hash tables
lots of ways to hold data
not all of those ways are actually efficient, as it turns out
if the books aren't sorted
O(n) linear time
if the books are sorted
O(log n) logarithmic time
look up a book instantly?
array
hash function
+
hash table
holds our data
decides where our data should live
hash functions create a mapping and assign data to an index in the
array
data
3
sets of pairs: (k, v)
k = index in the array
v = value that lives at that index
hash buckets
function bookHashing(bookTitle, hashTableSize) {
// Remove any spaces from book title.
var strippedBookTitle = bookTitle.replace(/\s/g, '')
// Divide the length of the title by the hash table size.
// Return the remainder.
return strippedBookTitle.length % hashTableSize;
}
bookHashing("The Grapes of Wrath", 12);
// 4
bookHashing("The Sound and the Fury", 12);
// 6
collisions 💣
two elements inserted into the same hash bucket
no hash function will return an unique hash bucket value for every item
the size of our dataset will usually be larger than the size of our hash table
uses the 1st letter of input data to generate index key
alphabetization: 26 buckets
phonebook, dictionary
equally distributed data amongst the hash buckets
hash functions should
- always return the same key
- avoid collisions
- be easy to compute
- avoid clustering of data
