Perl code: Songlations post formatting

#!/pkg/bin/perl -w
# Formatting script for songlations.livejournal.com posts.
# Place the raw song text into song_input.txt
# Run program. Edited text will be in song_output.txt
# Author: Cairaguas Gonzalez <cairaguas@gmail.com>
# Last modified: November 15, 2016
 
use warnings; #this replaces -w in Perl 5.6.x and above
 
use Win32::Console::ANSI
    # http://stackoverflow.com/questions/700187/unicode-utf-ascii-ansi-format-differences
 
print "\nSONG_TRANSFORM by Cairaguas: \nIf WordPad.exe is not available, choose NO for all prompts to \nopen files and instead manually open and edit the text files \nin the folder with the Perl script.\n\n";
print "Open song_input.txt now? (y=1/n=0)\n";
$opennow = <>;
 
######################################################################
# INITIAL SONG INPUT AND DECISION OF TRANSFORMATION TYPE
 
    if ($opennow == 1)
        {
        system("start wordpad.exe song_input.txt"); # This opens the input file in the foreground using WordPad.
        print "Opening... \n";
        }
 
    else
        { print "Continuing with existing song_input.txt.\n";   
        }
         
print "Input song text, save, and close WordPad.
    \nType 1 for simple transformation (just capitalization).\nType 3 for Songlations format transformation.\nType 5 for changing special characters to HTML code.\n"; 
     
    $continue = <>; #This requests input from user.
                    #Press ENTER when input is finished.
 
open (IN, "<:encoding(Windows-1252)", "song_input.txt") or die("Could not open file song_input.txt from Perl folder.");  #open file for reading
open (OUT, ">:encoding(Windows-1252)", "song_output.txt") or die("Could not open song_output.txt from Perl folder.");        #open file for writing
 
######################################################################
# SIMPLE TRANSFORMATION
 
if ($continue == 1) 
{
    while ($line = <IN>)  
    {
        chomp $line; #removes line break at end of string
             
        if ($line =~ m/^http/) 
            {print OUT "$line\n";}      
            # do not capitalize url
            # ^ is an anchor character, requires match to be at the beginning
             
        elsif (($line =~ m/^[^a-zA-Z]/) && ($line !~ m/^\<[ibu]\>/))  
            # first char not letter character, not tag
            # could also use \W to match a non-word character
            {
            $first = substr($line,0,1); # first character
            $rest = substr($line,1);    # rest of the string
            print OUT "$first\u$rest\n";
            }
             
        elsif ($line =~ m/^\<[ibu]\>[^a-zA-Z]/)       
            # if line begins with tag preceding non-letter character
            # not currently working for quotation marks and not sure why
            {
            $first = substr($line,0,4);     # grab the tag and next character
            $rest = substr($line,4);        # grab the rest
            print OUT "$first\u$rest\n";        
            }   
         
        elsif ($line =~ m/^\<[ibu]\>[a-zA-Z]/)        
            # if line begins with italics or bold or underline tag
            {
            $first = substr($line,0,3);     # grab the <i> or <b> or <u> tag
            $rest = substr($line,3);        # grab the rest
            print OUT "$first\u$rest\n";        
            }
             
        else
            {print OUT "\u$line\n";}    # \u makes next letter uppercase
    }
}
 
######################################################################
# COMPLEX TRANSFORMATION
     
elsif ($continue == 3)
{
    print OUT ", English translation of lyrics \n\n";                           
    print OUT "<b>\"\"<\/b> \n";
    print OUT "Album: <i><\/i> \n";
    print OUT "Style: \n";
    print OUT "Country:  \n\n";
     
    print OUT "<b><u>Listen:<\/u><\/b>\r\n\n\n";
     
    print OUT "<lj-spoiler text=\"Expand embedded video\"><\/lj-spoiler>\n\n";
 
    print OUT "<b><u>Translation:<\/u><\/b> \n\n \n\n";
     
    while ($line = <IN>)  
    {
        if (($prevline !~ m/[a-zA-Z]/) && ($line !~ m/(Chorus|chorus)/) && ($prevline !~ m/^(---)/)) 
            #If the previous line was empty, the current line isn't a chorus line, or following a dash
            #This prevents double <i> after a dash line 
            { print OUT "<i>";
            }
 
