#!/usr/bin/perl -w
# Convert BibTeX files to dictionary format, using comments
# Author   : Lyndon Hill
# Last Edit: 2008.02.28
#
# (c) Lyndon Hill, 2008.
# This script is free software under terms of the GNU General Public License version 3.
# Please see http://www.gnu.org/licenses/ for more details.

use strict;

## User definable variables ##

# location of your bibtex files, make sure abbreviations are first	 
my $bibtexdir = "/home/username/Documents/BibTex";

# location of your comments; plain text files named after each citation key
my $bibsdir = "/home/username/Documents/Bibs";

# My comment files have the title and author on the first few lines.
# I want to skip these lines for the dictionary file. $titlelength is
# the number of lines to skip
my $titlelength = 3;

# the name of the dictionary file to create
my $output = "bibs-src.bedic";

# your email address
my $email = "My Name <email" . "@" . "domain.com>";

## End of user definable variables. ##

# Get individual BibTeX filenames 
my $source = `ls -1 $bibtexdir`;
my @bibtexfiles = split /\n/, $source;

# statistics
my $entries = 0;
my $bibs = 0;

my $line = "";
my $abbr = "";
my $full = "";

my %abbrhash = ();

open(OUTPUT, ">$output") || die "Can't open file $output for writing.\n";

my $filehandle = *OUTPUT;
printheader($filehandle);

foreach my $dic (@bibtexfiles)
{
  # open file
  open(INPUT, "$bibtexdir/$dic") || die "Can't open bibtex file $bibtexdir/$dic.\n";

  print "Opened $dic\n";

  # parse file
  while(<INPUT>)
  {
    $line = $_; chop $line;

    $abbr = ""; $full = "";

    if(lc(substr($line, 0, 7)) eq ("@" . "string"))
    {
      ($abbr, $full) = split /=/, $line; 

      substr($abbr, 0, 8) = "";

      $abbr =~ s/\{//g; 
      $abbr =~ s/\}//g; 

      while(substr($abbr, length($abbr)-1, 1) eq " ") { chop $abbr; } # eat trailing spaces

      $full =~ s/\{//g; 
      $full =~ s/\}//g; 
      while(substr($full, 0, 1) eq " ") { substr($full, 0, 1) = ""; } # eat preceding spaces

      $abbrhash{$abbr} = $full;
    }
    elsif(substr($line, 0, 1) eq "@")
    {
      $entries++;
      my $type = "";
      my $ref = "";
      ($type, $ref) = split /\{/, $line;
      chop $ref;
      $type = lc(substr($type, 1, length($type)-1));
      # print "Found reference: $ref type $type\n";

      print OUTPUT "$ref\n";
      print OUTPUT "{s}\n";
      print OUTPUT " {ct}$type" . "{/ct}\n";

      my $comment = "";

      # read comments
      if( -e "$bibsdir/$ref")
      {
        $bibs++;
        my $linecounter = 0;
        my $linebreak = 0;   # Last iteration inserted a linebreak for paragraphisation
        open(COMMENT, "$bibsdir/$ref");
     
        while(<COMMENT>)
        {
          $line = $_; chop $line;
          $linecounter++;

          # remove special brackets and deal with html
          $line =~ s/\</\&lt;/g; 
          $line =~ s/\>/\&gt;/g; 
          $line =~ s/\{/\(/g; 
          $line =~ s/\}/\)/g; 
          $line =~ s/\&/\&amp;/g; 

          # add lines to comment, if blank line add a line feed
          if($linecounter > $titlelength)
          {
            if($line eq "")
            {
              if($linebreak == 0) { $comment .= "{br/}\n{br/}\n"; }
              else { $comment .= "{br/}\n"; }
              $linebreak = 1;
            }
            elsif(length($line) < 40 && $line ne "")
            {
              $comment .= $line . "{br/}\n";
              $linebreak = 1;
            }
            else
            {
              $comment .= $line . " ";
              $linebreak = 0;
            }
          }
        }
        close(COMMENT);
      }

      # parse the rest of the entry
      my %entryhash = ();
      my $author = "";

      while(<INPUT>)
      {
        $line = $_; chop $line;

        my $tag = "";
        my $data = "";
        ($tag, $data) = split /\=/, $line;
  
        if(!defined($tag)) { $tag = ""; }
        if(!defined($data)) { $data = ""; }
 
        # remove whitespace from tag
        $tag =~ s/\ //g;
        $tag =~ s/\t//g;

        $tag = lc($tag);

        # remove speech marks, curly braces and final comma from data
        $data =~ s/"//g;
        $data =~ s/\{//g;
        $data =~ s/\}//g;
        $data =~ s/\\//g;
        while(substr($data, length($data)-1, 1) eq " ") { chop $data; } # eat trailing spaces
        if(substr($data, length($data)-1, 1) eq ",") { chop $data; }

        # remove white space from certain data
        if($tag eq "year" || $tag eq "volume" || $tag eq "number" || $tag eq "pages")
        {
          $data =~ s/\ //g;
          $data =~ s/\t//g;
        }

        while(substr($data, 0, 1) eq " ") { substr($data, 0, 1) = ""; } # eat preceding spaces

