Changes between Initial Version and Version 13 of Ticket #17252


Ignore:
Timestamp:
10/30/14 06:09:50 (5 years ago)
Author:
aschilling
Comment:

Looks good now!

Anne

Legend:

Unmodified
Added
Removed
Modified
  • Ticket #17252

    • Property Status changed from new to needs_review
    • Property Cc tscrim added
    • Property Priority changed from minor to major
    • Property Branch changed from to public/combinat/zabrocki/fixstrongtableaux/17252
    • Property Reviewers changed from to Anne Schilling
    • Property Keywords tableaux added
    • Property Commit changed from to 381acd835802d712e3a4a7ba547c7aeb2393b94a
  • Ticket #17252 – Description

    initial v13  
    88This is exactly what happens in this case because if you apply the sequence of transpositions t_{2,4} on t_{3,4} t_{4,5} T then in fact t_{2,4} does reduce the length of shape of the tableau by 1, but there are no cells of content 2 and 3 in the tableau t_{3,4} t_{4,5} T.
    99
    10 It looks like there are a couple of ways of catching this condition.  It might be sufficient to check that applying the transposition t_{2,4} is not valid (because it really should be applying t_{-2,0} instead).  That fix might or might not be sufficient (and a counterexample will be difficult to identify).  To be more thorough, I will verify that if t_{ij} decreases the length of the core by 1, then the labels of the cells in range(i,j) that are removed include exactly one negative label.
     10The fix was to implement the method `to_transposition_sequence` recursively by checking that the result of applying `t_{i,j}` can reduced to the empty core.