use strict;
use warnings;
-use XML::Twig;
-use YAML::Tiny;
use Getopt::Long;
+use Equinox::Migration::MapDrivenMARCXMLProc;
+use Equinox::Migration::MARCXMLSampler;
-my $conf = initialize();
-my $marcxml = shift;
+my $c = initialize();
-open HOLDINGS, '>', $conf->{output};
-open X, '>', $conf->{pubnotesfile};
-open Z, '>', $conf->{privnotesfile};
-my $holdings = {};
-my %sample = ( x => {}, z => {} ); # hash of all subfields in all 852s
-my $copyid = 0;
+# run samples if we've been asked for them
+run_samples($c) if ($c->{sample} or $c->{samplemap} or $c->{samplestr});
-$| = 1;
-my $count = 0;
-my $total = `grep -c '<record' $marcxml`;
-my $percent = 0;
-my $prevper = -1;
-
-my $yaml = YAML::Tiny->new;
-my $t = XML::Twig->new( twig_handlers => { record => \&record } );
-$t->parsefile($marcxml);
-$yaml->[0] = \%sample;
-$yaml->write('holdings.sample');
-print "\n\n";
-
-sub record {
- my($t, $r)= @_;
- $holdings = { copies => [] };
-
- my @dfields = $r->children('datafield');
- for my $d (@dfields)
- { process_datafields($d) }
-
- for my $copy (@{$holdings->{copies}})
- { print_reports($copy) }
- $r->purge;
-
- $count++;
- $percent = int(($count / $total) * 100);
- print "\r$percent% done ($count)" if ($percent != $prevper);
- $prevper = $percent;
-}
-
-sub process_datafields {
- my ($d) = @_;
- # get 903
- if ($d->{'att'}->{'tag'} == 903) {
- my $s = $d->first_child('subfield');
- $holdings->{id} = $s->text;;
+sub run_samples {
+ my ($c) = @_;
+ my $s;
+ if ($c->{samplemap}) {
+ $s = Equinox::Migration::MARCXMLSampler->new( marcfile => $c->{marcfile},
+ mapfile => $c->{samplemap});
+ } elsif ($c->{samplestr}) {
+ $s = Equinox::Migration::MARCXMLSampler->new( marcfile => $c->{marcfile},
+ mapstring => $c->{samplestr});
+ } else {
+ $s = Equinox::Migration::MARCXMLSampler->new( marcfile => $c->{marcfile} );
}
+ $s->parse_records;
- # and holdings data
- if ($d->{'att'}->{'tag'} == $conf->{tag}) {
- push @{$holdings->{copies}}, { x =>[], z => [] };
- $holdings->{copies}[-1]{copyid} = $copyid;
- my @subs = $d->children('subfield');
- for my $s (@subs)
- { process_subs($s) }
- $copyid++;
- }
+ dump_sample_overview($c, $s) if $c->{sample};
+ dump_sample_detail($c, $s) if ($c->{samplemap} or $c->{samplestr});
}
-sub process_subs {
- my ($s) = @_;
- my $copy = $holdings->{copies}[-1];
-
- my $code = $s->{'att'}->{'code'};
- my $value = $s->text;
-
- if ($code eq $conf->{pubnotes} or $code eq $conf->{privnotes}) {
- push @{$copy->{$code}}, $value;
- my ($k,$v) = split /:/, $value;
- $sample{$code}{$k} = $v;
- } else {
- $copy->{$code} = $value;
- $sample{$code} = $value;
+sub dump_sample_detail {
+ my ($c, $s) = @_;
+ my $tags = $s->{data}{samp};
+ my $count = $s->{data}{rcnt};
+
+ open DETAIL, '>', ($c->{prefix} . "-HOLDINGS-DETAIL.txt");
+ select DETAIL;
+ for my $tag (sort keys %{$tags}) {
+ print ">>>>> TAG $tag\n";
+ for my $subkey (sort keys %{$tags->{$tag}}) {
+ my $sub = $tags->{$tag}{$subkey};
+ print " Subfield: $subkey\n";
+ print " Sample: '", $sub->{value}, "'\n";
+ print " Count: ", $sub->{count}, " in ", $sub->{tcnt}, " tags\n\n";
+ #print "(", int($sub->{count} / $sub->{rcnt}), "%)\n";
+ }
}
+ select STDOUT;
+ close DETAIL;
}
-sub print_reports {
- return unless defined $holdings->{id};
