A hashing function, I think! From http://stackoverflow.com/a/14366546/479554

hashing_function.R

library(digest)
set.seed(1)

(x <- sample(1e9, size=6))
# [1] 265508664 372123900 572853364 908207790 201681932 898389685

## To hash R's internal representation of these numbers
strtoi(substr(sapply(x, digest), 28, 32), 16L) %% 1e3
# [1] 552 511 233 293 607 819

## Or, for a hash mapping that's comparable to other programs' md5 hash 
## implementations
strtoi(substr(sapply(as.character(x), digest, serialize=FALSE),28,32),16L) %% 1e3
# [1] 153 180 892 294 267 807

someWords <- apply(matrix(sample(LETTERS, 50000, replace = TRUE), ncol = 10), 1, paste, collapse = "")
someWords <- sort(someWords)
table(table(someWords))
(hash1 <- strtoi(substr(sapply(someWords, digest), 28, 32), 16L) %% 1e2)
(hash2 <- strtoi(substr(sapply(as.character(someWords), digest, serialize=FALSE),28,32),16L) %% 1e2)
table(table(hash1))
table(table(hash2))


(y <- c("wub", "wub2", "wuc", "testing", "asdfkljwefmklemf,.msp"))
strtoi(substr(sapply(y, digest), 28, 32), 16L) %% 1e3
strtoi(substr(sapply(as.character(y), digest, serialize=FALSE),28,32),16L) %% 1e3