package readPowerPoint;

use strict;
use Win32::OLE qw(in with);
use Win32::OLE::Const 'Microsoft PowerPoint';
use Win32::OLE::Const 'Microsoft Office';
use Data::Dumper;

require Exporter;
use vars qw(@ISA @EXPORT);

@ISA    = qw(Exporter);
@EXPORT = qw(readActivePPT);

sub readActivePPT
# this method returns the name of the active PPT, its content and a hash 
#  the PPT's properties we consider to be useful. Hence it is ready for use
{
   
   if ( my $powerpoint = Win32::OLE->GetActiveObject('Powerpoint.Application')) # connect to powerpoint application
   {
      if (defined (my $activePPT=$powerpoint->ActivePresentation))
      {
         my $PPTName = $activePPT->Name; # retrieve PPT name i.e. subject
         my $PPTProperties = $activePPT->{BuiltInDocumentProperties};

         my %PPTContent = (); # read PPT content - see below for structure details
         
         for my $slideNumber(1..$activePPT->{Slides}->Count)
         {
            my $total = $slideNumber / $activePPT->{Slides}->Count;
            print "\n---------------- Please wait (".100*$total."% done)----------------";          
            my $slide = $activePPT->slides($slideNumber);
            if ($slide->{Shapes}->Count)
            {
               for my $shapeNumber(1..$slide->{Shapes}->Count)
               {                  
                  if ( ($slide->shapes($shapeNumber)->HasTextFrame()==msoCTrue)
                     or($slide->shapes($shapeNumber)->HasTextFrame()==msoTrue))
                  {
		     my $wordGroup=$slide->shapes($shapeNumber)->TextFrame->{TextRange};
		     if (defined $wordGroup)
		     {
		        my $text = "";
		        foreach my $palabre (in $wordGroup->Words) # prints out every single word
		        { 
		           $text.= lc($palabre->{Text})." "; # lc stands for lower case
                        }
                        $PPTContent{("slide".$slideNumber)}{("shape".$shapeNumber)} = $text;
                     }
                  }                  
               }
            }
         }
         return ($PPTName,$PPTProperties,%PPTContent);
      }
   }
   return (undef,undef,undef);
}

# structure of $PPTContent will be a hash of hashes :
# first hash will contain all the slides and each slide will be a hash itself containing all
# the text areas found.
# to access do this $PPTContent{slide1}{shape1}