.aut
and α : R → A as RX.action
. An appropriate name
for RX
is chosen automatically. Continuing the previous example,
M is k4
and R is cyclic of order 3.
gap> RX := RModuleXMod( R ); Crossed module [v4->PermSubAut(v4)] gap> XModPrint( RX ); Crossed module [v4->PermSubAut(v4)] : Source group has parent s4 and has generators: [ (1,2)(3,4), (1,3)(2,4) ] : Range group = PermSubAut(v4) has generators: [ (1,2,3) ] : Boundary homomorphism maps source generators to: [ () ] : Action homomorphism maps range generators to automorphisms: (1,2,3) --> { source gens --> [ (1,3)(2,4), (1,4)(2,3) ] } This automorphism generates the group of automorphisms.
XModSelect( size [, gpnum, type, norm] )
Here the parameter size may take any value up to 47, gpnum refers
to the isomorphism class of groups of order size as ordered in the
GAP3 library. The norm parameter is only used in the case
"conj"
and specifies the position of the source group in the list of
normal subgroups of the range R. The list Cat1List
is used to
store the data for these groups. The allowable types are "conj"
for normal inclusions with conjugation, "aut"
for automorphism
groups and "rmod"
for Rmodules. If type is not specified the
default is "conj"
. If norm is not specified, then the
AutomorphismXMod
of R is returned.
In the following example the fourteenth class of groups of size 24 is
a special linear group sl(2,3)
and a double cover of a4
. The
third normal subgroup of sl(2,3)
is a quaternion group, and a
conjugation crossed module is returned.
gap> SX := XModSelect( 24, 14, "conj", 3 ); Crossed module [N3->sl(2,3)] gap> XModPrint( SX ); Crossed module [N3->sl(2,3)] :- : Source group has parent ( sl(2,3) ) and has generators: [ (1,2,3,4)(5,8,7,6), ( 1, 5, 3, 7)( 2, 6, 4, 8) ] : Range group = sl(2,3) has generators: [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8), (2,5,6)(4,7,8)(9,10,11) ] : Boundary homomorphism maps source generators to: [ (1,2,3,4)(5,8,7,6), ( 1, 5, 3, 7)( 2, 6, 4, 8) ] : Action homomorphism maps range generators to automorphisms: (1,2,3,4)(5,8,7,6) --> { source gens --> [ (1,2,3,4)(5,8,7,6), ( 1, 7, 3, 5)( 2, 8, 4, 6) ] } (1,5,3,7)(2,6,4,8) --> { source gens --> [ (1,4,3,2)(5,6,7,8), ( 1, 5, 3, 7)( 2, 6, 4, 8) ] } ( 2, 5, 6)( 4, 7, 8)( 9,10,11) --> { source gens --> [ ( 1, 5, 3, 7)( 2, 6, 4, 8), ( 1, 6, 3, 8)( 2, 7, 4, 5) ] } These 3 automorphisms generate the group of automorphisms.
80.15 Operations for crossed modules
Special operations defined for crossed modules are stored in the
record structure XModOps
. Every crossed module X
has
X.operations := XModOps;
.
gap> RecFields( XModOps ); [ "name", "operations", "Elements", "IsFinite", "Size", "=", "<", "in", "IsSubset", "Intersection", "Union", "IsParent", "Parent", "Difference", "Representative", "Random", "Print", "Kernel", "IsAspherical", "IsSimplyConnected", "IsConjugation", "IsTrivialAction", "IsCentralExtension", "DirectProduct", "IsAutomorphismXMod", "IsZeroBoundary", "IsRModule", "InclusionMorphism", "WhiteheadPermGroup", "Whitehead", "Norrie", "Lue", "Actor", "InnerMorphism", "Centre", "InnerActor", "AutomorphismPermGroup", "IdentityMorphism", "InnerAutomorphism", ]
Crossed modules X,Y
are considered equal if they have the same
source, boundary, range, and action. The remaining functions are
discussed below and following section About actors.
80.16 Print for crossed modules
XModOps.Print( X )
This function is the special print command for crossed modules,
producing a single line of output, and is called automatically when a
crossed module is displayed. For more detail use XModPrint( X )
.
gap> CX; Crossed module [v4->a4]
80.17 Size for crossed modules
XModOps.Size( X )
This function returns a 2-element list containing the sizes of the
source and the range of X
.
gap> Size( CX ); [ 4, 12 ]
80.18 Elements for crossed modules
XModOps.Elements( X )
This function returns a 2-element list of lists of elements of the
source and range of X
.
gap> Elements( CX ); [ [ (), (1,2)(3,4), (1,3)(2,4), (1,4)(2,3) ], [ (), (2,3,4), (2,4,3), (1,2)(3,4), (1,2,3), (1,2,4), (1,3,2), (1,3,4), (1,3)(2,4), (1,4,2), (1,4,3), (1,4)(2,3) ] ]
80.19 IsConjugation for crossed modules
XModOps.IsConjugation( X )
This Boolean function checks that the source is a normal subgroup of the range and that the boundary is an inclusion.
gap> IsConjugation( CX ); true
XModOps.IsAspherical( X )
This Boolean function checks that the boundary map is monomorphic.
gap> IsAspherical( CX ); true
XModOps.IsSimplyConnected( X )
This Boolean function checks that the boundary map is surjective. The corresponding groupoid then has a single connected component.
gap> IsSimplyConnected( EX ); true
XModOps.IsCentralExtension( X )
This Boolean function checks that the boundary is surjective with kernel central in the source.
gap> IsCentralExtension( EX ); true
XModOps.IsAutomorphismXMod( X )
This Boolean function checks that the range group is a subgroup of the automorphism group of the source group containing the group of inner automorphisms, and that the boundary and action homomorphisms are of the correct form.
gap> IsAutomorphismXMod( AX ); true
XModOps.IsTrivialAction( X )
This Boolean function checks that the action is the zero map.
gap> IsTrivialAction( TX ); true
XModOps.IsZeroBoundary( X )
This Boolean function checks that the boundary is the zero map.
gap> IsZeroBoundary( EX ); false
80.26 IsRModule for crossed modules
XModOps.IsRModule( X )
This Boolean function checks that the boundary is the zero map and that the source is abelain.
gap> IsRModule( RX ); true
WhatTypeXMod( X )
This function checks whether the crossed module X
is one or more of
the six standard type listed above.
gap> WhatTypeXMod( EX ); [ " extn, " ]
80.28 DirectProduct for crossed modules
XModOps.DirectProduct( X,Y )
The direct product of crossed modules X,Y
has as source and range
the direct products of the sources and ranges of X
and Y
. The
boundary map is the product of the two boundaries. The range of X
acts trivially on the source of Y
and conversely. Because the
standard DirectProduct
function requires the two parameters to be
groups, the XModOps.
prefix must be used (at least for GAP33.4.3).
gap> DX := XModOps.DirectProduct( CX, CX ); Crossed module [v4xv4->a4xa4] gap> XModPrint( DX ); Crossed module [v4xv4->a4xa4] :- : Source group v4xv4 has generators: [ (1,2)(3,4), (1,3)(2,4), (5,6)(7,8), (5,7)(6,8) ] : Range group = a4xa4 has generators: [ (1,2,3), (2,3,4), (5,6,7), (6,7,8) ] : Boundary homomorphism maps source generators to: [ (1,2)(3,4), (1,3)(2,4), (5,6)(7,8), (5,7)(6,8) ] : Action homomorphism maps range generators to automorphisms: (1,2,3) --> { source gens --> [ (1,4)(2,3), (1,2)(3,4), (5,6)(7,8), (5,7)(6,8) ] } (2,3,4) --> { source gens --> [ (1,3)(2,4), (1,4)(2,3), (5,6)(7,8), (5,7)(6,8) ] } (5,6,7) --> { source gens --> [ (1,2)(3,4), (1,3)(2,4), (5,8)(6,7), (5,6)(7,8) ] } (6,7,8) --> { source gens --> [ (1,2)(3,4), (1,3)(2,4), (5,7)(6,8), (5,8)(6,7) ] } These 4 automorphisms generate the group of automorphisms.
XModMorphism( X, Y, homs )
A morphism of crossed modules is a pair of homomorphisms [ sourceHom,
rangeHom ]
, where sourceHom
, rangeHom
are respectively
homomorphisms between the sources and ranges of X and Y, which
commute with the two boundary maps and which are morphisms for the two
actions.
In this implementation a morphism of crossed modules mor
is a record
beg-tabularll
mor.source
, & the source crossed module X
,
mor.range
, & the range crossed module Y
,
mor.sourceHom
, & a homomorphism from X.source
to Y.source
,
mor.rangeHom
, & a homomorphism from X.range
to Y.range
,
mor.isXModMorphism
, & a Boolean flag, normally true
,
mor.operations
, & a special set of operations XModMorphismOps
(see \refOperations for morphisms of crossed modules),
mor.name
, & a concatenation of the names of X
and Y
.
end-tabular
The function XModMorphism
requires as parameters two crossed modules
and a two-element list containing the source and range homomorphisms.
It sets up the required fields for mor
, but does not check the
axioms. The IsXModMorphism
function should be used to perform these
checks. Note that the XModMorphismPrint
function is needed to print
out the morphism in detail.
gap> smor := GroupHomomorphismByImages( q8, k4, genq8, genk4 ); GroupHomomorphismByImages( q8, v4, [(1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8)], [(1,2)(3,4), (1,3)(2,4)] ) gap> IsHomomorphism(smor); true gap> sl23 := SX.range;; gap> gensl23 := sl23.generators; [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8), (2,5,6)(4,7,8)(9,10,11) ] gap> images := [ (1,2)(3,4), (1,3)(2,4), (2,3,4) ];; gap> rmor := GroupHomomorphismByImages( sl23, a4, gensl23, images );; gap> IsHomomorphism(rmor); true gap> mor := XModMorphism( SX, CX, [ smor, rmor ] ); Morphism of crossed modules <[N3->sl(2,3)] >-> [v4->a4]>
IsXModMorphism( mor )
This Boolean function checks that mor includes homomorphisms between
the corresponding source and range crossed modules, and that these
homomorphisms commute with the two actions. In the example we
increase the value of XModPrintLevel
to show the effect of such an
increase in a simple case.
gap> XModPrintLevel := 3;; gap> IsXModMorphism( mor ); Checking that the diagram commutes :- Y.boundary(morsrc(x)) = morrng(X.boundary(x)) Checking: morsrc(x2^x1) = morsrc(x2)^(morrng(x1)) true gap> XModPrintLevel := 1;;
XModMorphismPrint( mor )
This function is used to display the main fields of a crossed module.
gap> XModMorphismPrint( mor ); Morphism of crossed modules :- : Source = Crossed module [N3->sl(2,3)] with generating sets [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8) ] [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8), (2,5,6)(4,7,8)(9,10,11) ] : Range = Crossed module [v4->a4] with generating sets [ (1,2)(3,4), (1,3)(2,4) ]] [ (1,2,3), (2,3,4) ] : Source Homomorphism maps source generators to: [ (1,2)(3,4), (1,3)(2,4) ] : Range Homomorphism maps range generators to: [ (1,2)(3,4), (1,3)(2,4), (2,3,4) ] : isXModMorphism? true
XModMorphismName( mor )
Whenever the names of the source or range crossed module are changed,
this function may be used to produce the new standard form
<mor.source.name >-> mor.range.name>
for the name of mor. This
function is automatically called by XModMorphismPrint
.
gap> k4.name := "k4";; XModName( CX );; gap> XModMorphismName( mor ); <[N3->sl(2,3)] >-> [k4->a4]>
80.33 Operations for morphisms of crossed modules
Special operations defined for morphisms of crossed modules are stored
in the record structure XModMorphismOps
which is based on
MappingOps
. Every crossed module morphism mor
has field
mor.operations
set equal to XModMorphismOps;
.
gap> IsMonomorphism( mor ); false gap> IsEpimorphism( mor ); true gap> IsIsomorphism( mor ); false gap> IsEndomorphism( mor ); false gap> IsAutomorphism( mor ); false
IdentitySubXMod( X )
Every crossed module X
has an identity sub-crossed module whose
source and range are the identity subgroups of the source and range.
gap> IdentitySubXMod( CX ); Crossed module [Id[k4->a4]]
SubXMod( X, subS, subR )
A sub-crossed module of a crossed module X
has as source a subgroup
subS
of X.source
and as range a subgroup subR
of X.range
. The
boundary map and the action are the appropriate restrictions. In the
following example we construct a sub-crossed module of SX
with range
q8
and source a cyclic group of order 4.
gap> q8 := SX.source;; genq8 := q8.generators;; gap> q8.name := "q8";; XModName( SX );; gap> c4 := Subgroup( q8, [ genq8[1] ] ); Subgroup( sl(2,3), [ (1,2,3,4)(5,8,7,6) ] ) gap> c4.name := "c4";; gap> subSX := SubXMod( SX, c4, q8 ); Crossed module [c4->q8] gap> XModPrint( subSX ); Crossed module [c4->q8] :- : Source group has parent ( sl(2,3) ) and has generators: [ (1,2,3,4)(5,8,7,6) ] : Range group has parent ( sl(2,3) ) and has generators: [ (1,2,3,4)(5,8,7,6), ( 1, 5, 3, 7)( 2, 6, 4, 8) ] : Boundary homomorphism maps source generators to: [ ( 1, 2, 3, 4)( 5, 8, 7, 6) ] : Action homomorphism maps range generators to automorphisms: (1,2,3,4)(5,8,7,6) --> {source gens --> [ (1,2,3,4)(5,8,7,6) ]} (1,5,3,7)(2,6,4,8) --> {source gens --> [ (1,4,3,2)(5,6,7,8) ]} These 2 automorphisms generate the group of automorphisms.
IsSubXMod( X,S )
This boolean function checks that S
is a sub-crossed module of X
.
gap> IsSubXMod( SX, subSX ); true
80.37 InclusionMorphism for crossed modules
InclusionMorphism( S,X )
This function constructs the inclusion of a sub-crossed module S
of
X
. When S = X
the identity morphism is returned.
gap> inc := InclusionMorphism( subSX, SX ); Morphism of crossed modules <[c4->q8] >-> [q8->sl(2,3)]> gap> IsXModMorphism( inc ); true gap> XModMorphismPrint( inc ); Morphism of crossed modules :- : Source = Crossed module [c4->q8] with generating sets: [ (1,2,3,4)(5,8,7,6) ] [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8) ] : Range = Crossed module [q8->sl(2,3)] with generating sets: [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8) ] [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8), (2,5,6)(4,7,80(9,10,11) ] : Source Homomorphism maps source generators to: [ (1,2,3,4)(5,8,7,6) ] : Range Homomorphism maps range generators to: [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8) ] : isXModMorphism? true
IsNormalSubXMod( X,Y )
A sub-crossed module Y=(N->M)
is normal in X=(S->R)
when
• N,M are normal subgroups of S,R respectively,
• nr ∈ N for all n ∈ N, r ∈ R,
• s-1 sm ∈ N for all m ∈ M, s ∈ S.
