Shell Programming and Scripting

# read through the tagged document
while (<TOPIC_ORIG>) {
        if (/<title>(.+?)<\/title>/) {
                
                print TOPIC_TEMP "<title>";

                # file to store text between <title> tags prior to stemming
                $INFILE = ">$HOME/bin/stem.in";
                open INFILE or die;
                 
                # file to store output of stemmer
                $OUTFILE = ">$HOME/bin/stem.out";
                open OUTFILE or die;

                # store pre-processed text in stem.in
                print INFILE "$1";
                 
                 # run stemmer
                 system "$HOME/bin/porter-stemmer $HOME/bin/stem.in > $HOME/bin/stem.out";

                # retrieve stemmed text in output file, write to TOPIC_TEMP
                # between title tags.
                while (<OUTFILE>) {
                            print TOPIC_TEMP;
                } 

                # print end-tag
                print TOPIC_TEMP "</title>";

                # close temp files
                close INFILE;
                close OUTFILE;
        } else {
                 #output all other tags and contents
                print TOPIC_TEMP;

        }
}

close TOPIC_ORIG;
close TOPIC_TEMP;

<doc>
<title>some title</title>
..........
</doc>