# For licensing info read LICENSE file in the TWiki root.
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details, published at 
# http://www.gnu.org/copyleft/gpl.html


package TWiki::Contrib::SearchEngineKinoSearchAddOn::StringifyPlugins::Tika;
use base 'TWiki::Contrib::SearchEngineKinoSearchAddOn::StringifyBase';
use TWiki::Contrib::SearchEngineKinoSearchAddOn::Stringifier;
use File::Temp qw/tmpnam/;

# use full pathname if tika2txt not in cron's path
# Only if program exists, I register myself for various file types.
if (__PACKAGE__->_programExists("tika2txt")){
    __PACKAGE__->register_handler("application/word",  ".doc");
    __PACKAGE__->register_handler("text/docx",         ".docx");
    __PACKAGE__->register_handler("text/html",         ".html");
    __PACKAGE__->register_handler("application/pdf",   ".pdf");
    __PACKAGE__->register_handler("text/ppt",          ".ppt");
    __PACKAGE__->register_handler("text/pptx",         ".pptx");
    __PACKAGE__->register_handler("application/excel", ".xls");
    __PACKAGE__->register_handler("application/vnd.openxmlformats-officedocument.spreadsheet.sheet", ".xlsx");
}

sub stringForFile {
    my ($self, $filename) = @_;
    my $tmp_file = tmpnam();
    
    # use full path - cron may not have /usr/local/bin in path
    my $cmd = "/usr/local/bin/tika2txt '$filename' > $tmp_file 2>/dev/null";
    return "" unless ((system($cmd) == 0) && (-f $tmp_file));

    # use the text stringifier to rework encoding (from utf to iso-8859-15)
    # tika2txt handles this, so don't repeat
#    my $text = TWiki::Contrib::SearchEngineKinoSearchAddOn::Stringifier->stringFor($tmp_file);

    my $in;
    open $in, $tmp_file;
    my $text = join(" ", <$in>);
    close($in);
    unlink($tmp_file);

    # remove all numbers
#    $text =~ s/(?<![a-zA-Z])\d+\s*//g;

    return $text;
}

# alternate version to use Tika server
# (not well implemented in Tika, not yet working in TWiki)
sub stringForFileServer {
  my ($self, $filename) = @_;
  my ($in, $socket, $data, $text);
  undef $/; # read all input as one scalar
  open $in, $filename;
  $socket = new IO::Socket::INET(PeerHost => '127.0.0.1', PeerPort => '20000', Proto => 'tcp')
    or die "Couldn't connect to Tika server\n";
  $data = <$in>;
  print $socket $data;
  $text = <$socket>;
  close($socket);
  return $text;
}

1;