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; |
11 |
#use GDBM_File; |
#use GDBM_File; |
12 |
use Fcntl; # for O_RDWR |
use Fcntl; # for O_RDWR |
13 |
use TDB_File; |
use TDB_File; |
14 |
|
use Carp; |
15 |
|
|
16 |
$|=1; |
$|=1; |
17 |
|
|
23 |
my $config; |
my $config; |
24 |
|
|
25 |
#use index_DBI; # default DBI module for index |
#use index_DBI; # default DBI module for index |
26 |
use index_DBI_cache; # faster DBI module using memory cache |
#use index_DBI_cache; # faster DBI module using memory cache |
27 |
|
use index_DBI_filter; # filter support for indexes |
28 |
my $index; |
my $index; |
29 |
|
|
30 |
my %opts; |
my %opts; |
63 |
'isis' => 'isis', |
'isis' => 'isis', |
64 |
'excel' => 'column', |
'excel' => 'column', |
65 |
'marc' => 'marc', |
'marc' => 'marc', |
66 |
'feed' => 'feed' |
'feed' => 'feed', |
67 |
); |
); |
68 |
|
|
69 |
my $cache; # for cacheing |
my $cache; # for cacheing |
76 |
|
|
77 |
my $last_field_name; # cache to prevent repeated fields |
my $last_field_name; # cache to prevent repeated fields |
78 |
|
|
79 |
|
my $broken_cdata = XMLin('<foo><![CDATA[<bar>]]></foo>') eq '<bar>>'; |
80 |
|
warn "XML::Simple on this system seems broken with <![CDATA[..]]>.\n" if ($broken_cdata); |
81 |
|
|
82 |
sub data2xml { |
sub data2xml { |
83 |
|
|
84 |
use xmlify; |
use xmlify; |
116 |
$cache->{tags_by_order} = \@sorted_tags; |
$cache->{tags_by_order} = \@sorted_tags; |
117 |
} |
} |
118 |
|
|
119 |
|
if (! @sorted_tags) { |
120 |
|
print STDERR "WARNING: no tags for this type found in import_xml file!\n"; |
121 |
|
} |
122 |
|
|
123 |
# lookup key |
# lookup key |
124 |
my $lookup_key; |
my $lookup_key; |
125 |
|
|
129 |
delete $cache->{swish_exact_data}; |
delete $cache->{swish_exact_data}; |
130 |
delete $cache->{index_data}; |
delete $cache->{index_data}; |
131 |
delete $cache->{index_delimiter}; |
delete $cache->{index_delimiter}; |
132 |
|
delete $cache->{distinct}; |
133 |
my @page_fields; # names of fields |
my @page_fields; # names of fields |
134 |
|
|
135 |
|
|
149 |
} else { |
} else { |
150 |
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
print STDERR "WARNING: field '$field' doesn't have 'name' attribute!"; |
151 |
} |
} |
152 |
|
|
153 |
if ($field_name) { |
if ($field_name) { |
154 |
|
$field_name = x($field_name); |
155 |
if (! $last_field_name) { |
if (! $last_field_name) { |
156 |
$last_field_name = x($field_name); |
$last_field_name = $field_name; |
157 |
return $last_field_name; |
return $last_field_name; |
158 |
} elsif ($field_name ne $last_field_name) { |
} elsif ($field_name ne $last_field_name) { |
159 |
$last_field_name = x($field_name); |
$last_field_name = $field_name; |
160 |
return $last_field_name; |
return $last_field_name; |
161 |
} |
} |
162 |
} |
} |
210 |
delete $x->{value}; |
delete $x->{value}; |
211 |
delete $x->{delimiter}; |
delete $x->{delimiter}; |
212 |
$x->{content} = $v; |
$x->{content} = $v; |
213 |
|
$d =~ s#>$## if ($d && $broken_cdata); |
214 |
$x->{delimiter} = $d; |
$x->{delimiter} = $d; |
215 |
} |
} |
216 |
return $x; |
return $x; |
230 |
|
|
231 |
my ($swish,$display); |
my ($swish,$display); |
232 |
|
|
233 |
