# Soundex Algorithm in AWK: Michael Sanders 2024
# example usage: awk -f soundex.awk < words.txt
# see also: https://en.wikipedia.org/wiki/Soundex

{ print $0 " : " soundex($0) }

function soundex(word, i, code, c, firstLetter, lastCode, buf) {
word = toupper(word) # convert word to uppercase
firstLetter = substr(word, 1, 1)
code = buf = ""

# map of letters to soundex digits
for (i = 2; i <= length(word); i++) {
c = substr(word, i, 1)
if (c ~ /[BFPV]/) code = "1"
else if (c ~ /[CGJKQSXZ]/) code = "2"
else if (c ~ /[DT]/) code = "3"
else if (c ~ /[L]/) code = "4"
else if (c ~ /[MN]/) code = "5"
else if (c ~ /[R]/) code = "6"
else code = "" # skip A, E, I, O, U, H, W, Y

# ignore consecutive identical codes
if (code != lastCode && code != "") {
buf = buf code
lastCode = code
}
}

# combine 1st letter with buf, pad with zeros or truncate to 4 characters
return substr(firstLetter buf "000", 1, 4)
}

# eof

--
:wq
Mike Sanders

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)