PM-Master/policy-machine-go

improve prohibition decision algorithm

Opened this issue · 0 comments

joshua-roberts commented

Problem

  • The prohibitions decision algorithm tracks every target node visited, which could be a lot of nodes for big graphs.

Solution

  • Retrieve prohibitions for the user BEFORE processing the target DAG, and store the targets of those prohibitions. During traversal, store the visited node ONLY if it is in the prohibitions target list.