pwn2maps.awk

# https://github.com/ekaf/ski/pwn2maps.awk
# (c)2017 Eric Kafe, CC BY 4.0, https://creativecommons.org/licenses/by/4.0/

# Produce synset mappings between all Princeton WordNet versions 

$0!~"^\\#"{
# For each synset key, split its set of synset ids:
  z=split($2,a,",")
# Map from each synset id
  for(i=1;i<z;i++){
    v1=a[i]
    sub(":.*","",v1)
# map to each subsequent WordNet version:
    for(j=i+1;j<=z;j++){
      v2=a[j]
      sub(":.*","",v2)
# simple mapping strategy:
# TODO: alternative mapping strategies, f. ex. select most frequent target
      print a[i],a[j] | "sort -u > " db "-map-pwn" v1 "-pwn" v2 ".txt"
# Optionally, print the inverse mapping:
#      print a[j],a[i] | "sort -u > " db "-map-pwn" v2 "-pwn" v1 ".txt"
    }
  }
}