Package io.datarouter.model.key.primary

Interface PrimaryKey<PK extends PrimaryKey<PK>>

All Superinterfaces:
Comparable<FieldSet<PK>>, Fielder<PK>, FieldSet<PK>, Key<PK>, PrimaryKeyFielder<PK>, UniqueKey<PK>
All Known Subinterfaces:
EntityPrimaryKey<EK,​PK>, FieldlessIndexEntryPrimaryKey<IK,​PK,​D>, RegularPrimaryKey<PK>
All Known Implementing Classes:
BaseEntityPrimaryKey, BasePrimaryKey, BaseRegularPrimaryKey
public interface PrimaryKey<PK extends PrimaryKey<PK>>
extends UniqueKey<PK>, PrimaryKeyFielder<PK>
A primary key is an ordered set of fields that uniquely identify a Databean among others of the same type. It corresponds to MySQL's primary key or a Memcached key. The PrimaryKey defines the hashCode(), equals() and compareTo() methods that distinguish databeans. The user should generally not override these methods. They are based on the PK fields, in order, and allow java collections to mimic the behavior of the underlying datastore. For example, adding a Databean to a TreeSet will insert it in the same order that MySQL inserts it into the database. If the Databean already exists in the set, then calling treeSet.put(databean) will overwrite the existing Databean in the set, similar to updating the Databean in the database. Having a strongly-typed PrimaryKey defined for each table makes it easier to construct compound primary keys. PK fields and their ordering is an important design decision for a large database table, and the PK class aims to support compound primary keys without adding more work down the line for comparison, SQL generation, etc. To keep the application portable, avoid relying on the underlying datastore's automatic id generation. Unless you need incrementing primary keys, it's usually more flexible to generate random IDs or rely on dedicated ID generator.