my $tag = $type2tag{$type} || die "can't find which tag to use for type $type"; |
my $tag = $cfg->val($database, 'import_xml_tag') || $type2tag{$type} || die "can't find which tag to use for type $type"; |
234 |
|
|
235 |
# is this field page-by-page? |
# is this field page-by-page? |
236 |
my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page}; |
my $iterate_by_page = $config->{indexer}->{$field}->{iterate_by_page}; |
239 |
# default line_delimiter if using |
# default line_delimiter if using |
240 |
my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>'; |
my $page_line_delimiter = $config->{indexer}->{$field}->{page_line_delimiter} || '<br/>'; |
241 |
$cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter}; |
$cache->{index_delimiter}->{$field} = $config->{indexer}->{$field}->{index_delimiter}; |
242 |
|
my $distinct = $config->{indexer}->{$field}->{distinct}; |
243 |
|
if ($distinct && !$iterate_by_page) { |
244 |
|
warn "WARNING: distinct is currently not supported without iterate_by_page!\n"; |
245 |
|
$distinct = 0; |
246 |
|
} |
247 |
|
|
248 |
my $format_name = $config->{indexer}->{$field}->{format_name}; |
my $format_name = $config->{indexer}->{$field}->{format_name}; |
249 |
my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter}; |
my $format_delimiter = $config->{indexer}->{$field}->{format_delimiter}; |
273 |
# init vars so that we go into while... |
# init vars so that we go into while... |
274 |
($swish,$display) = (1,1); |
($swish,$display) = (1,1); |
275 |
|
|
|
# placeholder for all repeatable entries for index |
|
|
|
|
276 |
sub mkformat($$) { |
sub mkformat($$) { |
277 |
my $x = shift || die "mkformat needs tag reference"; |
my $x = shift || die "mkformat needs tag reference"; |
278 |
my $data = shift || return; |
my $data = shift || return; |
422 |
$ldel = " " if ($append); |
$ldel = " " if ($append); |
423 |
#print STDERR "line delimiter: ",Dumper($ldel) if ($ldel); |
#print STDERR "line delimiter: ",Dumper($ldel) if ($ldel); |
424 |
if (! $cache->{$what}->{$field}->[$page]) { |
if (! $cache->{$what}->{$field}->[$page]) { |
425 |
$cache->{$what}->{$field}->[$page] = $data; |
push @{$cache->{$what}->{$field}->[$page]}, { |
426 |
} else { |
data => $data, |
427 |
$cache->{$what}->{$field}->[$page] .= $ldel.$data; |
delimiter => $ldel, |
428 |
|
}; |
429 |
} |
} |
430 |
} |
} |
431 |
|
|
492 |
if ($val) { |
if ($val) { |
493 |
$display_data .= $delimiter.$val if ($d); |
$display_data .= $delimiter.$val if ($d); |
494 |
$swish_data .= " ".$val if ($s); |
$swish_data .= " ".$val if ($s); |
495 |
$index->insert($field, $val, $path) if ($i); |
$index->insert($field, $val, $val, $path) if ($i); |
496 |
} |
} |
497 |
|
|
498 |
if ($iterate_by_page) { |
if ($iterate_by_page) { |
500 |
# on first page!!! |
# on first page!!! |
501 |
my $page = 0; |
my $page = 0; |
502 |
if ($display_data) { |
if ($display_data) { |
503 |
$cache->{display_data}->{$field}->[$page] = $display_data; |
push @{$cache->{display_data}->{$field}->[$page]}, { data => $display_data }; |
504 |
$display_data = ""; |
$display_data = ""; |
505 |
} |
} |
506 |
if ($swish_data) { |
if ($swish_data) { |
507 |
