Oh venerable monks,

being merely a self-taught and clueless linguist trying to get perl to work with my text corpus, I have come across the following problem: My corpus consists of annotated text, where each line of text is followed by a line of part-of-speech terms, each word in the text line corresponding to exactly one term in the part-of-speech line.

Eventually, I would like to be able to search for a given item in a text line, followed by an item which is glossed as, say "n" in the pos-line.

So far, I have created two two-dimensional arrays: each text line is an array of words and the totality of the text lines is a superordinate array; thus, each word is assigned the number of its line and its number within the line as in $word[linenumber][wordnumber]. And the same goes for the pos elements.

The trouble is that I don't know how to articulate a condition like "If $word of an arbitrary $linenumber and $wordnumber matches "x" and $pos of THE SAME $linenumber and $wordnumber matches "y" then print $line of $linenumber."

My own humble attempts have gone no further than this:

open(my $in, "<", "Texts.txt") or die "impossible: $!";

#$morph=0;

LINE: while (<$in>){

    
    if(/\\tx/){
    push (@tx, $_);    
    push @txtoken, [split];
    }

    if(/\\ps/){
    push (@ps, $_);    
    push @pstoken, [split];
    }
    }
            foreach($txtoken[$j][$i] = "temeli" and $pstoken[$j][$i]= 
+"n"){                    
            print "$tx[$j]\n" ;
            }
}
[download]