4 |
_set_rec _set_lookup |
_set_rec _set_lookup |
5 |
_get_ds _clean_ds |
_get_ds _clean_ds |
6 |
_debug |
_debug |
7 |
|
_pack_subfields_hash |
8 |
|
|
9 |
tag search display |
tag search display |
10 |
marc marc_indicators marc_repeatable_subfield |
marc marc_indicators marc_repeatable_subfield |
11 |
marc_compose marc_leader |
marc_compose marc_leader |
12 |
marc_duplicate marc_remove |
marc_duplicate marc_remove |
13 |
|
marc_original_order |
14 |
|
|
15 |
rec1 rec2 rec |
rec1 rec2 rec |
16 |
regex prefix suffix surround |
regex prefix suffix surround |
24 |
|
|
25 |
#use base qw/WebPAC::Common/; |
#use base qw/WebPAC::Common/; |
26 |
use Data::Dump qw/dump/; |
use Data::Dump qw/dump/; |
|
use Encode qw/from_to/; |
|
27 |
use Storable qw/dclone/; |
use Storable qw/dclone/; |
28 |
|
|
29 |
# debugging warn(s) |
# debugging warn(s) |
36 |
|
|
37 |
=head1 VERSION |
=head1 VERSION |
38 |
|
|
39 |
Version 0.10 |
Version 0.19 |
40 |
|
|
41 |
=cut |
=cut |
42 |
|
|
43 |
our $VERSION = '0.10'; |
our $VERSION = '0.19'; |
44 |
|
|
45 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
46 |
|
|
70 |
row => $row, |
row => $row, |
71 |
rules => $normalize_pl_config, |
rules => $normalize_pl_config, |
72 |
marc_encoding => 'utf-8', |
marc_encoding => 'utf-8', |
73 |
|
config => $config, |
74 |
); |
); |
75 |
|
|
76 |
Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all |
Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all |
92 |
no strict 'subs'; |
no strict 'subs'; |
93 |
_set_lookup( $arg->{lookup} ); |
_set_lookup( $arg->{lookup} ); |
94 |
_set_rec( $arg->{row} ); |
_set_rec( $arg->{row} ); |
95 |
|
_set_config( $arg->{config} ); |
96 |
_clean_ds( %{ $arg } ); |
_clean_ds( %{ $arg } ); |
97 |
eval "$arg->{rules}"; |
eval "$arg->{rules}"; |
98 |
die "error evaling $arg->{rules}: $@\n" if ($@); |
die "error evaling $arg->{rules}: $@\n" if ($@); |
114 |
$rec = shift or die "no record hash"; |
$rec = shift or die "no record hash"; |
115 |
} |
} |
116 |
|
|
117 |
|
=head2 _set_config |
118 |
|
|
119 |
|
Set current config hash |
120 |
|
|
121 |
|
_set_config( $config ); |
122 |
|
|
123 |
|
Magic keys are: |
124 |
|
|
125 |
|
=over 4 |
126 |
|
|
127 |
|
=item _ |
128 |
|
|
129 |
|
Code of current database |
130 |
|
|
131 |
|
=item _mfn |
132 |
|
|
133 |
|
Current MFN |
134 |
|
|
135 |
|
=back |
136 |
|
|
137 |
|
=cut |
138 |
|
|
139 |
|
my $config; |
140 |
|
|
141 |
|
sub _set_config { |
142 |
|
$config = shift; |
143 |
|
} |
144 |
|
|
145 |
=head2 _get_ds |
=head2 _get_ds |
146 |
|
|
147 |
Return hash formatted as data structure |
Return hash formatted as data structure |
357 |
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
warn "## saved/3 ", dump( $field ),$/ if ($debug); |
358 |
} |
} |
359 |
|
|
360 |
return @m; |
return \@m; |
361 |
} |
} |
362 |
|
|
363 |
=head2 _debug |
=head2 _debug |
469 |
foreach (@_) { |
foreach (@_) { |
470 |
my $v = $_; # make var read-write for Encode |
my $v = $_; # make var read-write for Encode |
471 |
next unless (defined($v) && $v !~ /^\s*$/); |
