Repository navigation
Hashtables
Billy Bunn edited this page Apr 25, 2019
·
3 revisions
For the purposes of hashtables, a hash is a unique identifying number that's used to look up specific data. This is done in key value pairs in a hashtable, where each value's key is a unique hash. It's a numeric representation of some form of data.
A table is a collection of data (usually with columns and rows).
A hash table is really just an array. And in this array, each index contains a bucket. Each bucket holds values.
Buckets are key value pairs. The key of each bucket is a hash, and the value can be anything.
All wiki pages are summaries of the work of others. See the top of each page for the source blogs/articles/books and their authors. All other content by Billy Bunn.