$cache->{swish_data}->{$field}->[$page] = $swish_data; |
push @{$cache->{swish_data}->{$field}->[$page]}, { data => $swish_data }; |
508 |
$swish_data = ""; |
$swish_data = ""; |
509 |
} |
} |
510 |
if ($swish_exact_data) { |
if ($swish_exact_data) { |
511 |
$cache->{swish_exact_data}->{$field}->[$page] = $swish_exact_data; |
push @{$cache->{swish_exact_data}->{$field}->[$page]}, { data => $swish_exact_data }; |
512 |
$swish_exact_data = ""; |
$swish_exact_data = ""; |
513 |
} |
} |
514 |
} |
} |
519 |
my $nr_pages = $page_max{$field} || next; |
my $nr_pages = $page_max{$field} || next; |
520 |
#print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n"; |
#print STDERR "field '$field' iterate over ",($nr_pages || 0)," pages...\n"; |
521 |
#print STDERR Dumper($cache->{display_data}); |
#print STDERR Dumper($cache->{display_data}); |
522 |
|
my $seen; # used for distinct |
523 |
for (my $page=0; $page <= $nr_pages; $page++) { |
for (my $page=0; $page <= $nr_pages; $page++) { |
524 |
my $display_data; |
my $display_data; |
525 |
if ($cache->{format}->{$field}) { |
my $delimiter = ''; |
526 |
my $tmp = mkformat($cache->{format}->{$field},$cache->{display_data}->{$field}->[$page]); |
foreach my $element (@{ $cache->{display_data}->{$field}->[$page] }) { |
527 |
$display_data=$tmp if ($tmp); |
my $data = $element->{data}; |
528 |
} else { |
die "BUG! no data in element?" unless ($data); |
529 |
$display_data = $cache->{display_data}->{$field}->[$page]; |
|
530 |
|
if ($distinct) { |
531 |
|
next if ($cache->{distinct}->{$field}->{ $data }); |
532 |
|
$cache->{distinct}->{$field}->{ $data } = 1; |
533 |
|
} |
534 |
|
|
535 |
|
if ($cache->{format}->{$field}) { |
536 |
|
my $tmp = mkformat($cache->{format}->{$field},$data); |
537 |
|
$display_data .= $delimiter . $tmp if ($tmp); |
538 |
|
} else { |
539 |
|
$display_data .= $delimiter . $data; |
540 |
|
} |
541 |
|
$delimiter = $element->{delimiter} if ($element->{delimiter}); |
542 |
} |
} |
543 |
|
|
544 |
if ($display_data) { # default |
if ($display_data) { # default |
545 |
if ($field eq "headline") { |
if ($field eq "headline") { |
546 |
$xml .= xmlify("headline", $display_data); |
$xml .= xmlify("headline", $display_data); |
552 |
} |
} |
553 |
} |
} |
554 |
|
|
555 |
my $swish_data = $cache->{swish_data}->{$field}->[$page]; |
my $swish_data = join(" ",map { $_->{data} } @{ $cache->{swish_data}->{$field}->[$page] }); |
556 |
if ($swish_data) { |
if ($swish_data) { |
557 |
# remove extra spaces |
# remove extra spaces |
558 |
$swish_data =~ s/ +/ /g; |
$swish_data =~ s/ +/ /g; |
561 |
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
$xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data)); |
562 |
} |
} |
563 |
|
|
564 |
my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page]; |
my $swish_exact_data = join(" ", map { $_->{data} } @{ $cache->{swish_exact_data}->{$field}->[$page] }); |
565 |
if ($swish_exact_data) { |
if ($swish_exact_data) { |
566 |
$swish_exact_data =~ s/ +/ /g; |
$swish_exact_data =~ s/ +/ /g; |
567 |
$swish_exact_data =~ s/ +$//g; |
$swish_exact_data =~ s/ +$//g; |
663 |
|
|
664 |
foreach my $database ($cfg->Sections) { |
