=head1 BOPHScraper

=head1 NAME

Webg::BOPHScraper - Scraper para el Boletín Oficial de la Provincia de Huesca

=head1 SYNOPSIS

use Webg::BOPHScraper->new();
my $boph = Webg::BOPHScraper->new();
$boph->download("2008","05");
$boph->doit("La Puebla de Castro","contenidos/boletin-lapuebla.html","LAPUEBLA DE CASTRO","LA PUEBLA");

=head1 DESCRIPTION

Downloads announcements for different Major Councils from www.dphuesca.es

=cut

package Webg::BOPHScraper;
use strict;

sub new {
  my $type = shift;
  my %params =  @_;
  my $self = {};
  my @fields = qw( url_path );
  foreach my $field (@fields) {
    (exists $params{$field} ) ? 
      $self->{$field} = $params{$field} : ();
  }
  unless ( defined $self->{url_path} ) {
    $self->{url_path} = "http://www.dphuesca.es/boph/pdfs";
  }
  @{$self->{theme_change}} =
    ('COMARCAS',
     'AYUNTAMIENTO',
     'ENTIDADES LOCALES',
     'ENTIDAD LOCAL',
     'ADMINISTRACIÓN DEL ESTADO',
     'Administración del Estado',
     'SOCIEDAD DE GESTIÓN');

  @{$self->{days}} =
    ( "Lunes",
      "Martes",
      "Miércoles",
      "Jueves",
      "Viernes",
      "Sábado",
      "Domingo" );
  $self->{init_an} = "AYUNTAMIENTO DE";

  bless $self, $type;
}

sub download {
  my $self = shift;
  my $year = shift;
  my $month = shift;
  unless ( -e "txt" && -d "txt" ) {
    mkdir("txt");
  }
  unless ( -e "pdf" && -d "pdf" ) {
    mkdir("pdf");
  }
  foreach my $day ( 1..31 ) {
    $day < 10 ? $day = "0$day" : ();
    unless ( -e "pdf/BOPH${year}${month}$day.pdf" ) {
      print "# Getting $self->{url_path}/BOPH${year}${month}$day.pdf\n";
      system("wget --quiet $self->{url_path}/BOPH${year}${month}$day.pdf");
      if ( -e "BOPH${year}${month}$day.pdf") {
	system("mv BOPH${year}${month}$day.pdf pdf/");
      }
    }
    unless ( -e "txt/BOPH${year}${month}$day.txt" ) {
      print "# Translating BOPH${year}${month}$day.pdf\n";
      system("pdftotext -q -eol unix -enc Latin1 -nopgbrk -raw pdf/BOPH${year}${month}$day.pdf");
      if ( -e "pdf/BOPH${year}${month}$day.txt" ) {
	system("mv pdf/BOPH${year}${month}$day.txt txt/");
      }
    }
  }
}

sub doit {
  my $self = shift;
  my $title = shift;
  my $filename = shift;
  my $rss_filename = shift;
  my @pattern = @_;
  my $pattern = join("|",@pattern);
  my @anouncements = ();
  my @records = ();
  foreach my $file ( <txt/*.txt> ) {
    push(@records,$self->parse($pattern,$file));
  }
  my @sort = reverse @records;
  $self->dump_as_html($filename,$title,@sort);
  if ( $rss_filename =~ m/.*\.rss\Z/i ) {
    $self->dump_as_rss($rss_filename,$title,@sort);
  }
}

sub dump_as_rss {
  my $self = shift;
  my $filename = shift;
  my $title = shift;
  my @records = @_;
  open(RSS,">$filename");

  print RSS <<EOF;
<?xml version="1.0" encoding="ISO-8859-1" ?>
 <rss version="0.91">

 <channel>
 <title>Anuncios B.O.P.H. Ayuntamiento de $title</title>
 <link>http://www.lapuebladecastro.com/boletin/</link>
 <description>Histórico de anuncios publicados en el 
  Boletín Oficial de la Provincia de Huesca</description>
 <language>es</language>

 <image>
 <title>La Puebla de Castro</title>
 <url>http://www.lapuebladecastro.com/gif/ban/t_tescudo.gif</url>
 <link>http://www.lapuebladecastro.com</link>
 <width>30</width>
 <height>30</height>
 </image>

EOF

  foreach my $post ( @records ) {
    my ($name) = $post->{file} =~ m/(BOPH.*)\.txt/gi;
    my $url = "$self->{url_path}/$name.pdf";
    print RSS <<EOF;


