Fixed attribute error when calling WreathProduct

[b-reinke]

I tried to reproduce the example session in chapter 2 of the FR documentation. The command

K := NormalClosure(Bm,Group(c));

produced

Error, no method found! For debugging hints type ?Recovery from NoMethodFound
Error, no 1st choice method found for 'WreathProduct' on 4 arguments called from
WreathProduct( data.group, Range( i ), InverseGeneralMapping( i ), l ) at /usr/lib/gap/pkg/fr-2.3.6/gap/group.gi:3159 called from
HOMOMORPHISMGERMPCPGROUP@fr( G, data, i ) at /usr/lib/gap/pkg/fr-2.3.6/gap/group.gi:3249 called from
func( C[i] ) at /usr/lib/gap/lib/coll.gi:746 called from
List( [ 1 .. n ], function ( i )
return HOMOMORPHISMGERMPCPGROUP@fr( G, data, i );
end ) at /usr/lib/gap/pkg/fr-2.3.6/gap/group.gi:3249 called from
d.pifunc( arg[1], d.level ) at /usr/lib/gap/pkg/fr-2.3.6/gap/group.gi:147 called from
... at line 1 of stdin
you can 'quit;' to quit to outer loop, or
you can 'return;' to continue

It seems that InverseGeneralMapping( i ) didn't have the attribute IsMapping set. The fix calls IsMapping explicitly before calling WreathProduct.

[laurentbartholdi]

I don't think the bug is in FR; consider the following:

gap> i1 := IsomorphismPcpGroup(Group(()));; i1 := InverseGeneralMapping(i1);
[  ] -> [  ]
gap> HasIsTotal(i1); HasIsSingleValued(i1);
false
false
gap> i2 := IsomorphismPcpGroup(Group((1,2)));; i2 := InverseGeneralMapping(i2);
[ g1 ] -> [ (1,2) ]
gap> HasIsTotal(i2); HasIsSingleValued(i2);
true
true

and I would prefer polycyclic / inversegeneralmapping to be consistent wrt the attributes of its arguments.

[b-reinke]

You are right, it was not a bug of fr, but indeed a bug in the main gap system. It turns out InvertGeneralMapping is consistent wrt. the attributes of its argument, but IsomorphismPcpGroup was not.