Contents

• floatvec
  • Author

floatvec

This Postgres extension provides basic arithmetic functions for operating on arrays (treated as vectors). It supports addition, subtraction, multiplication, division, and raising to a power. Everything (so far) is element-by-element, so no dot products or other matrixy stuff. Every function lets you pass either two arrays of the same length, or an array and a scalar, or a scalar and an array.

Despite the name, we support arguments of any numeric type (SMALLINT, INTEGER, BIGINT, REAL, and FLOAT aka DOUBLE PRECISION). The two arguments must be of the same type, and the result will be of that type too.

In general if an array argument is NULL, then the whole result will be NULL. If an array element is NULL, then the corresponding result element will be NULL. If a scalar argument is NULL, then you’ll get an array of the same length as the input, but all elements will be NULL.

In theory we could define custom operators so instead of vec_add(a, b) you could say a + b, but I’m reluctant to risk conflicts with the widely-used intarray extension. If you really want this feature let me know.

If you like this extension you might also like aggs_for_vecs and aggs_for_arrays, which support more stats-focused activity.

The functions are:

anyarray vec_add(anyarray, anyarray)

anyarray vec_add(anyarray, anyelement)

anyarray vec_add(anyelement, anyarray)

anyarray vec_sub(anyarray, anyarray)

anyarray vec_sub(anyarray, anyelement)

anyarray vec_sub(anyelement, anyarray)

anyarray vec_mul(anyarray, anyarray)

anyarray vec_mul(anyarray, anyelement)

anyarray vec_mul(anyelement, anyarray)

anyarray vec_div(anyarray, anyarray)

anyarray vec_div(anyarray, anyelement)

anyarray vec_div(anyelement, anyarray)

anyarray vec_pow(anyarray, anyarray)

anyarray vec_pow(anyarray, anyelement)

anyarray vec_pow(anyelement, anyarray)

Author

Copyright 2017 Paul A. Jungwirth