<item>
<title>Anuncio $title B.O.P.H. $post->{date}</title>
<link>$url</link>
<description>
$post->{text}
</description>
</item>


EOF
  }
print HTML <<EOF;
</channel>
</rss>
EOF
  close(HTML);


}
sub dump_as_html {
  my $self = shift;
  my $filename = shift;
  my $title = shift;
  my @records = @_;
  open(HTML,">$filename");
  print HTML <<EOF;
<h3>Anuncios del Ayto. de 
$title en el Boletín Oficial de la Provincia de Huesca</h3>
<p><i><font size=\"-1\" face=\"Arial\">Haz click sobre los 
enlaces a las fechas para obtener la edición del
Boletín Original en formato PDF</font></i></p>
EOF
  foreach my $post ( @records ) {
    my $text = $post->{text};
    $text =~ s/\A(\d+)/<font color=\"red\" style=\"font-weight:bold\">$1<\/font>/gi;
    my ($name) = $post->{file} =~ m/(BOPH.*)\.txt/gi;
    my $url = "$self->{url_path}/$name.pdf";
    print HTML "<hr/>";
    print HTML "<h3><a href=\"$url\" target=\"_blank\">$post->{date}</a> ";
    print HTML "(<i>Extracto del B.O.P.H.</i>)</h3>\n";
    print HTML "<p style=\"width:550px\"><font size=\"-1\" face=\"Arial\">";
    print HTML "$text</font></p>\n\n\n";
  }
print HTML <<EOF;
EOF
  close(HTML);
}

sub parse {
  my $self = shift;
  my $pattern = shift;
  my $file = shift;
  my @lines = $self->get_lines_from_file($file);
  my @anuncios = $self->ayuntamiento($pattern,$file,@lines);
  @anuncios;
}

sub ayuntamiento {
  my $self = shift;
  my $pattern = shift;
  my $file = shift;
  my @lines = @_;
  my @records = ();
  my $date = $lines[6];
  my @dias = @{$self->{days}};
  my $dias = join("|",@dias);
  foreach my $i ( 0..15 ) {
    if ( $lines[$i] =~ m/($dias)/gi ) {
      $date = $lines[$i];
    }
  }
  chomp($date);
  my ($end_line,@codes) = $self->code_search($pattern,@lines);
  my $regexp = join("|",@codes);
  my @theme_change = @{$self->{theme_change}};
  push(@theme_change,@codes);
  my $theme_change = join("|",@theme_change);
  unless ( @codes ) {
    return ();
  }
  foreach my $i ( $end_line..$#lines ) {
    my $line = $lines[$i];
    chomp($line);
    if ( ($#records >= 0) && ($line =~ m/\A($theme_change)/g)) {
      unless ( defined $records[$#records]->{fline} ) {
	$records[$#records]->{fline} = $i-1;
      }
    } 
    if ( $line =~ m/\A$regexp\s*\Z/ ) {
      push(@records,{ iline => $i });
    }
  }
  foreach my $record ( @records ) {
    $record->{date} = $date;
    $record->{file} = $file;
    foreach my $i ( $record->{iline} .. $record->{fline} ) {
      $record->{text} .= $lines[$i];
    }
  }
  @records;
}

sub code_search {
  my $self = shift;
  my $pattern = shift;
  my @lines = @_;
  my @codes = ();
  my $end_line;
  foreach my $i ( 0..$#lines ) {
    my $line = $lines[$i];
    my $code = "";
    if ( $line =~ m/B\.\s*O\.\s*P\.\s*HU\.-\s* N/gi ) {
      $end_line = $i;
      last;
    }
    if ( (($code)
	  = $line
	  =~ m/\A(\d+)\s*\.*\s*$self->{init_an} ($pattern)\s*\.*/gi) ) {
      push(@codes,$code);
    }
  }
  ($end_line,@codes);
}

sub get_lines_from_file {
  my $self = shift;
  my $file = shift;
  open(TXT,$file);
  my @lines = <TXT>;
  close(TXT);
  @lines;
}

1;

=head1 LICENSE AND COPYRIGHT

Copyright: Alfonso Egio 2008.

This library is free software; you can redistribute it and/or modify
it under the same terms as Perl itself, either Perl version 5.000 or,
at your option, any later version of Perl 5 you may have available.

The full text of the licenses can be found at the following url:
http://www.opensource.org/licenses/artistic-license-2.0.php

=head1 AUTHOR

Alfonso Egio <webmaster@lapuebladecastro.com>