next unless (defined($v) && $v !~ /^\s*$/); |
|
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
|
472 |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
473 |
if (defined $sf) { |
if (defined $sf) { |
474 |
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
525 |
'c', rec('200','c') |
'c', rec('200','c') |
526 |
); |
); |
527 |
|
|
528 |
|
If you specify C<+> for subfield, value will be appended |
529 |
|
to previous defined subfield. |
530 |
|
|
531 |
=cut |
=cut |
532 |
|
|
533 |
sub marc_compose { |
sub marc_compose { |
537 |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
538 |
my $m = [ $f, $i1, $i2 ]; |
my $m = [ $f, $i1, $i2 ]; |
539 |
|
|
540 |
|
warn "### marc_compose input subfields = ", dump(@_),$/ if ($debug > 2); |
541 |
|
|
542 |
|
if ($#_ % 2 != 1) { |
543 |
|
die "ERROR: marc_compose",dump($f,@_)," not valid (must be even).\nDo you need to add first() or join() around some argument?\n"; |
544 |
|
} |
545 |
|
|
546 |
while (@_) { |
while (@_) { |
547 |
my $sf = shift or die "marc_compose $f needs subfield"; |
my $sf = shift; |
548 |
my $v = shift; |
my $v = shift; |
549 |
|
|
550 |
next unless (defined($v) && $v !~ /^\s*$/); |
next unless (defined($v) && $v !~ /^\s*$/); |
|
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
|
|
push @$m, ( $sf, $v ); |
|
551 |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
552 |
|
if ($sf ne '+') { |
553 |
|
push @$m, ( $sf, $v ); |
554 |
|
} else { |
555 |
|
$m->[ $#$m ] .= $v; |
556 |
|
} |
557 |
} |
} |
558 |
|
|
559 |
warn "## marc_compose(d) ", dump( $m ),$/ if ($debug > 1); |
warn "## marc_compose current marc = ", dump( $m ),$/ if ($debug > 1); |
560 |
|
|
561 |
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
562 |
} |
} |
607 |
|
|
608 |
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
609 |
|
|
610 |
foreach my $i ( 0 .. $#{ $marc } ) { |
my $i = 0; |
611 |
|
foreach ( 0 .. $#{ $marc } ) { |
612 |
last unless (defined $marc->[$i]); |
last unless (defined $marc->[$i]); |
613 |
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
614 |
if ($marc->[$i]->[0] eq $f) { |
if ($marc->[$i]->[0] eq $f) { |
634 |
} |
} |
635 |
} |
} |
636 |
} |
} |
637 |
|
$i++; |
638 |
} |
} |
639 |
|
|
640 |
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
644 |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
645 |
} |
} |
646 |
|
|
647 |
|
=head2 marc_original_order |
648 |
|
|
649 |
|
Copy all subfields preserving original order to marc field. |
650 |
|
|
651 |
|
marc_original_order( marc_field_number, original_input_field_number ); |
652 |
|
|
653 |
|
Please note that field numbers are consistent with other commands (marc |
654 |
|
field number first), but somewhat counter-intuitive (destination and then |
655 |
|
source). |
656 |
|
|
657 |
|
You might want to use this command if you are just renaming subfields or |
658 |
|
using pre-processing modify_record in C<config.yml> and don't need any |
659 |
|
post-processing or want to preserve order of original subfields. |
660 |
|
|
661 |
|
|
662 |
|
=cut |