These axioms are sufficient to ensure that M \semidirect N is a normal subgroup of R \semidirect S. They also ensure that the inclusion morphism of a normal sub-crossed module forms a conjugation crossed square, analogous to the construction of a conjugation crossed module.
gap> IsNormalSubXMod( SX, subSX ); false
NormalSubXMods( X )
This function takes pairs of normal subgroups from the source and
range of X
and constructs a normal sub-crossed module whenever the
axioms are satisfied. Appropriate names are chosen where possible.
gap> NSX := NormalSubXMods( SX ); [ Crossed module [Id[q8->sl(2,3)]], Crossed module [I->?], Crossed module [Sub[q8->sl(2,3)]], Crossed module [?->q8], Crossed module [?->q8], Crossed module [q8->sl(2,3)] ]
FactorXMod( X, subX )
The quotient crossed module of a crossed module by a normal sub-crossed module has quotient groups as source and range, with the obvious action.
gap> Size( NSX[3] ); [ 2, 2 ] gap> FX := FactorXMod( SX, NSX[3] ); Crossed module [?->?] gap> Size( FX ); [ 4, 12 ]
80.41 Kernel of a crossed module morphism
Kernel( mor )
The kernel of a morphism mor
: X → Y of crossed modules is the
normal sub-crossed module of X whose source is the kernel of
mor.sourceHom
and whose range is the kernel of mor.rangeHom
. An
appropriate name for the kernel is chosen automatically. A field
.kernel
is added to mor
.
gap> XModMorphismName( mor );; gap> KX := Kernel( mor ); Crossed module Ker<[q8->sl(2,3)] >-> [k4->a4]> gap> XModPrint( KX ); Crossed module Ker<[q8->SL(2,3)] >-> [k4->a4]> :- : Source group has parent ( sl(2,3) ) and has generators: [ (1,3)(2,4)(5,7)(6,8) ] : Range group has parent ( sl(2,3) ) and has generators: [ ( 1, 3)( 2, 4)( 5, 7)( 6, 8) ] : Boundary homomorphism maps source generators to: [ (1,3)(2,4)(5,7)(6,8) ] : The automorphism group is trivial. gap> IsNormalSubXMod( SX, KX ); true
80.42 Image for a crossed module morphism
ImageXModMorphism( mor, S )
The image of a sub-crossed module S
of X
under a morphism mor
: X → Y of crossed modules is the sub-crossed module of Y whose
source is the image of S.source
under mor.sourceHom
and whose
range is the image of S.range
under mor.rangeHom
. An appropriate
name for the image is chosen automatically. A field .image
is added
to mor
. Note that thjis function should be named
XModMorphismOps.Image
, but the command J := Image( mor, S );
does
not work with version 3 of GAP3.
gap> subSX; Crossed module [c4->q8] gap> JX := ImageXModMorphism( mor, subSX ); Crossed module [Im([c4->q8]) by <[q8->sl(2,3)] >-> [k4->a4]>] gap> RecFields( mor ); [ "sourceHom", "rangeHom", "source", "range", "name", "isXModMorphism", "domain", "kernel", "image", "isMonomorphism", "isEpimorphism", "isIsomorphism", "isEndomorphism", "isAutomorphism", "operations" ] gap> XModPrint( JX ); Crossed module [Im([c4->q8]) by <[q8->sl(2,3)] >-> [k4->a4]>] :- : Source group has parent ( s4 ) and has generators: [ (1,2)(3,4) ] : Range group has parent ( s4 ) and has generators: [ (1,2)(3,4), (1,3)(2,4) ] : Boundary homomorphism maps source generators to: [ (1,2)(3,4) ] : The automorphism group is trivial.
80.43 InnerAutomorphism of a crossed module
InnerAutomorphism( X, r )
Each element r
of X.range
determines an automorphism of X
in
which the automorphism of X.source
is given by the image of
X.action
on r
and the automorphism of X.range
is conjugation by
r
. The command InnerAutomorphism( X, r );
does not work with
version 3 of GAP3.
gap> g := Elements( q8 )[8]; (1,8,3,6)(2,5,4,7) gap> psi := XModOps.InnerAutomorphism( subSX, g ); Morphism of crossed modules <[c4->q8] >-> [c4->q8]> gap> XModMorphismPrint( psi ); Morphism of crossed modules :- : Source = Crossed module [c4->q8] with generating sets: [ (1,2,3,4)(5,8,7,6) ] [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8) ] : Range = Crossed module [c4->q8] with generating sets: [ (1,2,3,4)(5,8,7,6) ] [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8) ] : Source Homomorphism maps source generators to: [ ( 1,4,3,2)(5,6,7,8) ] : Range Homomorphism maps range generators to: [ ( 1,4,3,2)(5,6,7,8), (1,7,3,5)(2,8,4,6) ] isXModMorphism? true
80.44 Order of a crossed module morphism
XModMorphismOps.Order( mor )
This function calculates the order of an automorphism of a crossed module.
gap> XModMorphismOps.Order( psi ); 2
80.45 CompositeMorphism for crossed modules
CompositeMorphism( mor1, mor2 )
Morphisms μ1 : X → Y and μ2 : Y → Z have a composite μ = μ2 o μ1 : X → Z whose source and range homomorphisms are the composites of those of μ1 and μ2.
In the following example we compose psi
with the inc
obtained previously.
gap> xcomp := XModMorphismOps.CompositeMorphism( psi, inc ); Morphism of crossed modules <[c4->q8] >-> [q8->sl(2,3)]> gap> XModMorphismPrint( xcomp ); Morphism of crossed modules :- : Source = Crossed module [c4->q8] with generating sets: [ (1,2,3,4)(5,8,7,6) ] [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8) ] : Range = Crossed module [q8->sl(2,3)] with generating sets: [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8) ] [ (1,2,3,4)(5,8,7,6), (1,5,3,7)(2,6,4,8), (2,5,6)(4,7,8)(9,10,11) ] : Source Homomorphism maps source generators to: [ (1,4,3,2)(5,6,7,8) ] : Range Homomorphism maps range generators to: [ (1,4,3,2)(5,6,7,8), (1,7,3,5)(2,8,4,6) ] : isXModMorphism? true
SourceXModXPModMorphism( mor )
When crossed modules X,Y have a common range P and mor
is a
morphism from X to Y whose range homomorphism is the identity
homomorphism, then mor.sourceHom : X.source -> Y.source)
is a
crossed module.
gap> c2 := Subgroup( q8, [ genq8[1]^2 ] ); Subgroup( sl(2,3), [ (1,3)(2,4)(5,7)(6,8) ] ) gap> c2.name := "c2";; gap> sub2 := SubXMod( subSX, c2, q8 ); Crossed module [c2->q8] gap>inc2 := InclusionMorphism( sub2, subSX ); Morphism of crossed modules <[c2->q8] >-> [c4->q8]> gap> PX := SourceXModXPModMorphism( inc2 ); Crossed module [c2->c4] gap> IsConjugation( PX ); true
\newpage
In xmodL1 Loday reformulated the notion of a crossed module as a cat1-group, namely a group G with a pair of homomorphisms t,h : G → G having a common image R and satisfying certain axioms. We find it convenient to define a cat1-group C = (e;t,h : G → R ) as having source group G, range group R, and three homomorphisms: two surjections t,h : G → R and an embedding e : R → G satisfying:
The maps t,h are often referred to as the source and target, but we choose to call them the tail and head of C, because source is the GAP3 term for the domain of a function.
A morphism C1 → C2 of cat1-groups is a pair (γ, ρ) where γ : G1 → G2 and ρ : R1 → R2 are homomorphisms satisfying
h2 γ = ρ h1, t2 γ = ρ t1, e2 ρ = γ e1, |
In this implementation a cat1-group C is a record with the following fields:
beg-tabularll
C.source
, & the source G
,
C.range
, & the range R
,
C.tail
, & the tail homomorphism t
,
C.head
, & the head homomorphism h
,
C.embedRange
, & the embedding of R
in G
,
C.kernel
, & a permutation group isomorphic to the kernel of t
,
C.embedKernel
,& the inclusion of the kernel in G
,
C.boundary
, & the restriction of h
to the kernel,
C.isDomain
, & set true
,
C.operations
, & a special set of operations Cat1Ops
(see \refOperations for cat1-groups,
C.name
, & a concatenation of the names of the source and range.
C.isCat1
& a boolean flag, normally true
.
end-tabular
The following listing shows a simple example:
gap> s3c4gen := s3c4.generators; [ (1,2), (2,3), (4,5,6,7) ] gap> t1 := GroupHomomorphismByImages( s3c4, s3, s3c4gen, > [ (1,2), (2,3), () ] );; gap> C1 := Cat1( s3c4, t1, t1 ); cat1-group [s3c4 ==> s3] gap> Cat1Print( C1 ); cat1-group [s3c4 ==> s3] :- : source group has generators: [ (1,2), (2,3), (4,5,6,7) ] : range group has generators: [ (1,2), (2,3) ] : tail homomorphism maps source generators to: [ ( 1, 2), ( 2, 3), () ] : head homomorphism maps source generators to: [ ( 1, 2), ( 2, 3), () ] : range embedding maps range generators to: [ (1,2), (2,3) ] : kernel has generators: [ (4,5,6,7) ] : boundary homomorphism maps generators of kernel to: [ () ] : kernel embedding maps generators of kernel to: [ (4,5,6,7) ]
The category of crossed modules is equivalent to the category of cat1-groups, and the functors between these two categories may be described as follows.
Starting with the crossed module X = (∂ : S → R) the group G is defined as the semidirect product G = R \semidirect S using the action from X. The structural morphisms are given by
t(r,s) = r, h(r,s) = r (∂ s), er = (r,1). |
On the other hand, starting with a cat1-group C = (e;t,h : G → R) we define S = ker t, the range R remains unchanged and ∂ = h|S . The action of R on S is conjugation in S via the embedding of R in G.
gap> X1; Crossed module [c5->PermAut(c5)] gap> CX1 := Cat1XMod(X1); cat1-group [Perm(PermAut(c5) |X c5) ==> PermAut(c5)] gap> CX1.source.generators; [ (2,3,5,4), (1,2,3,4,5) ] gap> gap> XC1 := XModCat1( C1 ); Crossed module [ker([s3c4 ==> s3])->s3] gap> WhatTypeXMod( XC1 ); [ " triv, ", " zero, ", " RMod, " ]
Cat1( G, t, h )
This function constructs a cat1-group C from a group G and
a pair of endomorphisms, the tail and head of C. The example
uses the holomorph of c5
, a group of size 20, which was the source
group in XC1
in \refAbout cat1-groups. Note that when t = h the
boundary is the zero map.
gap> h20 := Group( (1,2,3,4,5), (2,3,5,4) );; gap> h20.name := "h20";; gap> genh20 := h20.generators;; gap> imh20 := [ (), (2,3,5,4) ];; gap> h := GroupHomomorphismByImages( h20, h20, genh20, imh20 );; gap> t := h;; gap> C := Cat1( h20, t, h ); cat1-group [h20 ==> R]
IsCat1( C )
This function checks that the axioms of a cat1-group are satisfied and that the main fields of a cat1-group record exist.
gap> IsCat1(C); true
Cat1Print( C )
This function is used to display the main fields of a cat1-group.
gap> Cat1Print(C); cat1-group [h20 ==> R] :- : source group has generators: [ (1,2,3,4,5), (2,3,5,4) ] : range group has generators: [ ( 2, 3, 5, 4) ] : tail homomorphism maps source generators to: [ (), ( 2, 3, 5, 4) ] : head homomorphism maps source generators to: [ (), ( 2, 3, 5, 4) ] : range embedding maps range generators to: [ ( 2, 3, 5, 4) ] : kernel has generators: [ (1,2,3,4,5) ] : boundary homomorphism maps generators of kernel to: [ () ] : kernel embedding maps generators of kernel to: [ ( 1, 2, 3, 4, 5) ]
Cat1Name( C )
Whenever the names of the source or the range of C
are changed, this
function may be used to produce the new standard form
[<C.source.name> ==> <C.range.name>]
for the name of C
. This
function is called automatically by Cat1Print
. Note the use of =
,
rather than -
in the arrow shaft, to indicate the pair of maps.
gap> C.range.name := "c4";; Cat1Name( C ); "[h20 ==> c4]"
\newpage
ConjugationCat1( R, S )
When S is a normal subgroup of a group R form the semi-direct
product G = R \semidirect S to R and take this as the source, with
R as the range. The tail and head homomorphisms are defined by
t(r,s) = r(∂ s), h(r,s) = r. In the example h20
is the
range, rather than the source.
gap> c5 := Subgroup( h20, [(1,2,3,4,5)] );; gap> c5.name := "c5";; gap> CC := ConjugationCat1( h20, c5 ); cat1-group [Perm(h20 |X c5) ==> h20] gap> Cat1Print( CC ); cat1-group [Perm(h20 |X c5) ==> h20] :- : source group has generators: [ ( 6, 7, 8, 9,10), ( 2, 3, 5, 4)( 7, 8,10, 9), (1,2,3,4,5) ] : range group has generators: [ (1,2,3,4,5), (2,3,5,4) ] : tail homomorphism maps source generators to: [ ( 1, 2, 3, 4, 5), ( 2, 3, 5, 4), () ] : head homomorphism maps source generators to: [ ( 1, 2, 3, 4, 5), ( 2, 3, 5, 4), ( 1, 2, 3, 4, 5) ] : range embedding maps range generators to: [ ( 6, 7, 8, 9,10), ( 2, 3, 5, 4)( 7, 8,10, 9) ] : kernel has generators: [ (1,2,3,4,5) ] : boundary homomorphism maps generators of kernel to: [ ( 1, 2, 3, 4, 5) ] : kernel embedding maps generators of kernel to: [ ( 1, 2, 3, 4, 5) ] : associated crossed module is Crossed module [c5->h20] gap> ct := CC.tail;; gap> ch := CC.head;; gap> CG := CC.source;; gap> genCG := CG.generators;; gap> x := genCG[2] * genCG[3]; ( 1, 2, 4, 3 )( 7, 8,10, 9 ) gap> tx := Image( ct, x ); ( 2, 3, 5, 4) gap> hx := Image( ch, x ); ( 1, 2, 4, 3) gap> RecFields( CC ); [ "source", "range", "tail", "head", "embedRange", "kernel", "boundary", "embedKernel", "isDomain", "operations", "isCat1", "name", "xmod" ]
80.53 Operations for cat1-groups
Special operations defined for crossed modules are stored in the
record structure Cat1Ops
based on DomainOps
. Every cat1-group C
has C.operations := Cat1Ops;
.
gap> RecFields( Cat1Ops ); [ "name", "operations", "Elements", "IsFinite", "Size", "=", "<", "in", "IsSubset", "Intersection", "Union", "IsParent", "Parent", "Difference", "Representative", "Random", "Print", "Actor", "InnerActor", "InclusionMorphism", "WhiteheadPermGroup" ]
Cat1-groups are considered equal if they have the same source, range, tail, head and embedding. The remaining functions are described below.
