1 |
#!/usr/bin/perl -w |
#!/usr/bin/perl -w |
2 |
|
|
3 |
use strict; |
=head1 NAME |
|
use OpenIsis; |
|
|
use Getopt::Std; |
|
|
use Data::Dumper; |
|
|
use XML::Simple; |
|
|
use Text::Unaccent 1.02; # 1.01 won't compile on my platform, |
|
|
use Text::Iconv; |
|
|
use Config::IniFiles; |
|
|
use Encode; |
|
|
|
|
|
$|=1; |
|
|
|
|
|
my $config_file = $0; |
|
|
$config_file =~ s/\.pl$/.conf/; |
|
|
die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file); |
|
|
|
|
|
my $config; |
|
|
|
|
|
use index_DBI; # there is no other, right now ;-) |
|
|
|
|
|
my $index = new index_DBI(); # open index |
|
|
|
|
|
my %opts; |
|
|
|
|
|
# usage: |
|
|
# -d directory name |
|
|
# -m multiple directories |
|
|
# -q quiet |
|
|
# -s run swish |
|
|
|
|
|
getopts('d:m:qs', \%opts); |
|
|
|
|
|
my $db_dir; |
|
|
|
|
|
Text::Iconv->raise_error(1); # Conversion errors raise exceptions |
|
4 |
|
|
5 |
# this is encoding of all files on disk, including import_xml/*.xml file and |
all2all.pl - basic script for all WebPAC needs |
|
# filter/*.pm files! It will be used to store strings in perl internally! |
|
|
my $codepage = 'ISO-8859-2'; |
|
6 |
|
|
7 |
my $utf2cp = Text::Iconv->new('UTF-8',$codepage); |
=cut |
|
# this function will convert data from XML files to local encoding |
|
|
sub x { |
|
|
return $utf2cp->convert($_[0]); |
|
|
} |
|
8 |
|
|
9 |
# decode isis import codepage |
use strict; |
10 |
my $isis2cp; |
use Data::Dumper; |
11 |
|
use Carp; |
12 |
|
|
13 |
# outgoing xml must be in UTF-8 |
use lib './lib'; |
14 |
my $cp2utf = Text::Iconv->new($codepage,'UTF-8'); |
use WebPAC; |
15 |
|
|
16 |
sub isis2xml { |
my $webpac = new WebPAC( |
17 |
|
code_page => 'ISO-8859-2', |
18 |
|
limit_mfn => 500, |
19 |
|
) || die; |
20 |
|
|
21 |
use xmlify; |
$|=1; |
|
|
|
|
my $row = shift @_; |
|
|
my $add_xml = shift @_; |
|
|
|
|
|
my $xml; |
|
|
|
|
|
use parse_format; |
|
|
|
|
|
my $html = ""; # html formatted display output |
|
|
|
|
|
my %field_usage; # counter for usage of each field |
|
|
|
|
|
# sort subrouting using order="" attribute |
|
|
sub by_order { |
|
|
return 0 if (! $config->{indexer}->{$a}->{order}); |
|
|
return 0 if (! $config->{indexer}->{$b}->{order}); |
|
|
|
|
|
return $config->{indexer}->{$a}->{order} <=> |
|
|
$config->{indexer}->{$b}->{order} ; |
|
|
} |
|
|
|
|
|
foreach my $field (sort by_order keys %{$config->{indexer}}) { |
|
|
|
|
|
$field=x($field); |
|
|
|
|
|
$field_usage{$field}++; |
|
|
|
|
|
my $swish_data = ""; |
|
|
my $display_data = ""; |
|
|
my $line_delimiter; |
|
|
|
|
|
my ($swish,$display); |
|
|
|
|
|
foreach my $x (@{$config->{indexer}->{$field}->{isis}}) { |
|
|
|
|
|
my $format = x($x->{content}); |
|
|
my $delimiter = x($x->{delimiter}) || ' '; |
|
|
|
|
|
my $isis_i = 0; # isis repeatable offset |
|
|
|
|
|
my ($s,$d,$i) = (1,1,0); # swish, display default |
|
|
$s = 0 if (lc($x->{type}) eq "display"); |
|
|
$d = 0 if (lc($x->{type}) eq "swish"); |
|
|
($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index"); |
|
|
|
|
|
# what will separate last line from this one? |
|
|
if ($display_data && $x->{append} && $x->{append} eq "1") { |
|
|
$line_delimiter = ' '; |
|
|
} elsif ($display_data) { |
|
|
$line_delimiter = '<br/>'; |
|
|
} |
|
|
|
|
|
# init vars so that we go into while... |
|
|
($swish,$display) = (1,1); |
|
|
|
|
|
while ($swish || $display) { |
|
|