663 |
|
|
664 |
|
sub marc_original_order { |
665 |
|
|
666 |
|
my ($to, $from) = @_; |
667 |
|
die "marc_original_order needs from and to fields\n" unless ($from && $to); |
668 |
|
|
669 |
|
return unless defined($rec->{$from}); |
670 |
|
|
671 |
|
my $r = $rec->{$from}; |
672 |
|
die "record field $from isn't array\n" unless (ref($r) eq 'ARRAY'); |
673 |
|
|
674 |
|
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
675 |
|
warn "## marc_original_order($to,$from) source = ", dump( $r ),$/ if ($debug > 1); |
676 |
|
|
677 |
|
foreach my $d (@$r) { |
678 |
|
|
679 |
|
if (! defined($d->{subfields}) && ref($d->{subfields}) ne 'ARRAY') { |
680 |
|
warn "# marc_original_order($to,$from): field $from doesn't have subfields specification\n"; |
681 |
|
next; |
682 |
|
} |
683 |
|
|
684 |
|
my @sfs = @{ $d->{subfields} }; |
685 |
|
|
686 |
|
die "field $from doesn't have even number of subfields specifications\n" unless($#sfs % 2 == 1); |
687 |
|
|
688 |
|
warn "#--> d: ",dump($d), "\n#--> sfs: ",dump(@sfs),$/ if ($debug > 2); |
689 |
|
|
690 |
|
my $m = [ $to, $i1, $i2 ]; |
691 |
|
|
692 |
|
while (my $sf = shift @sfs) { |
693 |
|
|
694 |
|
warn "#--> sf: ",dump($sf), $/ if ($debug > 2); |
695 |
|
my $offset = shift @sfs; |
696 |
|
die "corrupted sufields specification for field $from\n" unless defined($offset); |
697 |
|
|
698 |
|
my $v; |
699 |
|
if (ref($d->{$sf}) eq 'ARRAY') { |
700 |
|
$v = $d->{$sf}->[$offset] if (defined($d->{$sf}->[$offset])); |
701 |
|
} elsif ($offset == 0) { |
702 |
|
$v = $d->{$sf}; |
703 |
|
} else { |
704 |
|
die "field $from subfield '$sf' need occurence $offset which doesn't exist", dump($d->{$sf}); |
705 |
|
} |
706 |
|
push @$m, ( $sf, $v ) if (defined($v)); |
707 |
|
} |
708 |
|
|
709 |
|
if ($#{$m} > 2) { |
710 |
|
push @{ $marc_record->[ $marc_record_offset ] }, $m; |
711 |
|
} |
712 |
|
} |
713 |
|
|
714 |
|
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
715 |
|
} |
716 |
|
|
717 |
|
|
718 |
=head1 Functions to extract data from input |
=head1 Functions to extract data from input |
719 |
|
|
720 |
This function should be used inside functions to create C<data_structure> described |
This function should be used inside functions to create C<data_structure> described |
721 |
above. |
above. |
722 |
|
|
723 |
|
=head2 _pack_subfields_hash |
724 |
|
|
725 |
|
@values = _pack_subfields_hash( $h, $include_subfields ) |
726 |
|
|
727 |
|
=cut |
728 |
|
|
729 |
|
sub _pack_subfields_hash { |
730 |
|
|
731 |
|
warn "## _pack_subfields_hash( ",dump(@_), " )\n" if ($debug > 1); |
732 |
|
|
733 |
|
my ($h,$include_subfields) = @_; |
734 |
|
|
735 |
|
if ( defined($h->{subfields}) ) { |
736 |
|
my $sfs = delete $h->{subfields} || die "no subfields?"; |
737 |
|
my @out; |
738 |
|
while (@$sfs) { |
739 |
|
my $sf = shift @$sfs; |
740 |
|
push @out, '^' . $sf if ($include_subfields); |
741 |
|
my $o = shift @$sfs; |
742 |
|
if ($o == 0 && ref( $h->{$sf} ) ne 'ARRAY' ) { |
743 |
|
# single element subfields are not arrays |
744 |
|
