Return to makemeta-lib.pl CVS log

Up to [Repository] / foxridge-archiver

updated to Ubuntu Perl paths.

#!/usr/bin/perl -w

use strict;
use XML::LibXML;

use lib '/usr/local/mpiwg/archive';
use MPIWGStor;

# make output unbuffered
$|=1;

# program version
my $version = "0.2.6 (11.12.2006 ROC)";
my $help = 
"use: makemeta-lib [options] file.xml
options:
  -debug  show debugging info
  -dry-run  simulate, dont'do anything
  -online-mode  mode for creating online/permanent files
  -online-base=dir  base directory for online ids (for online mode)
  -cw-mode  mode for copying einstein_cw archive documents
  -digifiles-mode  mode for copying files from digifiles
  -map-file=mapfile.xml  digilib mapping file (for digifiles mode)
  -access=free  adds free access tag (use access=mpiwg for restricted access)
";
logger("INFO", "makemeta-lib $version");

###########################################
# mappings

# generic mappings at top level
my %gen_map = (
    'Device' => 'meta/image-acquisition/device',
    'Image_Type' => 'meta/image-acquisition/image-type',
    'Production_Comment' => 'meta/image-acquisition/production-comment',
    'Postproduction' => 'meta/image-acquisition/production-comment',
    'Language' => 'meta/lang'
    );
# sub type switch tag
my %type_map = (
    'Reference_Type' => 'meta/bib@type'
    );
# sub type mappings
my %subtype_map = (
    'Book' => {
	'_name' => 'book',
	'Author' => 'meta/bib/author',
	'Title' => 'meta/bib/title',
	'Year' => 'meta/bib/year',
	'Place_Published' => 'meta/bib/city',
	'Publisher' => 'meta/bib/publisher',
	'Edition' => 'meta/bib/edition'
    },
    'Journal Article' => {
	'_name' => 'journal-article',
	'Author' => 'meta/bib/author',
	'Title' => 'meta/bib/title',
	'Year' => 'meta/bib/year',
	'Secondary_Title' => 'meta/bib/journal',
	'Volume' => 'meta/bib/volume',
	'Number' => 'meta/bib/issue',
	'Pages' => 'meta/bib/pages'
    },
    'In Book' => {
	'_name' => 'inbook',
	'Author' => 'meta/bib/author',
	'Title' => 'meta/bib/title',
	'Year' => 'meta/bib/year',
	'Secondary_Title' => 'meta/bib/book-title',
	'Pages' => 'meta/bib/pages'
    },
    'Newspaper Article' => {
	'_name' => 'newspaper-article',
	'Author' => 'meta/bib/author',
	'Title' => 'meta/bib/title',
	'Year' => 'meta/bib/year',
	'Secondary_Title' => 'meta/bib/newspaper',
	'Place_Published' => 'meta/bib/city',
	'Number' => 'meta/bib/issue-date',
	'Pages' => 'meta/bib/pages'
    },
    'Edited Book' => {
	'_name' => 'edited-book',
	'Author' => 'meta/bib/editor',
	'Title' => 'meta/bib/title',
	'Year' => 'meta/bib/year',
	'Place_Published' => 'meta/bib/city',
	'Publisher' => 'meta/bib/publisher',
	'Edition' => 'meta/bib/edition'
    },
    'Manuscript' => {
	'_name' => 'manuscript',
	'Author' => 'meta/bib/author',
	'Title' => 'meta/bib/title',
	'Year' => 'meta/bib/year',
	'Place_Published' => 'meta/bib/location',
    }
    );
# language element
my $lang_field = 'Language';
# languages to iso codes
my %lang_map = (
    'German' => 'de',
    'English' => 'en',
    'Italian' => 'it',
    'French' => 'fr',
    'Latin' => 'la',
    'Japanese' => 'ja',
    'Dutch' => 'nl',
    'Spanish' => 'es',
    'Swedish' => 'sv'
    );
# storage fields
my $arch_id_field = 'ID_Archive';
my $online_url_field = 'URL';
my $online_id_field = 'ID_OnlinePermanent';

#######################################################
# internal parameters
#

# storage
my $lib_arch_dir = '/mpiwg/archive/data/library';
my $lib_online_dir = '/mpiwg/online/permanent';
my $lib_digilib_path = 'permanent';
my $digifiles_dir = "/net/digifiles.mpiwg-berlin.mpg.de/Volumes/raid";

# read command line parameters
my $args = MPIWGStor::parseargs;
if (! scalar(%$args)) {
    print $help, "\n";
    exit 1;
}

# debug level
$debug = (exists $$args{'debug'}) ? $$args{'debug'} : 0;

# simulate action only
my $dry_run = (exists $$args{'dry-run'}) ? $$args{'dry-run'} : 0;
logger('DEBUG', "dry-run: $dry_run");

