1 |
#!/usr/bin/perl -w |
#!/usr/bin/perl -w |
2 |
|
|
3 |
use strict; |
use strict; |
4 |
use OpenIsis; |
use Biblio::Isis; |
5 |
use Getopt::Std; |
use Getopt::Std; |
6 |
use Data::Dumper; |
use Data::Dumper; |
7 |
use XML::Simple; |
use XML::Simple; |
|
use Text::Unaccent 1.02; # 1.01 won't compile on my platform, |
|
8 |
use Text::Iconv; |
use Text::Iconv; |
9 |
use Config::IniFiles; |
use Config::IniFiles; |
10 |
use Encode; |
use Encode; |
16 |
|
|
17 |
my $config_file = $0; |
my $config_file = $0; |
18 |
$config_file =~ s/\.pl$/.conf/; |
$config_file =~ s/\.pl$/.conf/; |
19 |
$config_file = $ARGV[0] if (-f $ARGV[0]); |
$config_file = $ARGV[0] if ($ARGV[0] && -f $ARGV[0]); |
20 |
die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file); |
die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file); |
21 |
|
|
22 |
my $config; |
my $config; |
23 |
|
|
24 |
#use index_DBI; # default DBI module for index |
#use index_DBI; # default DBI module for index |
25 |
use index_DBI_cache; # faster DBI module using memory cache |
#use index_DBI_cache; # faster DBI module using memory cache |
26 |
|
use index_DBI_filter; # filter support for indexes |
27 |
my $index; |
my $index; |
28 |
|
|
29 |
my %opts; |
my %opts; |
140 |
} else { |
} else { |
141 |
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
142 |
} |
} |
143 |
|
|
144 |
if ($field_name) { |
if ($field_name) { |
145 |
|
$field_name = x($field_name); |
146 |
if (! $last_field_name) { |
if (! $last_field_name) { |
147 |
$last_field_name = x($field_name); |
$last_field_name = $field_name; |
148 |
return $last_field_name; |
return $last_field_name; |
149 |
} elsif ($field_name ne $last_field_name) { |
} elsif ($field_name ne $last_field_name) { |
150 |
$last_field_name = x($field_name); |
$last_field_name = $field_name; |
151 |
return $last_field_name; |
return $last_field_name; |
152 |
} |
} |
153 |
} |
} |
171 |
($s,$se,$d,$i) = (0,1,0,0); |
($s,$se,$d,$i) = (0,1,0,0); |
172 |
} elsif (lc($type) =~ /^lookup/) { |
} elsif (lc($type) =~ /^lookup/) { |
173 |
($s,$se,$d,$i,$il) = (0,1,0,0,1); |
($s,$se,$d,$i,$il) = (0,1,0,0,1); |
174 |
|
} elsif ($type) { |
175 |
|
print STDERR "WARNING: unknown type: $type\n"; |
176 |
} |
} |
177 |
return ($s,$se,$d,$i,$il); |
return ($s,$se,$d,$i,$il); |
178 |
} |
} |
258 |
# init vars so that we go into while... |
# init vars so that we go into while... |
259 |
($swish,$display) = (1,1); |
($swish,$display) = (1,1); |
260 |
|
|
|
# placeholder for all repeatable entries for index |
|
|
|
|
261 |
sub mkformat($$) { |
sub mkformat($$) { |
262 |
my $x = shift || die "mkformat needs tag reference"; |
my $x = shift || die "mkformat needs tag reference"; |
263 |
my $data = shift || return; |
my $data = shift || return; |
528 |
$swish_data =~ s/ +/ /g; |
$swish_data =~ s/ +/ /g; |
529 |
$swish_data =~ s/ +$//g; |
$swish_data =~ s/ +$//g; |
530 |
|
|
531 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
532 |
} |
} |
533 |
|
|
534 |
my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page]; |
my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page]; |
538 |
|
|
539 |
# add delimiters before and after word. |
# add delimiters before and after word. |
540 |
# That is required to produce exact match |
# That is required to produce exact match |
541 |
