Here we assign a cost according to how far keys are located away from each other. However, in the previous implementation, the matrix_position was actually used instead of the real distance (=`position). This PR fixes this issue.
Slightly off-topic:
This might be the place where the distance functions may have been useful. However, I wouldn't know where to apply them anyway, as there are two different calculations going on: (dx + dy - 3.0) and 1.0 + dy * dy.
Why do we exclude_rows: [3] in movement_pattern_same_row and trigram_rolls?
Here we assign a cost according to how far keys are located away from each other. However, in the previous implementation, the
matrix_positionwas actually used instead of the real distance (=`position). This PR fixes this issue.Slightly off-topic:
distancefunctions may have been useful. However, I wouldn't know where to apply them anyway, as there are two different calculations going on:(dx + dy - 3.0)and1.0 + dy * dy.exclude_rows: [3]inmovement_pattern_same_rowandtrigram_rolls?