Cat1Ops.Size( C )
This function returns a two-element list containing the sizes of the
source and range of C
.
gap> Size( C ); [ 20, 4 ]
80.55 Elements for cat1-groups
Cat1Ops.Elements( C )
This function returns the two-element list of lists of elements of the source and range of C.
gap> Elements( C ); [ [ (), (2,3,5,4), (2,4,5,3), (2,5)(3,4), (1,2)(3,5), (1,2,3,4,5), (1,2,4,3), (1,2,5,4), (1,3,4,2), (1,3)(4,5), (1,3,5,2,4), (1,3,2,5), (1,4,5,2), (1,4,3,5), (1,4)(2,3), (1,4,2,5,3), (1,5,4,3,2), (1,5,3,4), (1,5,2,3), (1,5)(2,4) ], [ (), (2,3,5,4), (2,4,5,3), (2,5)(3,4) ] ]
XModCat1( C )
This function acts as the functor from the category of cat1-groups to the category of crossed modules.
gap> XC := XModCat1( C ); Crossed module [ker([h20 ==> c4])->c4] gap> XModPrint( XC ); Crossed module [ker([h20 ==> c4])->c4] :- : Source group has parent ( h20 ) and has generators: [ (1,2,3,4,5) ] : Range group has parent ( h20 ) and has generators: [ ( 2, 3, 5, 4) ] : Boundary homomorphism maps source generators to: [ () ] : Action homomorphism maps range generators to automorphisms: (2,3,5,4) --> { source gens --> [ (1,3,5,2,4) ] } This automorphism generates the group of automorphisms. : Associated cat1-group = cat1-group [h20 ==> c4]
Cat1XMod( X )
This function acts as the functor from the category of crossed modules
to the category of cat1-groups. A permutation representation of the
semidirect product R \semidirect S is constructed for G. See
section \refSemidirectCat1XMod for a version where G is a
semidirect product group. The example uses the crossed module CX
constructed in section
\refConjugationXMod.
gap> CX; Crossed module [k4->a4] gap> CCX := Cat1XMod( CX ); cat1-group [a4.k4 ==> a4] gap> Cat1Print( CCX ); cat1-group [a4.k4 ==> a4] :- : source group has generators: [ (2,4,3)(5,6,7), (2,3,4)(6,7,8), (1,2)(3,4), (1,3)(2,4) ] : range group has generators: [ (1,2,3), (2,3,4) ] : tail homomorphism maps source generators to: [ ( 1, 2, 3), ( 2, 3, 4), (), () ] : head homomorphism maps source generators to: [ ( 1, 2, 3), ( 2, 3, 4), ( 1, 2)( 3, 4), ( 1, 3)( 2, 4) ] : range embedding maps range generators to: [ ( 2, 4, 3)( 5, 6, 7), ( 2, 3, 4)( 6, 7, 8) ] : kernel has generators: [ (1,2)(3,4), (1,3)(2,4) ] : boundary homomorphism maps generators of kernel to: [ (1,2)(3,4), (1,3)(2,4) ] : kernel embedding maps generators of kernel to: [ (1,2)(3,4), (1,3)(2,4) ] : associated crossed module is Crossed module [k4->a4]
SemidirectCat1XMod( X )
This function is similar to the previous one, but a permutation representation for R \semidirect S is not constructed.
gap> Unbind( CX.cat1 ); gap> SCX := SemidirectCat1XMod( CX ); cat1-group [a4 |X k4 ==> a4] gap> Cat1Print( SCX ); cat1-group [a4 |X k4 ==> a4] :- : source group has generators: [ SemidirectProductElement( (1,2,3), GroupHomomorphismByImages( k4, k4, [(1,3)(2,4), (1,4)(2,3)], [(1,2)(3,4), (1,3)(2,4)] ), () ), SemidirectProductElement( (2,3,4), GroupHomomorphismByImages( k4, k4, [(1,4)(2,3), (1,2)(3,4)], [(1,2)(3,4), (1,3)(2,4)] ), () ), SemidirectProductElement( (), IdentityMapping( k4 ), (1,2)(3,4) ), SemidirectProductElement( (), IdentityMapping( k4 ), (1,3)(2,4) ) ] : range group has generators: [ (1,2,3), (2,3,4) ] : tail homomorphism maps source generators to: [ (1,2,3), (2,3,4), (), () ] : head homomorphism maps source generators to: [ (1,2,3), (2,3,4), (1,2)(3,4), (1,3)(2,4) ] : range embedding maps range generators to: [ SemidirectProductElement( (1,2,3), GroupHomomorphismByImages( k4, k4, [(1,3)(2,4), (1,4)(2,3)], [(1,2)(3,4), (1,3)(2,4)] ), () ), SemidirectProductElement( (2,3,4), GroupHomomorphismByImages( k4, k4, [(1,4)(2,3), (1,2)(3,4)], [(1,2)(3,4), (1,3)(2,4)] ), () ) ] : kernel has generators: [ (1,2)(3,4), (1,3)(2,4) ] : boundary homomorphism maps generators of kernel to: [ (1,2)(3,4), (1,3)(2,4) ] : kernel embedding maps generators of kernel to: [ SemidirectProductElement( (), IdentityMapping( k4 ), (1,2)(3,4) ), SemidirectProductElement( (), IdentityMapping( k4 ), (1,3)(2,4) ) ] : associated crossed module is Crossed module [k4->a4]
Cat1List
is a list containing data on all cat1-structures on groups
of size up to 47. The list is used by Cat1Select
to construct these
small examples of cat1-groups.
gap> Length( Cat1List ); 198 gap> Cat1List[8]; [ 6, 2, [ (1,2), (2,3) ], "s3", [ [ [ (2,3), (2,3) ], "c3", "c2", [ (2,3), (2,3) ], [ (2,3), (2,3) ] ] ] ]
Cat1Select( size, [gpnum, num] )
All cat-structures on groups of order up to 47 are stored in a list
Cat1List
and may be obtained from the list using this function.
Global variables Cat1ListMaxSize := 47
and
NumbersOfIsomorphismClasses
are also stored. The example
illustrated is the first case in which t ≠ h and the associated
conjugation crossed module is given by the normal subgroup c3
of
s3
.
gap> Cat1ListMaxSize; 47 gap> NumbersOfIsomorphismClasses[18]; 5 gap> Cat1Select( 18 ); Usage: Cat1Select( size, gpnum, num ) [ "c6c3", "c18", "d18", "s3c3", "c3^2|Xc2" ] gap> Cat1Select( 18, 5 ); There are 4 cat1-structures for the group c3^2|Xc2. [ [range generators], [tail.genimages], [head.genimages] ] :- [ [ (1,2,3), (4,5,6), (2,3)(5,6) ], tail = head = identity mapping ] [ [ (2,3)(5,6) ], "c3^2", "c2", [ (), (), (2,3)(5,6) ], [ (), (), (2,3)(5,6) ] ] [ [ (4,5,6), (2,3)(5,6) ], "c3", "s3", [ (), (4,5,6), (2,3)(5,6) ], [ (), (4,5,6), (2,3)(5,6) ] ] [ [ (4,5,6), (2,3)(5,6) ], "c3", "s3", [ (4,5,6),(4,5,6),(2,3)(5,6) ], [ (), (4,5,6), (2,3)(5,6) ] ] Usage: Cat1Select( size, gpnum, num ) Group has generators [ (1,2,3), (4,5,6), (2,3)(5,6) ] gap> SC := Cat1Select( 18, 5, 4 ); cat1-group [c3^2|Xc2 ==> s3] gap> Cat1Print( SC ); cat1-group [c3^2|Xc2 ==> s3] :- : source group has generators: [ (1,2,3), (4,5,6), (2,3)(5,6) ] : range group has generators: [ (4,5,6), (2,3)(5,6) ] : tail homomorphism maps source generators to: [ ( 4, 5, 6), ( 4, 5, 6), ( 2, 3)( 5, 6) ] : head homomorphism maps source generators to: [ (), ( 4, 5, 6), ( 2, 3)( 5, 6) ] : range embedding maps range generators to: [ ( 4, 5, 6), ( 2, 3)( 5, 6) ] : kernel has generators: [ ( 1, 2, 3)( 4, 6, 5) ] : boundary homomorphism maps generators of kernel to: [ ( 4, 6, 5) ] : kernel embedding maps generators of kernel to: [ ( 1, 2, 3)( 4, 6, 5) ] gap> XSC := XModCat1( SC ); Crossed module [c3->s3]
For each group G
the first cat1-structure is the identity
cat1-structure (id;id,id : G -> G)
with trivial kernel. The
corresponding crossed module has as boundary the inclusion map of the
trivial subgroup.
gap> AC := Cat1Select( 12, 5, 1 ); cat1-group [a4 ==> a4]
Cat1Morphism( C, D, L )
A morphism of cat1-groups is a pair of homomorphisms [ sourceHom,
rangeHom ]
, where sourceHom
, rangeHom
are respectively
homomorphisms between the sources and ranges of C and D, which
commute with the two tail homomorphisms with the two head
homomorphisms and with the two embeddings.
In this implementation a morphism of cat1-groups mu
is a record with
fields:
beg-tabularll
mu.source
, & the source cat1-group C
,
mu.range
, & the range cat1-group D
,
mu.sourceHom
, & a homomorphism from C.source
to D.source
,
mu.rangeHom
, & a homomorphism from C.range
to D.range
,
mu.isCat1Morphism
, & a Boolean flag, normally true
,
mu.operations
, & a special set of operations Cat1MorphismOps
,
mu.name
, & a concatenation of the names of C
and D
.
end-tabular
The function Cat1Morphism
requires as parameters two cat1-groups and
a two-element list containing the source and range homomorphisms. It
sets up the required fields for mu
, but does not check the axioms.
The IsCat1Morphism
function should be used to perform these checks.
Note that the Cat1MorphismPrint
function is needed to print out the
morphism in detail.
gap> GCCX := CCX.source; Perm(a4 |X k4) gap> GAC := AC.source; a4 gap> genGAC := GAC.generators; [ (1,2,3), (2,3,4) ] gap> im := Sublist( GCCX.generators, [1..2] ); [ (2,4,3)(5,6,7), (2,3,4)(6,7,8) ] gap> musrc := GroupHomomorphismByImages( GAC, GCCX, genGAC, im );; gap> murng := InclusionMorphism( a4, a4 );; gap> mu := Cat1Morphism( AC, CCX, [ musrc, murng ] ); Morphism of cat1-groups <[a4 ==> a4]-->[Perm(a4 |X k4) ==> a4]>
IsCat1Morphism( mu )
This Boolean function checks that μ includes homomorphisms between the corresponding source and range groups, and that these homomorphisms commute with the pairs of tail and head homomorphisms.
gap> IsCat1Morphism( mu ); true
Cat1MorphismName( mu )
This function concatenates the names of the source and range of a morphism of cat1-groups.
gap> CCX.source.name := "a4.k4";; Cat1Name( CCX ); "[a4.k4 ==> a4]" gap> Cat1MorphismName( mu ); "<[a4 ==> a4]-->[a4.k4 ==> a4]>"
Cat1MorphismPrint( mu )
This printing function for cat1-groups is one of the special functions
in Cat1MorphismOps
.
gap> Cat1MorphismPrint( mu ); Morphism of cat1-groups := : Source = cat1-group [a4 ==> a4] : Range = cat1-group [a4.k4 ==> a4] : Source homomorphism maps source generators to: [ (2,4,3)(5,6,7), (2,3,4)(6,7,8) ] : Range homomorphism maps range generators to: [ (1,2,3), (2,3,4) ]
80.65 Operations for morphisms of cat1-groups
Special operations defined for morphisms of cat1-groups are stored in
the record structure Cat1MorphismOps
which is based on MappingOps
.
Every morphism of cat1-groups mor
has field mor.operations
set
equal to Cat1MorphismOps;
.
gap> IsMonomorphism( mu ); true gap> IsEpimorphism( mu ); false gap> IsIsomorphism( mu ); false gap> IsEndomorphism( mu ); false gap> IsAutomorphism( mu ); false
80.66 Cat1MorphismSourceHomomorphism
Cat1MorphismSourceHomomorphism ( C, D, phi )
Given a homomorphism from the source of C
to the source of D
, this
function defines the corresponding cat1-group morphism.
gap> GSC := SC.source;; gap> homsrc := GroupHomomorphismByImages( a4, GSC, > [(1,2,3),(2,3,4)],[(4,5,6),(4,6,5)]);; gap> musrc := Cat1MorphismSourceHomomorphism( AC, SC, homsrc ); Morphism of cat1-groups <[a4 ==> a4]-->[c3^2|Xc2 ==> s3]> gap> IsCat1Morphism( musrc ); true gap> Cat1MorphismPrint( musrc ); Morphism of cat1-groups := : Source = cat1-group [a4 ==> a4] : Range = cat1-group [c3^2|Xc2 ==> s3] : Source homomorphism maps source generators to: [ (4,5,6), (4,6,5) ] : Range homomorphism maps range generators to: [ (4,5,6), (4,6,5) ]
ReverseCat1( C )
The reverse of a cat1-group is an isomorphic cat1-group with the same source, range and embedding, but with the tail and head interchanged (see xmodAW1, section 2).
gap> revCC := ReverseCat1( CC ); cat1-group [h20 |X c5 ==> h20]
ReverseIsomorphismCat1( C )
gap> revmu := ReverseIsomorphismCat1( CC ); Morphism of cat1-groups <[Perm(h20 |X c5) ==> h20]-->[h20 |X c5 ==> h20]> gap> IsCat1Morphism( revmu ); true
80.69 Cat1MorphismXModMorphism
Cat1MorphismXModMorphism( mor )
If C1, C2
are the cat1-groups produced from X1, X2
by the function
Cat1XMod
, then for any mor : X1 -> X2
there is an associated
mu : C1 -> C2. The result is stored as
mor.cat1Morphism
.
