Close

Règle d'automatisation Jira pour le merge d'une pull request

Posted by: AJ Welch

For a relational database like PostgreSQL, it could widely be considered a sin among developers not to include a primary key in every table. It is therefore crucial that you do your utmost to add that all-important primary key column to every table, and thankfully Postgres provides two methods for accomplishing this task.


Au-delà d'Agile


By far the simplest and most common technique for adding a primary key in Postgres is by using the SERIAL or BIGSERIAL data types when CREATING a new table. As indicated in the official documentationSERIAL is not a true data type, but is simply shorthand notation that tells Postgres to create a auto incremented, unique identifier for the specified column.

Below we’ll create our simple books table with an appropriate SERIAL data type for the primary key.

CREATE TABLE books (
  id              SERIAL PRIMARY KEY,
  title           VARCHAR(100) NOT NULL,
  primary_author  VARCHAR(100) NULL
);

By simply setting our id column as SERIAL with PRIMARY KEY attached, Postgres will handle all the complicated behind-the-scenes work and automatically increment our id column with a unique, primary key value for every INSERT.

Au-delà d'Agile


In some rare cases, the standard incremental nature built into the SERIAL and BIGSERIAL data types may not suit your needs. In these cases, you can perform the same auto incremented primary key functionality for your column by creating a custom SEQUENCE, similar to the method used in older version of Oracle.

Perhaps we’re particularly fond of even numbers but also have a strong distaste for anything smaller than 100, so we only want our primary key to be incremented by two starting at 100 for every insert. This can be accomplished with a custom SEQUENCE like so:

CREATE SEQUENCE books_sequence
  start 2
  increment 2;

Now when we INSERT a new record into our books table, we need to evaluate the the next value of our sequence with nextval('books_sequence') and use that as our id.

INSERT INTO books
  (id, title, primary_author)
VALUES
  (nextval('books_sequence'), 'The Hobbit', 'Tolkien');

SEQUENCES can be spiced up even more if desired, with options like minvalue and maxvalue to of course indicate extreme values, and even CYCLE, which allows the sequence to “loop around” once it reaches the maxvalue, returning back to the start value and beginning the climb all over again. Far more information can be found in the official documentation.


Thème suivant