gap-packages / wedderga

Wedderburn Decomposition of Group Algebras
https://gap-packages.github.io/wedderga/
GNU General Public License v2.0
3 stars 6 forks source link

Improve tests reproducibility to work with all three sets of packages loaded #34

Closed olexandr-konovalov closed 5 years ago

olexandr-konovalov commented 6 years ago

The next task is to look which examples look differently dependently on which GAP packages are loaded. At the moment, in the GAP master branch all test pass when GAP is started with default packages, but there are two other modes - with no packages and with all packages loaded , which show harmless diffs dues to randomised algorithms. For example,

########> Diff in /circa/scratch/gap-jenkins/workspace/GAP-master-test/GAPCOPT\
S/64build/GAPTARGET/packages/label/kovacs/GAP-master-snapshot/pkg/wedderga-4.9\
.0/tst/wedderga08.tst:32
# Input is:
QHK:=Image(epi,H);
# Expected output:
Group([ f1, f1 ])
# But found:
Group([ <identity> of ..., f1 ])
########

or

########> Diff in /circa/scratch/gap-jenkins/workspace/GAP-master-test/GAPCOPT\
S/64build/GAPTARGET/packages/label/kovacs/GAP-master-snapshot/pkg/wedderga-4.9\
.0/tst/wedderga02.tst:136
# Input is:
List(chi4,x->SimpleAlgebraByCharacterInfo(QG,x));
# Expected output:
[ [ 2, NF(8,[ 1, 3 ]), 8, [ 2, 3, 0 ] ], [ 2, NF(8,[ 1, 3 ]), 8, [ 2, 3, 0 ] ]\
, 
  [ 2, NF(8,[ 1, 3 ]), 8, [ 2, 3, 4 ] ], [ 2, NF(8,[ 1, 3 ]), 8, [ 2, 3, 4 ] ]\
 ]

# But found:
[ [ 2, NF(8,[ 1, 3 ]), 8, [ 2, 3, 0 ] ], [ 2, NF(8,[ 1, 3 ]), 8, [ 2, 3, 0 ] ]
    , [ 2, NF(8,[ 1, 3 ]), 8, [ 2, 3, 0 ] ], 
  [ 2, NF(8,[ 1, 3 ]), 8, [ 2, 3, 0 ] ] ]
########
olexandr-konovalov commented 6 years ago

Consider extending Jenkins tests to run them with other sets of packages loaded.

olexandr-konovalov commented 5 years ago

Tests are in a good state now.