9 |
marc marc_indicators marc_repeatable_subfield |
marc marc_indicators marc_repeatable_subfield |
10 |
marc_compose marc_leader |
marc_compose marc_leader |
11 |
marc_duplicate marc_remove |
marc_duplicate marc_remove |
12 |
|
marc_original_order |
13 |
|
|
14 |
rec1 rec2 rec |
rec1 rec2 rec |
15 |
regex prefix suffix surround |
regex prefix suffix surround |
36 |
|
|
37 |
=head1 VERSION |
=head1 VERSION |
38 |
|
|
39 |
Version 0.11 |
Version 0.17 |
40 |
|
|
41 |
=cut |
=cut |
42 |
|
|
43 |
our $VERSION = '0.11'; |
our $VERSION = '0.17'; |
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 |
|
$config = shift; |
143 |
|
} |
144 |
|
|
145 |
=head2 _get_ds |
=head2 _get_ds |
146 |
|
|
147 |
Return hash formatted as data structure |
Return hash formatted as data structure |
526 |
'c', rec('200','c') |
'c', rec('200','c') |
527 |
); |
); |
528 |
|
|
529 |
|
If you specify C<+> for subfield, value will be appended |
530 |
|
to previous defined subfield. |
531 |
|
|
532 |
=cut |
=cut |
533 |
|
|
534 |
sub marc_compose { |
sub marc_compose { |
538 |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
539 |
my $m = [ $f, $i1, $i2 ]; |
my $m = [ $f, $i1, $i2 ]; |
540 |
|
|
541 |
|
warn "### marc_compose input subfields = ", dump(@_),$/ if ($debug > 2); |
542 |
|
|
543 |
|
if ($#_ % 2 != 1) { |
544 |
|
die "ERROR: marc_compose",dump($f,@_)," not valid (must be even).\nDo you need to add first() or join() around some argument?\n"; |
545 |
|
} |
546 |
|
|
547 |
while (@_) { |
while (@_) { |
548 |
my $sf = shift or die "marc_compose $f needs subfield"; |
my $sf = shift; |
549 |
my $v = shift; |
my $v = shift; |
550 |
|
|
551 |
next unless (defined($v) && $v !~ /^\s*$/); |
next unless (defined($v) && $v !~ /^\s*$/); |
552 |
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
|
push @$m, ( $sf, $v ); |
|
553 |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
554 |
|
if ($sf ne '+') { |
555 |
|
push @$m, ( $sf, $v ); |
556 |
|
} else { |
557 |
|
$m->[ $#$m ] .= $v; |
558 |
|
} |
559 |
} |
} |
560 |
|
|
561 |
warn "## marc_compose(d) ", dump( $m ),$/ if ($debug > 1); |
warn "## marc_compose current marc = ", dump( $m ),$/ if ($debug > 1); |
562 |
|
|
563 |
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
564 |
} |
} |
646 |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
647 |
} |
} |
648 |
|
|
649 |
|
=head2 marc_original_order |
650 |
|
|
651 |
|
Copy all subfields preserving original order to marc field. |
652 |
|
|
653 |
|
marc_original_order( marc_field_number, original_input_field_number ); |
654 |
|
|
655 |
|
Please note that field numbers are consistent with other commands (marc |
656 |
|
field number first), but somewhat counter-intuitive (destination and then |
657 |
|
source). |
658 |
|
|
659 |
|
You might want to use this command if you are just renaming subfields or |
660 |
|
using pre-processing modify_record in C<config.yml> and don't need any |
661 |
|
post-processing or want to preserve order of original subfields. |
662 |
|
|
663 |
|
|
664 |
|
=cut |
665 |
|
|
666 |
|
sub marc_original_order { |
667 |
|
|
668 |
|
my ($to, $from) = @_; |
669 |
|
die "marc_original_order needs from and to fields\n" unless ($from && $to); |
670 |
|
|
671 |
|
return unless defined($rec->{$from}); |
672 |
|
|
673 |
|
my $r = $rec->{$from}; |
674 |
|
die "record field $from isn't array\n" unless (ref($r) eq 'ARRAY'); |
675 |
|
|
676 |
|
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
677 |
|
warn "## marc_original_order($to,$from) source = ", dump( $r ),$/ if ($debug > 1); |
