source: trunk/gsdl/perllib/plugins/BibTexPlug.pm@ 1903

###########################################################################
#
# BibTexPlug.pm - a plugin for bibliography records in BibTex format
#
# A component of the Greenstone digital library software
# from the New Zealand Digital Library Project at the 
# University of Waikato, New Zealand.
#
# Copyright 2000 Gordon W. Paynter
# Copyright 1999-2000 New Zealand Digital Library Project
#
# 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.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
#
###########################################################################


# BibTexPlug reads bibliography files in BibTex format.
#
# by Gordon W. Paynter ([email protected]), November 2000
# Based on ReferPlug.  See ReferPlug for geneology.
#
# BibTexPlug creates a document object for every reference a the file.
# It is a subclass of SplitPlug, so if there are multiple records, all
# are read.


package BibTexPlug;

use SplitPlug;


# BibTexPlug is a sub-class of BasPlug.
sub BEGIN {
    @ISA = ('SplitPlug');
}

# This plugin processes files with the suffix ".bib"
sub get_default_process_exp {
    return q^(?i)\.bib$^;
}

# This plugin splits the input text at blank lines
sub get_default_split_exp {
    return q^\n+(?=@)^;
}


# The process function reads a single bibliographic record and stores
# it as a new document.

sub process {
    my $self = shift (@_);
    my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_;
    my $outhandle = $self->{'outhandle'};

    # Check that we're dealing with a valid BibTex record
    return undef unless ($$textref =~ /^@\w+\{.*\}/s);

    # Ignore things we can't use
    return 0 if ($$textref =~ /^\@String/);

    # Report that we're processing the file
    print $outhandle "BibTexPlug: processing $file\n"
    if ($self->{'verbosity'}) > 1;


    # This hash translates BibTex field names into metadata names.  The
    # BibTex names are taken from the "Local Guide to Latex" Graeme
    # McKinstry.  Metadata names are consistabnt with ReferPlug.

    my %field = (
         'address', 'PublisherAddress',
         'author', 'Creator',
         'booktitle', 'Booktitle',
         'chapter', 'Chapter',
         'edition', 'Edition',
         'editor', 'Editor',
         'institution', 'Publisher',
         'journal', 'Journal',
         'month', 'Month',
         'number', 'Number',
         'pages', 'Pages',
         'publisher', 'Publisher',
         'school', 'Publisher',
         'title', 'Title',
         'volume', 'Volume',
         'year', 'Date',

         'keywords', 'Keywords',
         'abstract', 'Abstract',
         'copyright', 'Copyright');

    # Metadata fields 
    my %metadata;
    my ($EntryType, $EntryID, $Creator, $Keywords, $text);

    # Make sure the text has exactly one entry per line
    my $lines = $$textref;
    $lines =~ s/,\s*\n/=====/g;
    $lines =~ s/\s+/ /g;
    $lines =~ s/\s*=====\s*/\n/g;
    my @lines = split(/\n+/, $lines);
    
    # Read and process each line in the bib file.
    my ($id, $name, $value, $line);
    foreach $line (@lines) {
    
    # Add each line.  Most lines consist of a field identifer and
    # then data, and we simply store them, though we treat some
    # of the fields a bit differently.

    $line =~ s/\s+/ /g;
    $text .= "$line\n";

    # The first line is special, it contains the reference type and OID
    if ($line =~ /\@(\w+)\{([\w\d]*),/) {
        $EntryType = $1;
        $EntryID = $2;
        print "** $EntryType - $EntryID \n";
        next;
    }
    
    # otherwise, parse the metadata out of this line
    next unless ($line =~ /^\s*(\w+)\s+=\s+(.*)/);
    $id = lc($1);
    $value = $2;
    
    # Add this line of metadata
    $metadata{$id} .= "$value\n";
    }

    # Add the Entry type as metadata
    $doc_obj->add_metadata ($cursection, "EntryType", $EntryType);

    # Add the various field as metadata
    foreach my $id (keys %metadata) {
    
    next unless (defined $field{$id});
    next unless (defined $metadata{$id});   
    
