2 |
use Exporter 'import'; |
use Exporter 'import'; |
3 |
@EXPORT = qw/ |
@EXPORT = qw/ |
4 |
_set_rec _set_lookup |
_set_rec _set_lookup |
5 |
|
_set_load_row |
6 |
_get_ds _clean_ds |
_get_ds _clean_ds |
7 |
_debug |
_debug |
8 |
|
_pack_subfields_hash |
9 |
|
|
10 |
tag search display |
tag search display |
11 |
marc marc_indicators marc_repeatable_subfield |
marc marc_indicators marc_repeatable_subfield |
16 |
rec1 rec2 rec |
rec1 rec2 rec |
17 |
regex prefix suffix surround |
regex prefix suffix surround |
18 |
first lookup join_with |
first lookup join_with |
19 |
|
save_into_lookup |
20 |
|
|
21 |
split_rec_on |
split_rec_on |
22 |
|
|
23 |
|
get set |
24 |
|
count |
25 |
/; |
/; |
26 |
|
|
27 |
use warnings; |
use warnings; |
29 |
|
|
30 |
#use base qw/WebPAC::Common/; |
#use base qw/WebPAC::Common/; |
31 |
use Data::Dump qw/dump/; |
use Data::Dump qw/dump/; |
|
use Encode qw/from_to/; |
|
32 |
use Storable qw/dclone/; |
use Storable qw/dclone/; |
33 |
|
use Carp qw/confess/; |
34 |
|
|
35 |
# debugging warn(s) |
# debugging warn(s) |
36 |
my $debug = 0; |
my $debug = 0; |
42 |
|
|
43 |
=head1 VERSION |
=head1 VERSION |
44 |
|
|
45 |
Version 0.16 |
Version 0.26 |
46 |
|
|
47 |
=cut |
=cut |
48 |
|
|
49 |
our $VERSION = '0.16'; |
our $VERSION = '0.26'; |
50 |
|
|
51 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
52 |
|
|
72 |
Return data structure |
Return data structure |
73 |
|
|
74 |
my $ds = WebPAC::Normalize::data_structure( |
my $ds = WebPAC::Normalize::data_structure( |
75 |
lookup => $lookup->lookup_hash, |
lookup => $lookup_hash, |
76 |
row => $row, |
row => $row, |
77 |
rules => $normalize_pl_config, |
rules => $normalize_pl_config, |
78 |
marc_encoding => 'utf-8', |
marc_encoding => 'utf-8', |
79 |
config => $config, |
config => $config, |
80 |
|
load_row_coderef => sub { |
81 |
|
my ($database,$input,$mfn) = shift; |
82 |
|
$store->load_row( database => $database, input => $input, id => $mfn ); |
83 |
|
}, |
84 |
); |
); |
85 |
|
|
86 |
Options C<lookup>, C<row>, C<rules> and C<log> are mandatory while all |
Options C<row>, C<rules> and C<log> are mandatory while all |
87 |
other are optional. |
other are optional. |
88 |
|
|
89 |
|
C<load_row_coderef> is closure only used when executing lookups, so they will |
90 |
|
die if it's not defined. |
91 |
|
|
92 |
This function will B<die> if normalizastion can't be evaled. |
This function will B<die> if normalizastion can't be evaled. |
93 |
|
|
94 |
Since this function isn't exported you have to call it with |
Since this function isn't exported you have to call it with |
96 |
|
|
97 |
=cut |
=cut |
98 |
|
|
99 |
|
my $load_row_coderef; |
100 |
|
|
101 |
sub data_structure { |
sub data_structure { |
102 |
my $arg = {@_}; |
my $arg = {@_}; |
103 |
|
|
105 |
die "need normalisation argument" unless ($arg->{rules}); |
die "need normalisation argument" unless ($arg->{rules}); |
106 |
|
|
107 |
no strict 'subs'; |
no strict 'subs'; |
108 |
_set_lookup( $arg->{lookup} ); |
_set_lookup( $arg->{lookup} ) if defined($arg->{lookup}); |
109 |
_set_rec( $arg->{row} ); |
