PostgreSQL's array data type Show archive.org snapshot is pretty useful, but manipulating values of arrays can be awkward because of its syntax.
Consider the following users table which each example below will start from:
| name | topics |
|---|---|
| Alice | {cats,dogs} |
| Bob | {llamas} |
(PostgreSQL uses curly braces to represent arrays, true story.)
Adding values
Use the array_cat function, or the || operator.
These calls will add the values "cats" and "mice" to users.topics:
UPDATE users SET topics = array_cat(topics, '{cats,mice}');
UPDATE users SET topics = topics || '{cats,mice}';
Note that this may produce duplicates.
| name | topics |
|---|---|
| Alice | {cats,dogs,cats,mice} |
| Bob | {llamas,cats,mice} |
Removing values
Use the array_remove function (Postgres 9.2+) to remove 1 value from the array (all its occurences, should it appear more than once):
UPDATE users SET topics = array_remove(topics, 'cats');
| name | topics |
|---|---|
| Alice | {dogs} |
| Bob | {llamas} |
Replacing values
Postgres 9.3+ offers array_replace. It will find 1 value (again all occurences) and replace it with 1 given other value.
UPDATE users SET topics = array_replace(topics, 'dogs', 'mice');
| name | topics |
|---|---|
| Alice | {cats,mice} |
| Bob | {llamas} |
Replacing multiple values
To replace multiple values, you need to combine what you learned above as array_replace won't do the trick. Here is one way:
UPDATE users
SET topics = array_remove(topics, 'cats') || '{mice,pandas}'
WHERE topics @> ARRAY['cats'];
Note the condition. It's important as we want to append "mice" and "pandas" only to those arrays that include "cats" in the first place.
| name | topics |
|---|---|
| Alice | {dogs,mice,pandas} |
| Bob | {llamas} |
Notes on working with String arrays in PostgreSQL
- Prefer the
'{one,two}'syntax. If you use theARRAY[]syntax, you'll need to add a type in order to perform operations with e.g. a string array column:ARRAY['one', 'two']::varchar[]. - Avoid working with plain strings, always wrap them as arrays when you are working with arrays. E.g. bad:
WHERE topics @> 'value', better:WHERE topics @> '{value}'
Check the PostgreSQL documentation on Array Functions and Operators for more information, and other useful functions.