($swish,$display) = parse_format($format,$row,$isis_i++,$isis2cp); |
|
|
|
|
|
# filter="name" ; filter this field through |
|
|
# filter/[name].pm |
|
|
my $filter = $x->{filter}; |
|
|
if ($filter) { |
|
|
require "filter/".$filter.".pm"; |
|
|
} |
|
|
# type="swish" ; field for swish |
|
|
if ($s && $swish) { |
|
|
if ($filter) { |
|
|
no strict 'refs'; |
|
|
$swish_data .= join(" ",&$filter($swish)); |
|
|
print STDERR "#### $swish_data\n"; |
|
|
} else { |
|
|
$swish_data .= $swish; |
|
|
} |
|
|
} |
|
|
|
|
|
# type="display" ; field for display |
|
|
if ($d && $display) { |
|
|
if ($line_delimiter && $display_data) { |
|
|
$display_data .= $line_delimiter; |
|
|
undef $line_delimiter; |
|
|
} |
|
|
if ($filter) { |
|
|
no strict 'refs'; |
|
|
$display_data .= join($delimiter,&$filter($display)); |
|
|
} else { |
|
|
if ($display_data) { |
|
|
$display_data .= $delimiter.$display; |
|
|
} else { |
|
|
$display_data .= $display; |
|
|
} |
|
|
} |
|
|
} |
|
|
|
|
|
# type="index" ; insert into index |
|
|
if ($i && $display) { |
|
|
my $index_data = $display; |
|
|
if ($filter) { |
|
|
no strict 'refs'; |
|
|
foreach my $d (&$filter($index_data)) { |
|
|
$index->insert($field, $d, $db_dir); |
|
|
} |
|
|
} else { |
|
|
$index->insert($field, $index_data, $db_dir); |
|
|
} |
|
|
} |
|
|
} |
|
|
} |
|
|
|
|
|
|
|
|
if ($display_data) { |
|
|
|
|
|
if ($field eq "headline") { |
|
|
$xml .= xmlify("headline", $display_data); |
|
|
} else { |
|
|
|
|
|
# find field name (signular, plural) |
|
|
my $field_name = ""; |
|
|
if ($config->{indexer}->{$field}->{name_singular} && $field_usage{$field} == 1) { |
|
|
$field_name = $config->{indexer}->{$field}->{name_singular}."#-#"; |
|
|
} elsif ($config->{indexer}->{$field}->{name_plural}) { |
|
|
$field_name = $config->{indexer}->{$field}->{name_plural}."#-#"; |
|
|
} elsif ($config->{indexer}->{$field}->{name}) { |
|
|
$field_name = $config->{indexer}->{$field}->{name}."#-#"; |
|
|
} else { |
|
|
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
|
|
} |
|
|
if ($field_name) { |
|
|
$html .= x($field_name); |
|
|
} |
|
|
$html .= $display_data."###\n"; |
|
|
} |
|
|
} |
|
|
if ($swish_data) { |
|
|
# remove extra spaces |
|
|
$swish_data =~ s/ +/ /g; |
|
|
$swish_data =~ s/ +$//g; |
|
|
|
|
|
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
|
|
} |
|
|
|
|
|
|
|
|
} |
|
|
|
|
|
# dump formatted output in <html> |
|
|
if ($html) { |
|
|
$xml .= xmlify("html",$html); |
|
|
} |
|
|
|
|
|
if ($xml) { |
|
|
$xml .= $add_xml if ($add_xml); |
|
|
return "<xml>\n$xml</xml>\n"; |
|
|
} else { |
|
|
return; |
|
|
} |
|
|
} |
|
22 |
|
|
23 |
########################################################################## |
print "reading database\n"; |
24 |
|
my $maxmfn = $webpac->open_isis( |
25 |
|
filename => shift @ARGV || '/data/hidra/THS/THS', |
26 |
|
lookup => [ |
27 |
|
{ 'key' => 'd:v900', 'val' => 'v250^a' }, |
28 |
|
# { 'eval' => '"v901^a" eq "Područje"', 'key' => 'pa:v561^4:v562^4:v461^1', 'val' => 'v900' }, |
29 |
|
# { 'eval '=> '"v901^a" eq "Mikrotezaurus"', 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' }, |
30 |
|
# { 'eval' => '"v901^a" eq "Deskriptor"', 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' }, |
31 |
|
{ 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' }, |
32 |
|
], |
33 |
|
); |
34 |
|
|
35 |
|
print "rows: $maxmfn\n\n"; |
36 |