678 |
|
|
679 |
|
foreach my $d (@$r) { |
680 |
|
|
681 |
|
if (! defined($d->{subfields}) && ref($d->{subfields}) ne 'ARRAY') { |
682 |
|
warn "# marc_original_order($to,$from): field $from doesn't have subfields specification\n"; |
683 |
|
next; |
684 |
|
} |
685 |
|
|
686 |
|
my @sfs = @{ $d->{subfields} }; |
687 |
|
|
688 |
|
die "field $from doesn't have even number of subfields specifications\n" unless($#sfs % 2 == 1); |
689 |
|
|
690 |
|
warn "#--> d: ",dump($d), "\n#--> sfs: ",dump(@sfs),$/ if ($debug > 2); |
691 |
|
|
692 |
|
my $m = [ $to, $i1, $i2 ]; |
693 |
|
|
694 |
|
while (my $sf = shift @sfs) { |
695 |
|
|
696 |
|
warn "#--> sf: ",dump($sf), $/ if ($debug > 2); |
697 |
|
my $offset = shift @sfs; |
698 |
|
die "corrupted sufields specification for field $from\n" unless defined($offset); |
699 |
|
|
700 |
|
my $v; |
701 |
|
if (ref($d->{$sf}) eq 'ARRAY') { |
702 |
|
$v = $d->{$sf}->[$offset] if (defined($d->{$sf}->[$offset])); |
703 |
|
} elsif ($offset == 0) { |
704 |
|
$v = $d->{$sf}; |
705 |
|
} else { |
706 |
|
die "field $from subfield '$sf' need occurence $offset which doesn't exist", dump($d->{$sf}); |
707 |
|
} |
708 |
|
push @$m, ( $sf, $v ) if (defined($v)); |
709 |
|
} |
710 |
|
|
711 |
|
if ($#{$m} > 2) { |
712 |
|
push @{ $marc_record->[ $marc_record_offset ] }, $m; |
713 |
|
} |
714 |
|
} |
715 |
|
|
716 |
|
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
717 |
|
} |
718 |
|
|
719 |
|
|
720 |
=head1 Functions to extract data from input |
=head1 Functions to extract data from input |
721 |
|
|
722 |
This function should be used inside functions to create C<data_structure> described |
This function should be used inside functions to create C<data_structure> described |
740 |
if (ref($rec->{$f}) eq 'ARRAY') { |
if (ref($rec->{$f}) eq 'ARRAY') { |
741 |
return map { |
return map { |
742 |
if (ref($_) eq 'HASH') { |
if (ref($_) eq 'HASH') { |
743 |
values %{$_}; |
my $h = $_; |
744 |
|
if ( defined($h->{subfields}) ) { |
745 |
|
my $sfs = delete $h->{subfields} || die "no subfields?"; |
746 |
|
my @out; |
747 |
|
while (@$sfs) { |
748 |
|
my $sf = shift @$sfs; |
749 |
|
my $o = shift @$sfs; |
750 |
|
if ($o == 0 && ref( $h->{$sf} ) ne 'ARRAY' ) { |
751 |
|
# single element subfields are not arrays |
752 |
|
push @out, $h->{$sf}; |
753 |
|
} else { |
754 |
|
warn "====> $f $sf $o $#$sfs ", dump( $sfs ), "\n"; |
755 |
|
push @out, $h->{$sf}->[$o]; |
756 |
|
} |
757 |
|
} |
758 |
|
return @out; |
759 |
|
} else { |
760 |
|
# FIXME this should probably be in alphabetical order instead of hash order |
761 |
|
values %{$h}; |
762 |
|
} |
763 |
} else { |
} else { |
764 |
$_; |
$_; |
765 |
} |
} |
781 |
my $f = shift; |
my $f = shift; |
782 |
return unless (defined($rec && $rec->{$f})); |
return unless (defined($rec && $rec->{$f})); |
783 |
my $sf = shift; |
my $sf = shift; |
784 |
return map { $_->{$sf} } grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
warn "rec2($f,$sf) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
785 |
|
return map { |
786 |
|
if (ref($_->{$sf}) eq 'ARRAY') { |
787 |
|
@{ $_->{$sf} }; |
788 |
|
} else { |
789 |
|
$_->{$sf}; |
790 |
|
} |
791 |
|
} grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
792 |
} |
} |
793 |
|
|
794 |
=head2 rec |
=head2 rec |
801 |
=cut |
=cut |
802 |
|
|
803 |
sub rec { |
sub rec { |
804 |
|
my @out; |
805 |
if ($#_ == 0) { |
if ($#_ == 0) { |
806 |
return rec1(@_); |
@out = rec1(@_); |
807 |
} elsif ($#_ == 1) { |
} elsif ($#_ == 1) { |
808 |