warn "====> $sf $o / $#$sfs ", dump( $sfs, $h->{$sf} ), "\n"; |
745 |
|
|
746 |
|
push @out, $h->{$sf}; |
747 |
|
} else { |
748 |
|
warn "====> $sf $o / $#$sfs ", dump( $sfs, $h->{$sf} ), "\n"; |
749 |
|
push @out, $h->{$sf}->[$o]; |
750 |
|
} |
751 |
|
} |
752 |
|
if ($include_subfields) { |
753 |
|
return join('', @out); |
754 |
|
} else { |
755 |
|
return @out; |
756 |
|
} |
757 |
|
} else { |
758 |
|
if ($include_subfields) { |
759 |
|
my $out = ''; |
760 |
|
foreach my $sf (keys %$h) { |
761 |
|
if (ref($h->{$sf}) eq 'ARRAY') { |
762 |
|
$out .= '^' . $sf . join('^' . $sf, @{ $h->{$sf} }); |
763 |
|
} else { |
764 |
|
$out .= '^' . $sf . $h->{$sf}; |
765 |
|
} |
766 |
|
} |
767 |
|
return $out; |
768 |
|
} else { |
769 |
|
# FIXME this should probably be in alphabetical order instead of hash order |
770 |
|
values %{$h}; |
771 |
|
} |
772 |
|
} |
773 |
|
} |
774 |
|
|
775 |
=head2 rec1 |
=head2 rec1 |
776 |
|
|
777 |
Return all values in some field |
Return all values in some field |
788 |
return unless (defined($rec) && defined($rec->{$f})); |
return unless (defined($rec) && defined($rec->{$f})); |
789 |
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
790 |
if (ref($rec->{$f}) eq 'ARRAY') { |
if (ref($rec->{$f}) eq 'ARRAY') { |
791 |
return map { |
my @out; |
792 |
if (ref($_) eq 'HASH') { |
foreach my $h ( @{ $rec->{$f} } ) { |
793 |
values %{$_}; |
if (ref($h) eq 'HASH') { |
794 |
|
push @out, ( _pack_subfields_hash( $h ) ); |
795 |
} else { |
} else { |
796 |
$_; |
push @out, $h; |
797 |
} |
} |
798 |
} @{ $rec->{$f} }; |
} |
799 |
|
return @out; |
800 |
} elsif( defined($rec->{$f}) ) { |
} elsif( defined($rec->{$f}) ) { |
801 |
return $rec->{$f}; |
return $rec->{$f}; |
802 |
} |
} |
814 |
my $f = shift; |
my $f = shift; |
815 |
return unless (defined($rec && $rec->{$f})); |
return unless (defined($rec && $rec->{$f})); |
816 |
my $sf = shift; |
my $sf = shift; |
817 |
return map { $_->{$sf} } grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
warn "rec2($f,$sf) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
818 |
|
return map { |
819 |
|
if (ref($_->{$sf}) eq 'ARRAY') { |
820 |
|
@{ $_->{$sf} }; |
821 |
|
} else { |
822 |
|
$_->{$sf}; |
823 |
|
} |
824 |
|
} grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
825 |
} |
} |
826 |
|
|
827 |
=head2 rec |
=head2 rec |
834 |
=cut |
=cut |
835 |
|
|
836 |
sub rec { |
sub rec { |
837 |
|
my @out; |
838 |
if ($#_ == 0) { |
if ($#_ == 0) { |
839 |
return rec1(@_); |
@out = rec1(@_); |
840 |
} elsif ($#_ == 1) { |
} elsif ($#_ == 1) { |
841 |
return rec2(@_); |
@out = rec2(@_); |
842 |
|
} |
843 |
|
if (@out) { |
844 |
|
return @out; |
845 |
|
} else { |
846 |
|
return ''; |
847 |
} |
} |
848 |
} |
} |
849 |
|
|
876 |
=cut |
=cut |
877 |
|
|
878 |
sub prefix { |
sub prefix { |
879 |
my $p = shift or die "prefix needs string as first argument"; |
my $p = shift or return; |
880 |
return map { $p . $_ } grep { defined($_) } @_; |
return map { $p . $_ } grep { defined($_) } @_; |
881 |
} |
} |
882 |
|
|
939 |
} |
} |
940 |
} |
} |
941 |
|
|
942 |