_set_rec( $arg->{row} ); |
110 |
_set_config( $arg->{config} ); |
_set_config( $arg->{config} ) if defined($arg->{config}); |
111 |
_clean_ds( %{ $arg } ); |
_clean_ds( %{ $arg } ); |
112 |
|
$load_row_coderef = $arg->{load_row_coderef}; |
113 |
|
|
114 |
eval "$arg->{rules}"; |
eval "$arg->{rules}"; |
115 |
die "error evaling $arg->{rules}: $@\n" if ($@); |
die "error evaling $arg->{rules}: $@\n" if ($@); |
116 |
|
|
167 |
|
|
168 |
=cut |
=cut |
169 |
|
|
170 |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators); |
my ($out, $marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $leader); |
171 |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
my ($marc_record_offset, $marc_fetch_offset) = (0, 0); |
172 |
|
|
173 |
sub _get_ds { |
sub _get_ds { |
184 |
|
|
185 |
sub _clean_ds { |
sub _clean_ds { |
186 |
my $a = {@_}; |
my $a = {@_}; |
187 |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators) = (); |
($out,$marc_record, $marc_encoding, $marc_repeatable_subfield, $marc_indicators, $leader) = (); |
188 |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
($marc_record_offset, $marc_fetch_offset) = (0,0); |
189 |
$marc_encoding = $a->{marc_encoding}; |
$marc_encoding = $a->{marc_encoding}; |
190 |
} |
} |
203 |
$lookup = shift; |
$lookup = shift; |
204 |
} |
} |
205 |
|
|
206 |
|
=head2 _get_lookup |
207 |
|
|
208 |
|
Get current lookup hash |
209 |
|
|
210 |
|
my $lookup = _get_lookup(); |
211 |
|
|
212 |
|
=cut |
213 |
|
|
214 |
|
sub _get_lookup { |
215 |
|
return $lookup; |
216 |
|
} |
217 |
|
|
218 |
|
=head2 _set_load_row |
219 |
|
|
220 |
|
Setup code reference which will return L<data_structure> from |
221 |
|
L<WebPAC::Store> |
222 |
|
|
223 |
|
_set_load_row(sub { |
224 |
|
my ($database,$input,$mfn) = @_; |
225 |
|
$store->load_row( database => $database, input => $input, id => $mfn ); |
226 |
|
}); |
227 |
|
|
228 |
|
=cut |
229 |
|
|
230 |
|
sub _set_load_row { |
231 |
|
my $coderef = shift; |
232 |
|
confess "argument isn't CODE" unless ref($coderef) eq 'CODE'; |
233 |
|
|
234 |
|
$load_row_coderef = $coderef; |
235 |
|
} |
236 |
|
|
237 |
=head2 _get_marc_fields |
=head2 _get_marc_fields |
238 |
|
|
239 |
Get all fields defined by calls to C<marc> |
Get all fields defined by calls to C<marc> |
490 |
my ($offset,$value) = @_; |
my ($offset,$value) = @_; |
491 |
|
|
492 |
if ($offset) { |
if ($offset) { |
493 |
$out->{' leader'}->{ $offset } = $value; |
$leader->{ $offset } = $value; |
494 |
} else { |
} else { |
495 |
return $out->{' leader'}; |
return $leader; |
496 |
} |
} |
497 |
} |
} |
498 |
|
|
517 |
foreach (@_) { |
foreach (@_) { |
518 |
my $v = $_; # make var read-write for Encode |
my $v = $_; # make var read-write for Encode |
519 |
next unless (defined($v) && $v !~ /^\s*$/); |
next unless (defined($v) && $v !~ /^\s*$/); |
|
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
|
520 |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$f}) ? @{ $marc_indicators->{$f} } : (' ',' '); |
521 |
if (defined $sf) { |
if (defined $sf) { |
522 |
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