foreach my $database ($cfg->Sections) { |
665 |
|
|
666 |
|
# save database name in global variable path for later |
667 |
|
# (need for index filter creation) |
668 |
|
$path = $database; |
669 |
|
|
670 |
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"; |
671 |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
my $add_xml = $cfg -> val($database, 'xml'); # optional |
672 |
|
|
692 |
print STDERR "opening lookup file '$lookup_file'\n"; |
print STDERR "opening lookup file '$lookup_file'\n"; |
693 |
} |
} |
694 |
|
|
695 |
print STDERR "reading ./import_xml/$type.xml\n"; |
my $import_xml_type = $cfg->val($database, 'import_xml_file') || $type; |
696 |
|
my $import_xml_file = "./import_xml/$import_xml_type.xml"; |
697 |
|
|
698 |
|
if (! -r $import_xml_file) { |
699 |
|
print STDERR "ERROR: file $import_xml_file not readable skipping!\n"; |
700 |
|
next; |
701 |
|
} |
702 |
|
|
703 |
|
print STDERR "reading $import_xml_file\n"; |
704 |
|
|
705 |
# extract just type basic |
# extract just type basic |
706 |
my $type_base = $type; |
my $type_base = $type; |
707 |
$type_base =~ s/_.+$//g; |
$type_base =~ s/_.+$//g; |
708 |
|
|
709 |
$config=XMLin("./import_xml/$type.xml", ForceArray => [ $type2tag{$type_base}, 'config', 'format' ], ForceContent => 1 ); |
my $tag = $cfg->val($database, 'import_xml_tag') || $type2tag{$type_base} || die "can't find which tag to use for type $type"; |
710 |
|
$config=XMLin($import_xml_file, ForceArray => [ $tag, 'config', 'format' ], ForceContent => 1 ); |
711 |
|
|
712 |
|
# check for broken XML::Simple |
713 |
|
if ( $broken_cdata ) { |
714 |
|
map { |
715 |
|
$config->{format}->{$_}->{content} =~ s#>$##; |
716 |
|
} keys %{ $config->{format} }; |
717 |
|
} |
718 |
|
|
719 |
|
# helper for progress bar |
720 |
|
sub fmt_time { |
721 |
|
my $t = shift || 0; |
722 |
|
my $out = ""; |
723 |
|
|
724 |
|
my ($ss,$mm,$hh) = gmtime($t); |
725 |
|
$out .= "${hh}h" if ($hh); |
726 |
|
$out .= sprintf("%02d:%02d", $mm,$ss); |
727 |
|
$out .= " " if ($hh == 0); |
728 |
|
return $out; |
729 |
|
} |
730 |
|
|
731 |
# output current progress indicator |
# output current progress indicator |
732 |
my $last_p = 0; |
my $last_p = 0; |
733 |
|
my $start_t = time(); |
734 |
sub progress { |
sub progress { |
735 |
return if (! $show_progress); |
return if (! $show_progress); |
736 |
my $current = shift; |
my $current = shift; |
737 |
my $total = shift || 1; |
my $total = shift || 1; |
738 |
my $p = int($current * 100 / $total); |
my $p = int($current * 100 / $total); |
739 |
if ($p != $last_p) { |
if ($p < $last_p || $current == 1) { |
740 |
printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$current,$total,"=" x ($p/2).">", $p ); |
$start_t = time(); |
741 |
|
$last_p = 0; |
742 |
|
} elsif ($p != $last_p) { |
743 |
|
my $rate = ($current / (time() - $start_t || 1)); |
744 |
|
my $eta = ($total-$current) / ($rate || 1); |
745 |
|
printf STDERR ("%5d [%-38s] %-5d %0.1f/s %s\r",$current,"=" x ($p/3)."$p%>", $total, $rate, fmt_time($eta)); |
746 |
$last_p = $p; |
$last_p = $p; |
747 |
} |
} |
748 |
} |
} |
749 |
|
|
750 |
my $fake_dir = 1; |
my $fake_dir = 1; |
751 |
|
my $fake_pos = 0; |