- my ($copy) = @_;
- my $note = 0;
- for (@{$copy->{x}}) {
- print X join("\t", $holdings->{id}, $copy->{copyid}, $note, $_), "\n";
- $note++;
- }
- $note = 0;
- for (@{$copy->{z}}) {
- print Z join("\t", $holdings->{id}, $copy->{copyid}, $note, $_), "\n";
- $note++;
+sub dump_sample_overview {
+ my ($c, $s) = @_;
+ my $tags = $s->{data}{tags};
+ my $count = $s->{data}{rcnt};
+
+ my @tagsbyname = sort keys %{$tags};
+ my @tagsbycount = reverse sort { $tags->{$a} <=> $tags->{$b} } keys %{$tags};
+
+ open SAMPLE, '>', ($c->{prefix} . "-HOLDINGS-OVERVIEW.txt");
+ select SAMPLE;
+ print "FOUND TAGS (BY TAG) FOUND TAGS (BY COUNT)\n";
+ print "------------------- ---------------------\n";
+ for my $i (0 .. @tagsbyname - 1) {
+ print $tagsbyname[$i], (" " x (14 - length $tags->{ $tagsbyname[$i] })),
+ $tags->{ $tagsbyname[$i] }, " ";
+ print $tagsbycount[$i], (" " x (16 - length $tags->{ $tagsbycount[$i] })),
+ $tags->{ $tagsbycount[$i] }, "\n";
}
- my @fields = ();
- for ( @{$conf->{fields}} )
- { $copy->{$_} = '' unless defined $copy->{$_}; push @fields, $copy->{$_} }
- print HOLDINGS join("\t", $holdings->{id}, $copy->{copyid}, @fields), "\n";
+ select STDOUT;
+ close SAMPLE;
}
-
-#------------------------------------------------
-
+#--------------------------
sub initialize {
my $c = {};
binmode(STDIN, ':utf8');
my $rc = GetOptions( $c,
- 'fields|f=s',
- 'output|o=s',
+ 'sample|s',
+ 'samplemap|sm=s',
+ 'samplestr|ss=s',
+ 'marcfile|m=s',
+ 'ils|i=s',
+ 'library|l=s',
'prefix|p=s',
- 'pubnotes|pub=i',
- 'pubnotesfile=s',
- 'privnotes|priv=s',
- 'privnotesfile=s',
- 'tag|t=i',
'help|h',
);
show_help() unless $rc;
show_help() if ($c->{help});
- # set defaults
- $c->{prefix} = (defined $c->{prefix}) ? ($c->{prefix} . '.') : '';
- $c->{tag} = $c->{tag} || '852';
- $c->{output} =
- $c->{output} || join('', $c->{prefix}, "holdings.pg");
- $c->{pubnotes} = $c->{pubnotes} || 'x';
- $c->{pubnotesfile} =
- $c->{pubnotesfile} || join('', $c->{prefix}, "holdings.pubnote.pg");
- $c->{privnotes} = $c->{privnotes} || 'z';
- $c->{privnotesfile} =
- $c->{privnotesfile} || join('', $c->{prefix}, "holdings.privnote.pg");
-
my @keys = keys %{$c};
- show_help() unless (@ARGV and @keys);
- for my $key ('fields', 'tag')
+ for my $key ('prefix', 'ils')
{ push @missing, $key unless $c->{$key} }
if (@missing) {
print "Required option: ", join(', ', @missing), " missing!\n";
show_help();
}
- # explode and validate fields string
- process_fields($c);
return $c;
}
-sub process_fields {
- my ($c) = @_;
- my @holdings_fields = split /,/, $c->{fields};
- for (@holdings_fields) {
- die "Field names must be alphanumeric!\n" if /\W/;
- die "Field names must be single characters!\n"
- if /\w{2,}/;
- }
- $c->{fields} = \@holdings_fields;
-}
-
sub show_help {
print <<HELP;
-Usage is: extract_holdings MARCXML_FILE HOLDINGS_TAG
+Usage is: extract_holdings -p PREFIX -m MAPFILE -i ILS [OPTIONS] <MARCXML>
+ prefix p Prefix string for output filenames
+ map m Subfield map file to use
+ ils i Legacy ILS name (affects SQL generation)
+
+ sample s Comma-delineated list of datafield tags to be sampled
+ library l Legacy library name (affects SQL generation)
HELP
exit;
}
projects / migration-tools.git / blobdiff