$xml .= xmlify($field."_swish_exact", unac_string($codepage,$swish_exact_data)); |
$xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data)); |
542 |
} |
} |
543 |
|
|
544 |
my $idel = $cache->{index_delimiter}->{$field}; |
my $idel = $cache->{index_delimiter}->{$field}; |
571 |
$swish_data =~ s/ +/ /g; |
$swish_data =~ s/ +/ /g; |
572 |
$swish_data =~ s/ +$//g; |
$swish_data =~ s/ +$//g; |
573 |
|
|
574 |
$xml .= xmlify($field."_swish", unac_string($codepage,$swish_data)); |
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
575 |
} |
} |
576 |
|
|
577 |
if ($swish_exact_data) { |
if ($swish_exact_data) { |
580 |
|
|
581 |
# add delimiters before and after word. |
# add delimiters before and after word. |
582 |
# That is required to produce exact match |
# That is required to produce exact match |
583 |
$xml .= xmlify($field."_swish_exact", unac_string($codepage,$swish_exact_data)); |
$xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data)); |
584 |
} |
} |
585 |
} |
} |
586 |
} |
} |
617 |
|
|
618 |
my $show_progress = $cfg_global->val('global', 'show_progress'); |
my $show_progress = $cfg_global->val('global', 'show_progress'); |
619 |
|
|
620 |
my $unac_filter = $cfg_global->val('global', 'unac_filter'); |
my $my_unac_filter = $cfg_global->val('global', 'my_unac_filter'); |
621 |
if ($unac_filter) { |
if ($my_unac_filter) { |
622 |
require $unac_filter; |
print STDERR "using $my_unac_filter to filter characters for search\n"; |
623 |
|
require $my_unac_filter; |
624 |
|
} else { |
625 |
|
print STDERR "### fallback to default my_unac_string!\n"; |
626 |
|
eval q{ |
627 |
|
sub main::my_unac_string($$) { |
628 |
|
my ($charset, $string) = (@_); |
629 |
|
return $string; |
630 |
|
} |
631 |
|
}; |
632 |
} |
} |
633 |
|
|
634 |
foreach my $database ($cfg->Sections) { |
foreach my $database ($cfg->Sections) { |
635 |
|
|
636 |
|
# save database name in global variable path for later |
637 |
|
# (need for index filter creation) |
638 |
|
$path = $database; |
639 |
|
|
640 |
my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined"; |
my $type = lc($cfg -> val($database, 'type')) || die "$database doesn't have 'type' defined"; |
641 |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
642 |
|
|
644 |
my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional |
my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional |
645 |
if ($lookup_file) { |
if ($lookup_file) { |
646 |
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644; |
#tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644; |
647 |
|
if (! -e $lookup_file) { |
648 |
|
open(LOOKUP, "> $lookup_file") || die "can't create $lookup_file': $!"; |
649 |
|
close(LOOKUP); |
650 |
|
} |
651 |
tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644; |
tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644; |
652 |
print STDERR "creating lookup file '$lookup_file'\n"; |
print STDERR "creating lookup file '$lookup_file'\n"; |
653 |
# delete memory cache for lookup file |
# delete memory cache for lookup file |
670 |
|
|
671 |
$config=XMLin("./import_xml/$type.xml", ForceArray => [ $type2tag{$type_base}, 'config', 'format' ], ForceContent => 1 ); |
