16 |
marc_template |
marc_template |
17 |
|
|
18 |
rec1 rec2 rec |
rec1 rec2 rec |
19 |
frec |
frec frec_eq frec_ne |
20 |
regex prefix suffix surround |
regex prefix suffix surround |
21 |
first lookup join_with |
first lookup join_with |
22 |
save_into_lookup |
save_into_lookup |
48 |
|
|
49 |
=cut |
=cut |
50 |
|
|
51 |
our $VERSION = '0.33'; |
our $VERSION = '0.34'; |
52 |
|
|
53 |
=head1 SYNOPSIS |
=head1 SYNOPSIS |
54 |
|
|
815 |
return unless defined($rec->{$from}); |
return unless defined($rec->{$from}); |
816 |
|
|
817 |
my $r = $rec->{$from}; |
my $r = $rec->{$from}; |
818 |
die "record field $from isn't array\n" unless (ref($r) eq 'ARRAY'); |
die "record field $from isn't array ",dump( $rec ) unless (ref($r) eq 'ARRAY'); |
819 |
|
|
820 |
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
my ($i1,$i2) = defined($marc_indicators->{$to}) ? @{ $marc_indicators->{$to} } : (' ',' '); |
821 |
warn "## marc_original_order($to,$from) source = ", dump( $r ),$/ if ($debug > 1); |
warn "## marc_original_order($to,$from) source = ", dump( $r ),$/ if ($debug > 1); |
873 |
die "$_ not ARRAY" if ref($args->{$_}) ne 'ARRAY'; |
die "$_ not ARRAY" if ref($args->{$_}) ne 'ARRAY'; |
874 |
} |
} |
875 |
|
|
876 |
my $r = $rec->{ $args->{from} }; # || return; |
my $r = $rec->{ $args->{from} } || return; |
877 |
die "record field ", $args->{from}, " isn't array\n" unless (ref($r) eq 'ARRAY'); |
die "record field ", $args->{from}, " isn't array ",dump( $rec ) unless (ref($r) eq 'ARRAY'); |
878 |
|
|
879 |
my @subfields_rename = @{ $args->{subfields_rename} }; |
my @subfields_rename = @{ $args->{subfields_rename} }; |
880 |
# warn "### subfields_rename [$#subfields_rename] = ",dump( @subfields_rename ); |
# warn "### subfields_rename [$#subfields_rename] = ",dump( @subfields_rename ); |
933 |
|
|
934 |
my ( $new_r, $from_count, $to_count ); |
my ( $new_r, $from_count, $to_count ); |
935 |
foreach my $sf ( keys %{$r} ) { |
foreach my $sf ( keys %{$r} ) { |
936 |
|
# skip everything which isn't one char subfield (e.g. 'subfields') |
937 |
|
next unless $sf =~ m/^\w$/; |
938 |
my $nr = $from_count->{$sf}++; |
my $nr = $from_count->{$sf}++; |
939 |
my $rename_to = $subfields_rename->{ $sf }; # || |
my $rename_to = $subfields_rename->{ $sf } || |
940 |
# die "can't find subfield rename for $sf/$nr in ", dump( $subfields_rename ); |
die "can't find subfield rename for $sf/$nr in ", dump( $subfields_rename ); |
941 |
warn "### rename $sf/$nr to ", dump( $rename_to->[$nr] ), $/; |
warn "### rename $sf/$nr to ", dump( $rename_to->[$nr] ), $/; |
942 |
my ( $to_sf, $to_nr ) = @{ $rename_to->[$nr] }; |
my ( $to_sf, $to_nr ) = @{ $rename_to->[$nr] }; |
943 |
$new_r->{ $to_sf }->[ $to_nr ] = [ $sf => $nr ]; |
$new_r->{ $to_sf }->[ $to_nr ] = [ $sf => $nr ]; |
959 |
|
|
960 |
$fill_in = {}; |
$fill_in = {}; |
961 |
|
|
962 |
foreach my $sf ( split(/\|/, $template ) ) { |
my @templates = split(/\|/, $template ); |
963 |
|
@templates = ( $template ); |
964 |
|
|
965 |
|
foreach my $sf ( @templates ) { |
966 |
sub fill_in { |
sub fill_in { |
967 |
my ( $r, $sf, $nr ) = @_; |
my ( $r, $sf, $nr ) = @_; |
968 |
my ( $from_sf, $from_nr ) = @{ $new_r->{ $sf }->[ $nr ] }; |
my ( $from_sf, $from_nr ) = @{ $new_r->{ $sf }->[ $nr ] }; |
989 |
my ( $sf, $nr ) = @$sf; |
my ( $sf, $nr ) = @$sf; |
990 |
my $v = $fill_in->{$sf}->[$nr] || die "can't find fill_in $sf/$nr"; |
my $v = $fill_in->{$sf}->[$nr] || die "can't find fill_in $sf/$nr"; |
991 |
warn "++ $sf/$nr |$v|\n"; |
warn "++ $sf/$nr |$v|\n"; |
992 |
push @$m, [ $sf, $v ]; |
push @$m, ( $sf, $v ); |
993 |
} |
} |
994 |
|
|
995 |
warn "#### >>>> created marc: ", dump( $m ); |
warn "#### >>>> created marc: ", dump( $m ); |
998 |
} |
} |
999 |
|
|
1000 |
warn "### marc_template produced: ",dump( @marc_out ); |
warn "### marc_template produced: ",dump( @marc_out ); |
1001 |
return @marc_out; |
|
1002 |
|
foreach my $marc ( @marc_out ) { |
1003 |
|
warn "+++ ",dump( $marc ); |
1004 |
|
push @{ $marc_record->[ $marc_record_offset ] }, $marc; |
1005 |
|
} |
1006 |
} |
} |
1007 |
|
|
1008 |
=head2 marc_count |
=head2 marc_count |
1177 |
return shift @out; |
return shift @out; |
1178 |
} |
} |
1179 |
|
|
1180 |
|
=head2 frec_eq |
1181 |
|
|
1182 |
|
=head2 frec_ne |
1183 |
|
|
1184 |
|
Check if first values from two fields are same or different |
1185 |
|
|
1186 |
|
if ( frec_eq( 900 => 'a', 910 => 'c' ) ) { |
1187 |
|
# values are same |
1188 |
|
} else { |
1189 |
|
# values are different |
1190 |
|
} |
1191 |
|
|
1192 |
|
Strictly speaking C<frec_eq> and C<frec_ne> wouldn't be needed if you |
1193 |
|
could write something like: |
1194 |
|
|
1195 |
|
if ( frec( '900','a' ) eq frec( '910','c' ) ) { |
1196 |
|
# yada tada |
1197 |
|
} |
1198 |
|
|
1199 |
|
but you can't since our parser L<WebPAC::Parser> will remove all whitespaces |
1200 |
|
in order to parse text and create invalid function C<eqfrec>. |
1201 |
|
|
1202 |
|
=cut |
1203 |
|
|
1204 |
|
sub frec_eq { |
1205 |
|
my ( $f1,$sf1, $f2, $sf2 ) = @_; |
1206 |
|
return (rec( $f1, $sf1 ))[0] eq (rec( $f2, $sf2 ))[0]; |
1207 |
|
} |
1208 |
|
|
1209 |
|
sub frec_ne { |
1210 |
|
return ! frec_eq( @_ ); |
1211 |
|
} |
1212 |
|
|
1213 |
=head2 regex |
=head2 regex |
1214 |
|
|
1215 |
Apply regex to some or all values |
Apply regex to some or all values |