gap> CX.Cat1 := CCX;; gap> CSX := Cat1XMod( SX ); cat1-group [Perm(sl(2,3) |X q8) ==> sl(2,3)] gap> mor; Morphism of crossed modules <[q8->sl(2,3)] >-> [k4->a4]> gap> catmor := Cat1MorphismXModMorphism( mor ); Morphism of cat1-groups <[Perm(sl(2,3) |X q8) ==> sl(2,3)]-->[Perm(a4 |X k4) ==> a4]> gap> IsCat1Morphism( catmor ); true gap> Cat1MorphismPrint( catmor ); Morphism of cat1-groups := : Source = cat1-group [Perm(sl(2,3) |X q8) ==> sl(2,3)] : Range = cat1-group [Perm(a4 |X k4) ==> a4] : Source homomorphism maps source generators to: [ (5,6)(7,8), (5,7)(6,8), (2,3,4)(6,7,8), (1,2)(3,4), (1,3)(2,4) ] : Range homomorphism maps range generators to: [ (1,2)(3,4), (1,3)(2,4), (2,3,4) ]
80.70 XModMorphismCat1Morphism
XModMorphismCat1Morphism ( mu )
If X1,X2
are the two crossed modules produced from C1,C2
by the
function XModCat1
, then for any mu : C1 -> C2
there is an
associated morphism of crossed modules from X1
to X2
. The result
is stored as mu.xmodMorphism
.
gap> mu; Morphism of cat1-groups <[a4 ==> a4]-->[a4.k4 ==> a4]> gap> xmu := XModMorphismCat1Morphism( mu ); Morphism of crossed modules <[a4->a4] >-> [k4->a4]>
80.71 CompositeMorphism for cat1-groups
Cat1MorphismOps.CompositeMorphism( mu1,mu2 )
Morphisms μ1 : C → D and μ2 : D → E have a composite μ = μ2 o μ1 : C → E whose source and range homomorphisms are the composites of those of μ1 and μ2. The example corresponds to that in \refCompositeMorphism for crossed modules.
gap> psi; Morphism of crossed modules <[c4->q8] >-> [c4->q8]> gap> inc; Morphism of crossed modules <[c4->q8] >-> [q8->sl(2,3)]> gap> mupsi := Cat1MorphismXModMorphism( psi ); Morphism of cat1-groups <[Perm(q8 |X c4) ==> q8]-->[Perm(q8 |X c4) ==> q8]> gap> muinc := Cat1MorphismXModMorphism( inc ); Morphism of cat1-groups <[Perm(q8 |X c4) ==> q8]-->[Perm(sl(2,3) |X q8) ==> sl(2,3)]> gap> mucomp := Cat1MorphismOps.CompositeMorphism( mupsi, muinc ); Morphism of cat1-groups <[Perm(q8 |X c4) ==> q8]-->[Perm(sl(2,3) |X q8) ==> sl(2,3)]> gap> muxcomp := Cat1MorphismXModMorphism( xcomp );; gap> mucomp = muxcomp; true
IdentitySubCat1( C )
Every cat1-group C has an identity sub-cat1-group whose source and range are the identity subgroups of the source and range of C.
gap> IdentitySubCat1( SC ); cat1-group [Id[c3^2|Xc2 ==> s3]]
SubCat1( C, H )
When H
is a subgroup of C.source
and the restrictions of C.tail
and C.head
to H
have a common image, these homomorphisms determine
a sub-cat1-group of C
.
gap> d20 := Subgroup( h20, [ (1,2,3,4,5), (2,5)(3,4) ] );; gap> subC := SubCat1( C, d20 ); cat1-group [Sub[h20 ==> c4]] gap> Cat1Print( subC ); cat1-group [Sub[h20 ==> c4]] :- : source group has generators: [ (1,2,3,4,5), (2,5)(3,4) ] : range group has generators: [ ( 2, 5)( 3, 4) ] : tail homomorphism maps source generators to: [ (), ( 2, 5)( 3, 4) ] : head homomorphism maps source generators to: [ (), ( 2, 5)( 3, 4) ] : range embedding maps range generators to: [ ( 2, 5)( 3, 4) ] : kernel has generators: [ (1,2,3,4,5) ] : boundary homomorphism maps generators of kernel to: [ () ] : kernel embedding maps generators of kernel to: [ ( 1, 2, 3, 4, 5) ]
80.74 InclusionMorphism for cat1-groups
InclusionMorphism( S, C )
This function constructs the inclusion morphism S -> C
of a
sub-cat1-group S
of a cat1-group C
.
gap> InclusionMorphism( subC, C ); Morphism of cat1-groups <[Sub[h20 ==> c4]]-->[h20 ==> c4]>
NormalSubCat1s( C )
This function takes pairs of normal subgroups from the source and range of C and constructs a normal sub-cat1-group whenever the axioms are satisfied.
gap> NormalSubCat1s( SC ); [ cat1-group [Sub[c3^2|Xc2 ==> s3]] , cat1-group [Sub[c3^2|Xc2 ==> s3]] , cat1-group [Sub[c3^2|Xc2 ==> s3]] , cat1-group [Sub[c3^2|Xc2 ==> s3]] ]
AllCat1s( G )
By a cat1-structure on G we mean a cat1-group C where
R is a subgroup of G and e is the inclusion map. For such a
structure to exist, G must contain a normal subgroup S with G/S
≅ R. Furthermore, since t,h are respectively the identity and
zero maps on S, we require R ∩ S = { 1G }. This function
uses EndomorphismClasses( G, 3 )
(see \refEndomorphismClasses,
\refIdempotentImages) to construct idempotent endomorphisms of G
as potential tails and heads. A backtrack procedure then tests to see
which pairs of idempotents give cat1-groups. A non-documented
function AreIsomorphicCat1s
is called in order that the function
returns representatives for isomorphism classes of cat1-structures on
G
. See xmodAW1 for all cat1-structures on groups of order up to
30.
gap> AllCat1s( a4 ); There are 1 endomorphism classes. Calculating idempotent endomorphisms. # idempotents mapping to lattice class representatives: [ 1, 0, 1, 0, 1 ] Isomorphism class 1 : kernel of tail = [ "2x2" ] : range group = [ "3" ] Isomorphism class 2 : kernel of tail = [ "1" ] : range group = [ "A4" ] [ cat1-group [a4 ==> a4.H3] , cat1-group [a4 ==> a4] ]
The first class has range c3
and kernel k4
.
The second class contails all cat1-groups
C = (α-1; α, α : G → G)
where α is an automorphism of G.
\newpage
80.77 About derivations and sections
The Whitehead monoid \rmDer( X) of X was defined in xmodW2 to be the monoid of all derivations from R to S, that is the set of all maps R → S, with composition o , satisfying
The zero map is the identity for this composition. Invertible elements in the monoid are called regular. The Whitehead group of X is the group of regular derivations in \rmDer( X ). In section \refAbout actors the actor of X is defined as a crossed module whose source and range are permutation representations of the Whitehead group and the automorphism group of X.
The construction for cat1-groups equivalent to the derivation of a crossed module is the section. The monoid of sections of C is the set of group homomorphisms ξ : R → G, with composition o , satisfying:
Derivations are stored like group homomorphisms by specifying the
images of a generating set. Images of the remaining elements may then
be obtained using axiom \textbfDer 1. The function IsDerivation
is automatically called to check that this procedure is well-defined.
gap> X1; Crossed module [c5->PermAut(c5)] gap> chi1 := XModDerivationByImages( X1, [ () ] ); XModDerivationByImages( PermAut(c5), c5, [ (1,2,4,3) ], [ () ] ) gap> IsDerivation( chi1 ); true
A derivation is stored as a record chi
with fields:
beg-tabularll
chi.source
, & the range group R of X,
chi.range
, & the source group S of X,
chi.generators
, & a fixed generating set for R,
chi.genimages
, & the chosen images of the generators,
chi.xmod
, & the crossed module X,
chi.operations
, & special set of operations XModDerivationByImagesOps
,
chi.isDerivation
,& a boolean flag, normally true
.
end-tabular
Sections \emphare group homomorphisms, and are stored as such, but
with a modified set of operations Cat1SectionByImagesOps
which
includes a special .Print
function to display the section in the
manner shown below. Functions SectionDerivation
and
DerivationSection
convert derivations to sections, and vice-versa,
calling Cat1XMod
and XModCat1
automatically.
The equation ξ r = (er)(χ r) determines a section ξ of \mathcalC, given a derivation χ of \mathcalX, and conversely.
gap> xi1 := SectionDerivation( chi1 ); Cat1SectionByImages( PermAut(c5), Perm(PermAut(c5) |X c5), [ (1,2,4,3) ], [ (2,3,5,4) ] ) gap> xi1.cat1; cat1-group [Perm(PermAut(c5) |X c5) ==> PermAut(c5)]
There are two functions to determine all the elements of the Whitehead
group and the Whitehead monoid of X, namely
RegularDerivations
and AllDerivations
. If the whole monoid is
needed at some stage, then the latter function should be used. A
field D = X.derivations
is created which stores all the required
information:
beg-tabularll
D.areDerivations
, & a boolean flag, normally true
,
D.isReg
, & true
when only the regular derivations are known,
D.isAll
, & true
when all the derivations have been found,
D.generators
, & a \emphcopy of R.generators
,
D.genimageList
, & a list of .genimages
lists for the derivations,
D.regular
, & the number of regular derivations (if known),
D.xmod
, & the crossed module X,
D.operations
, & a special set of operations XModDerivationsOps
.
end-tabular
Using our standard example X1
we find that there are just five
derivations, all of them regular, so the associated group is cyclic of
size 5.
gap> RegularDerivations( X1 ); RegularDerivations record for crossed module [c5->PermAut(c5)], : 5 regular derivations, others not found. gap> AllDerivations( X1 ); AllDerivations record for crossed module [c5->PermAut(c5)], : 5 derivations found but unsorted. gap> DerivationsSorted( X1 ); true gap> imder1 := X1.derivations.genimageList; [ [()], [(1,2,3,4,5)], [(1,3,5,2,4)], [(1,4,2,5,3)], [(1,5,4,3,2)] ]
The functions RegularSections
and AllSections
perform
corresponding tasks for a cat1-group. Two strategies for calculating
derivations and sections are implemented, see xmodAW1. The default
method for AllDerivations
is to search for all possible sets of
images using a backtracking procedure, and when all the derivations
are found it is not known which are regular. The function
DerivationsSorted
sorts the .genImageList
field, placing the
regular ones at the top of the list and adding the .regular
field.
The default method for AllSections( C )
computes all endomorphisms
on the range group R
of C
as possibilities for the composite
hξ. A backtrack method then finds possible images for such a
section. When either the set of derivations or the set of sections
already exists, the other set is computed using SectionDerivation
or
DerivationSection
.
gap> CX1 := Cat1XMod( X1 ); cat1-group [Perm(PermAut(c5) |X c5) ==> PermAut(c5)] gap> CX1.source.name := "Hol(c5)";; Cat1Name( CX1 ); gap> RegularSections( CX1 ); RegularSections record for cat1-group [Hol(c5) ==> PermAut(c5)], : 5 regular sections, others not found. gap> CX1.sections.genimageList; [ [(2,3,5,4)], [(1,2,4,3)], [(1,3,2,5)], [(1,4,5,2)], [(1,5,3,4)] ]
The derivation images and the composition table may be listed as follows.
gap> chi2 := XModDerivationByImages( X1, imder1[2] ); XModDerivationByImages( PermAut(c5), c5, [(1,2,4,3)], [(1,2,3,4,5)] ) gap> DerivationImage( chi2, (1,4)(2,3) ); ( 1, 4, 2, 5, 3) gap> DerivationImages( chi2 ); [ 1, 2, 3, 4 ] gap> PrintList( DerivationTable( X1 ) ); [ 1, 1, 1, 1 ] [ 1, 2, 3, 4 ] [ 1, 3, 5, 2 ] [ 1, 4, 2, 5 ] [ 1, 5, 4, 3 ] gap> PrintList( WhiteheadGroupTable( X1 ) ); [ 1, 2, 3, 4, 5 ] [ 2, 3, 4, 5, 1 ] [ 3, 4, 5, 1, 2 ] [ 4, 5, 1, 2, 3 ] [ 5, 1, 2, 3, 4 ]
Each χ or ξ determines endomorphisms of R, S, G, X and C, namely:
When these endomorphisms are automorphisms, the derivation is regular. When the boundary of X is the zero map, both σ and ρ are identity homomorphisms, and every derivation is regular, which is the case in this example.
gap> sigma2 := SourceEndomorphismDerivation( chi2 ); GroupHomomorphismByImages( c5, c5, [ (1,2,3,4,5) ], [ (1,2,3,4,5) ] ) gap> rho2 := RangeEndomorphismDerivation( chi2 ); GroupHomomorphismByImages( PermAut(c5), PermAut(c5), [ (1,2,4,3) ], [ (1,2,4,3) ] ) gap> xi2 := SectionDerivation( chi2 );; gap> gamma2 := SourceEndomorphismSection( xi2 ); GroupHomomorphismByImages( Hol(c5), Hol(c5), [(2,3,5,4),(1,2,3,4,5)], [(2,3,5,4),(1,2,3,4,5)] ) gap> mor2 := XModMorphism( X1, X1, [sigma2,rho2] ); Morphism of crossed modules <[c5->PermAut(c5)] >-> [c5->PermAut(c5)]> gap> mu2 := Cat1Morphism( CX1, CX1, [gamma2,rho2] ); Morphism of cat1-groups <[Hol(c5) ==> PermAut(c5)]--> [Hol(c5) ==> PermAut(c5)]>
XModDerivationByImages( X, im )
This function takes a list of images in S = X.source
for the
generators of R = X.range
and constructs a map χ : R → S
which is then tested to see whether the axioms of a derivation are
satisfied.
gap> XSC; Crossed module [c3->s3] gap> imchi := [ (1,2,3)(4,6,5), (1,2,3)(4,6,5) ];; gap> chi := XModDerivationByImages( XSC, imchi ); XModDerivationByImages( s3, c3, [ (4,5,6), (2,3)(5,6) ], [ (1,2,3)(4,6,5), (1,2,3)(4,6,5) ] )
IsDerivation( X, im )
IsDerivation( chi )
This function may be called in two ways, and tests that the derivation given by the images of its generators is well-defined.
gap> im0 := [ (1,3,2)(4,5,6), () ];; gap> IsDerivation( XSC, im0 ); true
DerivationImage( chi, r )
This function returns χ(r) ∈ S when χ is a derivation.
gap> DerivationImage( chi, (4,6,5) ); (1,3,2)(4,5,6)
DerivationImages( chi )
All the images of the elements of R are found using
DerivationImage
and their positions in S.elements
is returned as a
list.
gap> XSC.source.elements; [ (), ( 1, 2, 3)( 4, 6, 5), ( 1, 3, 2)( 4, 5, 6) ] gap> DerivationImages(chi); [ 1, 2, 3, 2, 3, 1 ]
InnerDerivation( X, s )
When S,R are respectively the source and range of X
, each s ∈
S defines a derivation ηs : R → S, r → sr s-1.
