dmeoli / PBeFS / 40430288617

push

github

BucketOpenList keeps its arrays by g only while every g is below DenseG = 256, and at the first larger g makes every layer a binary heap by g and by the order of the pushes, so that the pops come in the same order and a layer costs its entries and not the range of g: on the grids, where g grows to thousands, a worker of HDA* rebuilt the array of its least layer each time it emptied and filled again, which took most of the time of the search; the case open checks both OPEN lists against an ordered set, with the least f moving back and g below and above DenseG, and the page of best-first search tells why

41 of 41 new or added lines in 1 file covered. (100.0%)

3601 of 3782 relevant lines covered (95.21%)

16821353.28 hits per line