push @{ $marc_record->[ $marc_record_offset ] }, [ $f, $i1, $i2, $sf => $v ]; |
587 |
|
|
588 |
warn "### marc_compose input subfields = ", dump(@_),$/ if ($debug > 2); |
warn "### marc_compose input subfields = ", dump(@_),$/ if ($debug > 2); |
589 |
|
|
590 |
|
if ($#_ % 2 != 1) { |
591 |
|
die "ERROR: marc_compose",dump($f,@_)," not valid (must be even).\nDo you need to add first() or join() around some argument?\n"; |
592 |
|
} |
593 |
|
|
594 |
while (@_) { |
while (@_) { |
595 |
my $sf = shift or die "marc_compose $f needs subfield"; |
my $sf = shift; |
596 |
my $v = shift; |
my $v = shift; |
597 |
|
|
598 |
next unless (defined($v) && $v !~ /^\s*$/); |
next unless (defined($v) && $v !~ /^\s*$/); |
|
from_to($v, 'iso-8859-2', $marc_encoding) if ($marc_encoding); |
|
599 |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
warn "## ++ marc_compose($f,$sf,$v) ", dump( $m ),$/ if ($debug > 1); |
600 |
if ($sf ne '+') { |
if ($sf ne '+') { |
601 |
push @$m, ( $sf, $v ); |
push @$m, ( $sf, $v ); |
638 |
|
|
639 |
This will erase field C<200> or C<200^a> from current MARC record. |
This will erase field C<200> or C<200^a> from current MARC record. |
640 |
|
|
641 |
|
marc_remove('*'); |
642 |
|
|
643 |
|
Will remove all fields in current MARC record. |
644 |
|
|
645 |
This is useful after calling C<marc_duplicate> or on it's own (but, you |
This is useful after calling C<marc_duplicate> or on it's own (but, you |
646 |
should probably just remove that subfield definition if you are not |
should probably just remove that subfield definition if you are not |
647 |
using C<marc_duplicate>). |
using C<marc_duplicate>). |
659 |
|
|
660 |
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
warn "### marc_remove before = ", dump( $marc ), $/ if ($debug > 2); |
661 |
|
|
662 |
my $i = 0; |
if ($f eq '*') { |
663 |
foreach ( 0 .. $#{ $marc } ) { |
|
664 |
last unless (defined $marc->[$i]); |
delete( $marc_record->[ $marc_record_offset ] ); |
665 |
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
666 |
if ($marc->[$i]->[0] eq $f) { |
|
667 |
if (! defined $sf) { |
} else { |
668 |
# remove whole field |
|
669 |
splice @$marc, $i, 1; |
my $i = 0; |
670 |
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
foreach ( 0 .. $#{ $marc } ) { |
671 |
$i--; |
last unless (defined $marc->[$i]); |
672 |
} else { |
warn "#### working on ",dump( @{ $marc->[$i] }), $/ if ($debug > 3); |
673 |
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
if ($marc->[$i]->[0] eq $f) { |
674 |
my $o = ($j * 2) + 3; |
if (! defined $sf) { |
675 |
if ($marc->[$i]->[$o] eq $sf) { |
# remove whole field |
676 |
# remove subfield |
splice @$marc, $i, 1; |
677 |
splice @{$marc->[$i]}, $o, 2; |
warn "#### slice \@\$marc, $i, 1 = ",dump( @{ $marc }), $/ if ($debug > 3); |
678 |
warn "#### slice \@{\$marc->[$i]}, $o, 2 = ", dump( @{ $marc }), $/ if ($debug > 3); |
$i--; |
679 |
# is record now empty? |
} else { |
680 |
if ($#{ $marc->[$i] } == 2) { |
foreach my $j ( 0 .. (( $#{ $marc->[$i] } - 3 ) / 2) ) { |
681 |
splice @$marc, $i, 1; |
my $o = ($j * 2) + 3; |
682 |
