Context Navigation

Changes between Initial Version and Version 1 of Ticket #1707, comment 31

-              initial
+              v1
 It doesn't 'break' the iterators - leaving in 'invalid' entries is intentional in order to get the speed boost. Once we start reusing entity handles, this solution will be perfect :)
+It doesn't 'break' the iterators per se - leaving in 'invalid' entries is intentional in order to get the speed boost. Once we start reusing entity handles, this solution would be quite perfect :)
+Only issue might be current simulation code that expects size = size - 1 after erase(), which doesn't happen right now. That is easy to fix. Just search for all references with your favorite IDE and all is well.
+When we iterate over the entitymap, we can do 2 things to fix it:
+) Write a custom iterator (easy) that skips handles with '-1' (invalid).
+) Let the code itself handle it: if(it->first == -1) continue;
+Another issue might be current simulation code that expects size = size - 1 after erase(), which doesn't happen right now. That is easy to fix. Just search for all references with your favorite IDE and all is well.
 I don't see why we can't use this solution though - it's pretty much the best we've got so far. I don't really care about wasting a bit of memory either. This vector never grows too big :)