$config=XMLin("./import_xml/$type.xml", ForceArray => [ $type2tag{$type_base}, 'config', 'format' ], ForceContent => 1 ); |
672 |
|
|
673 |
|
# helper for progress bar |
674 |
|
sub fmt_time { |
675 |
|
my $t = shift || 0; |
676 |
|
my $out = ""; |
677 |
|
|
678 |
|
my ($ss,$mm,$hh) = gmtime($t); |
679 |
|
$out .= "${hh}h" if ($hh); |
680 |
|
$out .= sprintf("%02d:%02d", $mm,$ss); |
681 |
|
$out .= " " if ($hh == 0); |
682 |
|
return $out; |
683 |
|
} |
684 |
|
|
685 |
# output current progress indicator |
# output current progress indicator |
686 |
my $last_p = 0; |
my $last_p = 0; |
687 |
|
my $start_t = time(); |
688 |
sub progress { |
sub progress { |
689 |
return if (! $show_progress); |
return if (! $show_progress); |
690 |
my $current = shift; |
my $current = shift; |
691 |
my $total = shift || 1; |
my $total = shift || 1; |
692 |
my $p = int($current * 100 / $total); |
my $p = int($current * 100 / $total); |
693 |
if ($p != $last_p) { |
if ($p < $last_p || $current == 1) { |
694 |
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p ); |
$start_t = time(); |
695 |
|
$last_p = 0; |
696 |
|
} elsif ($p != $last_p) { |
697 |
|
my $rate = ($current / (time() - $start_t || 1)); |
698 |
|
my $eta = ($total-$current) / ($rate || 1); |
699 |
|
printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$current,"=" x ($p/3)."$p%>", $total, $rate, fmt_time($eta)); |
700 |
$last_p = $p; |
$last_p = $p; |
701 |
} |
} |
702 |
} |
} |
703 |
|
|
704 |
my $fake_dir = 1; |
my $fake_dir = 1; |
705 |
|
my $fake_pos = 0; |
706 |
|
my $last_fake_t = time(); |
707 |
sub fakeprogress { |
sub fakeprogress { |
708 |
return if (! $show_progress); |
return if (! $show_progress); |
709 |
my $current = shift @_; |
my $current = shift @_; |
710 |
|
|
711 |
my @ind = ('-','\\','|','/','-','\\','|','/', '-'); |
my @ind = ('-','\\','|','/','-','\\','|','/'); |
712 |
|
|
713 |
|
if ($current < $fake_pos) { |
714 |
|
$start_t = time(); |
715 |
|
$last_fake_t = 0; |
716 |
|
$fake_dir = 1; |
717 |
|
$fake_pos = 0; |
718 |
|
} |
719 |
|
|
720 |
$last_p += $fake_dir; |
if (time()-$last_fake_t >= 1) { |
721 |
$fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0); |
$last_fake_t = time(); |
722 |
if ($last_p % 10 == 0) { |
$fake_pos += $fake_dir; |
723 |
printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]); |
$fake_dir = -$fake_dir if ($fake_pos > 38); |
724 |
|
} |
725 |
|
|
726 |
|
if ($current % 10 == 0) { |
727 |
|
my $rate = ($current / (time() - $start_t || 1)); |
728 |
|
printf STDERR ("%5d [%-38s] %0.1f/s\r",$current, " " x $fake_pos .$ind[($current / 10) % 8], $rate); |
729 |
} |
} |
730 |
} |
} |
731 |
|
|
740 |
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!"; |
my $isis_db = $cfg -> val($database, 'isis_db') || die "$database doesn't have 'isis_db' defined!"; |
741 |
|
|
742 |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
743 |
my $db = OpenIsis::open( $isis_db ); |
my $db = new Biblio::Isis( isisdb => $isis_db ); |
|
|
|
|
# check if .txt database for OpenIsis is zero length, |
|
|
# if so, erase it and re-open database |
|
|
sub check_txt_db { |
|
|
my $isis_db = shift || die "need isis database name"; |
|
|
my $reopen = 0; |
|
|
|
|
|
if (-e $isis_db.".TXT") { |
|
|
print STDERR "WARNING: removing $isis_db.TXT OpenIsis database...\n"; |
|
|