warn "#### slice \@\$marc, $i, 1 = ", dump( @{ $marc }), $/ if ($debug > 3); |
if ($marc->[$i]->[$o] eq $sf) { |
683 |
$i--; |
# remove subfield |
684 |
}; |
splice @{$marc->[$i]}, $o, 2; |
685 |
|
warn "#### slice \@{\$marc->[$i]}, $o, 2 = ", dump( @{ $marc }), $/ if ($debug > 3); |
686 |
|
# is record now empty? |
687 |
|
if ($#{ $marc->[$i] } == 2) { |
688 |
|
splice @$marc, $i, 1; |
689 |
|
warn "#### slice \@\$marc, $i, 1 = ", dump( @{ $marc }), $/ if ($debug > 3); |
690 |
|
$i--; |
691 |
|
}; |
692 |
|
} |
693 |
} |
} |
694 |
} |
} |
695 |
} |
} |
696 |
|
$i++; |
697 |
} |
} |
|
$i++; |
|
|
} |
|
698 |
|
|
699 |
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
warn "### marc_remove($f", $sf ? ",$sf" : "", ") after = ", dump( $marc ), $/ if ($debug > 2); |
700 |
|
|
701 |
$marc_record->[ $marc_record_offset ] = $marc; |
$marc_record->[ $marc_record_offset ] = $marc; |
702 |
|
} |
703 |
|
|
704 |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
warn "## full marc_record = ", dump( @{ $marc_record }), $/ if ($debug > 1); |
705 |
} |
} |
745 |
|
|
746 |
die "field $from doesn't have even number of subfields specifications\n" unless($#sfs % 2 == 1); |
die "field $from doesn't have even number of subfields specifications\n" unless($#sfs % 2 == 1); |
747 |
|
|
748 |
warn "#--> d: ",dump($d), "\n#--> sfs: ",dump(@sfs),$/; |
warn "#--> d: ",dump($d), "\n#--> sfs: ",dump(@sfs),$/ if ($debug > 2); |
749 |
|
|
750 |
my $m = [ $to, $i1, $i2 ]; |
my $m = [ $to, $i1, $i2 ]; |
751 |
|
|
752 |
while (my $sf = shift @sfs) { |
while (my $sf = shift @sfs) { |
753 |
warn "#--> sf: ",dump($sf), $/; |
|
754 |
|
warn "#--> sf: ",dump($sf), $/ if ($debug > 2); |
755 |
my $offset = shift @sfs; |
my $offset = shift @sfs; |
756 |
die "corrupted sufields specification for field $from\n" unless defined($offset); |
die "corrupted sufields specification for field $from\n" unless defined($offset); |
757 |
|
|
772 |
} |
} |
773 |
|
|
774 |
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
warn "## marc_record = ", dump( $marc_record ),$/ if ($debug > 1); |
|
|
|
|
warn "# marc_original_order is partly implemented"; |
|
775 |
} |
} |
776 |
|
|
777 |
|
|
780 |
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 |
781 |
above. |
above. |
782 |
|
|
783 |
|
=head2 _pack_subfields_hash |
784 |
|
|
785 |
|
@subfields = _pack_subfields_hash( $h ); |
786 |
|
$subfields = _pack_subfields_hash( $h, 1 ); |
787 |
|
|
788 |
|
Return each subfield value in array or pack them all together and return scalar |
789 |
|
with subfields (denoted by C<^>) and values. |
790 |
|
|
791 |
|
=cut |
792 |
|
|
793 |
|
sub _pack_subfields_hash { |
794 |
|
|
795 |
|
warn "## _pack_subfields_hash( ",dump(@_), " )\n" if ($debug > 1); |
796 |
|
|
797 |
|
my ($h,$include_subfields) = @_; |
798 |
|
|
799 |
|
if ( defined($h->{subfields}) ) { |
800 |
|
my $sfs = delete $h->{subfields} || die "no subfields?"; |
801 |
|
my @out; |
802 |
|
while (@$sfs) { |
803 |
|
my $sf = shift @$sfs; |
804 |
|
push @out, '^' . $sf if ($include_subfields); |
805 |
|
my $o = shift @$sfs; |
806 |
|