    $name = $field{$id};
    $value = $metadata{$id};

    # Get rid of silly Latex stuff
    if ($value =~ /\"(.*)\"/) {
        $value = $1;
    }
    if ($value =~ /\{(.*)\}/) {
        $value = $1;
    }
        
    # Add the various field as metadata 
    $value = &text_into_html($value);
    $doc_obj->add_metadata ($cursection, $name, $value);

    # Several special operatons on metadata follow
    
    # Add individual keywords.
    # The full set of keywords will be added, in due course, as "Keywords".
    # However, we also want to add them as individual "Keyword" metadata elements.
    if ($id eq "keywords") {
        my @keywordlist = split(/,/, $value);
        foreach my $k (@keywordlist) {
        $k = lc($k);
        $k =~ s/\s*$//; 
        $k =~ s/^\s*//; 
        if ($k =~ /\w/) {
            $k = &text_into_html($k);
            $doc_obj->add_metadata ($cursection, "Keyword", $k);
        }
        } 
    }
    
    # Add individual authors
    # The author metadata will be stored as one "Creator" entry, but we
    # also want to split it into several individual "Author" fields in
    # "Lastename, Firstnames" format so we can browse it.
    if ($id eq "author") {
        
        my @authorlist = split(/(,|and)/, $value);
        foreach $a (@authorlist) {
        $a =~ s/\s*$//; 
        $a =~ s/^\s*//; 
        
        # Reformat and add author name
        my @words = split(/ /, $a);
        my $lastname = pop @words;
        my $firstname = join(" ",  @words);
        my $fullname = $lastname . ", " . $firstname;
        
        # Add each name to set of Authors
        if ($fullname =~ /\w+, \w+/) {
            $fullname = &text_into_html($fullname);
            $doc_obj->add_metadata ($cursection, "Author", $fullname);
        }
        }
    }

    # Books and Journals are additionally marked for display purposes
    if ($id eq "booktitle") {
        $doc_obj->add_metadata($cursection, "BookConfOnly", 1);
    } elsif ($id eq "journal") {
        $doc_obj->add_metadata($cursection, "JournalsOnly", 1); 
    }

    }

    # Add the text in BibTex format (all fields)
    if ($text =~ /\w/) {
    $text = &text_into_html($text);
    $doc_obj->add_text ($cursection, $text);
    }

    return 1;
}

1;


# Convert a text string into HTML.

# The HTML is going to be inserted into a GML file, so we have to be
# careful not to use symbols like ">", which ocurs frequently in email
# messages (and use &gt instead.

# This function also turns URLs and email addresses into links, and
# replaces carriage returns with <BR> tags (and multiple carriage returns
# with <P> tags).


sub text_into_html {
    my ($text) = @_;


    # Convert problem charaters into HTML symbols
    $text =~ s/&/&amp;/g;
    $text =~ s/</&lt;/g;
    $text =~ s/>/&gt;/g;
    $text =~ s/\"/&quot;/g;
    $text =~ s/\'/ /g;
    $text =~ s/\+/ /g;
    $text =~ s/\(/ /g;
    $text =~ s/\)/ /g;

    # convert email addresses and URLs into links
    $text =~ s/([\w\d\.\-]+@[\w\d\.\-]+)/<a href=\"mailto:$1\">$1<\/a>/g;
    $text =~ s/(http:\/\/[\w\d\.\-]+[\/\w\d\.\-]*)/<a href=\"$1">$1<\/a>/g;

    # Clean up whitespace and convert \n charaters to <BR> or <P>
    $text =~ s/ +/ /g;
    $text =~ s/\s*$//; 
    $text =~ s/^\s*//; 
    $text =~ s/\n/\n<BR>/g;
    $text =~ s/<BR>\s*<BR>/<P>/g;

    return $text;
}