unlink $isis_db.".TXT" || warn "FATAL: unlink error on '$isis_db.TXT': $!"; |
|
|
$reopen++; |
|
|
} |
|
|
if (-e $isis_db.".PTR") { |
|
|
print STDERR "WARNING: removing $isis_db.PTR OpenIsis database...\n"; |
|
|
unlink $isis_db.".PTR" || warn "FATAL: unlink error on '$isis_db.PTR': $!"; |
|
|
$reopen++; |
|
|
} |
|
|
return OpenIsis::open( $isis_db ) if ($reopen); |
|
|
} |
|
|
|
|
|
# EOF error |
|
|
if ($db == -1) { |
|
|
$db = check_txt_db($isis_db); |
|
|
if ($db == -1) { |
|
|
print STDERR "FATAL: OpenIsis can't open zero size file $isis_db\n"; |
|
|
next; |
|
|
} |
|
|
} |
|
744 |
|
|
745 |
# OpenIsis::ERR_BADF |
if (! $db) { |
746 |
if ($db == -4) { |
print STDERR "FATAL: can't read ISIS database: $isis_db, skipping...\n"; |
|
print STDERR "FATAL: OpenIsis can't find file $isis_db\n"; |
|
|
next; |
|
|
# OpenIsis::ERR_IO |
|
|
} elsif ($db == -5) { |
|
|
print STDERR "FATAL: OpenIsis can't access file $isis_db\n"; |
|
|
next; |
|
|
} elsif ($db < 0) { |
|
|
print STDERR "FATAL: OpenIsis unknown error $db with file $isis_db\n"; |
|
747 |
next; |
next; |
748 |
} |
} |
749 |
|
|
750 |
my $max_rowid = OpenIsis::maxRowid( $db ); |
my $max_rowid = $db->count if ($db); |
|
|
|
|
# if 0 records, try to rease isis .txt database |
|
|
if ($max_rowid == 0) { |
|
|
# force removal of database |
|
|
$db = check_txt_db($isis_db); |
|
|
$max_rowid = OpenIsis::maxRowid( $db ); |
|
|
} |
|
751 |
|
|
752 |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
753 |
|
|
|
my $path = $database; |
|
|
|
|
754 |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
755 |
my $row = OpenIsis::read( $db, $row_id ); |
my $row = $db->to_hash( $row_id ); |
756 |
if ($row && $row->{mfn}) { |
if ($row) { |
757 |
|
|
758 |
|
$row->{mfn} = $row_id; |
759 |
|
$row->{record} = $db->{record}; |
760 |
|
|
761 |
progress($row->{mfn}, $max_rowid); |
progress($row->{mfn}, $max_rowid); |
762 |
|
|
763 |
my $swishpath = $path."#".int($row->{mfn}); |
my $swishpath = $path."#".int($row->{mfn}); |
771 |
} |
} |
772 |
} |
} |
773 |
} |
} |
|
# for this to work with current version of OpenIsis (0.9.0) |
|
|
# you might need my patch from |
|
|
# http://www.rot13.org/~dpavlin/projects/openisis-0.9.0-perl_close.diff |
|
|
OpenIsis::close($db); |
|
774 |
print STDERR "\n"; |
print STDERR "\n"; |
775 |
|
|
776 |
} elsif ($type_base eq "excel") { |
} elsif ($type_base eq "excel") { |
804 |
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
for(my $iC = $oWorksheet->{MinCol} ; defined $oWorksheet->{MaxCol} && $iC <= $oWorksheet->{MaxCol} ; $iC++) { |
805 |
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
my $cell = $oWorksheet->{Cells}[$iR][$iC]; |
806 |
if ($cell) { |
if ($cell) { |
807 |
$row->{int2col($iC)} = $cell->Value; |
# this conversion is a cludge. |
808 |
|
# Files from Excell could have |
809 |
|
# characters which don't fit into |
810 |
|
# destination encoding. |
811 |
|
$row->{int2col($iC)} = $utf2cp->convert($cell->Value) || $cell->Value; |
812 |
} |
} |
813 |
} |
} |
814 |
|
|
832 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
833 |
} |
} |
834 |
} |
} |
835 |
|
|
836 |
|
print STDERR "\n"; |
837 |
|
|
838 |
} elsif ($type_base eq "marc") { |
} elsif ($type_base eq "marc") { |
839 |
|
|
840 |
require MARC; |
require MARC::File::USMARC; |
