Skip to contents

Hash composite join keys to 32-bit integers (stable & type-aware)

hash_keys32.Rd

Produces a deterministic 32-bit integer per row from one or more key columns. The hash respects column types, canonicalizes -0 to +0, normalizes strings to UTF-8, and maps all NA/NaN variants to a single value. Factors are hashed by their level strings so that factor/character equality works as expected.

Usage

hash_keys32(x, by, normalize_strings = TRUE)

Arguments

x

A data.frame or tibble containing the key columns.

by

Character vector of column names forming the key, in order.

normalize_strings

Logical; if TRUE (default) strings are normalized to UTF-8 before hashing.

Value

An integer vector of length nrow(x) containing deterministic 32-bit hashes. Values may be negative.

Details

Prior to hashing, POSIXlt columns are converted to POSIXct. When the optional bit64 package is installed, integer64 columns are converted to their exact decimal string representations so they hash losslessly. Without bit64, integer64 inputs fall back to as.character() with the same equality semantics.

Examples

if (interactive()) {
  df <- data.frame(
    id = c(1L, 2L, NA_integer_),
    ts = as.POSIXct("2020-01-01", tz = "UTC") + 0:2,
    value = c("a", "b", NA_character_),
    stringsAsFactors = FALSE
  )
  hash_keys32(df, by = c("id", "ts", "value"))
}