These inner derivations are often called principal
derivations in the literature.
gap> InnerDerivation( XSC, (1,2,3)(4,6,5) ); XModDerivationByImages( s3, c3, [ (4,5,6), (2,3)(5,6) ], [ (), (1,2,3)(4,6,5) ] )
ListInnerDerivations( X )
This functions applies InnerDerivation
to every element of
X.source
and outputs a list of genimages for the resulting
derivations. This list is stored as .innerImageList
in the
derivations record.
gap> PrintList( ListInnerDerivations( XSC ) ); [ (), () ] [ (), ( 1, 2, 3)( 4, 6, 5) ] [ (), ( 1, 3, 2)( 4, 5, 6) ]
80.84 Operations for derivations
The operations record for derivations is XModDerivationByImagesOps
.
gap> RecFields( chi.operations ); [ "name", "operations", "IsMapping", "IsInjective", "IsSurjective", "IsBijection", "IsHomomorphism", "IsMonomorphism", "IsEpimorphism", "IsIsomorphism", "IsEndomorphism", "IsAutomorphism", "=", "<", "*", "/", "mod", "Comm", "^", "ImageElm", "ImagesElm", "ImagesSet", "ImagesSource", "ImagesRepresentative", "PreImageElm", "PreImagesSet", "PreImagesRange", "PreImagesRepresentative", "PreImagesElm", "CompositionMapping", "PowerMapping", "IsGroupHomomorphism", "KernelGroupHomomorphism", "IsFieldHomomorphism", "KernelFieldHomomorphism", "InverseMapping", "Print", "IsRegular" ]
Cat1SectionByImages( C, im )
This function takes a list of images in G = C.source
for the
generators of R = C.range
and constructs a homomorphism ξ : R
→ G which is then tested to see whether the axioms of a section are
satisfied.
gap> SC; cat1-group [c3^2|Xc2 ==> s3] gap> imxi := [ (1,2,3), (1,2)(4,6) ];; gap> xi := Cat1SectionByImages( SC, imxi ); Cat1SectionByImages( s3, c3^2|Xc2, [ (4,5,6), (2,3)(5,6) ], [ (1,2,3), (1,2)(4,6) ] )
IsSection( C, im )
IsSection( xi )
This function may be called in two ways, and tests that the section given by the images of its generators is well-defined.
gap> im0 := [ (1,2,3), (2,3)(4,5) ];; gap> IsSection( SC, im0 ); false
80.87 IsRegular for Crossed Modules
IsRegular( chi )
This function tests a derivation or a section to see whether it is invertible in the Whitehead monoid.
gap> IsRegular( chi ); false gap> IsRegular( xi ); false
The operations record for sections is Cat1SectionByImagesOps
.
gap> RecFields( xi.operations ); [ "name", "operations", "IsMapping", "IsInjective", "IsSurjective", "IsBijection", "IsHomomorphism", "IsMonomorphism", "IsEpimorphism", "IsIsomorphism", "IsEndomorphism", "IsAutomorphism", "=", "<", "*", "/", "mod", "Comm", "^", "ImageElm", "ImagesElm", "ImagesSet", "ImagesSource","ImagesRepresentative", "PreImageElm", "PreImagesElm", "PreImagesSet", "PreImagesRange", "PreImagesRepresentative", "CompositionMapping", "PowerMapping", "InverseMapping", "IsGroupHomomorphism", "CoKernel", "KernelGroupHomomorphism", "MakeMapping", "Print", "IsRegular" ]
RegularDerivations( X [,"back" \rm or "cat1"] )
By default, this function uses a backtrack search to find all the regular derivations of X. The result is stored in a derivations record. The alternative strategy, for which "cat1" option should be specified is to calculate the regular sections of the associated cat1-group first, and convert these to derivations.
gap> regXSC := RegularDerivations( XSC ); RegularDerivations record for crossed module [c3->s3], : 6 regular derivations, others not found. gap> PrintList( regXSC.genimageList ); [ (), () ] [ (), ( 1, 2, 3)( 4, 6, 5) ] [ (), ( 1, 3, 2)( 4, 5, 6) ] [ ( 1, 3, 2)( 4, 5, 6), () ] [ ( 1, 3, 2)( 4, 5, 6), ( 1, 2, 3)( 4, 6, 5) ] [ ( 1, 3, 2)( 4, 5, 6), ( 1, 3, 2)( 4, 5, 6) ] gap> RecFields( regXSC ); [ "areDerivations", "isReg", "isAll", "genimageList", "operations", "xmod", "generators", "regular" ]
AllDerivations( X [,"back" \rm or "cat1"] )
This function calculates all the derivations of X and
overwrites any existing subfields of X.derivations
.
gap> allXSC := AllDerivations( XSC ); AllDerivations record for crossed module [c3->s3], : 9 derivations found but unsorted.
DerivationsSorted( D )
This function tests the derivations in the derivation record D
to
see which are regular; sorts the list D.genimageList
, placing the
regular images first; and stores the number of regular derivations in
D.regular
. The function returns true
on successful completion.
gap> DerivationsSorted( allXSC ); true gap> PrintList( allXSC.genimageList ); [ (), () ] [ (), ( 1, 2, 3)( 4, 6, 5) ] [ (), ( 1, 3, 2)( 4, 5, 6) ] [ ( 1, 3, 2)( 4, 5, 6), () ] [ ( 1, 3, 2)( 4, 5, 6), ( 1, 2, 3)( 4, 6, 5) ] [ ( 1, 3, 2)( 4, 5, 6), ( 1, 3, 2)( 4, 5, 6) ] [ ( 1, 2, 3)( 4, 6, 5), () ] [ ( 1, 2, 3)( 4, 6, 5), ( 1, 2, 3)( 4, 6, 5) ] [ ( 1, 2, 3)( 4, 6, 5), ( 1, 3, 2)( 4, 5, 6) ]
DerivationTable( D )
The function DerivationImages
in \refDerivationImages is applied
to each derivation in the current derivations record and a list of
positions of images in S is returned.
gap> PrintList( DerivationTable( allXSC ) ); [ 1, 1, 1, 1, 1, 1 ] [ 1, 1, 1, 2, 2, 2 ] [ 1, 1, 1, 3, 3, 3 ] [ 1, 3, 2, 1, 3, 2 ] [ 1, 3, 2, 2, 1, 3 ] [ 1, 3, 2, 3, 2, 1 ] [ 1, 2, 3, 1, 2, 3 ] [ 1, 2, 3, 2, 3, 1 ] [ 1, 2, 3, 3, 1, 2 ]
AreDerivations( D )
This function checks that the record D
has the correct fields for a
derivations record (regular or all).
gap> AreDerivations( regXSC ); true
RegularSections( C [,"endo" \rm or "xmod"] )
By default, this function computes the set of idempotent automorphisms
from R → R and takes these as possible choices for hξ. A
backtrack procedure then calculates possible images for such a
section. The result is stored in a sections record C.sections
with
fields similar to those of a serivations record. The alternative
strategy, for which "xmod" option should be specified is to
calculate the regular derivations of the associated crossed module
first, and convert the resulting derivations to sections.
gap> Unbind( XSC.derivations ); gap> regSC := RegularSections( SC ); RegularSections record for cat1-group [c3^2|Xc2 ==> s3], : 6 regular sections, others not found.
AllSections( C [,"endo" \rm or "xmod"] )
By default, this function computes the set of idempotent endomorphisms
from R → R (see sections \refEndomorphismClasses,
\refIdempotentImages) and takes these as possible choices for the
composite homomorphism hξ. A backtrack procedure then calculates
possible images for such a section. This function calculates all the
sections of C and overwrites any existing subfields of
C.sections
.
gap> allSC := AllSections( SC ); AllSections record for cat1-group [c3^2|Xc2 ==> s3], : 6 regular sections, 3 irregular ones found. gap> RecFields( allSC ); [ "areSections", "isReg", "isAll", "regular", "genimageList", "generators", "cat1", "operations" ] gap> PrintList( allSC.genimageList ); [ ( 4, 5, 6), ( 2, 3)( 5, 6) ] [ ( 4, 5, 6), ( 1, 3)( 4, 5) ] [ ( 4, 5, 6), ( 1, 2)( 4, 6) ] [ ( 1, 3, 2)( 4, 6, 5), ( 2, 3)( 5, 6) ] [ ( 1, 3, 2)( 4, 6, 5), ( 1, 3)( 4, 5) ] [ ( 1, 3, 2)( 4, 6, 5), ( 1, 2)( 4, 6) ] [ ( 1, 2, 3), ( 2, 3)( 5, 6) ] [ ( 1, 2, 3), ( 1, 2)( 4, 6) ] [ ( 1, 2, 3), ( 1, 3)( 4, 5) ] gap> allXSC := AllDerivations( XSC, "cat1" ); AllDerivations record for crossed module [c3->s3], : 6 regular derivations, 3 irregular ones found.
AreSections( S )
This function checks that the record S
has the correct fields for a
sections record (regular or all).
gap> AreSections( allSC ); true
SectionDerivation( D, i )
This function converts a derivation of X
to a section of the
associated cat1-group C
. This function is inverse to
DerivationSection
. In the following examples we note that allXSC
has been obtained using allSC
, so the derivations and sections
correspond in the same order.
gap> chi8 := XModDerivationByImages( XSC, allXSC.genimageList[8] ); XModDerivationByImages( s3, c3, [ (4,5,6), (2,3)(5,6) ], [ ( 1,2,3)(4,6,5), (1,2,3)(4,6,5) ] ) gap> xi8 := SectionDerivation( chi8 ); GroupHomomorphismByImages( s3, c3^2|Xc2, [ (4,5,6), (2,3)(5,6) ], [ (1,2,3), (1,2)(4,6) ] )
DerivationSection( C, xi )
This function converts a section of C
to a derivation of the
associated crossed module X
. This function is inverse to
SectionDerivation
.
gap> xi4 := Cat1SectionByImages( SC, allSC.genimageList[4] ); Cat1SectionByImages( s3, c3^2|Xc2, [ (4,5,6), (2,3)(5,6) ], [ (1,3,2)(4,6,5), (2,3)(5,6) ] ) gap> chi4 := DerivationSection( xi4 ); XModDerivationByImages( s3, c3, [ (4,5,6), (2,3)(5,6) ], [ (1,3,2)(4,5,6), () ] )
CompositeDerivation( chi, chj )
This function applies the Whitehead product to two derivations and
returns the composite. In the example, derivations chi4
, chi8
correspond to sections xi4
and xi8
.
gap> chi48 := CompositeDerivation( chi4, chi8 ); XModDerivationByImages( s3, c3, [ (4,5,6), (2,3)(5,6) ], [ (1,2,3)(4,6,5), (1,3,2)(4,5,6) ] )
CompositeSection( xi, xj )
This function applies the Whitehead composition to two sections and returns the composite.
gap> xi48 := CompositeSection( xi4, xi8 ); Cat1SectionByImages( s3, c3^2|Xc2, [ (4,5,6), (2,3)(5,6) ], [ ( 1,2,3), (1,3)(4,5) ] ) gap> SectionDerivation( chi48 ) = xi48; true
WhiteheadGroupTable( X )
This function applies CompositeDerivation
to all pairs of regular
derivations, producing the Whitehead group multiplication table. A
field .groupTable
is added to D
.
gap> WGT := WhiteheadGroupTable( XSC );; PrintList( WGT ); returning existing ALL derivations [ 1, 2, 3, 4, 5, 6 ] [ 2, 3, 1, 5, 6, 4 ] [ 3, 1, 2, 6, 4, 5 ] [ 4, 6, 5, 1, 3, 2 ] [ 5, 4, 6, 2, 1, 3 ] [ 6, 5, 4, 3, 2, 1 ]
WhiteheadMonoidTable( X )
The derivations of X
form a monoid with the first derivation as
identity. This function applies CompositeDerivation
to all pairs of
derivations and produces the multiplication table as a list of lists.
A field .monoidTable
is added to D
. In our example there are 9
derivations and the three irregular ones, numbers 7,8,9, are all left
zeroes.
gap> WMT := WhiteheadMonoidTable( XSC );; PrintList(WMT ); [ 1, 2, 3, 4, 5, 6, 7, 8, 9 ] [ 2, 3, 1, 5, 6, 4, 9, 7, 8 ] [ 3, 1, 2, 6, 4, 5, 8, 9, 7 ] [ 4, 6, 5, 1, 3, 2, 7, 9, 8 ] [ 5, 4, 6, 2, 1, 3, 9, 8, 7 ] [ 6, 5, 4, 3, 2, 1, 8, 7, 9 ] [ 7, 7, 7, 7, 7, 7, 7, 7, 7 ] [ 8, 8, 8, 8, 8, 8, 8, 8, 8 ] [ 9, 9, 9, 9, 9, 9, 9, 9, 9 ]
InverseDerivations( X, i )
When T[i]
is a regular derivation, this function returns the
position j
such that T[j]
is the inverse of T[i]
in the
Whitehead group. When T[i]
is not regular, a list of values j
is
returned such that the inverse semigroup condition xyx = x, yxy =
y is satisfied, where x = T[i], y = T[j]. Notice that derivation 8
has order 3 and derivation 15 as inverse.
gap> inv4 := InverseDerivations( chi4 ); [ 4 ] gap> inv8 := InverseDerivations( chi8 ); [ 7, 8, 9 ]
ListInverseDerivations( X )
This function applies InverseDerivations
to all the derivations. A
field .inverses
is added to D
.
gap> inv := ListInverseDerivations( XSC ); [ [ 1 ], [ 3 ], [ 2 ], [ 4 ], [ 5 ], [ 6 ], [ 7, 8, 9 ], [ 7, 8, 9 ], [ 7, 8, 9 ] ]
80.105 SourceEndomorphismDerivation
SourceEndomorphismDerivation( chi )
Each derivation χ determines an endomorphism σ of S
such
that σ s = s (χ ∂ s). This construction defines a
homomorphism from the Whitehead group to Aut(S) which forms the
action homomorphism of the Whitehead crossed module described in
section
\refWhitehead crossed module.
gap> sigma8 := SourceEndomorphismDerivation( chi8 ); GroupHomomorphismByImages( c3, c3, [ (1,2,3)(4,6,5) ], [ () ] ) gap> sigma4 := SourceEndomorphismDerivation( chi4 ); GroupHomomorphismByImages( c3, c3, [ (1,2,3)(4,6,5) ], [ (1,3,2)(4,5,6) ] )
80.106 TableSourceEndomorphismDerivations
TableSourceEndomorphismDerivations( X )
Applying SourceEndomorphismDerivation
to every derivation produces a
list of endomorphisms of S = X.source
. This function returns a list
of .genimages
for these endomorphisms. Note that, in this example,
S = c3
and the irregular derivations produce zero maps.
gap> TSE := TableSourceEndomorphismDerivations( XSC );; gap> PrintList( TSE ); [ ( 1, 2, 3)( 4, 6, 5) ] [ ( 1, 2, 3)( 4, 6, 5) ] [ ( 1, 2, 3)( 4, 6, 5) ] [ ( 1, 3, 2)( 4, 5, 6) ] [ ( 1, 3, 2)( 4, 5, 6) ] [ ( 1, 3, 2)( 4, 5, 6) ] [ () ] [ () ] [ () ]
80.107 RangeEndomorphismDerivation
RangeEndomorphismDerivation( chi )
Each derivation χ determines an endomorphism ρ of R
such
that ρ r = r (∂ χ r). This construction defines a
homomorphism from the Whitehead group to Aut(R).