if ($o == 0 && ref( $h->{$sf} ) ne 'ARRAY' ) { |
807 |
|
# single element subfields are not arrays |
808 |
|
#warn "====> $sf $o / $#$sfs ", dump( $sfs, $h->{$sf} ), "\n"; |
809 |
|
|
810 |
|
push @out, $h->{$sf}; |
811 |
|
} else { |
812 |
|
#warn "====> $sf $o / $#$sfs ", dump( $sfs, $h->{$sf} ), "\n"; |
813 |
|
push @out, $h->{$sf}->[$o]; |
814 |
|
} |
815 |
|
} |
816 |
|
if ($include_subfields) { |
817 |
|
return join('', @out); |
818 |
|
} else { |
819 |
|
return @out; |
820 |
|
} |
821 |
|
} else { |
822 |
|
if ($include_subfields) { |
823 |
|
my $out = ''; |
824 |
|
foreach my $sf (sort keys %$h) { |
825 |
|
if (ref($h->{$sf}) eq 'ARRAY') { |
826 |
|
$out .= '^' . $sf . join('^' . $sf, @{ $h->{$sf} }); |
827 |
|
} else { |
828 |
|
$out .= '^' . $sf . $h->{$sf}; |
829 |
|
} |
830 |
|
} |
831 |
|
return $out; |
832 |
|
} else { |
833 |
|
# FIXME this should probably be in alphabetical order instead of hash order |
834 |
|
values %{$h}; |
835 |
|
} |
836 |
|
} |
837 |
|
} |
838 |
|
|
839 |
=head2 rec1 |
=head2 rec1 |
840 |
|
|
841 |
Return all values in some field |
Return all values in some field |
852 |
return unless (defined($rec) && defined($rec->{$f})); |
return unless (defined($rec) && defined($rec->{$f})); |
853 |
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
warn "rec1($f) = ", dump( $rec->{$f} ), $/ if ($debug > 1); |
854 |
if (ref($rec->{$f}) eq 'ARRAY') { |
if (ref($rec->{$f}) eq 'ARRAY') { |
855 |
return map { |
my @out; |
856 |
if (ref($_) eq 'HASH') { |
foreach my $h ( @{ $rec->{$f} } ) { |
857 |
values %{$_}; |
if (ref($h) eq 'HASH') { |
858 |
|
push @out, ( _pack_subfields_hash( $h ) ); |
859 |
} else { |
} else { |
860 |
$_; |
push @out, $h; |
861 |
} |
} |
862 |
} @{ $rec->{$f} }; |
} |
863 |
|
return @out; |
864 |
} elsif( defined($rec->{$f}) ) { |
} elsif( defined($rec->{$f}) ) { |
865 |
return $rec->{$f}; |
return $rec->{$f}; |
866 |
} |
} |
895 |
@v = rec('200') |
@v = rec('200') |
896 |
@v = rec('200','a') |
@v = rec('200','a') |
897 |
|
|
898 |
|
If rec() returns just single value, it will |
899 |
|
return scalar, not array. |
900 |
|
|
901 |
=cut |
=cut |
902 |
|
|
903 |
sub rec { |
sub rec { |
907 |
} elsif ($#_ == 1) { |
} elsif ($#_ == 1) { |
908 |
@out = rec2(@_); |
@out = rec2(@_); |
909 |
} |
} |
910 |
if (@out) { |
if ($#out == 0 && ! wantarray) { |
911 |
|
return $out[0]; |
912 |
|
} elsif (@out) { |
913 |
return @out; |
return @out; |
914 |
} else { |
} else { |
915 |
return ''; |
return ''; |
993 |
|
|
994 |
Consult lookup hashes for some value |
Consult lookup hashes for some value |
995 |
|
|
996 |
@v = lookup( $v ); |
@v = lookup( |
997 |
@v = lookup( @v ); |
sub { |
998 |
|
'ffkk/peri/mfn'.rec('000') |
999 |
|
}, |
1000 |
|
'ffkk','peri','200-a-200-e', |
1001 |
|
sub { |
1002 |
|
first(rec(200,'a')).' '.first(rec('200','e')) |
1003 |
|
} |
1004 |
|
); |
1005 |
|
|
1006 |
|
Code like above will be B<automatically generated> using L<WebPAC::Parse> from |
1007 |
|
normal lookup definition in C<conf/lookup/something.pl> which looks like: |
1008 |
|
|
1009 |
|
lookup( |
1010 |
|
