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 0.15 |
40 |
|
|
41 |
=cut |
=cut |
42 |
|
|
43 |
our $VERSION = '0.11'; |
our $VERSION = '0.15'; |
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 |
while (@_) { |
while (@_) { |
544 |
my $sf = shift or die "marc_compose $f needs subfield"; |
my $sf = shift or die "marc_compose $f needs subfield"; |
545 |
my $v = shift; |
my $v = shift; |
546 |
|
|
547 |
next unless (defined($v) && $v !~ /^\s*$/); |
next unless (defined($v) && $v !~ /^\s*$/); |
548 |
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, $v ); |
|
549 |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
550 |
|
if ($sf ne '+') { |
551 |
|
push @$m, ( $sf, $v ); |
552 |
|
} else { |
553 |
|
$m->[ $#$m ] .= $v; |
554 |
|
} |
555 |
} |
} |
556 |
|
|
557 |
warn "## marc_compose(d) ", dump( $m ),$/ if ($debug > 1); |
warn "## marc_compose current marc = ", dump( $m ),$/ if ($debug > 1); |
558 |
|
|
559 |
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
push @{ $marc_record->[ $marc_record_offset ] }, $m if ($#{$m} > 2); |
560 |
} |
} |
642 |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
643 |
} |
} |
644 |
|
|
645 |
|
=head2 marc_original_order |
646 |
|
|
647 |
|
Copy all subfields preserving original order to marc field. |
648 |
|
|
649 |
|
marc_original_order(210, 260); |
650 |
|
|
651 |
|
You might want to use this command if you are just renaming subfields or |
652 |
|
using pre-processing modify_record in C<config.yml> and don't need any |
653 |
|
post-processing or want to preserve order of original subfields. |
654 |
|
|
655 |
|
=cut |
656 |
|
|
657 |
|
sub marc_original_order { |
658 |
|
|
659 |
|
my ($from, $to) = @_; |
660 |
|
die "marc_original_order needs from and to fields\n" unless ($from && $to); |
661 |
|
|
662 |
|
my $r = $rec->{$from} || return; |
663 |
|
die "record field $from isn't array\n" unless (ref($r) eq 'ARRAY'); |
664 |
|
|
665 |
|
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
666 |
|
warn "## marc_original_order($from,$to) source = ", dump( $r ),$/ if ($debug > 1); |
667 |
|
|
668 |
|
foreach my $d (@$r) { |
669 |
|
|
670 |
|
if (! defined($d->{subfields}) && ref($d->{subfields}) ne 'ARRAY') { |
671 |
|
warn "# marc_original_order($from,$to): field $from doesn't have subfields specification\n"; |
672 |
|
next; |
673 |
|
} |
674 |
|
|
675 |
|
my @sfs = @{ $d->{subfields} }; |
676 |
|
|
677 |
|
die "field $from doesn't have even number of subfields specifications\n" unless($#sfs % 2 == 1); |
678 |
|
|
679 |
|
warn "#--> d: ",dump($d), "\n#--> sfs: ",dump(@sfs),$/; |
680 |
|
|
681 |
|
my $m = [ $to, $i1, $i2 ]; |
682 |
|
|
683 |
|
while (my $sf = shift @sfs) { |
684 |
|
warn "#--> sf: ",dump($sf), $/; |
685 |
|
my $offset = shift @sfs; |
686 |
|
die "corrupted sufields specification for field $from\n" unless defined($offset); |
687 |
|
|
688 |
|
my $v; |
689 |
|
if (ref($d->{$sf}) eq 'ARRAY') { |
690 |
|
$v = $d->{$sf}->[$offset] if (defined($d->{$sf}->[$offset])); |
691 |
|
} elsif ($offset == 0) { |
692 |
|
$v = $d->{$sf}; |
693 |
|
} else { |
694 |
|
die "field $from subfield '$sf' need occurence $offset which doesn't exist", dump($d->{$sf}); |
695 |
|
} |
696 |
|
push @$m, ( $sf, $v ) if (defined($v)); |
697 |
|
} |
698 |
|
|
699 |
|
if ($#{$m} > 2) { |
700 |
|
push @{ $marc_record->[ $marc_record_offset ] }, $m; |
701 |
|
} |
702 |
|
} |
703 |
|
|
704 |
|
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
705 |
|
|
706 |
|
warn "# marc_original_order is partly implemented"; |
707 |
|
} |
708 |
|
|
709 |
|
|
710 |
=head1 Functions to extract data from input |
=head1 Functions to extract data from input |
711 |
|
|
712 |
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 |
752 |
my $f = shift; |
my $f = shift; |
753 |
return unless (defined($rec && $rec->{$f})); |
return unless (defined($rec && $rec->{$f})); |
754 |
my $sf = shift; |
my $sf = shift; |
755 |
return map { $_->{$sf} } grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
warn "rec2($f,$sf) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
