count() — PostgreSQL aggregate aggregate function

count(): number of input rows for which the input expression is not null. PostgreSQL aggregate functions — signature, volatility, version applicability and an illustrative example.

The Sev-1 Admin
The Sev-1 Admin  ·  2 min read

Summary

count is a PostgreSQL built-in aggregate function in the Aggregate Functions group. PostgreSQL’s system catalog (pg_proc) describes it as: “number of input rows for which the input expression is not null”.

Signature

count has 2 documented overloaded forms:

count(any) → bigint
count() → bigint

Argument and return types are taken from the pg_proc catalog; internal type names are shown using their readable SQL spellings (for example int4 is shown as integer). (Derived from the catalog — see the linked reference for the canonical documentation.)

Classification

  • Category: Aggregate Functions
  • Kind: Aggregate function
  • Volatility: IMMUTABLE — Marked IMMUTABLE — it always returns the same result for the same arguments and can be used in indexes and other contexts that require immutability.
  • Returns: bigint

Example

Illustrative form (replace placeholder values with your own data):

SELECT count(col) FROM your_table;

The example above is illustrative and is meant to show calling syntax only; consult the linked PostgreSQL documentation for exact semantics, edge cases and accepted argument combinations.

Version applicability

count is present across the surveyed releases (PostgreSQL 15, 16, 17, 18, 19). On older major versions, behaviour may differ in detail — always check the documentation for the version you run.

Related & references

Reference: PostgreSQL documentation — Aggregate Functions.