X-Git-Url: http://git.equinoxoli.org/?p=migration-tools.git;a=blobdiff_plain;f=fingerprinter;h=f388174a447f8bc0554d384aa58b9389b71a4eb2;hp=ef5a29300881df327c620995f7a774fbbdaa67fb;hb=2b5820e71bb2c4b89c6460f63fb3a148807716b0;hpb=1609ab31adab3cf6d3beb8cbca64e1a6d8405f80 diff --git a/fingerprinter b/fingerprinter index ef5a293..f388174 100755 --- a/fingerprinter +++ b/fingerprinter @@ -1,4 +1,21 @@ #!/usr/bin/perl + +# Copyright 2009-2012, Equinox Software, Inc. +# +# This program is free software; you can redistribute it and/or +# modify it under the terms of the GNU General Public License +# as published by the Free Software Foundation; either version 2 +# of the License, or (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write to the Free Software +# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + use strict; use warnings; use open ':utf8'; @@ -7,6 +24,8 @@ use Getopt::Long; use MARC::Batch; use Unicode::Normalize; use MARC::File::XML ( BinaryEncoding => 'utf-8' ); +use Equinox::Migration::SubfieldMapper; +use Equinox::Migration::Utils qw/normalize_oclc_number/; my $conf = {}; # configuration hashref my $count = 0; my $scount = 0; @@ -15,18 +34,25 @@ $| = 1; initialize($conf); -open OF, '>', $conf->{output}; -open XF, '>', $conf->{exception}; +open OF, '>', $conf->{output} or die "$0: cannot open output file $conf->{output}: $!\n"; +open XF, '>', $conf->{exception} or die "$0: cannot open exception file $conf->{exception}: $!\n"; for my $file (@ARGV) { print XF "Processing $file\n"; - my $batch = undef; my $record = undef; - $batch = MARC::Batch->new($conf->{marctype}, $file); + my $batch = MARC::Batch->new($conf->{marctype}, $file); $batch->strict_off(); $batch->warnings_off(); - while ( $record = $batch->next ) { + my $record; + while ( 1 ) { + eval { $record = $batch->next; }; + if ($@) { + import MARC::File::XML; + print "skipping bad record: $@\n"; + next; + } + last unless $record; $count++; progress_ticker(); my $marc = undef; unless ( defined $record ) @@ -41,6 +67,8 @@ for my $file (@ARGV) { # populate and normalize marc $marc = populate_marc($record, $id); + # check for manual exclusion + next if this_record_is_excluded($record, $marc); normalize_marc($marc); unless (marc_isvalid($marc)) { dump_exception($marc); next; } @@ -74,29 +102,33 @@ sub populate_marc { # date1, date2 my $my_008 = $record->field('008'); - $my_008 = $my_008->as_string() if ($my_008); - if (defined $my_008) { - unless (length $my_008 == 40) { - $my_008 = $my_008 . (' ' x (40 - length($my_008))); - print XF ">> Short 008 padded to ",length($my_008)," at rec $count\n"; + $marc{tag008} = $my_008->as_string() if ($my_008); + if (defined $marc{tag008}) { + unless (length $marc{tag008} == 40) { + $marc{tag008} = $marc{tag008} . ('|' x (40 - length($marc{tag008}))); + print XF ">> Short 008 padded to ",length($marc{tag008})," at rec $count\n"; } - $marc{date1} = substr($my_008,7,4) if ($my_008); - $marc{date2} = substr($my_008,11,4) if ($my_008); # UNUSED + $marc{date1} = substr($marc{tag008},7,4) if ($marc{tag008}); + $marc{date2} = substr($marc{tag008},11,4) if ($marc{tag008}); # UNUSED } unless ($marc{date1} and $marc{date1} =~ /\d{4}/) { my $my_260 = $record->field('260'); - my $date1 = $my_260->subfield('c') if $my_260; - if (defined $date1 and $date1 =~ /\d{4}/) { - $marc{date1} = $date1; - print XF ">> using 260c as date1 at rec $count\n"; + if ($my_260 and $my_260->subfield('c')) { + my $date1 = $my_260->subfield('c'); + $date1 =~ s/\D//g; + if (defined $date1 and $date1 =~ /\d{4}/) { + $marc{date1} = $date1; + $marc{fudgedate} = 1; + print XF ">> using 260c as date1 at rec $count\n"; + } } } # item_form if ( $marc{record_type} =~ /[gkroef]/ ) { # MAP, VIS - $marc{item_form} = substr($my_008,29,1) if ($my_008); + $marc{item_form} = substr($marc{tag008},29,1) if ($marc{tag008}); } else { - $marc{item_form} = substr($my_008,23,1) if ($my_008); + $marc{item_form} = substr($marc{tag008},23,1) if ($marc{tag008}); } # isbns @@ -117,18 +149,31 @@ sub populate_marc { # oclc $marc{oclc} = []; - push @{ $marc{oclc} }, $record->field('001')->as_string() - if ($record->field('001') and $record->field('003') and - $record->field('003')->as_string() =~ /OCo{0,1}LC/); + if ($record->field('001') && + $record->field('003') && + $record->field('003')->as_string() =~ /OCo{0,1}LC/ && + defined normalize_oclc_number($record->field('001')->as_string())) { + push @{ $marc{oclc} }, normalize_oclc_number($record->field('001')->as_string()); + } for ($record->field('035')) { my $oclc = $_->subfield('a'); - push @{ $marc{oclc} }, $oclc - if (defined $oclc and $oclc =~ /\(OCoLC\)/ and $oclc =~/([0-9]+)/); + if (defined $oclc && + ($oclc =~ /\(OCoLC\)/ || $oclc =~ /(ocm|ocl7|ocn|on)/) && + defined normalize_oclc_number($oclc)) { + push @{ $marc{oclc} }, normalize_oclc_number($oclc); + } } - # "Accompanying material" (300e) - $marc{accomp} = $record->field('300')->subfield('e') - if $record->field('300'); + if ($record->field('999')) { + my $koha_bib_id = $record->field('999')->subfield('c'); + $marc{koha_bib_id} = $koha_bib_id if defined $koha_bib_id and $koha_bib_id =~ /^\d+$/; + } + + # "Accompanying material" and check for "copy" (300) + if ($record->field('300')) { + $marc{accomp} = $record->field('300')->subfield('e'); + $marc{tag300a} = $record->field('300')->subfield('a'); + } # issn, lccn, title, desc, pages, pub, pubyear, edition $marc{lccn} = $record->field('010')->subfield('a') if $record->field('010'); @@ -137,6 +182,10 @@ sub populate_marc { $marc{pages} = $1 if (defined $marc{desc} and $marc{desc} =~ /(\d+)/); $marc{title} = $record->field('245')->subfield('a') if $record->field('245'); + $marc{title} .= ' ' . $record->field('245')->subfield('b') + if ($record->field('245') and + $record->field('245')->subfield('b') and + not $conf->{ignoresubtitle}); $marc{edition} = $record->field('250')->subfield('a') if $record->field('250'); if ($record->field('260')) { @@ -208,69 +257,85 @@ sub marc_isvalid { Assign a score to the record based on various criteria. -Score is constructed by pushing elements onto a list. At the end of -the routine, the list is flattened into a string via join(); +Score is constructed by pushing elements onto a list, via a dispatch +table. This allows order of fingerprints in the output file to be +varied. =cut sub score_marc { my ($marc, $record) = @_; my @score = (); - my $chunk; - - # Is this an OCLC record? - if ($conf->{scores}{oclc}) { - if ($record->field('008')) { - $chunk = $record->field('008')->as_string(); - push @score, ( $chunk =~ /^o/i ? 