|
|
37 |
|
for (my $mfn = 1; $mfn <= $maxmfn; $mfn++) { |
38 |
|
my $rec = $webpac->{'data'}->{$mfn} || die "no record with mfn $mfn"; |
39 |
|
|
40 |
|
print "-- ",$webpac->fill_in($rec,'v250^a (v901^a)'),"\n"; |
41 |
|
|
42 |
|
my @t = $webpac->fill_in($rec,'v553^1;;v553^a'); |
43 |
|
print " Uži pojam: ",join("\t\n",@t),"\n" if (@t); |
44 |
|
|
45 |
my $cfg = new Config::IniFiles( -file => $config_file ); |
@t = $webpac->fill_in($rec,'[a:v251::];;[d:[a:v251::]]'); |
46 |
|
print " Područje: ",join("\t\n",@t),"\n" if (@t); |
47 |
|
|
48 |
foreach my $database ($cfg->Sections) { |
@t = $webpac->fill_in($rec,'[a:v561^4:v251:];;[d:[a:v561^4:v251:]]'); |
49 |
|
print " Mikrotezaurus: ",join("\t\n",@t),"\n" if (@t); |
50 |
|
|
51 |
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!"; |
@t = $webpac->fill_in($rec,'[a:v561^4:v562^4:v900];;[d:[a:v561^4:v562^4:v900]]'); |
52 |
my $type = $cfg -> val($database, 'type') || die "$database doesn't have 'type' defined"; |
print " Deskriptor: ",join("\t\n",@t),"\n" if (@t); |
|
my $add_xml = $cfg -> val($database, 'xml'); # optional |
|
|
|
|
|
$config=XMLin("./import_xml/$type.xml", forcearray => [ 'isis' ], forcecontent => 1); |
|
|
|
|
|
$isis2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
|
|
|
|
|
my $db = OpenIsis::open( $isis_db ); |
|
|
if (0) { |
|
|
# # FIX |
|
|
# if (! $db ) { |
|
|
print STDERR "WARNING: can't open '$isis_db'\n"; |
|
|
next ; |
|
|
} |
|
|
|
|
|
my $max_rowid = OpenIsis::maxRowid( $db ); |
|
|
|
|
|
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
|
|
|
|
|
my $path = $database; # was $isis_db |
|
|
|
|
|
my $last_p = 0; |
|
|
|
|
|
# { my $row_id = 4514; |
|
|
# FIX |
|
|
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
|
|
my $row = OpenIsis::read( $db, $row_id ); |
|
|
if ($row && $row->{mfn}) { |
|
|
# output current process indicator |
|
|
my $p = int($row->{mfn} * 100 / $max_rowid); |
|
|
if ($p != $last_p) { |
|
|
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$row->{mfn},$max_rowid,"=" x ($p/2).">", $p ) if (! $opts{q}); |
|
|
$last_p = $p; |
|
|
} |
|
|
|
|
|
if (my $xml = $cp2utf->convert(isis2xml($row,$add_xml))) { |
|
|
use bytes; # as opposed to chars |
|
|
print "Path-Name: $path#".int($row->{mfn})."\n"; |
|
|
print "Content-Length: ".(length($xml)+1)."\n"; |
|
|
print "Document-Type: XML\n\n$xml\n"; |
|
|
} |
|
|
} |
|
|
} |
|
|
print STDERR "\n"; |
|
53 |
} |
} |
54 |
|
|
55 |
# call this to commit index |
print "## lookup ",Dumper($webpac->{'lookup'}); |
56 |
$index->close; |
print "## data ",Dumper($webpac->{'data'}); |
|
|
|
|
1; |
|
|
__END__ |
|
|
########################################################################## |
|
|
|
|
|
=head1 NAME |
|
|
|
|
|
isis2xml.pl - read isis file and dump XML |
|
|
|
|
|
=head1 DESCRIPTION |
|
|
|
|
|
This command will read ISIS data file using OpenIsis perl module and |
|
|
create XML file for usage with I<SWISH-E> |
|
|
indexer. Dispite it's name, this script B<isn't general xml generator> |
|
|
from isis files (isis allready has something like that). Output of this |
|
|
script is tailor-made for SWISH-E. |
|
|
|
|
|
=head1 AUTHOR |
|
|
|
|
|
Dobrica Pavlinusic <dpavlin@rot13.org> |
|
|
|
|
|
=head1 COPYRIGHT |
|
|
|
|
|
GNU Public License (GPL) v2 or later |
|
|
|
|
|
=head1 SEE ALSO |
|
|
|
|
|
SWISH-E web site at http://www.swish-e.org |
|
|
|
|
|
=cut |
|