#!/bin/bash

IFS=$'\n'
for s in $*; do
    REGEX="$s"
    REGEX=$(echo "$REGEX" | sed -r 's/\s+/\\s\*/g')
    
    # Word Doc
    for d in *doc; do
        if antiword "$d" | grep -iqsP "($REGEX)"; then
            echo "Possible match: $d"
            echo "$d" >> msword.tmp
        fi
    done
    
    # Word Docx
    for d in *docx; do
        if unzip -p "$d" word/document.xml | grep -iqsP "($REGEX)"; then
            echo "Possible match: $d"
            echo "$d" >> msword.tmp
        fi
    done

done

# Cleanup results
if [[ -f msword.tmp ]]; then
    sort -u msword.tmp >> msword-matches.txt
fi