# which results to return from record recorded in lookup |
1011 |
|
sub { 'ffkk/peri/mfn' . rec('000') }, |
1012 |
|
# from which database and input |
1013 |
|
'ffkk','peri', |
1014 |
|
# such that following values match |
1015 |
|
sub { first(rec(200,'a')) . ' ' . first(rec('200','e')) }, |
1016 |
|
# if this part is missing, we will try to match same fields |
1017 |
|
# from lookup record and current one, or you can override |
1018 |
|
# which records to use from current record using |
1019 |
|
sub { rec('900','x') . ' ' . rec('900','y') }, |
1020 |
|
) |
1021 |
|
|
1022 |
|
You can think about this lookup as SQL (if that helps): |
1023 |
|
|
1024 |
|
select |
1025 |
|
sub { what } |
1026 |
|
from |
1027 |
|
database, input |
1028 |
|
where |
1029 |
|
sub { filter from lookuped record } |
1030 |
|
having |
1031 |
|
sub { optional filter on current record } |
1032 |
|
|
1033 |
|
Easy as pie, right? |
1034 |
|
|
1035 |
=cut |
=cut |
1036 |
|
|
1037 |
sub lookup { |
sub lookup { |
1038 |
my $k = shift or return; |
my ($what, $database, $input, $key, $having) = @_; |
1039 |
return unless (defined($lookup->{$k})); |
|
1040 |
if (ref($lookup->{$k}) eq 'ARRAY') { |
confess "lookup needs 5 arguments: what, database, input, key, having\n" unless ($#_ == 4); |
1041 |
return @{ $lookup->{$k} }; |
|
1042 |
|
warn "## lookup ($database, $input, $key)", $/ if ($debug > 1); |
1043 |
|
return unless (defined($lookup->{$database}->{$input}->{$key})); |
1044 |
|
|
1045 |
|
confess "lookup really need load_row_coderef added to data_structure\n" unless ($load_row_coderef); |
1046 |
|
|
1047 |
|
my $mfns; |
1048 |
|
my @having = $having->(); |
1049 |
|
|
1050 |
|
warn "## having = ", dump( @having ) if ($debug > 2); |
1051 |
|
|
1052 |
|
foreach my $h ( @having ) { |
1053 |
|
if (defined($lookup->{$database}->{$input}->{$key}->{$h})) { |
1054 |
|
warn "lookup for $database/$input/$key/$h return ",dump($lookup->{$database}->{$input}->{$key}->{$h}),"\n" if ($debug); |
1055 |
|
$mfns->{$_}++ foreach keys %{ $lookup->{$database}->{$input}->{$key}->{$h} }; |
1056 |
|
} |
1057 |
|
} |
1058 |
|
|
1059 |
|
return unless ($mfns); |
1060 |
|
|
1061 |
|
my @mfns = sort keys %$mfns; |
1062 |
|
|
1063 |
|
warn "# lookup loading $database/$input/$key mfn ", join(",",@mfns)," having ",dump(@having),"\n" if ($debug); |
1064 |
|
|
1065 |
|
my $old_rec = $rec; |
1066 |
|
my @out; |
1067 |
|
|
1068 |
|
foreach my $mfn (@mfns) { |
1069 |
|
$rec = $load_row_coderef->( $database, $input, $mfn ); |
1070 |
|
|
1071 |
|
warn "got $database/$input/$mfn = ", dump($rec), $/ if ($debug); |
1072 |
|
|
1073 |
|
my @vals = $what->(); |
1074 |
|
|
1075 |
|
push @out, ( @vals ); |
1076 |
|
|
1077 |
|
warn "lookup for mfn $mfn returned ", dump(@vals), $/ if ($debug); |
1078 |
|
} |
1079 |
|
|
1080 |
|
# if (ref($lookup->{$k}) eq 'ARRAY') { |
1081 |
|
# return @{ $lookup->{$k} }; |
1082 |
|
# } else { |
1083 |
|
# return $lookup->{$k}; |
1084 |
|
# } |
1085 |
|
|
1086 |
|
$rec = $old_rec; |
1087 |
|
|
1088 |
|
warn "## lookup returns = ", dump(@out), $/ if ($debug); |
1089 |
|
|
1090 |
|