1 : 0 ); - } else { - push @score, 0; - } - } + my $json = '{'; - # does 040a contain "dlc"? - if ($conf->{scores}{dlc}) { - if ($record->field('040') and $record->field('040')->subfield('a')) { - $chunk = $record->field('040')->subfield('a'); - push @score, ( $chunk =~ /dlc/i ? 1 : 0 ); - } else { - push @score, 0; - } - } + #---------------------------------- + # static criteria scoring + #---------------------------------- + $marc->{misc_score} = 999; + $marc->{age_score} = 999999999999; - # number of 650 datafields - # zero-padded to 4 digits with printf - if ($conf->{scores}{num_650}) { - if ($record->field('650')) { - my @tags = $record->field('650'); - push @score, scalar @tags; - } else { - push @score, '0000'; - } + # -1 if 008 has been padded, -2 if it doesn't exist + if ($marc->{tag008}) + { $marc->{misc_score}-- if ($marc->{tag008} =~ /\|$/) } + else + { $marc->{misc_score} -= 2 } + # -1 if date has been pulled from 260 + $marc->{misc_score}-- if $marc->{fudgedate}; + # -1 if this is a copy record + $marc->{misc_score}-- + if (defined $marc->{tag300a} and $marc->{tag300a} =~ /copy/i); + + # subtract record id if we want older records to win + #$marc->{age_score} -= $marc->{id} unless ($conf->{newwins}); + # handle arbitrary adjustments + $marc->{age_score} = 1; + if ($conf->{'arbitrarily-lose-above'}) { + $marc->{age_score} = 0 + if ($marc->{id} >= $conf->{'arbitrarily-lose-above'}); } - - # number of tags in total - # zero-padded to 4 digits with printf - if ($conf->{scores}{num_tags}) { - my @tags = $record->fields; - push @score, scalar @tags; + if ($conf->{'arbitrarily-lose-below'}) { + $marc->{age_score} = 0 + if ($marc->{id} <= $conf->{'arbitrarily-lose-below'}); } - # encoding level - if ($conf->{scores}{enc_lvl}) { - my $enc = substr($record->leader, 17, 1); + #---------------------------------- + # dynamic calculated scoring + #---------------------------------- + my %scores_code = ( + oclc => sub { return $marc->{oclc}[0] ? 1 : 0 }, + dlc => sub { + if ($record->field('040') and $record->field('040')->subfield('a')) + { return scalar($record->subfield( '040', 'a')) =~ /dlc/io ? 1 : 0 } + else { return 0 } + }, + num_650 => sub { + if ($record->field('650')) { + # can't say "scalar $record->field('650')"; MARC::Record + # behaves differently in list/scalar contexts + my @tags = $record->field('650'); + return sprintf("%04d", scalar @tags) + } else { return '0000' } + }, + num_tags=> sub { return sprintf( '%04d', scalar( $record->fields ) ) }, + enc_lvl => sub { + my $enc = substr($record->leader, 17, 1) || 'u'; my %levels = ( ' ' => 9, 1 => 8, 2 => 7, 3 => 6, 4 => 5, 5 => 4, 6 => 3, 7 => 2, 8 => 1, 'u' => 0, 'z' => 0 ); - if (defined $enc and $levels{$enc}) - { push @score, $levels{$enc} } - else - { push @score, 0 } + return $levels{$enc} || 0; + } + ); + + #---------------------------------- + # assemble and store scores + #---------------------------------- + for ( @{ $conf->{dyn_scores} } ) { + push @score, $scores_code{$_}->($marc, $record); + $json .= $_ . ':' . $score[-1] . ','; } + $json .