        if (($line !~ m/[a-zA-Z]/) && ($prevline =~ m/[a-zA-Z]/))   #If the line was empty.
            { print OUT "</i>\n$line";
            #to do: rewrite so it doesn't do this before Chorus line
            }
 
         
        else    #the current line isn't a line break
        {
            if ($prevline =~ m/[a-zA-Z]/)
                {print OUT "\n";}
                #adds line break as long as it's not after the <i> added above
                 
            chomp($line);
         
            if ($line =~ m/^(http|---)/) 
                {print OUT "\n$line\n";}
                # do not capitalize url; does not add italics to dash lines
                # ^ is an anchor character, requires match to be at the beginning
                 
            elsif ($line =~ m/^(Chorus|chorus)/)
                {
                    print OUT "\n<b>Chorus<\/b>\n<i>";
                }
                # encloses "chorus" line in bold tags and begins italics tag
                # warning: removes anything else on the same line
                 
            elsif (($line =~ m/^[^a-zA-Z]/) && ($line !~ m/^\<[ibu]\>/) && ($line !~ m/^(Chorus|chorus)/))    
                # first char not letter character, not tag
                # could also use \W to match a non-word character
                {
                $first = substr($line,0,1); # first character
                $rest = substr($line,1);    # rest of the string
                print OUT "$first\u$rest";
                }
                 
            elsif ($line =~ m/^\<[ibu]\>[^a-zA-Z]/)       
                # if line begins with tag preceding non-letter character
                # not currently working for quotation marks and not sure why
                {
                $first = substr($line,0,4);     # grab the tag and next character
                $rest = substr($line,4);        # grab the rest of the line
                print OUT "$first\u$rest";      
                }   
             
            elsif ($line =~ m/^\<[ibu]\>[a-zA-Z]/)        
                # if line begins with italics or bold or underline tag
                {
                $first = substr($line,0,3);     # grab the <i> or <b> or <u> tag
                $rest = substr($line,3);        # grab the rest
                print OUT "$first\u$rest";      
                }
                                 
            else
                {print OUT "\u$line";}      # \u makes next letter uppercase
             
        }
             
        $prevline = $line; #still inside the while loop
             
        last if eof(IN); #breaks the while loop
         
    }   #ends while loop
 
print OUT "<\/i>\n\n" . "<b><u>Translation Notes:<\/u><\/b>\n\n<\/lj-cut>";
     
} #ends complex transformation script block
 
######################################################################  
# ACCENT AND SPECIAL CHARACTER TO HTML CODE 
 
elsif ($continue == 5)
{
    while ($line = <IN>)  
    {
 
    $line =~ s/\xE1/\á\;/g;
    $line =~ s/\xE9/\é\;/g;
    $line =~ s/\xED/\í\;/g;
    $line =~ s/\xF3/\ó\;/g;
    $line =~ s/\xFA/\ú\;/g;
    $line =~ s/\xF1/\ñ\;/g;
    $line =~ s/\xA1/\¡\;/g;
     
    #see: https://en.wikipedia.org/wiki/Windows-1252
 
    print OUT $line;
    # last if eof(<IN>);
    }
}
 
else {print OUT "else was used 456"};
 
######################################################################
# FINISHED TRANSFORMATIONS  
     
print "\nTransforming...\nDone. \nTransformed text is located in song_output.txt. \n\n";
 
close (IN);
close (OUT);
 
print "Open song_output.txt now? (y=1/n=0).\n";
$opennow = <>;
    if ($opennow == 1)
        {
        system("start wordpad.exe song_output.txt");
        # This opens the ouput file in the foreground using WordPad.
        }
    else
        {
        print "Goodbye.\n";
        }
         
# http://perldoc.perl.org/perlre.html#Regular-Expressions
# http://www.pjb.com.au/comp/diacritics.html
# http://users.cs.cf.ac.uk/Dave.Marshall/PERL/node79.html
# http://www.w3schools.com/charsets/ref_utf_dingbats.asp
# different encodings: http://htmlpurifier.org/docs/enduser-utf8.html