752 |
|
my $last_fake_t = time(); |
753 |
sub fakeprogress { |
sub fakeprogress { |
754 |
return if (! $show_progress); |
return if (! $show_progress); |
755 |
my $current = shift @_; |
my $current = shift @_; |
756 |
|
|
757 |
my @ind = ('-','\\','|','/','-','\\','|','/', '-'); |
my @ind = ('-','\\','|','/','-','\\','|','/'); |
758 |
|
|
759 |
|
if ($current < $fake_pos) { |
760 |
|
$start_t = time(); |
761 |
|
$last_fake_t = 0; |
762 |
|
$fake_dir = 1; |
763 |
|
$fake_pos = 0; |
764 |
|
} |
765 |
|
|
766 |
$last_p += $fake_dir; |
if (time()-$last_fake_t >= 1) { |
767 |
$fake_dir = -$fake_dir if ($last_p > 1000 || $last_p < 0); |
$last_fake_t = time(); |
768 |
if ($last_p % 10 == 0) { |
$fake_pos += $fake_dir; |
769 |
printf STDERR ("%5d / %5s [%-51s]\r",$current,"?"," " x ($last_p/20).$ind[($last_p/20) % $#ind]); |
$fake_dir = -$fake_dir if ($fake_pos > 38); |
770 |
|
} |
771 |
|
|
772 |
|
if ($current % 10 == 0) { |
773 |
|
my $rate = ($current / (time() - $start_t || 1)); |
774 |
|
printf STDERR ("%5d [%-38s] %0.1f/s\r",$current, " " x $fake_pos .$ind[($current / 10) % 8], $rate); |
775 |
} |
} |
776 |
} |
} |
777 |
|
|
778 |
# now read database |
# now read database |
|
print STDERR "using: $type...\n"; |
|
779 |
|
|
780 |
# erase cache for tags by order in this database |
# erase cache for tags by order in this database |
781 |
delete $cache->{tags_by_order}; |
delete $cache->{tags_by_order}; |
785 |
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!"; |
786 |
|
|
787 |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{isis_codepage},$codepage); |
788 |
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; |
|
|
} |
|
|
} |
|
789 |
|
|
790 |
# OpenIsis::ERR_BADF |
if (! $db) { |
791 |
if ($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"; |
|
792 |
next; |
next; |
793 |
} |
} |
794 |
|
|
795 |
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 ); |
|
|
} |
|
796 |
|
|
797 |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; |
798 |
|
|
|
my $path = $database; |
|
|
|
|
799 |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { |
800 |
my $row = OpenIsis::read( $db, $row_id ); |
my $row = $db->to_hash( $row_id ); |
801 |
if ($row && $row->{mfn}) { |
if ($row) { |
802 |
|
|
803 |
|
$row->{mfn} = $row_id; |
804 |
|
$row->{record} = $db->{record}; |
805 |
|
|
806 |
progress($row->{mfn}, $max_rowid); |
progress($row->{mfn}, $max_rowid); |
807 |
|
|
808 |
my $swishpath = $path."#".int($row->{mfn}); |
my $swishpath = $path."#".int($row->{mfn}); |
816 |
} |
} |
817 |
} |
} |
818 |
} |
} |
|
# 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($db); |
|
819 |
print STDERR "\n"; |
print STDERR "\n"; |
820 |
|
|
821 |
} elsif ($type_base eq "excel") { |
} elsif ($type_base eq "excel") { |
827 |
my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!"; |
my $excel_file = $cfg -> val($database, 'excel_file') || die "$database doesn't have 'excel_file' defined!"; |
828 |
|
|
829 |
my $sheet = x($config->{sheet}) || die "no sheet in $type.xml"; |
my $sheet = x($config->{sheet}) || die "no sheet in $type.xml"; |
830 |
my $start_row = x($config->{start_row}) - 1 || die "no start_row in $type.xml"; |