        # Add tag to entry hash
        if($tag ne "") { $entryhash{$tag} = $data; }

        # end reading an entry if tag was empty
        if($tag eq "")
        {
          $line = "";

          my $potentialauthors = $entryhash{"author"};
          if(!defined($potentialauthors)) { $potentialauthors = ""; }
          my @authors = split /\ and\ /, $potentialauthors;
          my $numauthors = @authors;
          if($numauthors == 1) { $author = $entryhash{"author"}; }
          elsif($numauthors > 1)
          {
            if(substr($authors[0], 0, 1) eq " ") { substr($authors[0], 0, 1) = "" };
            $author = $authors[0]; 
            for($a = 1; $a < $numauthors - 1; $a++)
            {
              #substr($authors[$a], 0, 1) = "";
              #chop $authors[$a];
              $author .= ", " . $authors[$a];
            }
            #substr($authors[$numauthors-1], 0, 1) = "";
            $author .= " and " . $authors[$numauthors-1];
          }

          # Output entry details

          print OUTPUT " {ss}";
          if($author ne "") { print OUTPUT $author . ": "; }
          print OUTPUT $entryhash{"title"} . ".";

          # where published
          my $where = "";

          if($type eq "inproceedings" && defined($entryhash{"booktitle"})) 
          { 
            $where = $entryhash{"booktitle"};
            if(defined($abbrhash{$where})) { $where = $abbrhash{$where}; }
            print OUTPUT " <b>" . $where . "</b>";
          }
          if($type eq "article" && defined($entryhash{"journal"}))
          {
            $where = $entryhash{"journal"};
            if(defined($abbrhash{$where})) { $where = $abbrhash{$where}; }
            print OUTPUT " <b>" . $where . "</b>";
          }
          if($type eq "book" && defined($entryhash{"publisher"}))
          {
            $where = $entryhash{"publisher"};
            if(defined($abbrhash{$where})) { $where = $abbrhash{$where}; }
            print OUTPUT " <b>" . $where . "</b>";
          }
          if($type eq "techreport" && defined($entryhash{"institution"}))
          {
            $where = $entryhash{"institution"};
            if(defined($abbrhash{$where})) { $where = $abbrhash{$where}; }
            print OUTPUT " <b>" . $where . "</b>";
          }

          if($type eq "article" && defined($entryhash{"journal"})) { print "Citation $ref does not have a journal for it's article!\n"; }

          if($type eq "inproceedings" && defined($entryhash{"booktitle"})) { print "Citation $ref does not have a booktitle for it's proceedings!\n"; }

          # which issue

          my $issue = "";

          if(defined($entryhash{"volume"}) && defined($entryhash{"number"}))
          {
            $issue = $entryhash{"volume"} . "(" . $entryhash{"number"} . ")";
          }
          elsif(defined($entryhash{"volume"}))
          {
            $issue = "volume " . $entryhash{"volume"};
          }
          elsif(defined($entryhash{"number"}))
          {
            $issue = "number " . $entryhash{"number"};
          }

          if($where ne "" && $issue ne "") { print OUTPUT ", $issue"; }
          elsif($issue ne "") { print OUTPUT " " . $issue; }

           # important details
          my $pages = "";
          if(defined($entryhash{"pages"}))
          {
            $pages = $entryhash{"pages"};
            $pages =~ s/--/-/;
            if($issue ne "" || $where ne "") { print OUTPUT ", pages " . $pages; }
            else { print OUTPUT "Pages " . $pages; }
          }

          if(defined($entryhash{"year"}))
          {
            my $year = $entryhash{"year"};
            if(substr($year, length($year)-1, 1) eq ",") { chop $year; }
            if($issue ne "" || $pages ne "" || $where ne "") { print OUTPUT ", $year."; }
            else { print OUTPUT " $year."; } 
          }
          else
          {
            print "Citation $ref has no year!\n";
          }

          print OUTPUT "{/ss}\n{/s}\n";

          if(defined($entryhash{"note"}) || defined($entryhash{"url"}))
          {
            print OUTPUT "{s}\n";
            if(defined($entryhash{"note"}))
            {
              print OUTPUT " {ss}" . $entryhash{"note"} . "{/ss}\n";
            }

            if(defined($entryhash{"url"}))
            {
              print OUTPUT " {ss}" . $entryhash{"url"} . "{/ss}\n";
            }
            print OUTPUT "{/s}\n";
          }

          if($comment ne "")
          {
            print OUTPUT "{s}\n {ss}" . $comment . "{/ss}\n{/s}\n";
          }

          # entry separating linefeed
          print OUTPUT "\n";
          
          last;      
        }
      }

    }
  }

  close(INPUT);
}

close(OUTPUT);

print "$entries entries, $bibs with comments.\n";

exit;

# Output the header at the start of the dictionary file
sub printheader
{
  my $fh = shift;
  print $fh "id=BibTeX References\n";
  print $fh "description=My private BibTeX and notes.\n";
  print $fh "maintainer=$email\n";
  print $fh "char-precedence={ -,!/.()?*+=^&:;_[]<>\{\}%\"\#\'~}{0123456789}{aA}{bB}{cC}{dD}{eE}{fF}{gG}{hH}{iI}{jJ}{kK}{lL}{mM}{nN}{oO}{pP}{qQ}{rR}{sS}{tT}{uU}{vV}{wW}{xX}{yY}{zZ}\n\n";
}