return rec2(@_); |
@out = rec2(@_); |
809 |
|
} |
810 |
|
if (@out) { |
811 |
|
return @out; |
812 |
|
} else { |
813 |
|
return ''; |
814 |
} |
} |
815 |
} |
} |
816 |
|
|
843 |
=cut |
=cut |
844 |
|
|
845 |
sub prefix { |
sub prefix { |
846 |
my $p = shift or die "prefix needs string as first argument"; |
my $p = shift or return; |
847 |
return map { $p . $_ } grep { defined($_) } @_; |
return map { $p . $_ } grep { defined($_) } @_; |
848 |
} |
} |
849 |
|
|
906 |
} |
} |
907 |
} |
} |
908 |
|
|
909 |
|
=head2 config |
910 |
|
|
911 |
|
Consult config values stored in C<config.yml> |
912 |
|
|
913 |
|
# return database code (key under databases in yaml) |
914 |
|
$database_code = config(); # use _ from hash |
915 |
|
$database_name = config('name'); |
916 |
|
$database_input_name = config('input name'); |
917 |
|
$tag = config('input normalize tag'); |
918 |
|
|
919 |
|
Up to three levels are supported. |
920 |
|
|
921 |
|
=cut |
922 |
|
|
923 |
|
sub config { |
924 |
|
return unless ($config); |
925 |
|
|
926 |
|
my $p = shift; |
927 |
|
|
928 |
|
$p ||= ''; |
929 |
|
|
930 |
|
my $v; |
931 |
|
|
932 |
|
warn "### getting config($p)\n" if ($debug > 1); |
933 |
|
|
934 |
|
my @p = split(/\s+/,$p); |
935 |
|
if ($#p < 0) { |
936 |
|
$v = $config->{ '_' }; # special, database code |
937 |
|
} else { |
938 |
|
|
939 |
|
my $c = dclone( $config ); |
940 |
|
|
941 |
|
foreach my $k (@p) { |
942 |
|
warn "### k: $k c = ",dump($c),$/ if ($debug > 1); |
943 |
|
if (ref($c) eq 'ARRAY') { |
944 |
|
$c = shift @$c; |
945 |
|
warn "config($p) taking first occurence of '$k', probably not what you wanted!\n"; |
946 |
|
last; |
947 |
|
} |
948 |
|
|
949 |
|
if (! defined($c->{$k}) ) { |
950 |
|
$c = undef; |
951 |
|
last; |
952 |
|
} else { |
953 |
|
$c = $c->{$k}; |
954 |
|
} |
955 |
|
} |
956 |
|
$v = $c if ($c); |
957 |
|
|
958 |
|
} |
959 |
|
|
960 |
|
warn "## config( '$p' ) = ",dump( $v ),$/ if ($v && $debug); |
961 |
|
warn "config( '$p' ) is empty\n" if (! $v); |
962 |
|
|
963 |
|
return $v; |
964 |
|
} |
965 |
|
|
966 |
|
=head2 id |
967 |
|
|
968 |
|
Returns unique id of this record |
969 |
|
|
970 |
|
$id = id(); |
971 |
|
|
972 |
|
Returns C<42/2> for 2nd occurence of MFN 42. |
973 |
|
|
974 |
|
=cut |
975 |
|
|
976 |
|
sub id { |
977 |
|
my $mfn = $config->{_mfn} || die "no _mfn in config data"; |
978 |
|
return $mfn . $#{$marc_record} ? $#{$marc_record} + 1 : ''; |
979 |
|
} |
980 |
|
|
981 |
=head2 join_with |
=head2 join_with |
982 |
|
|
983 |
Joins walues with some delimiter |
Joins walues with some delimiter |
988 |
|
|
989 |
sub join_with { |
sub join_with { |
990 |
my $d = shift; |
my $d = shift; |
991 |
return join($d, grep { defined($_) && $_ ne '' } @_); |
warn "### join_with('$d',",dump(@_),")\n" if ($debug > 2); |
992 |
|
my $v = join($d, grep { defined($_) && $_ ne '' } @_); |
993 |
|
return '' unless defined($v); |
994 |
|
return $v; |
995 |
} |
} |
996 |
|
|
997 |
=head2 split_rec_on |
=head2 split_rec_on |
1019 |
my $v = shift @r; |
my $v = shift @r; |
1020 |
warn "### first rec($fld,$sf) = ",dump($v),$/ if ($debug > 2); |
warn "### first rec($fld,$sf) = ",dump($v),$/ if ($debug > 2); |
1021 |
|
|
1022 |
return '' if( ! defined($v) || $v =~ /^\s*$/); |
return '' if ( ! defined($v) || $v =~ /^\s*$/); |
1023 |
|
|
1024 |
my @s = split( $regex, $v ); |
my @s = split( $regex, $v ); |
1025 |
warn "## split_rec_on($fld,$sf,$regex,$part) = ",dump(@s),$/ if ($debug > 1); |
warn "## split_rec_on($fld,$sf,$regex,$part) = ",dump(@s),$/ if ($debug > 1); |