my $start_row = x($config->{start_row}) || die "no start_row in $type.xml"; |
831 |
|
$start_row--; |
832 |
|
|
833 |
my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'"; |
my $oBook = Spreadsheet::ParseExcel::Workbook->Parse($excel_file) || die "can't open Excel file '$excel_file'"; |
834 |
|
|
878 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
879 |
} |
} |
880 |
} |
} |
881 |
|
|
882 |
|
print STDERR "\n"; |
883 |
|
|
884 |
} elsif ($type_base eq "marc") { |
} elsif ($type_base eq "marc") { |
885 |
|
|
886 |
require MARC; |
require MARC::File::USMARC; |
887 |
|
|
888 |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
$import2cp = Text::Iconv->new($config->{marc_codepage},$codepage); |
889 |
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!"; |
890 |
|
|
891 |
# optional argument is format |
# optional argument is format |
892 |
my $format = x($config->{marc_format}) || 'usmarc'; |
warn "marc_format is no longer used!" if ($config->{marc_format}); |
|
|
|
893 |
print STDERR "Reading MARC file '$marc_file'\n"; |
print STDERR "Reading MARC file '$marc_file'\n"; |
894 |
|
|
895 |
my $marc = new MARC; |
my $marc = MARC::File::USMARC->in( $marc_file ); |
|
my $nr = $marc->openmarc({ |
|
|
file=>$marc_file, format=>$format |
|
|
}) || die "Can't open MARC file '$marc_file' with format '$format'"; |
|
896 |
|
|
897 |
# read MARC file in memory |
if (! $marc) { |
898 |
$marc->nextmarc(-1); |
print STDERR "FATAL: can't read MARC file: $marc_file, skipping...\n"; |
899 |
|
next; |
900 |
|
} |
901 |
|
|
902 |
|
# count records in MARC file |
903 |
|
sub marc_count { |
904 |
|
my $filename = shift || die; |
905 |
|
my $file = MARC::File::USMARC->in($filename) || return; |
906 |
|
my $count = 0; |
907 |
|
while ($file->skip()) { |
908 |
|
$count++; |
909 |
|
} |
910 |
|
return $count; |
911 |
|
} |
912 |
|
|
913 |
my $max_rec = $marc->marc_count(); |
my $count = marc_count($marc_file) || warn "no records in '$marc_file'?"; |
914 |
|
|
915 |
for(my $i=1; $i<=$max_rec; $i++) { |
my $i = 1; |
916 |
|
|
917 |
progress($i,$max_rec); |
while( my $rec = $marc->next() ) { |
918 |
|
|
919 |
# store value for marc_sf.pm |
progress($i,$count); |
|
$main::cache->{marc_record} = $i; |
|
920 |
|
|
921 |
my $swishpath = $database."#".$i; |
my $swishpath = $database."#".$i; |
922 |
|
|
923 |
if (my $xml = data2xml($type_base,$marc,$add_xml,$cfg,$database)) { |
if (my $xml = data2xml($type_base,$rec,$add_xml,$cfg,$database)) { |
924 |
$xml = $cp2utf->convert($xml); |
$xml = $cp2utf->convert($xml); |
925 |
use bytes; # as opposed to chars |
use bytes; # as opposed to chars |
926 |
print "Path-Name: $swishpath\n"; |
print "Path-Name: $swishpath\n"; |
927 |
print "Content-Length: ".(length($xml)+1)."\n"; |
print "Content-Length: ".(length($xml)+1)."\n"; |
928 |
print "Document-Type: XML\n\n$xml\n"; |
print "Document-Type: XML\n\n$xml\n"; |
929 |
} |
} |
930 |
|
|
931 |
|
$i++; |
932 |
} |
} |
933 |
|
|
934 |
print STDERR "\n"; |
print STDERR "\n"; |
973 |
} |
} |
974 |
# close lookup |
# close lookup |
975 |
untie %lhash if (%lhash); |
untie %lhash if (%lhash); |
976 |
|
|
977 |
|