if ($#out == 0) { |
1091 |
|
return $out[0]; |
1092 |
} else { |
} else { |
1093 |
return $lookup->{$k}; |
return @out; |
1094 |
} |
} |
1095 |
} |
} |
1096 |
|
|
1097 |
|
=head2 save_into_lookup |
1098 |
|
|
1099 |
|
Save value into lookup. It associates current database, input |
1100 |
|
and specific keys with one or more values which will be |
1101 |
|
associated over MFN. |
1102 |
|
|
1103 |
|
MFN will be extracted from first occurence current of field 000 |
1104 |
|
in current record, or if it doesn't exist from L<_set_config> C<_mfn>. |
1105 |
|
|
1106 |
|
my $nr = save_into_lookup($database,$input,$key,sub { |
1107 |
|
# code which produce one or more values |
1108 |
|
}); |
1109 |
|
|
1110 |
|
It returns number of items saved. |
1111 |
|
|
1112 |
|
This function shouldn't be called directly, it's called from code created by |
1113 |
|
L<WebPAC::Parser>. |
1114 |
|
|
1115 |
|
=cut |
1116 |
|
|
1117 |
|
sub save_into_lookup { |
1118 |
|
my ($database,$input,$key,$coderef) = @_; |
1119 |
|
die "save_into_lookup needs database" unless defined($database); |
1120 |
|
die "save_into_lookup needs input" unless defined($input); |
1121 |
|
die "save_into_lookup needs key" unless defined($key); |
1122 |
|
die "save_into_lookup needs CODE" unless ( defined($coderef) && ref($coderef) eq 'CODE' ); |
1123 |
|
|
1124 |
|
warn "## save_into_lookup rec = ", dump($rec), " config = ", dump($config), $/ if ($debug > 2); |
1125 |
|
|
1126 |
|
my $mfn = |
1127 |
|
defined($rec->{'000'}->[0]) ? $rec->{'000'}->[0] : |
1128 |
|
defined($config->{_mfn}) ? $config->{_mfn} : |
1129 |
|
die "mfn not defined or zero"; |
1130 |
|
|
1131 |
|
my $nr = 0; |
1132 |
|
|
1133 |
|
foreach my $v ( $coderef->() ) { |
1134 |
|
$lookup->{$database}->{$input}->{$key}->{$v}->{$mfn}++; |
1135 |
|
warn "# saved lookup $database/$input/$key [$v] $mfn\n" if ($debug > 1); |
1136 |
|
$nr++; |
1137 |
|
} |
1138 |
|
|
1139 |
|
return $nr; |
1140 |
|
} |
1141 |
|
|
1142 |
=head2 config |
=head2 config |
1143 |
|
|
1144 |
Consult config values stored in C<config.yml> |
Consult config values stored in C<config.yml> |
1263 |
} |
} |
1264 |
} |
} |
1265 |
|
|
1266 |
|
my $hash; |
1267 |
|
|
1268 |
|
=head2 set |
1269 |
|
|
1270 |
|
set( key => 'value' ); |
1271 |
|
|
1272 |
|
=cut |
1273 |
|
|
1274 |
|
sub set { |
1275 |
|
my ($k,$v) = @_; |
1276 |
|
warn "## set ( $k => ", dump($v), " )", $/ if ( $debug ); |
1277 |
|
$hash->{$k} = $v; |
1278 |
|
}; |
1279 |
|
|
1280 |
|
=head2 get |
1281 |
|
|
1282 |
|
get( 'key' ); |
1283 |
|
|
1284 |
|
=cut |
1285 |
|
|
1286 |
|
sub get { |
1287 |
|
my $k = shift || return; |
1288 |
|
my $v = $hash->{$k}; |
1289 |
|
warn "## get $k = ", dump( $v ), $/ if ( $debug ); |
1290 |
|
return $v; |
1291 |
|
} |
1292 |
|
|
1293 |
|
=head2 count |
1294 |
|
|
1295 |
|
if ( count( @result ) == 1 ) { |
1296 |
|
# do something if only 1 result is there |
1297 |
|
} |
1298 |
|
|
1299 |
|
=cut |
1300 |
|
|
1301 |
|
sub count { |
1302 |
|
warn "## count ",dump(@_),$/ if ( $debug ); |
1303 |
|
return @_ . ''; |
1304 |
|
} |
1305 |
|
|
1306 |
# END |
# END |
1307 |
1; |
1; |