[GAP Forum] DESIGN 1.3
Leonard Soicher
l.h.soicher at qmul.ac.uk
Fri Nov 10 17:00:29 GMT 2006
Dear All,
I am pleased to announce the release of version 1.3 of the DESIGN
package for GAP. This release fixes two small bugs in the Design 1.2
major upgrade (see below), and includes a new function `WittDesign'
for constructing the famous designs of Witt related to the Mathieu groups.
DESIGN 1.3 is available from both
http://designtheory.org/software/gap_design/ and
http://www.gap-system.org/Packages/design.html .
Best wishes,
Leonard
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Main changes from DESIGN 1.2 to DESIGN 1.3
------------------------------------------
(1) Added function: WittDesign.
(2) Fixed a bug which caused a break loop to be entered in certain cases
in function BlockDesigns when the least element of blockSizes was equal
to tSubsetStructure.t.
(3) Fixed a rarely occurring bug which caused no designs to be returned
in function BlockDesigns in certain cases when the designs had just one
block each.
(4) Small improvements made to the documentation and the code.
Main changes from DESIGN 1.1 to DESIGN 1.2
------------------------------------------
(1) Added functions:
BlockIntersectionPolynomial
BlockIntersectionPolynomialCheck
TDesignLambdas
TDesignLambdaMin
TDesignIntersectionTriangle
SteinerSystemIntersectionTriangle
TDesignBlockMultiplicityBound
ResolvableTDesignBlockMultiplicityBound
BlockDesignPoints
NrBlockDesignPoints
BlockDesignBlocks
NrBlockDesignBlocks
PairwiseBalancedLambda
AffineResolvableMu
DeletedPointsBlockDesign
DeletedBlocksBlockDesign
AddedPointBlockDesign
AddedBlocksBlockDesign
DerivedBlockDesign
ResidualBlockDesign
(2) The function BlockDesign can now (optionally) construct a block
design from base blocks and a group of automorphisms, and the list
of blocks given to BlockDesign need no longer be sorted.
(3) Improved documentation of the function TDesignFromTBD, and changed
the parameters (in a backward compatible way) to match the description
of the "*-construction" in the McSorley-Soicher paper.
(4) The (undocumented) resolveSimple option removed from functions
BlockDesigns and PartitionsIntoBlockDesigns, and the option b added
to these functions (to specify the number of blocks in a required block
design or partition part).
(5) Improvements made to the use of store (and time?) by the function
BlockDesigns.
More information about the Forum
mailing list