} elsif ($type_base eq "dbf") { |
978 |
|
|
979 |
|
my $dbf_file = $cfg -> val($database, 'dbf_file') || die "$database doesn't have 'dbf_file' defined!"; |
980 |
|
my $dbf_codepage = $cfg -> val($database, 'dbf_codepage') || die "$database doesn't have 'dbf_codepage' defined!"; |
981 |
|
my $dbf_mapping = $cfg -> val($database, 'dbf_mapping') || die "$database doesn't have 'dbf_mapping' defined!"; |
982 |
|
|
983 |
|
$import2cp = Text::Iconv->new($dbf_codepage,$codepage); |
984 |
|
require XBase; |
985 |
|
my $db = new XBase $dbf_file; |
986 |
|
|
987 |
|
if (! $db) { |
988 |
|
print STDERR "ERROR: can't read DBF database: $dbf_file, skipping...\n"; |
989 |
|
next; |
990 |
|
} |
991 |
|
|
992 |
|
my $max_rowid = $db->last_record; |
993 |
|
|
994 |
|
print STDERR "Reading database: $dbf_file [$max_rowid rows]\n"; |
995 |
|
|
996 |
|
my %dbf2iso; |
997 |
|
foreach my $m (split(/[\n\r]+/,$dbf_mapping)) { |
998 |
|
my ($col,$fld) = split(/\s+/,$m,2); |
999 |
|
$dbf2iso{$col} = $fld; |
1000 |
|
} |
1001 |
|
|
1002 |
|
#print STDERR "## dbf2iso: ",Dumper(\%dbf2iso),"\n## /dbf2iso\n"; |
1003 |
|
|
1004 |
|
# bad, bad... |
1005 |
|
require "to_hash.pm"; |
1006 |
|
|
1007 |
|
foreach my $row_id (0 .. $max_rowid) { |
1008 |
|
my $dbf_row = $db->get_record_as_hash($row_id); |
1009 |
|
if ($dbf_row) { |
1010 |
|
|
1011 |
|
#print STDERR "## dbf_row: ",Dumper($dbf_row),"\n## /dbf_row\n"; |
1012 |
|
# apply mapping from config file |
1013 |
|
# all unspecified records will get _ in |
1014 |
|
# front of them - _DELETE will be __DELETE |
1015 |
|
my $rec; |
1016 |
|
map { |
1017 |
|
my $new_fld = $dbf2iso{$_} || '_'.$_; |
1018 |
|
my $data = $dbf_row->{$_}; |
1019 |
|
push @{ $rec->{$new_fld} }, $data if ($data && $data !~ /^(?:\s+|\$a\.|)$/); |
1020 |
|
} keys %{$dbf_row}; |
1021 |
|
#print STDERR "## rec: ",Dumper($rec),"\n## /rec\n"; |
1022 |
|
my $row = to_hash($row_id+1, $rec); |
1023 |
|
|
1024 |
|
$row->{mfn} = $row_id+1; |
1025 |
|
$row->{record} = $rec; |
1026 |
|
|
1027 |
|
#print STDERR "## row: ",Dumper($row),"\n## /row\n"; |
1028 |
|
progress($row->{mfn}, $max_rowid); |
1029 |
|
|
1030 |
|
my $swishpath = $path."#".int($row->{mfn}); |
1031 |
|
|
1032 |
|
if (my $xml = data2xml($type_base,$row,$add_xml,$cfg,$database)) { |
1033 |
|
$xml = $cp2utf->convert($xml); |
1034 |
|
use bytes; # as opposed to chars |
1035 |
|
print "Path-Name: $swishpath\n"; |
1036 |
|
print "Content-Length: ".(length($xml)+1)."\n"; |
1037 |
|
print "Document-Type: XML\n\n$xml\n"; |
1038 |
|
} |
1039 |
|
} |
1040 |
|
} |
1041 |
|
print STDERR "\n"; |
1042 |
} |
} |
1043 |
} |
} |
1044 |
|
|
1059 |
|
|
1060 |
=head1 DESCRIPTION |
=head1 DESCRIPTION |
1061 |
|
|
1062 |
This command will read ISIS data file using OpenIsis perl module, MARC |
This command will read ISIS data file using Biblio::Isis perl module, MARC |
1063 |
records using MARC module and optionally Micro$oft Excel files to |
records using MARC::File module and optionally Micro$oft Excel files to |
1064 |
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, |
1065 |
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 |
1066 |
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. |