Miscellany

## Modules ##
use strict;
use lib '.'; # This looks for modules in the same folder as the script
use NCBI_PowerScripting; # NCBI eutils module
use XML::Twig; # Handle the complicated handlers, and parse xml
use XML::Simple; # Use to insert parsed xml data into file
use CGI;
use CGI::Carp qw(fatalsToBrowser);
## End Modules ##

# Global Constants

my $pCGI = new CGI;
print $pCGI->header;

my $sTerm = $pCGI->param('search');
# End

print "Searching for $sTerm...<br />\n";

# Now going to parse all "spaces" into "+" signs
$sTerm =~s/ /+/g;
$sTerm =~s/\+{2,}/\+/g; #just in case the user was smart and wrote in "+" signs
print "Converting to $sTerm...<br />\n";

# Information needed for ESearch
# Information passed from webbrower is $sTerm

my $PH_ESEARCH_PARAMS = {
	'db'         => 'pubmed',
	'term'       => "$sTerm",
	'rettype'    => 'uilist',
	'retmode'    => 'xml',
	'outfile'    => 'pdfids.xml',
	'tool'       => 'vbipatric'
};

select(STDOUT); $| = 1; # make STDOUT unbuffered (so prints are done imediately)

# Fetch Datas and create XML File
buildPMIDXMLFile($PH_ESEARCH_PARAMS);

# Strip file of unnecessary crap
# strip();

my $parse = "pdfids.xml";

# Building the Twig Object and parse it using the insertCitations Method
my $pTwig = new XML::Twig('TwigRoots' => {'IdList' => \&insertCitations});

if( !defined($pTwig) )
{
   print "Err: Can't get XML::Twig Object !\n";
   exit -2;
}

$pTwig->parsefile($parse);

print "All done \:\) \n";

exit;

sub buildPMIDXMLFile
{
   my $p_phESearchParams = shift;
   my $p_sTerm = shift;

   my %hESearchParams = esearch(('db' => $p_phESearchParams->{'db'}, 'term' => $p_phESearchParams->{'term'}));

   # Complete the %hESearchParams hashtable using required params from original $p_phESearchParams
   foreach my $sKey (keys %{$p_phESearchParams})
	{
	$hESearchParams{$sKey} = $p_phESearchParams->{$sKey};
	}

   $hESearchParams{'term'} = $p_sTerm;

   efetch_batch(%hESearchParams);

   print "<br />Successfully built pdfids.xml file<br />\n";
}

sub getPMID
{
   my $p_phXML = shift;

   my $paXMLPMID = getInfosFromXML($p_phXML, 'Id');

   return join(',', @{$paXMLPMID});

   print "Successfully grabbed all PMIDS and inserted into an array....proceeding \n"
}

sub getInfosFromXML
{
   my $p_phXML    = shift;
   my @p_aKeyList = @_;

   my $r_paXMLInfos;

   my $phTmpXML = $p_phXML;

   my @aKeyListSaved = @p_aKeyList;

   my $bReturnTmpXML = 1;

   foreach my $sKey (@p_aKeyList)
   {
      shift @aKeyListSaved;

      if( defined($phTmpXML) )
      {
         if( ref($phTmpXML) eq 'HASH' && defined($phTmpXML->{$sKey}) )
         {
            if( ref($phTmpXML->{$sKey}) eq 'HASH' )
            {
               $phTmpXML = $phTmpXML->{$sKey};
            }
            elsif( ref($phTmpXML->{$sKey}) eq 'ARRAY' )
            {
               foreach my $puTmp (@{$phTmpXML->{$sKey}})
               {
                  if( !defined($r_paXMLInfos) )
                  {
                     $r_paXMLInfos = [];
                  }

                  my $paXMLInfos = getInfosFromXML($puTmp, @aKeyListSaved);

                  if( defined($paXMLInfos) )
                  {
                     push @{$r_paXMLInfos}, @{$paXMLInfos};
                  }
               }
            }
            else
            {
               if( !defined($r_paXMLInfos) )
               {
                  $r_paXMLInfos = [];
               }

               push @{$r_paXMLInfos}, $phTmpXML->{$sKey};

               $bReturnTmpXML = 0;
               last;
            }
         }
         elsif( ref($phTmpXML) eq 'HASH' && !defined($phTmpXML->{$sKey}) )
         {
            $bReturnTmpXML = 0;
         }
      }
   }

   if( !defined($r_paXMLInfos) )
   {
      if( $bReturnTmpXML eq 1 )
      {
         $r_paXMLInfos = [$phTmpXML];
      }
      else
      {
         $r_paXMLInfos = [];
      }
   }

   return $r_paXMLInfos;

   print "Done grabbing all pmids.... \n";   
}

sub insertCitations
{
   my $p_pTwig    = shift;
   my $p_pIdList  = shift;

   # Save XML IdList into string
   my $sIdList = $p_pIdList->sprint;

   # release/purge Twig Datas (no more needed as we work on the $sIdList XML directly using XML::Simple
   $p_pTwig->purge;

   # Getting an XML::Simple Object from the IdList XML (using sprint), which makes things simple to parse...
   my $pXS = new XML::Simple();

   my $phXML = eval{$pXS->XMLin($sIdList, 'forcearray' => 0)};

   if( $@ )
   {
      print "Err: $@\n";
      return;
   }

   if( !defined($phXML) )
   {
      return;
   }

   # Getting PMID
   my $nPMID = getPMID($phXML);

   # 1112,111244,121423,1214314
   my @pmid = split (",",$nPMID);
   my $pmid;
   print "Grabbing all pmids .... \n";

   #if( !defined($nPMID) || $nPMID !~ /^\d+$/ )
   #{
   #   return;
   #}

   print "Writing list of PDF PMIDS that you'll need.... \n";
   open (PDFLIST,">/opt/lampp/htdocs/pdflist/pdflist-$sTerm.html");
   print PDFLIST "<html>";
   print PDFLIST "<body>";
   foreach $pmid (@pmid){
	print PDFLIST "<a href=\"http\:\/\/eutils.ncbi.nlm.nih.gov\/entrez\/eutils\/elink.fcgi\?dbfrom\=pubmed\&id\=$pmid\&retmode\=ref\&cmd\=prlinks\"\/>$pmid<\/a><br \> \n" || die "Could not write pdf list!!!"
	};
   print PDFLIST "</body>";
   print PDFLIST "</html>";
  # print "Check http\:\/\/localhost\/pdflist\/pdflist-$sTerm.html file for a list of the pmid.pdf you need to download for uploading to Textpresso<br>You need to download into a folder, zip the folder into a pdf-$sTerm.zip and then go to http:\/\/localhost\/cgi-bin\/uploadpdfs.cgi \n";
  # foreach $pmid (@pmid){
  # print "<a href=\"http\:\/\/eutils.ncbi.nlm.nih.gov\/entrez\/eutils\/elink.fcgi\?dbfrom\=pubmed\&id\=$pmid\&retmode\=ref\&cmd\=prlinks\"\/>$pmid<\/a><br \> \n" || die "Could not write pdf list!!!"};
}