Hello everybody, I think that I may have uncovered a bug in the weight distributions (spectrum) code for binary codes, though it may just be me misusing the function.
Problem: When a binary code is constructed using a matrix that doesn't have full rank then the spectrum calculation is incorrect. (maybe other computation are incorrect also) Example: (sage 4.7) sage: F Finite Field of size 2 sage: G [0 0 0 1 1] [1 0 1 0 1] [1 0 1 1 0] [1 1 1 0 0] sage: G.rank() 3 sage: C = LinearCode(G) sage: C.weight_distribution() [2, 0, 6, 6, 0, 2] sage: C.list() [(0, 0, 0, 0, 0), (1, 0, 1, 0, 1), (0, 1, 0, 0, 1), (1, 1, 1, 0, 0), (0, 0, 0, 1, 1), (1, 0, 1, 1, 0), (0, 1, 0, 1, 0), (1, 1, 1, 1, 1)] The assumption that the generating matrix is of full rank seems to lie in binary_code.pyx on line: 113 in Sage 4.7. binary_code.pyx: L113: cdef int i,j,k, dim=M.nrows(), deg=M.ncols() where the dimension is calculated as the number of rows. I don't know enough about the algorithm to submit a patch (if this is determined to actually be a bug) but if somebody can point me to some literature on this I could fix this. If G is assumed to be of full rank then maybe a warning can be put into either the documentation or printed out on the display. Thanks for making such a nice system. David Monarres 2 -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
