What methods do you use for collision handling in hash tables?

Hash tables typically use two main methods for collision handling:

1. Chaining — when a collision occurs, all elements with the same hash are stored in a linked list or another container within a single array cell. Searching or inserting involves traversing this list.

2. Open Addressing — when a collision occurs, the next free cell is sought according to a certain sequence (linear probing, quadratic probing, double hashing).

In Go, the standard map implementation uses chaining with optimizations. For example, in case of a collision, elements are stored in linked lists within buckets.

An example of chaining (simplified):

// Structure for an element
struct entry {
    key string
    value int
    next *entry
}

// Insertion with chaining
def insert(table []*entry, key string, value int) {
    index := hash(key) % len(table)
    e := &entry{key: key, value: value, next: table[index]}
    table[index] = e
}

Thus, collision handling is a way to maintain correctness and efficiency of the hash table when hashes collide.