|
=head2 config |
943 |
|
|
944 |
|
Consult config values stored in C<config.yml> |
945 |
|
|
946 |
|
# return database code (key under databases in yaml) |
947 |
|
$database_code = config(); # use _ from hash |
948 |
|
$database_name = config('name'); |
949 |
|
$database_input_name = config('input name'); |
950 |
|
$tag = config('input normalize tag'); |
951 |
|
|
952 |
|
Up to three levels are supported. |
953 |
|
|
954 |
|
=cut |
955 |
|
|
956 |
|
sub config { |
957 |
|
return unless ($config); |
958 |
|
|
959 |
|
my $p = shift; |
960 |
|
|
961 |
|
$p ||= ''; |
962 |
|
|
963 |
|
my $v; |
964 |
|
|
965 |
|
warn "### getting config($p)\n" if ($debug > 1); |
966 |
|
|
967 |
|
my @p = split(/\s+/,$p); |
968 |
|
if ($#p < 0) { |
969 |
|
$v = $config->{ '_' }; # special, database code |
970 |
|
} else { |
971 |
|
|
972 |
|
my $c = dclone( $config ); |
973 |
|
|
974 |
|
foreach my $k (@p) { |
975 |
|
warn "### k: $k c = ",dump($c),$/ if ($debug > 1); |
976 |
|
if (ref($c) eq 'ARRAY') { |
977 |
|
$c = shift @$c; |
978 |
|
warn "config($p) taking first occurence of '$k', probably not what you wanted!\n"; |
979 |
|
last; |
980 |
|
} |
981 |
|
|
982 |
|
if (! defined($c->{$k}) ) { |
983 |
|
$c = undef; |
984 |
|
last; |
985 |
|
} else { |
986 |
|
$c = $c->{$k}; |
987 |
|
} |
988 |
|
} |
989 |
|
$v = $c if ($c); |
990 |
|
|
991 |
|
} |
992 |
|
|
993 |
|
warn "## config( '$p' ) = ",dump( $v ),$/ if ($v && $debug); |
994 |
|
warn "config( '$p' ) is empty\n" if (! $v); |
995 |
|
|
996 |
|
return $v; |
997 |
|
} |
998 |
|
|
999 |
|
=head2 id |
1000 |
|
|
1001 |
|
Returns unique id of this record |
1002 |
|
|
1003 |
|
$id = id(); |
1004 |
|
|
1005 |
|
Returns C<42/2> for 2nd occurence of MFN 42. |
1006 |
|
|
1007 |
|
=cut |
1008 |
|
|
1009 |
|
sub id { |
1010 |
|
my $mfn = $config->{_mfn} || die "no _mfn in config data"; |
1011 |
|
return $mfn . $#{$marc_record} ? $#{$marc_record} + 1 : ''; |
1012 |
|
} |
1013 |
|
|
1014 |
=head2 join_with |
=head2 join_with |
1015 |
|
|
1016 |
Joins walues with some delimiter |
Joins walues with some delimiter |
1021 |
|
|
1022 |
sub join_with { |
sub join_with { |
1023 |
my $d = shift; |
my $d = shift; |
1024 |
return join($d, grep { defined($_) && $_ ne '' } @_); |
warn "### join_with('$d',",dump(@_),")\n" if ($debug > 2); |
1025 |
|
my $v = join($d, grep { defined($_) && $_ ne '' } @_); |
1026 |
|
return '' unless defined($v); |
1027 |
|
return $v; |
1028 |
} |
} |
1029 |
|
|
1030 |
=head2 split_rec_on |
=head2 split_rec_on |
1052 |
my $v = shift @r; |
my $v = shift @r; |
1053 |
warn "### first rec($fld,$sf) = ",dump($v),$/ if ($debug > 2); |
warn "### first rec($fld,$sf) = ",dump($v),$/ if ($debug > 2); |
1054 |
|
|
1055 |
return '' if( ! defined($v) || $v =~ /^\s*$/); |
return '' if ( ! defined($v) || $v =~ /^\s*$/); |
1056 |
|
|
1057 |
my @s = split( $regex, $v ); |
my @s = split( $regex, $v ); |
1058 |
warn "## split_rec_on($fld,$sf,$regex,$part) = ",dump(@s),$/ if ($debug > 1); |
warn "## split_rec_on($fld,$sf,$regex,$part) = ",dump(@s),$/ if ($debug > 1); |