756 |
|
return map { |
757 |
|
if (ref($_->{$sf}) eq 'ARRAY') { |
758 |
|
@{ $_->{$sf} }; |
759 |
|
} else { |
760 |
|
$_->{$sf}; |
761 |
|
} |
762 |
|
} grep { ref($_) eq 'HASH' && $_->{$sf} } @{ $rec->{$f} }; |
763 |
} |
} |
764 |
|
|
765 |
=head2 rec |
=head2 rec |
772 |
=cut |
=cut |
773 |
|
|
774 |
sub rec { |
sub rec { |
775 |
|
my @out; |
776 |
if ($#_ == 0) { |
if ($#_ == 0) { |
777 |
return rec1(@_); |
@out = rec1(@_); |
778 |
} elsif ($#_ == 1) { |
} elsif ($#_ == 1) { |
779 |
return rec2(@_); |
@out = rec2(@_); |
780 |
|
} |
781 |
|
if (@out) { |
782 |
|
return @out; |
783 |
|
} else { |
784 |
|
return ''; |
785 |
} |
} |
786 |
} |
} |
787 |
|
|
814 |
=cut |
=cut |
815 |
|
|
816 |
sub prefix { |
sub prefix { |
817 |
my $p = shift or die "prefix needs string as first argument"; |
my $p = shift or return; |
818 |
return map { $p . $_ } grep { defined($_) } @_; |
return map { $p . $_ } grep { defined($_) } @_; |
819 |
} |
} |
820 |
|
|
877 |
} |
} |
878 |
} |
} |
879 |
|
|
880 |
|
=head2 config |
881 |
|
|
882 |
|
Consult config values stored in C<config.yml> |
883 |
|
|
884 |
|
# return database code (key under databases in yaml) |
885 |
|
$database_code = config(); # use _ from hash |
886 |
|
$database_name = config('name'); |
887 |
|
$database_input_name = config('input name'); |
888 |
|
$tag = config('input normalize tag'); |
889 |
|
|
890 |
|
Up to three levels are supported. |
891 |
|
|
892 |
|
=cut |
893 |
|
|
894 |
|
sub config { |
895 |
|
return unless ($config); |
896 |
|
|
897 |
|
my $p = shift; |
898 |
|
|
899 |
|
$p ||= ''; |
900 |
|
|
901 |
|
my $v; |
902 |
|
|
903 |
|
warn "### getting config($p)\n" if ($debug > 1); |
904 |
|
|
905 |
|
my @p = split(/\s+/,$p); |
906 |
|
if ($#p < 0) { |
907 |
|
$v = $config->{ '_' }; # special, database code |
908 |
|
} else { |
909 |
|
|
910 |
|
my $c = dclone( $config ); |
911 |
|
|
912 |
|
foreach my $k (@p) { |
913 |
|
warn "### k: $k c = ",dump($c),$/ if ($debug > 1); |
914 |
|
if (ref($c) eq 'ARRAY') { |
915 |
|
$c = shift @$c; |
916 |
|
warn "config($p) taking first occurence of '$k', probably not what you wanted!\n"; |
917 |
|
last; |
918 |
|
} |
919 |
|
|
920 |
|
if (! defined($c->{$k}) ) { |
921 |
|
$c = undef; |
922 |
|
last; |
923 |
|
} else { |
924 |
|
$c = $c->{$k}; |
925 |
|
} |
926 |
|
} |
927 |
|
$v = $c if ($c); |
928 |
|
|
929 |
|
} |
930 |
|
|
931 |
|
warn "## config( '$p' ) = ",dump( $v ),$/ if ($v && $debug); |
932 |
|
warn "config( '$p' ) is empty\n" if (! $v); |
933 |
|
|
934 |
|
return $v; |
935 |
|
} |
936 |
|
|
937 |
|
=head2 id |
938 |
|
|
939 |
|
Returns unique id of this record |
940 |
|
|
941 |
|
$id = id(); |
942 |
|
|
943 |
|
Returns C<42/2> for 2nd occurence of MFN 42. |
944 |
|
|
945 |
|
=cut |
946 |
|
|
947 |
|
sub id { |
948 |
|
my $mfn = $config->{_mfn} || die "no _mfn in config data"; |
949 |
|
return $mfn . $#{$marc_record} ? $#{$marc_record} + 1 : ''; |
950 |
|
} |
951 |
|
|
952 |
=head2 join_with |
=head2 join_with |
953 |
|
|
954 |
Joins walues with some delimiter |
Joins walues with some delimiter |
959 |
|
|
960 |
sub join_with { |
sub join_with { |
961 |
my $d = shift; |
my $d = shift; |
962 |
return join($d, grep { defined($_) && $_ ne '' } @_); |
warn "### join_with('$d',",dump(@_),")\n" if ($debug > 2); |
963 |
|
my $v = join($d, grep { defined($_) && $_ ne '' } @_); |
964 |
|
return '' unless defined($v); |
965 |
|
return $v; |
966 |
} |
} |
967 |
|
|
968 |
=head2 split_rec_on |
=head2 split_rec_on |
990 |
my $v = shift @r; |
my $v = shift @r; |
991 |
warn "### first rec($fld,$sf) = ",dump($v),$/ if ($debug > 2); |
warn "### first rec($fld,$sf) = ",dump($v),$/ if ($debug > 2); |
992 |
|
|
993 |
return '' if( ! defined($v) || $v =~ /^\s*$/); |
return '' if ( ! defined($v) || $v =~ /^\s*$/); |
994 |
|
|
995 |
my @s = split( $regex, $v ); |
my @s = split( $regex, $v ); |
996 |
warn "## split_rec_on($fld,$sf,$regex,$part) = ",dump(@s),$/ if ($debug > 1); |
warn "## split_rec_on($fld,$sf,$regex,$part) = ",dump(@s),$/ if ($debug > 1); |