# use online mode
my $online_mode = (exists $$args{'online-mode'}) ? $$args{'online-mode'} : 0;
logger('DEBUG', "online_mode: $online_mode");
# online base dir
my $online_base_dir = (exists $$args{'online-base'}) ? $$args{'online-base'} : "";
logger('DEBUG', "online_base_dir: $online_base_dir");

# use einstein-cw mode
my $cw_mode = (exists $$args{'cw-mode'}) ? $$args{'cw-mode'} : 0;
logger('DEBUG', "cw_mode: $cw_mode");

# use digifiles mode
my $digifiles_mode = (exists $$args{'digifiles-mode'}) ? $$args{'digifiles-mode'} : 0;
logger('DEBUG', "digifiles_mode: $digifiles_mode");
# digilib mapping file
my $map_file_name = (exists $$args{'map-file'}) ? $$args{'map-file'} : "";
logger('DEBUG', "map_file_name: $map_file_name");
my $mapping_doc;
my $mapping_root;

# access type
my $access_type = (exists $$args{'access'}) ? $$args{'access'} : "";

# index.meta namespace (not really implemented!)
my $namespace = "";


my $xml_changed = 0;
my $errcnt = 0;
my $warncnt = 0;

#######################################################
# check parameters that were passed to the program
#
my $infile = $$args{'path'};
if (! $infile) {
    logger("ABORT", "no input file given!");
    exit 1;
}
# strip double slashes
$infile = sstrip($infile, 1);
if (! -f $infile) {
    logger("ABORT", "input file \'$infile\' doesn't exist!");
    exit 1;
}


#######################################################
# subroutines
#


sub add_digilib_mapping {
    my ($src_dir, $dest_dir) = @_;
    my $elem = $mapping_root->addNewChild($namespace, 'mapping');
    $elem->addChild($mapping_doc->createAttributeNS($namespace, 'link', $src_dir));
    $elem->addChild($mapping_doc->createAttributeNS($namespace, 'dir', $dest_dir));
    if ($map_file_name) {
	write_xml($mapping_doc, $map_file_name);
    } else {
	logger('ABORT', "unable to write mapping file!");
	exit 1;
    }
}

sub find_digifiles_dir {
    my ($input_node) = @_;
    my $digifiles_base = '/net/digifiles.mpiwg-berlin.mpg.de/Volumes/raid';
    my $src_dir = find_online_dir($input_node, $digifiles_base, '');
    if (! $src_dir) {
	logger('ERROR', "no online directory for digifiles entry");
	$errcnt++;
	return;
    }
    my $dest_id = sstrip($input_node->findvalue("fm:$online_id_field"));
    if (! $dest_id) {
	logger('ERROR', "no ID field for digifiles entry");
	$errcnt++;
	return;
    }
    my $dir = "$lib_online_dir/library/$dest_id";
    my $map_dir = "$lib_digilib_path/library/$dest_id";
    if ($dry_run) {
	logger('DEBUG', "would move $digifiles_base/$src_dir to $dir");
	add_digilib_mapping($src_dir, "$map_dir/pageimg");
	return $dir;
    } else {
	logger('INFO', "moving $digifiles_base/$src_dir to $dir");
	logger('DEBUG', "mkdir $dir/pageimg"); 
	if (system("mkdir -p $dir/pageimg && chmod -R 0775 $dir") == 0) {
	    logger('DEBUG', "cp $digifiles_base/$src_dir $dir/pageimg"); 
	    if (system("cp -rp $digifiles_base/$src_dir/* $dir/pageimg/") == 0) {
		if (-d "$dir/pageimg") {
		    logger('DEBUG', "directory $dir OK"); 
		    add_digilib_mapping($src_dir, "$map_dir/pageimg");
		    if (system("rm -rf $digifiles_base/$src_dir/* && rm -rf $digifiles_base/$src_dir") == 0) {
			logger('DEBUG', "directory $digifiles_base/$src_dir removed"); 
			return $dir;
		    } else {
			logger('ERROR', "unable to remove source directory $digifiles_base/$src_dir!");
			$errcnt++;
			return $dir;
		    }
		}
	    }
	}
	logger('ABORT', "unable to copy directory $src_dir to $dir!");
	exit 1;
    }
    return;
}

