Error in IsomorphismGroups on two pc groups of order 6272

[fingolfin]

Running the following code

G := PcGroupCode(2664057732308087629446918084678644731357911420184737672, 6272);
H := PcGroupCode(849169324138369034371257796152864931485540637536648, 6272);
iso:=IsomorphismGroups(G,H);

with the master branch leads to this error (after several minutes):

Error, no method found! For debugging hints type ?Recovery from NoMethodFound
The 1st argument is 'fail' which might point to an earlier problem
Error, no 1st choice method found for `^' on 2 arguments at /home/r/mhorn/gap/lib/methsel2.g:241 called from
act( pnt, id ) at /home/r/mhorn/gap/lib/oprt.gd:434 called from
TestIdentityAction( acts, pnt, act ) at /home/r/mhorn/gap/lib/oprt.gd:839 called from
CallFuncList( StabilizerFunc, arg ) at /home/r/mhorn/gap/lib/oprt.gi:2832 called from
Stabilizer( Image( yet ), Set( List( i, function ( x )
          return Position( m, x );
      end ) ), OnSets ) at /home/r/mhorn/gap/lib/grppcaut.gi:1223 called from
SpaceAndOrbitStabilizer( M.dimension, M.field, spaces, sporb ) at /home/r/mhorn/gap/lib/grppcaut.gi:1336 called from
AutomorphismGroupSolvableGroup( G ) at /home/r/mhorn/gap/lib/morpheus.gi:1944 called from
AutomorphismGroup( d ) at /home/r/mhorn/gap/lib/autsr.gi:920 called from
PatheticIsomorphism( G, H ) at /home/r/mhorn/gap/lib/morpheus.gi:2060 called from
IsomorphismGroups( tmp[i], tmp[j] ) at resolve-isos-helpers.g:80 called from
...

[hulpke]

This is fixed in
b5d8018
in #968

[fingolfin]

Excellent, thanks.

I noticed that you added this as a test case there. I wonder how effective doing that is on the long run: As the heuristics change, this particular test case might not trigger the same bugs anymore (i.e. even without your fix, it's quite possible that in a year, a different code path is used to handle that case, and it wouldn't error out anymore). In addition, these examples are all very long, which makes it impractical to run them frequently. But that in turn means regressions won't be caught.

So, how about instead directly adding tests for SpaceAndOrbitStabilizer? I extracted this test case from this error:

ospaces := Z(2)^0 * [
[[1,0,0,0,0,0,0,0],[0,1,0,0,0,0,0,0],[0,0,1,0,0,0,0,0],[0,0,0,0,1,0,0,0],[0,0,0,0,0,1,0,0],[0,0,0,0,0,0,1,0]],
[[0,1,0,0,0,0,0,0],[0,0,1,0,0,0,0,0],[0,0,0,1,0,0,0,0],[0,0,0,0,0,1,0,0],[0,0,0,0,0,0,1,0],[0,0,0,0,0,0,0,1]],
[[1,0,0,0,0,0,0,0],[0,0,1,0,0,0,0,0],[0,0,0,0,1,0,1,0],[0,0,0,0,0,1,0,0]],
[[0,0,1,0,0,0,0,0],[0,0,0,1,0,0,0,0],[0,0,0,0,0,0,1,0],[0,0,0,0,0,0,0,1]],
[[1,0,0,0,0,0,0,0],[0,1,0,0,0,0,0,0],[0,0,0,0,1,0,0,0],[0,0,0,0,0,1,0,0]],
[[1,1,0,0,0,0,0,0],[0,0,0,0,1,0,0,0]],
[[1,0,0,0,0,0,0,0],[0,1,0,0,0,0,1,0],[0,0,1,0,0,0,0,0],[0,0,0,0,1,0,1,0],[0,0,0,0,0,1,0,0]],
[[0,1,0,0,0,1,0,0],[0,0,1,0,0,0,0,0],[0,0,0,1,0,0,0,0],[0,0,0,0,0,0,1,0],[0,0,0,0,0,0,0,1]],
[[1,0,0,0,0,1,0,0],[0,1,0,0,0,1,0,0],[0,0,0,0,1,0,0,0]],[[1,1,0,0,1,0,0,0]]
];
osporb := Z(2)^0 * [
[[[1,0,0,0,0,0,0,0],[0,1,0,0,0,0,0,0],[0,0,1,0,0,0,0,0]],[[0,0,0,0,1,0,0,0],[0,0,0,0,0,1,0,0],[0,0,0,0,0,0,1,0]]],
[[[0,1,0,0,0,0,0,0],[0,0,1,0,0,0,0,0],[0,0,0,1,0,0,0,0]],[[0,0,0,0,0,1,0,0],[0,0,0,0,0,0,1,0],[0,0,0,0,0,0,0,1]]],
[[[1,0,0,0,0,0,0,0],[0,0,1,0,0,0,0,0]],[[0,0,0,0,1,0,1,0],[0,0,0,0,0,1,0,0]]],
[[[0,0,1,0,0,0,0,0],[0,0,0,1,0,0,0,0]],[[0,0,0,0,0,0,1,0],[0,0,0,0,0,0,0,1]]],
[[[1,0,0,0,0,0,0,0],[0,1,0,0,0,0,0,0]],[[0,0,0,0,1,0,0,0],[0,0,0,0,0,1,0,0]]],
[[[0,0,0,0,1,0,0,0]],[[1,1,0,0,0,0,0,0]]]
];
SpaceAndOrbitStabilizer(8,GF(2),ospaces,osporb);

It runs in just 1.5 seconds.