◄ prev contents next ►

⚠

The web version only has simple instructions since chapter 04, while the full book has detailed explanations and background info.

0204: CRUD

Add CRUD APIs

This step implements primary-key based CRUD (create, read, update, delete), matching insert, select, update, delete.

type DB struct {
    KV KV
}

func (db *DB) Open() error  { return db.KV.Open() }
func (db *DB) Close() error { return db.KV.Close() }

Use KV get, set, del to implement:

func (db *DB) Select(schema *Schema, row Row) (ok bool, err error)
func (db *DB) Insert(schema *Schema, row Row) (updated bool, err error)
func (db *DB) Upsert(schema *Schema, row Row) (updated bool, err error)
func (db *DB) Update(schema *Schema, row Row) (updated bool, err error)
func (db *DB) Delete(schema *Schema, row Row) (deleted bool, err error)

Requirements:

select, delete: input row contains the primary key.
insert, upsert, update: input row is complete.

For example, insert:

func (db *DB) Insert(schema *Schema, row Row) (updated bool, err error) {
    key := row.EncodeKey(schema)
    val := row.EncodeVal(schema)
    return db.KV.SetEx(key, val, ModeInsert)
}

For table create table t (a int64, b int64, primary key (b)), Select() takes a Row of length 2. Column 2 is the primary key. Column 1 is the value returned from KV.

Major Features of Relational Databases

Besides single-row access by primary key, later steps may add:

Full table scan.
Query by index.
Range query by index or primary key.
Filter query results.

ⓘ

CodeCrafters.io has similar courses in many programming languages, including build your own Redis, SQLite, Docker, etc. It’s worth checking out.

◄ prev contents next ►