sub find_cw_dir {
    my ($input_node) = @_;
    my $cw_base = '/mpiwg/archive/data/library/inbox/zwischen_backup';
    my $src_dir = find_online_dir($input_node, $cw_base, 'pageimg');
    my $dest_id = sstrip($input_node->findvalue("fm:$arch_id_field"));
    if (! $dest_id) {
	logger('ERROR', "no ID field for einstein-cw entry");
	$errcnt++;
	return;
    }
    my $dir = "$lib_arch_dir/$dest_id";
    if ($dry_run) {
	logger('DEBUG', "would move $cw_base/$src_dir to $dir");
	return $dir;
    } else {
	logger('DEBUG', "moving $cw_base/$src_dir to $dir");
	if (rename "$cw_base/$src_dir", $dir) {
	    if (-d $dir) {
		logger('DEBUG', "directory $dir OK"); 
		return $dir;
	    }
	} else {
	    logger('ABORT', "unable to rename directory $cw_base/$src_dir to $dir!");
	    exit 1;
	}
    }
    return;
}

sub find_permanent_dir {
    my ($input_node) = @_;
    my $online_id = sstrip($input_node->findvalue("fm:$online_id_field"));
    # try online_base_dir + online_id first
    if (($online_base_dir)&&($online_id)) {
	my $dir = sstrip("$online_base_dir/$online_id", 1);
	return $dir;
    }
    # then online_url
    my $online_base = '/mpiwg/online/permanent';
    my $online_dir = find_online_dir($input_node, $online_base, 'pageimg');
    if ((! $online_dir)) {
	logger('ERROR', "no ID or URL for online permanent entry");
	$errcnt++;
	return;
    }
    my $dir = sstrip("$online_base/$online_dir", 1);
    return $dir;
}

#
# $dir = find_online_dir($input_node, $base_dir, $page_dir)
#
# Takes the path from the $online_url_field of the $input_node document
# and looks in the directory $base_dir for it. Strips $page_dir from the end.
# Returns the directory path sans $base_dir if it exists
#
sub find_online_dir {
    my ($input_node, $base_dir, $page_dir) = @_;
    $base_dir = $lib_online_dir unless ($base_dir);

    my $online_url = $input_node->findvalue("fm:$online_url_field");
    logger('DEBUG', "checking URL: $online_url");
    my $online_dir;
    if ($online_url =~ /fn=permanent\/(.+)/) {
	# new style digilib URL
	$online_dir = $1;
    } elsif ($online_url =~ /\?([^\+]+)\+/) {
	# old style digilib URL
	$online_dir = $1;
    }
    #logger('DEBUG', "online_dir1: $online_dir");
    if ($online_dir) {
	$online_dir =~ s/\/$//; # strip ending slashes
	if ($page_dir) {
	    # strip page_dir
	    $online_dir =~ s/\/${page_dir}$//;
	}
	#logger("DEBUG", "dir: $base_dir/$online_dir");
	if (-d "$base_dir/$online_dir") {
	    logger('DEBUG', "directory $base_dir/$online_dir exists");
	    return $online_dir;
	}
    }
    return;
}

sub find_arch_dir {
    my ($input_node) = @_;
    my $dir = "";

    my $bib_dir = $input_node->findvalue("fm:$arch_id_field");
    #logger('DEBUG', "bibdir: $bib_dir");
    if ($bib_dir) {
	$dir = "$lib_arch_dir/$bib_dir";
	if (-d $dir) {
	    logger('DEBUG', "directory $dir exists"); 
	    return $dir;
	}
    }
    return;
}


sub convert_bib {
    my ($input_node, $index_root, $index_doc) = @_;
    my $cnt = 0;
    my $type = "";
    my $type_path = "";

    # process general stuff first
    foreach my $n ($input_node->getChildNodes()) {
	my $name = $n->nodeName();
	my $val = $n->textContent();
	#logger('DEBUG', "  NODE: $name = '$val'");
	if (exists $gen_map{$name}) {
	    # is a general field
	    if ($name eq $lang_field) {
		# language field -> convert to iso code
		if (exists $lang_map{$val}) {
		    $val = $lang_map{$val};
		} else {
		    logger('ERROR', "unknown language: $val! skipping...");
		    $errcnt++;
		    return 0;
		}
	    }
	    create_element_path($gen_map{$name}, $index_root, $namespace)
		->appendTextNode($val);
	    $cnt++;
	} elsif (exists $type_map{$name}) {
	    # is a type field
	    $type_path = $type_map{$name};
	    $type = $val;
	    # check with known types
	    if (exists $subtype_map{$val}) {
		my $indextype = $subtype_map{$val}->{'_name'};
		create_element_path("$type_path=$indextype", $index_root, $namespace);
		$cnt++;
	    } else {
		logger('ERROR', 'unknown bib type $val! skipping...');
		$errcnt++;
		return 0;
	    }
	}
    }
    # process sub type fields
    if ($type) {
	foreach my $n ($input_node->getChildNodes()) {
	    my $name = $n->nodeName();
	    my $val = $n->textContent();
	    #logger('DEBUG', "  NODE: $name = '$val'");
	    if (exists $subtype_map{$type}->{$name}) {
		create_element_path($subtype_map{$type}->{$name}, $index_root, $namespace)
		    ->appendTextNode($val);
		$cnt++;
	    }
	}
    }
    return $cnt;
}



