source: main/trunk/model-sites-dev/pei-jones/collect/written-works/perllib/plugins/PJPlugin.pm@ 32147

package PJPlugin;

use PagedImagePlugin;

use strict;
no strict 'refs'; # allow filehandles to be variables and viceversa
no strict 'subs';

sub BEGIN {
    @PJPlugin::ISA = ('PagedImagePlugin');
}

my $arguments = [];

my $options = { 'name'     => "PJPlugin",
                'desc'     => "PagedImagePlugin variant for pei-jones written-works collection",
                'abstract' => "no",
                'inherits' => "yes",
                'args'     => $arguments };

my $files_dir ="/nzdl-storage/other-projects/pei-jones/Jones_Collection/";

sub begin {
    my $self = shift (@_);
    my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;

    # Save base_dir for use in file cache                                                                                              
    $self->{'base_dir'} = $files_dir;
}


sub new {
    my ($class) = shift (@_);
    my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
    push(@$pluginlist, $class);

    push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
    push(@{$hashArgOptLists->{"OptList"}},$options);

    my $self = new PagedImagePlugin($pluginlist, $inputargs, $hashArgOptLists);

    return bless $self, $class;
}


sub process_item {
    my $self = shift (@_);
    my ($filename_full_path, $dir, $filename_no_path, $processor, $metadata) = @_;

    my $doc_obj = new doc ($filename_full_path, "indexed_doc", $self->{'file_rename_method'});
    $self->set_initial_doc_fields($doc_obj, $filename_full_path, $processor, $metadata);
    my $topsection = $doc_obj->get_top_section();
    # simple item files are always paged unless user specified
    if ($self->{'documenttype'} eq "auto") {
    $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", "paged");
    } else {
    $doc_obj->set_utf8_metadata_element ($topsection, "gsdlthistype", $self->{'documenttype'});
    }
    open (ITEMFILE, "<:encoding(UTF-8)", $filename_full_path) || die "couldn't open $filename_full_path\n";
    my $line = "";
    my $num = 0;
    while (defined ($line = <ITEMFILE>)) {
    
    next unless $line =~ /\w/;
    chomp $line;
    next if $line =~ /^#/; # ignore comment lines
    if ($line =~ /^<([^>]*)>\s*(.*?)\s*$/) {
        my $meta_name = $1;
        my $meta_value = $2;
        #if ($meta_name =~ /\./) {
        #   $meta_name = "ex.$meta_name";
        # }
        # PJ mod:
        if ($meta_value !~ /^unknown$/) {
        # don't add in unknown values
        # set all metadata at pj.
        $doc_obj->set_utf8_metadata_element ($topsection, "pj.".$meta_name, $meta_value);
        }
        #$meta->{$1} = $2;
    } else {
        $num++;
        # line should be like page:imagefilename:textfilename:r - the r is optional -> means rotate the image 180 deg
        $line =~ s/^\s+//; #remove space at the front
        $line =~ s/\s+$//; #remove space at the end
        my ($pagenum, $imgname, $txtname, $rotate) = split /:/, $line;

        # PJ: use jpg versions instead of tif versions
        #$imgname =~ s/tif/jpg/g;
        #print STDERR "new img name=$imgname\n";
        # create a new section for each image file
        my $cursection = $doc_obj->insert_section($doc_obj->get_end_child($topsection));
        # the page number becomes the Title
        $doc_obj->set_utf8_metadata_element($cursection, 'Title', $pagenum);
     
        print STDERR "image file $files_dir$imgname\n";
         
        # process the image for this page if there is one
        if (defined $imgname && $imgname ne "") {
        my $result1 = $self->process_image($files_dir.$imgname, $imgname, $doc_obj, $cursection, $rotate);
        if (!defined $result1)
        {
            print "PagedImagePlugin: couldn't process image \"$files_dir$imgname\" for item \"$filename_full_path\"\n";
        }
        }
        # process the text file if one is there
        if (defined $txtname && $txtname ne "") {
        my $result2 = $self->process_text ($files_dir.$txtname, $txtname, $doc_obj, $cursection);
               
        if (!defined $result2) {
            print "PagedImagePlugin: couldn't process text file \"$files_dir.$txtname\" for item \"$filename_full_path\"\n";
            $self->add_dummy_text($doc_obj, $cursection);
        }
        } else {
        # otherwise add in some dummy text 
        $self->add_dummy_text($doc_obj, $cursection);
        }
    }
    }
    
    close ITEMFILE;

    # add numpages metadata
    $doc_obj->set_utf8_metadata_element ($topsection, 'NumPages', "$num");

    $doc_obj->set_utf8_metadata_element($topsection,"MaxImageWidth",$self->{'MaxImageWidth'});
    $doc_obj->set_utf8_metadata_element($topsection,"MaxImageHeight",$self->{'MaxImageHeight'});
    $self->{'MaxImageWidth'} = undef;
    $self->{'MaxImageHeight'} = undef;


    return $doc_obj;
}

sub scan_item_for_files_to_block
{
    my $self = shift (@_);
    my ($filename_full_path, $dir, $block_hash) = @_;


    open (ITEMFILE, $filename_full_path) || die "couldn't open $filename_full_path to work out which files to block\n";
    my $line = "";
    while (defined ($line = <ITEMFILE>)) {
    next unless $line =~ /\w/;
    chomp $line;
    next if $line =~ /^#/; # ignore comment lines
    next if ($line =~ /^<([^>]*)>\s*(.*?)\s*$/); # ignore metadata lines 
    # line should be like page:imagefilename:textfilename:r 
    $line =~ s/^\s+//; #remove space at the front
    $line =~ s/\s+$//; #remove space at the end
    my ($pagenum, $imgname, $txtname, $rotate) = split /:/, $line;

    # PJ: use jpg versions instead of tif versions
    #$imgname =~ s/tif/jpg/g;

    # find the image file if there is one
    if (defined $imgname && $imgname ne "") {
        $self->block_raw_filename($block_hash, &FileUtils::filenameConcatenate( $dir,$imgname));
    }
    # find the text file if there is one
    if (defined $txtname && $txtname ne "") {
        $self->block_raw_filename($block_hash, &FileUtils::filenameConcatenate($dir,$txtname));
    }
    }
    close ITEMFILE;

}

1;