The thing is, rather than looking for a solution to this problem (storage of packet structures or pointers thereof with constant time access by id), you seem to have a solution (hash table) that's looking for a problem.Originally Posted by anya
A solution like uthash is designed for cases where you need a hash table. You don't need a hash table.
Look up a C++ Reference and learn How To Ask Questions The Smart Way
Actually, a more detailed description of the problem would be more enlightening for us. Are you really supposed to hash the packet.id, or are you supposed to hash the packet.buffer?
All the discussion so far, assumed that the key to be hashed is the packet.id... is that so?
Well, the requirement stated in post #1 does say: "search the packet in hash table by id and to delete it"
Look up a C++ Reference and learn How To Ask Questions The Smart Way
Yes, that's why I also assumed that the id is the hash key. But I also noticed that anya has marked the buffer as "value" in his/her original post. At first I thought it could be due to anya's inadequate understanding of hash-tables, but it was itching me if he/she is indeed supposed to use the buffer as the hash key.
Using the buffer for hasing and then using the id for retrieving, would be a bit odd and it would complicate a little bit the code, but I thought I should ask anyway.