sub process_all_fm_entries {
    my ($input_root) = @_;
    my $cnt = 0;

    foreach my $n ($input_root->findnodes('fm:ROW')) {
	logger('INFO', "processing entry $cnt ...");
	process_fm_entry($n);
	$cnt++;
    }
}    


sub process_fm_entry {
    my ($input_node) = @_;
    my $index_doc = XML::LibXML::Document->createDocument('1.0', 'UTF-8');
    my $index_root = $index_doc->createElementNS($namespace, 'resource');
    $index_root->addChild($index_doc->createAttributeNS($namespace, 'version', '1.1'));
    $index_root->addChild($index_doc->createAttributeNS($namespace, 'type', 'MPIWG'));
    $index_doc->setDocumentElement($index_root);
    my $derived_from = "";

    # try to find the document directory
    my $doc_dir = "";
    if ($online_mode) {
	$doc_dir = find_permanent_dir($input_node);
	$derived_from = find_arch_dir($input_node);
    } elsif ($cw_mode) {
	$doc_dir = find_cw_dir($input_node);
    } elsif ($digifiles_mode) {
	$doc_dir = find_digifiles_dir($input_node);
    } else {
	$doc_dir = find_arch_dir($input_node);
    }
    if (! $doc_dir) {
	logger('ERROR', "document directory not found! skipping...");
	$errcnt++;
	return;
    }

    # add standard stuff to index.meta
    my ($docname, $docpath) = split_file_path($doc_dir);
    # name and date
    create_text_path('name', $docname, $index_root, $namespace);
    create_text_path('archive-path', $doc_dir, $index_root, $namespace);
    create_text_path('archive-creation-date', stime(time), $index_root, $namespace);
    create_text_path('creator', 'digigroup', $index_root, $namespace);
    create_text_path('description', 'a scanned document', $index_root, $namespace);
    # acquisition
    create_text_path('meta/acquisition/date', stime(time), $index_root, $namespace);
    create_text_path('meta/acquisition/provider/provider-id', 'digigroup', $index_root, $namespace);
    create_text_path('meta/acquisition/provider/address', 'Max Planck Institute for the History of Science', $index_root, $namespace);
    # media
    create_text_path('media-type', 'image', $index_root, $namespace);
    create_text_path('meta/content-type', 'scanned document', $index_root, $namespace);
    # derived-from
    if ($derived_from) {
       create_text_path('derived-from/archive-path', $derived_from, $index_root, $namespace);
    }
    # access
    if ($access_type) {
	if ($access_type eq "free") {
	    create_element_path('meta/access-conditions/access@type=free', $index_root, $namespace);
	} else {
	    my $acc_tag = create_element_path('meta/access-conditions/access@type=institution', $index_root, $namespace);
	    create_text_path('name', $access_type, $acc_tag, $namespace);
	}
    }

    # convert bib entries
    my $cnt = convert_bib($input_node, $index_root, $index_doc);
    if ($cnt == 0) {
	# error or nothing to convert
	logger('ERROR', "no bibliographic metadata!");
	$errcnt++;
	return;
    }

    # write new index.meta file
    if ($dry_run) {
	logger('DEBUG', "would write $doc_dir/index.meta");
	logger('DEBUG', $index_doc->toString(1));
    } else {
	write_xml($index_doc, "$doc_dir/index.meta");
    }

}





#######################################################
# Main
#

# load filemaker xml dump
my ($input_doc, $input_root) = read_xml($infile);
# set namespace prefix
my $fm_namespace = $input_root->namespaceURI();
$input_root->setNamespace($fm_namespace, 'fm', 1);

# create digilib mapping file for digifiles mode
if ($digifiles_mode) {
    $mapping_doc = XML::LibXML::Document->createDocument('1.0', 'UTF-8');
    $mapping_root = $mapping_doc->createElementNS($namespace, 'digilib-aliases');
    $mapping_doc->setDocumentElement($mapping_root);
#<mapping link="exp1/archimedes_image_repository/archimedes_large/achil_propo_087_la_1545" dir="permanent/archimedes_repository/large/achil_propo_087_la_1545"/>

}

process_all_fm_entries($input_root);


logger("INFO", "$warncnt warnings");
logger("INFO", "$errcnt errors");
if ($errcnt > 0) {
    logger("ABORT", "there were errors!");
    exit 1;
} else {
    logger("DONE", "done something successfully!");
}