gap> rho8 := RangeEndomorphismDerivation( chi8 ); GroupHomomorphismByImages( s3, s3, [ (4,5,6), (2,3)(5,6) ], [ (), (2,3)(4,6) ] ) gap> rho4 := RangeEndomorphismDerivation( chi4 ); GroupHomomorphismByImages( s3, s3, [ (4,5,6), (2,3)(5,6) ], [ (4,6,5), (2,3)(5,6) ] )
80.108 TableRangeEndomorphismDerivations
TableRangeEndomorphismDerivations( X )
Applying RangeEndomorphismDerivation
to every derivation produces a
list of endomorphisms of R = X.range
. This function returns a list
of .genimages
for these endomorphisms. Note that, in this example,
the 3 irregular derivations map onto the 3 cyclic subgroups of order
2.
gap> TRE := TableRangeEndomorphismDerivations( XSC );; gap> PrintList( TRE ); [ (4,5,6), (2,3)(5,6) ] [ (4,5,6), (2,3)(4,5) ] [ (4,5,6), (2,3)(4 6) ] [ (4,6,5), (2,3)(5,6) ] [ (4,6,5), (2,3)(4,5) ] [ (4,6,5), (2,3)(4,6) ] [ (), (2,3)(5,6) ] [ (), (2,3)(4,6) ] [ (), (2,3)(4,5) ]
80.109 XModEndomorphismDerivation
XModEndomorphismDerivation( chi )
The endomorphisms sigma4
, rho4
together determine a pair which may
be used to construct an endomorphism of X. When the
derivation is regular, the resulting morphism is an automorphism, and
this construction determines a homomorphism from the Whitehead group
to the automorphism group of X
.
gap> phi4 := XModEndomorphismDerivation( chi4 ); Morphism of crossed modules <[c3->s3]->[c3->s3]>
80.110 SourceEndomorphismSection
SourceEndomorphismSection( xi )
Each section ξ determines an endomorphism γ of G
such
that
γ g = (e h ξ t g)(ξ t g-1) g (e h g-1) (ξ h g). |
gap> gamma4 := SourceEndomorphismSection( xi4 ); GroupHomomorphismByImages( c3^2|Xc2, c3^2|Xc2, [ (1,2,3), (4,5,6), (2,3)(5,6) ], [ (1,3,2), (4,6,5), (2,3)(5,6) ] )
80.111 RangeEndomorphismSection
RangeEndomorphismSection( xi )
Each derivation ξ determines an endomorphism ρ of R
such
that ρ r = h ξ r.
gap> rho4 := RangeEndomorphismSection( xi4 ); GroupHomomorphismByImages( s3, s3, [ (4,5,6), (2,3)(5,6) ], [ (4,6,5), (2,3)(5,6) ] )
80.112 Cat1EndomorphismSection
Cat1EndomorphismSection( xi )
The endomorphisms gamma4
, rho4
together determine a pair which may
be used to construct an endomorphism of C. When the
derivation is regular, the resulting morphism is an automorphism, and
this construction determines a homomorphism from the Whitehead group
to the automorphism group of C
.
gap> psi4 := Cat1EndomorphismSection( xi4 ); Morphism of cat1-groups <[c3^2|Xc2 ==> s3]-->[c3^2|Xc2 ==> s3]>
\newpage
The \emphactor of X is a crossed module (Δ : W( X) → Aut( X)) which was shown by Lue and Norrie, in xmodN2 and xmodN1 to give the automorphism object of a crossed module X. The source of the actor is a permutation representation W of the Whitehead group of regular derivations and the range is a permutation representation A of the automorphism group Aut( X) of X.
An automorphism ( σ, ρ ) of X
acts on the Whitehead monoid
by χ(σ,ρ) = σ-1 χ ρ, and this action
determines the action for the actor.
In fact the four groups R, S, W, A, the homomorphisms between them and the various actions, form five crossed modules:
beg-tabularrcll
X &:& S → R & the initial crossed module,
W(X) &:& S → W & the Whitehead crossed module of X,
L(X) &:& S → A & the Lue crossed module of X,
N(X) &:& R → A & the Norrie crossed module of X, and
Act( X) &:& W → A & the actor crossed module of X.
end-tabular
These 5 crossed modules, together with the evaluation W × R → S, (χ,r) → χ r, form a crossed square:
S ------ WX ------> W : \ : : \ : X LX ActX : \ : : \ : V \ V R ------ NX ------> A
in which pairs of boundaries or identity mappings provide six
morphisms of crossed modules. In particular, the boundaries of WX
and NX
form the \emphinner morphism of X
, mapping source
elements to inner derivations and range elements to inner
automorphisms. The image of X
under this morphism is the
\emphinner actor of X
, while the kernel is the \emphcentre of
X
.
In the example which follows, using the usual (X1 : c5 -> Aut(c5))
,
Act(X1)
is isomorphic to X1
and to LX1
while the Whitehead and Norrie boundaries are identity homomorphisms.
gap> X1; Crossed module [c5->PermAut(c5)] gap> WGX1 := WhiteheadPermGroup( X1 ); WG([c5->PermAut(c5)]) gap> WGX1.generators; [ (1,2,3,4,5) ] gap> AX1 := AutomorphismPermGroup( X1 ); PermAut([c5->PermAut(c5)]) gap> AX1.generators; [ (1,2,4,3) ] gap> XModMorphismAutoPerm( X1, AX1.generators[1] ); Morphism of crossed modules <[c5->PermAut(c5)] >-> [c5->PermAut(c5)]> gap> WX1 := Whitehead( X1 ); Crossed module Whitehead[c5->PermAut(c5)] gap> NX1 := Norrie( X1 ); Crossed module Norrie[c5->PermAut(c5)] gap> LX1 := Lue( X1 ); Crossed module Lue[c5->PermAut(c5)] gap> ActX1 := Actor( X1 );; gap> XModPrint( ActX1); Crossed module Actor[c5->PermAut(c5)] :- : Source group WG([c5->PermAut(c5)]) has generators: [ (1,2,3,4,5) ] : Range group has parent ( PermAut(c5)xPermAut(PermAut(c5)) ) and has generators: [ (1,2,4,3) ] : Boundary homomorphism maps source generators to: [ () ] : Action homomorphism maps range generators to automorphisms: (1,2,4,3) --> { source gens --> [ (1,3,5,2,4) ] } This automorphism generates the group of automorphisms. gap> InActX1 := InnerActor( X1 ); Crossed module Actor[c5->PermAut(c5)] gap> InActX1 = ActX1; true gap> InnerMorphism( X1 ); Morphism of crossed modules <[c5->PermAut(c5)] >-> Actor[c5->PermAut(c5)]> gap> Centre( X1 ); Crossed module Centre[c5->PermAut(c5)]
All of these constructions are stored in a sub-record X1.actorSquare
.
ActorSquareRecord( X )
ActorSquareRecord( C )
This function creates a new field .actorSquare
for the crossed
module or cat1-group, initially containing .isActorSquare := true
and .xmod
or .cat1
as appropriate. Components for the actor of
X
or C
are stored here when constructed.
gap> ActorSquareRecord( X1 ); rec( isActorSquare := true, xmod := Crossed module [c5->PermAut(c5)], WhiteheadPermGroup := WG([c5->PermAut(c5)]), automorphismPermGroup := PermAut([c5->PermAut(c5)]), Whitehead := Crossed module Whitehead[c5->PermAut(c5)], Norrie := Crossed module Norrie[c5->PermAut(c5)], Lue := Crossed module Lue[c5->PermAut(c5)], actor := Crossed module Actor[c5->PermAut(c5)], innerMorphism := Morphism of crossed modules <[c5->PermAut(c5)] >-> Actor[c5->PermAut(c5)]>, innerActor := Crossed module Actor[c5->PermAut(c5)] )
WhiteheadPermGroup( X )
This function first calls WhiteheadGroupTable
, see
\refWhiteheadGroupTable. These lists are then converted to
permutations, producing a permutation group which is effectively a
regular representation of the group. A field .WhiteheadPermGroup
is
added to X.actorSquare
and a field .genpos
is added to
D = X.derivations. The latter is a list of the positions in
D.genimageList
corresponding to the chosen generating elements. The
group is given the name WG(<name of X>)
.
For an example, we return to the crossed module XSC = [c3->s3]
obtained from the cat1-group SC
in section \refCat1Select which
has Whitehead group and automorphism group isomorphic to s3
.
gap> WG := WhiteheadPermGroup( XSC ); WG([c3->s3]) gap> XSC.derivations.genpos; [ 2, 4 ] gap> Elements( WG ); [ (), (1,2,3)(4,6,5), (1,3,2)(4,5,6), (1,4)(2,5)(3,6), (1,5)(2,6)(3,4), (1,6)(2,4)(3,5) ]
80.116 Whitehead crossed module
Whitehead( X )
This crossed module has the source of X
as source, and the Whitehead
group WX
as range. The boundary maps each element to the inner
derivation which it defines. The action uses
SourceEndomorphismDerivation
.
gap> WXSC := Whitehead( XSC ); Crossed module Whitehead[c3->s3] gap> XModPrint( WXSC ); Crossed module Whitehead[c3->s3] :- : Source group has parent ( c3^2|Xc2 ) and has generators: [ (1,2,3)(4,6,5) ] : Range group = WG([c3->s3]) has generators: [ (1,2,3)(4,6,5), (1,4)(2,5)(3,6) ] : Boundary homomorphism maps source generators to: [ (1,3,2)(4,5,6) ] : Action homomorphism maps range generators to automorphisms: (1,2,3)(4,6,5) --> { source gens --> [ (1,2,3)(4,6,5) ] } (1,4)(2,5)(3,6) --> { source gens --> [ (1,3,2)(4,5,6) ] } These 2 automorphisms generate the group of automorphisms.
80.117 AutomorphismPermGroup for crossed modules
XModOps.AutomorphismPermGroup( X )
This function constructs a permutation group PermAut(X)
isomorphic
to the group of automorphisms of the crossed module X
. First the
automorphism groups of the source and range of X
are obtained and
AutomorphismPair
used to obtain permutation representations of
these. The direct product of these permutation groups is constructed,
and the required automorphism group is a subgroup of this direct
product. The result is stored as X.automorphismPermGroup
which has
fields defining the various embeddings and projections.
gap> autXSC := AutomorphismPermGroup( XSC ); PermAut([c3->s3]) gap> autXSC.projsrc; GroupHomomorphismByImages( PermAut([c3->s3]), PermAut(c3), [ (5,6,7), (1,2)(3,4)(6,7) ], [ (), (1,2) ] ) gap> autXSC.projrng; GroupHomomorphismByImages( PermAut([c3->s3]), PermAut(s3), [ (5,6,7), (1,2)(3,4)(6,7) ], [ (3,4,5), (1,2)(4,5) ] ) gap> autXSC.embedSourceAuto; GroupHomomorphismByImages( PermAut(c3), PermAut(c3)xPermAut(s3), [ (1,2) ], [ (1,2) ] ) gap> autXSC.embedRangeAuto; GroupHomomorphismByImages( PermAut(s3), PermAut(c3)xPermAut(s3), [ (3,5,4), (1,2)(4,5) ], [ (5,7,6), (3,4)(6,7) ] ) gap> autXSC.autogens; [ [ GroupHomomorphismByImages( c3, c3, [ (1,2,3)(4,6,5) ], [ (1,2,3)(4,6,5) ] ), GroupHomomorphismByImages( s3, s3, [ (4,5,6), (2,3)(5,6) ], [ (4,5,6), (2,3)(4,5) ] ) ], [ GroupHomomorphismByImages( c3, c3, [ (1,2,3)(4,6,5) ], [ (1,3,2)(4,5,6) ] ), GroupHomomorphismByImages( s3, s3, [ (4,5,6), (2,3)(5,6) ], [ (4,6,5), (2,3)(5,6) ] ) ] ]
XModMorphismAutoPerm( X, perm )
Given the isomorphism between the automorphism group of X
and its
permutation representation PermAut(X)
, an element of the latter
determines an automorphism of X
.
gap> XModMorphismAutoPerm( XSC, (1,2)(3,4)(6,7) ); Morphism of crossed modules <[c3->s3] >-> [c3->s3]>
80.119 ImageAutomorphismDerivation
ImageAutomorphismDerivation( mor, chi )
An automorphism ( σ, ρ ) of X
acts on the left on the
Whitehead monoid by
(σ,ρ)χ = σ χ
ρ-1. This is converted to a right action on the
WhiteheadPermGroup
. In the example we see that phi4
maps chi8
to chi9
.
gap> chi8im := ImageAutomorphismDerivation( phi4, chi8 ); XModDerivationByImages( s3, c3, [ (4,5,6), (2,3)(5,6) ], [ (1,2,3)(4,6,5), (1,3,2)(4,5,6) ] ) gap> Position( allXSC.genimageList, chi8im.genimages ); 9
Norrie( X )
This crossed module has the range of X
as source and the
automorphism permutation group of X
as range.
gap> NXSC := Norrie( XSC ); Crossed module Norrie[c3->s3] gap> XModPrint( NXSC ); Crossed module Norrie[c3->s3] :- : Source group has parent ( c3^2|Xc2 ) and has generators: [ (4,5,6), (2,3)(5,6) ] : Range group has parent ( PermAut(c3)xPermAut(s3) ) and has generators: [ (5,6,7), (1,2)(3,4)(6,7) ] : Boundary homomorphism maps source generators to: [ (5,7,6), (1,2)(3,4)(6,7) ] : Action homomorphism maps range generators to automorphisms: (5,6,7) --> { source gens --> [ (4,5,6), (2,3)(4,5) ] } (1,2)(3,4)(6,7) --> { source gens --> [ (4,6,5), (2,3)(5,6) ] } These 2 automorphisms generate the group of automorphisms.
Lue( X )
This crossed module has the source of X
as source, and the
automorphism permutation group of X
as range.
gap> LXSC := Lue( XSC ); Crossed module Lue[c3->s3] gap> XModPrint( LXSC ); Crossed module Lue[c3->s3] :- : Source group has parent ( c3^2|Xc2 ) and has generators: [ (1,2,3)(4,6,5) ] : Range group has parent ( PermAut(c3)xPermAut(s3) ) and has generators: [ (5,6,7), (1,2)(3,4)(6,7) ] : Boundary homomorphism maps source generators to: [ (5,6,7) ] : Action homomorphism maps range generators to automorphisms: (5,6,7) --> { source gens --> [ (1,2,3)(4,6,5) ] } (1,2)(3,4)(6,7) --> { source gens --> [ (1,3,2)(4,5,6) ] } These 2 automorphisms generate the group of automorphisms.
