Re: thought on Lua's hash-table invariants

[Norman Ramsey <nr@...>]

 > I spent a bit of time worrying that in use cases where many keys are
 > inserted and removed from a table while the size of the table remains
 > more or less the same, like the hash table attached to a heap that
 > supports update-key in A* or Dijkstra or an LRU cache, the performance
 > may degrade. My mind was put at ease when Roberto shared that all this
 > won't matter anyway, because for use cases like these the table will
 > be regularly rehashed even if it does not change size.

What triggers the rehash?  Is it the movement of the `lastfree` pointer?


Norman

_______________________________________________
lua-l mailing list -- lua-l@lists.lua.org
To unsubscribe send an email to lua-l-leave@lists.lua.org