841 |
|
|
842 |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
843 |
my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!"; |
my $marc_file = $cfg -> val($database, 'marc_file') || die "$database doesn't have 'marc_file' defined!"; |
844 |
|
|
845 |
# optional argument is format |
# optional argument is format |
846 |
my $format = x($config->{marc_format}) || 'usmarc'; |
warn "marc_format is no longer used!" if ($config->{marc_format}); |
|
|
|
847 |
print STDERR "Reading MARC file '$marc_file'\n"; |
print STDERR "Reading MARC file '$marc_file'\n"; |
848 |
|
|
849 |
my $marc = new MARC; |
my $marc = MARC::File::USMARC->in( $marc_file ); |
850 |
my $nr = $marc->openmarc({ |
|
851 |
file=>$marc_file, format=>$format |
if (! $marc) { |
852 |
}) || die "Can't open MARC file '$marc_file' with format '$format'"; |
print STDERR "FATAL: can't read MARC file: $marc_file, skipping...\n"; |
853 |
|
next; |
854 |
|
} |
855 |
|
|
856 |
# read MARC file in memory |
# count records in MARC file |
857 |
$marc->nextmarc(-1); |
sub marc_count { |
858 |
|
my $filename = shift || die; |
859 |
|
my $file = MARC::File::USMARC->in($filename) || return; |
860 |
|
my $count = 0; |
861 |
|
while ($file->skip()) { |
862 |
|
$count++; |
863 |
|
} |
864 |
|
return $count; |
865 |
|
} |
866 |
|
|
867 |
my $max_rec = $marc->marc_count(); |
my $count = marc_count($marc_file) || warn "no records in '$marc_file'?"; |
868 |
|
|
869 |
for(my $i=1; $i<=$max_rec; $i++) { |
my $i = 1; |
870 |
|
|
871 |
progress($i,$max_rec); |
while( my $rec = $marc->next() ) { |
872 |
|
|
873 |
# store value for marc_sf.pm |
progress($i,$count); |
|
$main::cache->{marc_record} = $i; |
|
874 |
|
|
875 |
my $swishpath = $database."#".$i; |
my $swishpath = $database."#".$i; |
876 |
|
|
877 |
if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) { |
if (my $xml = data2xml($type_base,$rec,$add_xml,$cfg,$database)) { |
878 |
$xml = $cp2utf->convert($xml); |
$xml = $cp2utf->convert($xml); |
879 |
use bytes; # as opposed to chars |
use bytes; # as opposed to chars |
880 |
print "Path-Name: $swishpath\n"; |
print "Path-Name: $swishpath\n"; |
881 |
print "Content-Length: ".(length($xml)+1)."\n"; |
print "Content-Length: ".(length($xml)+1)."\n"; |
882 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
883 |
} |
} |
884 |
|
|
885 |
|
$i++; |
886 |
} |
} |
887 |
|
|
888 |
print STDERR "\n"; |
print STDERR "\n"; |
947 |
|
|
948 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
949 |
|
|
950 |
This command will read ISIS data file using OpenIsis perl module, MARC |
This command will read ISIS data file using Biblio::Isis perl module, MARC |
951 |
records using MARC module and optionally Micro$oft Excel files to |
records using MARC::File module and optionally Micro$oft Excel files to |
952 |
create one XML file for usage with I<SWISH-E> indexer. Dispite it's name, |
create one XML file for usage with I<SWISH-E> indexer. Dispite it's name, |
953 |
this script B<isn't general xml generator> from isis files (isis allready |
this script B<isn't general xml generator> from isis files (isis allready |
954 |
has something like that). Output of this script is tailor-made for SWISH-E. |
has something like that). Output of this script is tailor-made for SWISH-E. |