Actor( X )
The actor of a crossed module X
is a crossed module Act(X)
which
has the Whitehead group (of regular derivations) as source group and
the automorphism group PermAut(X)
of X
as range group. The
boundary of Act(X)
maps each derivation to the automorphism provided
by XModEndomorphismDerivation
. The action of an automorphism on a
derivation is that provided by ImageAutomorphismDerivation
.
gap> ActXSC := Actor( XSC ); Crossed module Actor[c3->s3] gap> XModPrint( ActXSC ); Crossed module Actor[c3->s3] :- : Source group WG([c3->s3]) has generators: [ (1,2,3)(4,6,5), (1,4)(2,5)(3,6) ] : Range group has parent ( PermAut(c3)xPermAut(s3) ) and has generators: [ (5,6,7), (1,2)(3,4)(6,7) ] : Boundary homomorphism maps source generators to: [ (5,7,6), (1,2)(3,4)(6,7) ] : Action homomorphism maps range generators to automorphisms: (5,6,7) --> { source gens --> [ (1,2,3)(4,6,5), (1,6)(2,4)(3,5) ] } (1,2)(3,4)(6,7) --> { source gens --> [ (1,3,2)(4,5,6), (1,4)(2,5)(3,6) ] } These 2 automorphisms generate the group of automorphisms.
80.123 InnerMorphism for crossed modules
InnerMorphism( X )
The boundary maps of WX
and NX
form a morphism from X
to its actor.
gap> innXSC := InnerMorphism( XSC ); Morphism of crossed modules <[c3->s3] >-> Actor[c3->s3]> gap> XModMorphismPrint( innXSC ); Morphism of crossed modules :- : Source = Crossed module [c3->s3] with generating sets: [ (1,2,3)(4,6,5) ] [ (4,5,6), (2,3)(5,6) ] : Range = Crossed module Actor[c3->s3] with generating sets: [ (1,2,3)(4,6,5), (1,4)(2,5)(3,6) ] [ (5,6,7), (1,2)(3,4)(6,7) ] : Source Homomorphism maps source generators to: [ (1,3,2)(4,5,6) ] : Range Homomorphism maps range generators to: [ (5,7,6), (1,2)(3,4)(6,7) ] : isXModMorphism? true
80.124 Centre for crossed modules
XModOps.Centre( X )
The kernel of the inner morphism X -> ActX
is called the centre of
X
, generalising the centre of a group G, which is the kernel of G
→ Aut(G), g → (h → hg). In this example the centre
is trivial.
gap> ZXSC := Centre( XSC ); Crossed module Centre[c3->s3]
80.125 InnerActor for crossed modules
InnerActor( X )
The inner actor of X
is the image of the inner morphism.
gap> InnActXSC := InnerActor( XSC ); Crossed module InnerActor[c3->s3] gap> XModPrint( InnActXSC ); Crossed module InnerActor[c3->s3] :- : Source group has parent ( WG([c3->s3]) ) and has generators: [ (1,3,2)(4,5,6) ] : Range group has parent ( PermAut(c3)xPermAut(s3) ) and has generators: [ (5,7,6), (1,2)(3,4)(6,7) ] : Boundary homomorphism maps source generators to: [ (5,6,7) ] : Action homomorphism maps range generators to automorphisms: (5,7,6) --> { source gens --> [ (1,3,2)(4,5,6) ] } (1,2)(3,4)(6,7) --> { source gens --> [ (1,2,3)(4,6,5) ] } These 2 automorphisms generate the group of automorphisms.
Actor( C )
The actor of a cat1-group C is the cat1-group associated to the actor crossed module of the crossed module X associated to C. Its range is the automorphism group A and its source is A \semidirect W where W is the Whitehead group.
gap> ActSC := Actor( SC );; gap> Cat1Print( ActSC ); cat1-group Actor[c3^2|Xc2 ==> s3] :- : source group has generators: [ (4,6,5), (2,3)(5,6), (1,2,3)(4,6,5), (1,4)(2,5)(3,6) ] : range group has generators: [ (5,6,7), (1,2)(3,4)(6,7) ] : tail homomorphism maps source generators to: [ (5,6,7), (1,2)(3,4)(6,7), (), () ] : head homomorphism maps source generators to: [ (5,6,7), (1,2)(3,4)(6,7), (5,7,6), (1,2)(3,4)(6,7) ] : range embedding maps range generators to: [ (4,6,5), (2,3)(5,6) ] : kernel has generators: [ (1,2,3)(4,6,5), (1,4)(2,5)(3,6) ] : boundary homomorphism maps generators of kernel to: [ (5,7,6), (1,2)(3,4)(6,7) ] : kernel embedding maps generators of kernel to: [ (1,2,3)(4,6,5), (1,4)(2,5)(3,6) ] : associated crossed module is Crossed module Actor[c3->s3]
\newpage
80.127 About induced constructions
A morphism of crossed modules (σ, ρ) : X1 → X2 factors uniquely through an induced crossed module ρ* X1 = (δ : ρ* S1 → R2). Similarly, a morphism of cat1-groups factors through an induced cat1-group. Calculation of induced crossed modules of X also provides an algebraic means of determining the homotopy 2-type of homotopy pushouts of the classifying space of X. For more background from algebraic topology see references in xmodBH1, xmodBW1, xmodBW2. Induced crossed modules and induced cat1-groups also provide the building blocks for constructing pushouts in the categories \textbfXMod and \textbfCat1.
Data for the cases of algebraic interest is provided by a conjugation crossed module X = (∂ : S → R) and a homomorphism ι from R to a third group Q. The output from the calculation is a crossed module ι* X = (δ : ι*S → Q) together with a morphism of crossed modules X → ι* X. When ι is a surjection with kernel K then ι* S = [S,K] (see xmodBH1). When ι is an inclusion the induced crossed module may be calculated using a copower construction xmodBW1 or, in the case when R is normal in Q, as a coproduct of crossed modules (xmodBW2, not yet implemented). When ι is neither a surjection nor an inclusion, ι is written as the composite of the surjection onto the image and the inclusion of the image in Q, and then the composite induced crossed module is constructed.
Other functions required by the induced crossed module construction
include a function to produce a common transversal for the left and
right cosets of a subgroup (see \refIsCommonTransversal and
\refCommonTransversal). Also, modifications to some of the Tietze
transformation routines in fptietze.g
are required. These have yet
to be released as part of the GAP3 library and so are made available
in this package in file felsch.g
, but are not documented here.
As a simple example we take for X the conjugation crossed module (∂ : c4 → d8) and for ι the inclusion of d8 in d16. The induced crossed module has c4 × c4 as source.
gap> d16 := DihedralGroup( 16 ); d16.name := "d16";; Group( (1,2,3,4,5,6,7,8), (2,8)(3,7)(4,6) ) gap> d8 := Subgroup( d16, [ (1,3,5,7)(2,4,6,8), (1,3)(4,8)(5,7) ] );; gap> c4 := Subgroup( d8, [ (1,3,5,7)(2,4,6,8) ] );; gap> d8.name := "d8";; c4.name := "c4";; gap> DX := ConjugationXMod( d8, c4 ); Crossed module [c4->d8] gap> iota := InclusionMorphism( d8, d16 );; gap> IDXincl := InducedXMod( DX, iota ); Action of RQ on generators of I :- (1,2,3,4,5,6,7,8) : (1,4)(2,3) (2,8)(3,7)(4,6) : (1,2)(3,4)
#
I Protecting the first 1 generators.
#
I there are 2 generators and 3 relators of total length 12 partitioning the generators: [ [ 2 ], [ 1 ] ] Simplified presentation for I :-
#
I generators: [ fI.1, fI.3 ]
#
I relators:
#
I 1. 4 [ 1, 1, 1, 1 ]
#
I 2. 4 [ 2, 2, 2, 2 ]
#
I 3. 4 [ 2, -1, -2, 1 ] I has Size: 16 **************** Group is abelian factor 1 is abelian with invariants: [ 4 ] factor 2 is abelian with invariants: [ 4 ] Image of I has index 4 in RQ and is generated by : [ ( 1, 3, 5, 7)( 2, 4, 6, 8), ( 1, 7, 5, 3)( 2, 8, 6, 4) ] gap> XModPrint( IDXincl ); Crossed module [i*(c4)->d16] :- : Source group i*(c4) has generators: [ ( 1, 2, 4, 7)( 3, 5, 8,11)( 6, 9,12,14)(10,13,15,16), ( 1, 3, 6,10)( 2, 5, 9,13)( 4, 8,12,15)( 7,11,14,16) ] : Range group = d16 has generators: [ (1,2,3,4,5,6,7,8), (2,8)(3,7)(4,6) ] : Boundary homomorphism maps source generators to: [ ( 1, 3, 5, 7)( 2, 4, 6, 8), ( 1, 7, 5, 3)( 2, 8, 6, 4) ] : Action homomorphism maps range generators to automorphisms: (1,2,3,4,5,6,7,8) --> { source gens --> [ ( 1,10, 6, 3)( 2,13, 9, 5)( 4,15,12, 8)( 7,16,14,11), ( 1, 7, 4, 2)( 3,11, 8, 5)( 6,14,12, 9)(10,16,15,13) ] } (2,8)(3,7)(4,6) --> { source gens --> [ ( 1, 7, 4, 2)( 3,11, 8, 5)( 6,14,12, 9)(10,16,15,13), ( 1,10, 6, 3)( 2,13, 9, 5)( 4,15,12, 8)( 7,16,14,11) ] } These 2 automorphisms generate the group of automorphisms. : Kernel of the crossed module has generators: [ ( 1, 5,12,16)( 2, 8,14,10)( 3, 9,15, 7)( 4,11, 6,13) ] : Induced XMod from Crossed module [c4->d8] with source morphism: [ (1,3,5,7)(2,4,6,8) ] --> [ ( 1, 2, 4, 7)( 3, 5, 8,11)( 6, 9,12,14)(10,13,15,16) ]
In some of the sections which follow the output is very lengthy and so has been pruned.
InducedXMod( X, iota )
InducedXMod( Q, P, M )
This function requires as data a conjugation crossed module X = (∂ : M → P) and a homomorphism ι : P → Q. This data may be specified using either of the two forms shown, where the latter form required Q ≥ P ≥ M.
In the first example, ι is a surjection from d8
to k4
.
gap> d8gen := d8.generators; [ (1,3,5,7)(2,4,6,8), (1,3)(4,8)(5,7) ] gap> k4gen := k4.generators; [ (1,2)(3,4), (1,3)(2,4) ] gap> DX; Crossed module [c4->d8] gap> iota := GroupHomomorphismByImages( d8, k4, d8gen, k4gen );; gap> IDXsurj := InducedXMod( DX, iota ); Crossed module [c4/ker->k4] gap> XModPrint( IDXsurj ); Crossed module [c4/ker->k4] :- : Source group c4/ker has generators: [ (1,2,3,4) ] : Range group has parent ( s4 ) and has generators: [ (1,2)(3,4), (1,3)(2,4) ] : Boundary homomorphism maps source generators to: [ ( 1, 2)( 3, 4) ] : Action homomorphism maps range generators to automorphisms: (1,2)(3,4) --> { source gens --> [ (1,2,3,4) ] } (1,3)(2,4) --> { source gens --> [ (1,4,3,2) ] } These 2 automorphisms generate the group of automorphisms. : Induced XMod from Crossed module [c4->d8] with source morphism: [ (1,3,5,7)(2,4,6,8) ] --> [ (1,2,3,4) ]
In a second example we take (c3 -> s3)
as the initial crossed module
and s3 -> s4
as the inclusion. The induced group turns out to be
the special linear group sl(2,3)
.
gap> s3 := Subgroup( s4, [ (2,3), (1,2,3) ] );; gap> c3 := Subgroup( s3, [ (1,2,3) ] ); gap> s3.name := "s3";; c3.name := "c3";; gap> InducedXMod( s4, s3, c3 ); Action of RQ on generators of I :- (1,2,3,4) : (1,7,6,3)(2,8,5,4) (1,2) : (1,2)(3,4)(5,8)(6,7)
#
I Protecting the first 1 generators.
#
I there are 2 generators and 3 relators of total length 12 Simplified presentation for I :-
#
I generators: [ fI.1, fI.5 ]
#
I relators:
#
I 1. 3 [ 2, 2, 2 ]
#
I 2. 3 [ 1, 1, 1 ]
#
I 3. 6 [ 2, -1, -2, 1, -2, -1 ] I has Size: 24 **************** Searching Solvable Groups Library: GroupId = rec( catalogue := [ 24, 14 ], names := [ "SL(2,3)" ], size := 24 ) Image of I has index 2 in RQ and is generated by : [ (1,2,3), (1,2,4), (1,4,3), (2,3,4) ] Crossed module [i*(c3)->s4]
AllInducedXMods( Q )
This function calculates InducedXMod( Q, P, M )
where P
runs over
all conjugacy classes of subgroups of Q
and M
runs over all normal
subgroups of P
.
gap> AllInducedXMods( d8 );⋅
Number of induced crossed modules calculated = 11
InducedCat1( C, iota )
When C is the induced cat1-group associated to X the
induced cat1-group may be obtained by construction the induced crossed
module and then using the Cat1XMod
function. An experimental,
alternative procedure is to calculate the induced cat1-group
ι* G = G *R Q directly. This has been implemented for
the case when C = ( e;t,h : G → R) and ι : R → Q
is an inclusion.
The output from the calculation is a cat1-group C* = (e*;t*, h* : ι*G → Q) together with a morphism of crossed modules C → C*.
In the example an induced cat1-group is constructed whose associated crossed module has source c4 × c4 and range d16, so the source of the cat1-group is d16 \semidirect (c4 × c4).
gap> CDX := Cat1XMod( DX ); cat1-group [Perm(d8 |X c4) ==> d8] gap> inc := InclusionMorphism( d8, d16 );; gap> ICDX := InducedCat1( CDX, inc );⋅
new perm group size 256 cat1-group <ICG([Perm(d8 |X c4) ==> d8])> gap> XICDX := XModCat1( ICDX ); Crossed module [ker(<ICG([Perm(d8 |X c4) ==> d8])>)->d16] gap> AbelianInvariants( XICDX.source ); [ 4, 4 ]
\newpage
By a utility function we mean a GAP3 function which is:
• needed by other functions in this package,
• not (as far as we know) provided by the standard GAP3 library,
• more suitable for inclusion in the main library than in this package.
The first two utilities give particular group homomorphisms,
InclusionMorphism(H,G)
and ZeroMorphism(G,H)
. We often prefer
gap> incs3 := InclusionMorphism( s3, s3 ); IdentityMapping( s3 ) gap> incs3.genimages; [ (1,2), (2,3) ]
to IdentityMapping(s3)
because the latter does not provide the
fields .generators
and the .genimages
which many of the functions
in this package expect homomorphisms to possess.
