1
0
mirror of https://github.com/jbranchaud/til synced 2026-01-03 07:08:01 +00:00

Add Postgre Does Not Support Unsigned Integers as a PostgreSQL TIL

This commit is contained in:
jbranchaud
2024-10-18 17:43:46 -05:00
parent f4ba6a9ef7
commit fb153f35bf
2 changed files with 32 additions and 1 deletions

View File

@@ -10,7 +10,7 @@ pairing with smart people at Hashrocket.
For a steady stream of TILs, [sign up for my newsletter](https://crafty-builder-6996.ck.page/e169c61186). For a steady stream of TILs, [sign up for my newsletter](https://crafty-builder-6996.ck.page/e169c61186).
_1473 TILs and counting..._ _1474 TILs and counting..._
--- ---
@@ -796,6 +796,7 @@ _1473 TILs and counting..._
- [Max Identifier Length Is 63 Bytes](postgres/max-identifier-length-is-63-bytes.md) - [Max Identifier Length Is 63 Bytes](postgres/max-identifier-length-is-63-bytes.md)
- [Open Heroku Database In Postico From Terminal](postgres/open-heroku-database-in-postico-from-terminal.md) - [Open Heroku Database In Postico From Terminal](postgres/open-heroku-database-in-postico-from-terminal.md)
- [pg Prefix Is Reserved For System Schemas](postgres/pg-prefix-is-reserved-for-system-schemas.md) - [pg Prefix Is Reserved For System Schemas](postgres/pg-prefix-is-reserved-for-system-schemas.md)
- [Postgres Does Not Support Unsigned Integers](postgres/postgres-does-not-support-unsigned-integers.md)
- [Prepare, Execute, And Deallocate Statements](postgres/prepare-execute-and-deallocate-statements.md) - [Prepare, Execute, And Deallocate Statements](postgres/prepare-execute-and-deallocate-statements.md)
- [Pretty Print Data Sizes](postgres/pretty-print-data-sizes.md) - [Pretty Print Data Sizes](postgres/pretty-print-data-sizes.md)
- [Pretty Printing JSONB Rows](postgres/pretty-printing-jsonb-rows.md) - [Pretty Printing JSONB Rows](postgres/pretty-printing-jsonb-rows.md)

View File

@@ -0,0 +1,30 @@
# Postgres Does Not Support Unsigned Integers
PostgreSQL has a variety of sizes of integer types, from `smallint` (2 bytes)
to `integer` (4 bytes) to `bigint` (8 bytes), as well as [other numeric
types](https://www.postgresql.org/docs/current/datatype-numeric.html).
It does _not_ however support unsigned versions of these numeric types.
That means, with an `integer` for instance, we can store numbers between
`-2147483648` and `+2147483647`. That's everything that can fit into 4 bytes.
In a system that supported 4 byte unsigned integers we'd be able to represent
from `0` all the way up to `4294967295`.
In PostgreSQL, we're limited to these _signed_ numeric types.
That means if we were hoping that the data type could essentially enforce a
non-negative restriction on the data in one of our columns, we're going to have
to be more creative. The obvious choice to me is to consider adding a [check
constraint](https://www.postgresql.org/docs/current/ddl-constraints.html#DDL-CONSTRAINTS-CHECK-CONSTRAINTS)
(e.g. `quantity integer check (quantity > 0)`).
Another option, as pointed out by [this StackOverflow
answer](https://stackoverflow.com/a/31833279/535590), is to create [a
user-defined _domain
type_](https://www.postgresql.org/docs/current/domains.html) that restricts
valid values. To me, the ergonomics of using a domain type are a bit awkward
and not worth the effort.
With either of these solutions, we are only approximating an unsigned integer
and do not actually have the same range of values available.