simdir

#!/bin/bash
shingle_size=4
feature_count=1024
if [[ ! -d $1 || ! -d $2 ]]; then
  echo "Usage: <dir1> <dir2>"
  exit 1
fi

simhash -f $feature_count -s $shingle_size -w $1/*.class
simhash -f $feature_count -s $shingle_size -w $2/*.class

for A in "$1"/*.sim; do
  match=""
  maxresult=0
  for B in "$2"/*.sim; do
    if [[ $A != $B ]]; then
      result=$(simhash -c $A $B)
      if (( result > maxresult )); then
        maxresult=$result
        match=$B
      fi
    fi
  done
  echo "${A#.sim} --> ${match#.sim} ($maxresult)"
done