The second set of utilities involve endomorphisms and automorphisms of groups. For example:
gap> end8 := EndomorphismClasses( d8 );; gap> RecFields( end8 ); [ "isDomain", "isEndomorphismClasses", "areNonTrivial", "classes", "intersectionFree", "group", "latticeLength", "latticeReps" ] gap> Length( end8.classes ); 11 gap> end8.classes[3]; rec( quotient := d8.Q3, projection := OperationHomomorphism( d8, d8.Q3 ), autoGroup := Group( IdentityMapping( d8.Q3 ) ), rangeNumber := 2, isomorphism := GroupHomomorphismByImages( d8.Q3, d8.H2, [ (1,2) ], [ (1,5)(2,6)(3,7)(4,8) ] ), conj := [ () ] ) gap> innd8 := InnerAutomorphismGroup( d8 ); Inn(d8) gap> innd8.generators; [ InnerAutomorphism( d8, (1,3,5,7)(2,4,6,8) ), InnerAutomorphism( d8, (1,3)(4,8)(5,7) ) ] gap> IsAutomorphismGroup( innd8 ); true
The third set of functions construct isomorphic pairs of groups, where a faithful permutation representation of a given type of group is constructed. Types covered include finitely presented groups, groups of automorphisms and semidirect products. A typical pair record includes the following fields:
beg-tabularll
.type & the given group G,
.perm & the permutation representation P,
.t2p & the isomorphism G → P,
.p2t & the inverse isomorphism P → G,
.isTypePair & a boolean flag, normally true
.
end-tabular
The inner automorphism group of the dihedral group d8
is isomorphic
to k4
:
gap> Apair := AutomorphismPair( innd8 ); rec( auto := Inn(d8), perm := PermInn(d8), a2p := OperationHomomorphism( Inn(d8), PermInn(d8) ), p2a := GroupHomomorphismByImages( PermInn(d8), Inn(d8), [ (1,3), (2,4) ], [ InnerAutomorphism( d8, (1,3,5,7)(2,4,6,8) ), InnerAutomorphism( d8, (1,3)(4,8)(5,7) ) ] ), isAutomorphismPair := true ) gap> IsAutomorphismPair( Apair ); true
The final set of functions deal with lists of subsets of [1..n]
and
construct systems of distinct and common representatives using simple,
non-recursive, combinatorial algorithms. The latter function returns
two lists: the set of representatives, and a permutation of the
subsets of the second list. It may also be used to provide a common
transversal for sets of left and right cosets of a subgroup H of a
group G, although a greedy algorithm is usually quicker.
gap> L := [ [1,4], [1,2], [2,3], [1,3], [5] ];; gap> DistinctRepresentatives( L ); [ 4, 2, 3, 1, 5 ] gap> M := [ [2,5], [3,5], [4,5], [1,2,3], [1,2,3] ];; gap> CommonRepresentatives( L, M ); [ [ 4, 1, 3, 1, 5 ], [ 3, 5, 2, 4, 1 ] ] gap> CommonTransversal( s4, c3 ); [ (), (3,4), (2,3), (1,3)(2,4), (1,2)(3,4), (2,4), (1,4), (1,4)(2,3) ]
InclusionMorphism( H, G )
This gives the inclusion map of a subgroup H of a group G. In the
case that H=G the IdentityMapping(G)
is returned, with fields
.generators
and .genimages
added.
gap> s4 := Group( (1,2,3,4), (1,2) );; s4.name:="s4";; gap> a4 := Subgroup( s4, [ (1,2,3), (2,3,4) ] );; a4.name:="a4";; gap> InclusionMorphism( a4, s4 ); GroupHomomorphismByImages( a4, s4, [ (1,2,3), (2,3,4) ], [ (1,2,3), (2,3,4) ] )
ZeroMorphism( G, H )
This gives the zero map from G to the identity subgroup of H.
gap> ZeroMorphism( s4, a4 ); GroupHomomorphismByImages( s4, a4, [ (1,2,3,4), (1,2) ], [ (), () ] )
EndomorphismClasses( G, case )
The monoid of endomorphisms is required when calculating the monoid of derivations of a crossed module and when determining all the cat1-structures on a group G (see sections \refAllDerivations and \refAllSections).
An endomorphism ε of R with image H′ is determined by
• a normal subgroup N of R and a permutation representation θ : R/N → Q of the quotient, giving a projection θ o ν : R → Q , where ν : R → R/N is the natural homomorphism;
• an automorphism α of Q;
• a subgroup H′ in a conjugacy class [H] of subgroups of R isomorphic to Q having representative H, an isomorphism φ : Q ≅ H, and a conjugating element c ∈ R such that Hc = H′, and takes values
ε r = (φ α θ ν r)c. |
Endomorphisms are placed in the same class if they have the same choice of N and [H], so the number of endomorphisms is
\|
\mathrmEnd(R)\|
= ∑classes
\|
\mathrmAut(Q)\|
\|
[H]\|
.
E = R.endomorphismClasses
and subfield
.classes
as shown below. Three cases are catered for as indicated
in the example.
gap> Ea4 := EndomorphismClasses( a4 , 7); Usage: EndomorphismClasses( G [, case] ); choose case = 1 to include automorphisms and zero, default case = 2 to exclude automorphisms and zero, case = 3 when N meet H is trivial, false gap> Ea4 := EndomorphismClasses( a4 ); rec( isDomain := true, isEndomorphismClasses := true, areNonTrivial := true, intersectionFree := false, classes := [ rec( quotient := a4.Q2, projection := OperationHomomorphism( a4, a4.Q2 ), autoGroup := Group( GroupHomomorphismByImages( a4.Q2, a4.Q2, [ (1,3,2) ], [ (1,2,3) ] ) ), rangeNumber := 3, isomorphism := GroupHomomorphismByImages( a4.Q2, a4.H3, [ (1,3,2) ], [ (2,3,4) ] ), conj := [ (), (1,3,2), (1,2)(3,4), (1,4,2) ] ) ], group := a4, latticeLength := 5, latticeReps := [ a4.id, a4.H2, a4.H3, a4.H4, a4 ] )
EndomorphismImages( G )
This returns the lists of images of the generators under the
endomorphisms, using the data in G.endomorphismClasses
. In this
example two trivial normal subgroups have been excluded. The
remaining normal subgroup of a4
is k4
, with quotient c3
and a4
has 8 elements of order 3 with which to generate a c3
, and hence 8
endomorphisms in this class.
gap> EndomorphismImages( a4 ); [ [ (2,3,4), (2,4,3) ], [ (2,4,3), (2,3,4) ], [ (1,2,4), (1,4,2) ], [ (1,4,2), (1,2,4) ], [ (1,4,3), (1,3,4) ], [ (1,3,4), (1,4,3) ], [ (1,3,2), (1,2,3) ], [ (1,2,3), (1,3,2) ] ]
IdempotentImages( G )
This return the images of idempotent endomorphisms. Various options are allowed.
gap> IdempotentImages( a4, 7 ); Usage: IdempotentImages( G [, case] ); where case = 1 for ALL idempotent images, case = 2 for all non-trivial images, case = 3 for case 2 and one group per conj class, case = 4 for case 3 and sorted into images. false gap> IdempotentImages( a4, 2 ); [ [ (2,4,3), (2,3,4) ], [ (1,4,2), (1,2,4) ], [ (1,3,4), (1,4,3) ], [ (1,2,3), (1,3,2) ] ] gap> IdempotentImages( a4, 3 ); [ [ (2,4,3), (2,3,4) ] ]
InnerAutomorphismGroup( G )
This creates the inner automorphism group of G
as the group
generated by the inner automorphisms by generators of G
. If a field
G.automorphismGroup
exists, it is specified as the parent of
Inn(G)
.
gap> inna4 := InnerAutomorphismGroup( a4 ); Inn(a4) gap> inna4.generators; [ InnerAutomorphism( a4, (1,2,3) ), InnerAutomorphism( a4, (2,3,4) ) ]
IsAutomorphismGroup( A )
This tests to see whether A
is a group of automorphisms.
gap> IsAutomorphismGroup( inna4 ); true
AutomorphismPair( A )
This returns a record pairA
containing a permutation group
isomorphic to the group A obtained using the OperationHomomorphism
function. The record contains A and pairA.auto
, P as
pairA.perm
. Isomorphisms in each direction are saved as pairA.p2a
and pairA.a2p
.
gap> ac3 := AutomorphismGroup( c3 ); Group( GroupHomomorphismByImages( c3, c3, [(1,2,3)], [(1,3,2)] ) ) gap> pairc3 := AutomorphismPair( ac3 ); rec( auto := Aut(c3), perm := PermAut(c3), a2p := OperationHomomorphism( Aut(c3), PermAut(c3) ), p2a := GroupHomomorphismByImages( PermAut(c3), Aut(c3), [(1,2)], [ GroupHomomorphismByImages( c3, c3, [(1,2,3)], [(1,3,2)] ) ] ), isAutomorphismPair := true ) gap> pc3 := pairc3.perm; PermAut(c3)
IsAutomorphismPair( pair )
This tests to see whether pair
is an (automorphism group, perm group) pair.
gap> IsAutomorphismPair( pairc3 ); true
AutomorphismPermGroup( G )
This combines AutomorphismGroup(G)
with the function
AutomorphismPair
and returns
G.automorphismGroup.automorphismPair.perm
. The name
PermAut(<G.name>)
is given automatically.
gap> P := AutomorphismPermGroup( a4 ); PermAut(a4) gap> P.generators; [ (1,8,4)(2,6,7), (3,6,7)(4,5,8), (1,2)(3,8)(4,7)(5,6) ]
FpPair( G )
When G is a finitely presented group, this function finds a faithful permutation representation P, which may be the regular representation, and sets up a pairing between G and P.
gap> f := FreeGroup( 2 );; gap> rels := [ f.1^3, f.2^3, (f.1*f.2)^2 ];; gap> g := f/rels;; gap> pairg := FpPair( g ); rec( perm := Group( (2,4,3), (1,3,2) ), fp := Group( f.1, f.2 ), f2p := GroupHomomorphismByImages( Group( f.1, f.2 ), Group( (2,4,3), (1,3,2) ), [ f.1, f.2 ], [ (2,4,3), (1,3,2) ] ), p2f := GroupHomomorphismByImages( Group( (2,4,3), (1,3,2) ), Group( f.1, f.2 ), [ (2,4,3), (1,3,2) ], [ f.1, f.2 ] ), isFpPair := true, isMinTransitivePair := true, generators := [ (2,4,3), (1,3,2) ], degree := 4, position := 3 )
When G is a permutation group, the function
PresentationViaCosetTable
is called to find a presentation for G
and hence a finitely presented group F isomorphic to G. When G
has a name, the name <name of G>Fp
is given automatically to F and
<name of G>Pair
to the pair.
gap> h20.generators; [ (1,2,3,4,5), (2,3,5,4) ] gap> pairh := FpPair( h20 ); rec( perm := h20, fp := h20Fp, f2p := GroupHomomorphismByImages( h20Fp, h20, [ f.1, f.2 ], [ (1,2,3,4,5), (2,3,5,4) ] ), p2f := GroupHomomorphismByImages( h20, h20Fp, [ (1,2,3,4,5), (2,3,5,4) ], [ f.1, f.2 ] ), isFpPair := true, degree := 5, presentation := << presentation with 2 gens and 3 rels of total length 14 >>, name := [ 'h', '2', '0', 'P', 'a', 'i', 'r' ] ) gap> pairh.fp.relators; [ f.2^4, f.1^5, f.1*f.2*f.1*f.2^-1*f.1 ]
IsFpPair( pair )
This tests to see whether pair
is an (Fp-group, perm group) pair.
gap> IsFpPair( pairh ); true
SemidirectPair( S )
When S
is a semidirect product, this function finds a faithful
permutation representation P and sets up a pairing between S and
P. The example illustrates c2|Xc3
≅s3
.
gap> agen := ac3.generators;; pgen := pc3.generators;; gap> a := GroupHomomorphismByImages( pc3, ac3, pgen, agen ); GroupHomomorphismByImages( PermAut(c3), Aut(c3), [ (1,2) ], [ GroupHomomorphismByImages( c3, c3, [ (1,2,3) ], [ (1,3,2) ] ) ] ) gap> G := SemidirectProduct( pc3, a, c3 );; gap> G.name := "G";; PG := SemidirectPair( G ); rec( perm := Perm(G), sdp := G, s2p := OperationHomomorphism( G, Perm(G) ), p2s := GroupHomomorphismByImages( Perm(G), G, [(1,2)(4,5), (3,5,4)], [ SemidirectProductElement( (1,2), GroupHomomorphismByImages ( c3, c3, [ (1,3,2) ], [ (1,2,3) ] ), () ), SemidirectProductElement( (), IdentityMapping(c3), (1,2,3) ) ] ))
IsSemidirectPair( pair )
This tests to see whether pair
is a (semidirect product, perm group) pair.
gap> IsSemidirectPair( PG ); true
PrintList( L )
This functions prints each of the elements of a list L
on a separate
line.
gap> J := [ [1,2,3], [3,4], [3,4], [1,2,4] ];; PrintList( J ); [ 1, 2, 3 ] [ 3, 4 ] [ 3, 4 ] [ 1, 2, 4 ]
80.147 DistinctRepresentatives
DistinctRepresentatives( L )
When L is a set of n subsets of [1..n] and the Hall condition is satisfied (the union of any k subsets has at least k elements), a standard algorithm for systems of distinct representatives is applied. (A backtrack algorithm would be more efficient.) If the elements of L are lists, they are converted to sets.
gap> DistinctRepresentatives( J ); [ 1, 3, 4, 2 ]
CommonRepresentatives( J, K )
When J
and K
are both lists of n
sets, the list L
is formed
where L[i] := { j : J[i] ∩ K[j] ≠ ∅ }. A
system of distinct represetatives reps
for L
provides a
permutation of the elements of K
such that J[i]
and K[i]
have
non-empty intersection. Taking the first element in each of these
intersections determines a system of common representatives com
.
The function returns the pair [ com, reps ]
. Note that there is no
requirement for the representatives to be distinct. See also the next
section.
gap> K := [ [3,4], [1,2], [2,3], [2,3,4] ];; gap> CommonRepresentatives( J, K ); [ [ 3, 3, 3, 1 ], [ 1, 3, 4, 2 ] ]
This has produced 3 ∈ J[1] ∩ K[1], 3 ∈ J[2] ∩ K[3], 3 ∈ J[3] ∩ K[4] and 1 ∈ J[4] ∩ K[2].
CommonTransversal( G, H )
The existence of a common transversal for the left and right cosets of a subgroup H of G is a special case of systems of common representatives.
gap> T := CommonTransversal( a4, c3 ); [ (), (1,3)(2,4), (1,2)(3,4), (1,4)(2,3) ]
IsCommonTransversal( G, H, T )
gap> IsCommonTransversal( a4, c3, T ); true
gap3-jm