= 'misc:' . $marc->{misc_score} . '}'; - # put score in marc hash - $marc->{score} = join('', '{oclc:', $score[0], ',dlc:', $score[1], - ',num_650:', $score[2], ',num_tags:', $score[3], - ',enc_lvl:', $score[4], '}'); + my $compact = join('', $marc->{age_score}, $marc->{misc_score}, @score); + $marc->{score} = "$compact\t$json"; } =head2 dump_fingerprints @@ -295,6 +360,14 @@ sub dump_fingerprints { } } + if ($conf->{fingerprints}{koha_bib_id} and exists $marc->{koha_bib_id}) { + print OF join("\t", $marc->{score}, $marc->{id}, "z_koha_bib_id", + $marc->{item_form}, $marc->{date1}, + $marc->{record_type}, + $marc->{bib_lvl}, $marc->{title}, + $marc->{koha_bib_id}), "\n"; + } + if ($conf->{fingerprints}{isbn}) { if ((scalar @{ $marc->{isbns} } > 0) and $marc->{pages}) { foreach my $isbn ( @{ $marc->{isbns}} ) { @@ -307,11 +380,11 @@ sub dump_fingerprints { } } - if ($conf->{fingerprints}{edition} and $marc->{edition}) { + if ($conf->{fingerprints}{edition} and $marc->{edition} and $marc->{author}) { print OF join("\t", $marc->{score}, $marc->{id}, "edition", $marc->{item_form}, $marc->{date1}, $marc->{record_type}, $marc->{bib_lvl}, - $marc->{title}, $marc->{edition}), "\n"; + $marc->{title}, $marc->{author}, $marc->{edition}), "\n"; } if ($conf->{fingerprints}{issn} and $marc->{issn}) { @@ -355,25 +428,67 @@ Write line of exception report =cut sub dump_exception { - my ($marc) = @_; + my ($marc, $msg) = @_; unless (defined $marc) { print XF "Undefined record at line $count; likely bad XML\n"; return; } - print XF "Record ", $marc->{id}, " did not make the cut: "; - print XF "Missing item_form. " unless ($marc->{item_form}); + + print XF "Record ", $marc->{id}, " excluded: "; + if (defined $msg) { + print XF "$msg\n"; + return + } + + print XF "missing item_form; " unless ($marc->{item_form}); unless (defined $marc->{date1}) - { print XF "Missing date1. " } + { print XF "missing date1; " } else - { print XF "Invalid date1: ", $marc->{date1}, " " + { print XF "invalid date1: '", $marc->{date1}, "'; " unless ($marc->{date1} =~ /\d{4}/); } - print XF "Missing record_type. " unless ($marc->{record_type}); - print XF "Missing bib_lvl. " unless ($marc->{bib_lvl}); - print XF "Missing title. " unless ($marc->{title}); + print XF "missing record_type; " unless ($marc->{record_type}); + print XF "missing bib_lvl; " unless ($marc->{bib_lvl}); + print XF "missing title " unless ($marc->{title}); print XF "\n"; } +=head2 this_record_is_excluded + +Returns 1 if the record B and 0 if the record B excluded, +according to the subfield mapping (generated via the C<--excludelist> +option). + +=cut + +sub this_record_is_excluded { + my ($rec, $marc) = @_; + return 0 unless defined $conf->{excludelist}; + + for my $tag (keys %{ $conf->{excludelist}->{tags} }) { + for my $sub (keys %{$conf->{excludelist}->{tags}{$tag}}) { + my $f = $conf->{excludelist}->field($tag, $sub); + + # if this record doesn't have the right tag/sub, it can't be + return 0 unless ($rec->field($tag) and $rec->field($tag)->subfield($sub)); + # but it does, so if there are no filters to check... + unless ($conf->{excludelist}->filters($f)) + { dump_exception($marc, "exclusion $tag$sub"); return 1 } + + my $sub_contents = $rec->field($tag)->subfield($sub); + for my $filter (@{ $conf->{excludelist}->filters($f)}) { + if ($sub_contents =~ /$filter/i) { + # filter matches. no fp. + dump_exception($marc, "exclusion $tag$sub '$filter'"); + return 1; + } + # no match, no exclude + return 0; + } + } + } +} + =head2 initialize Performs boring script initialization. Handles argument parsing, @@ -389,16 +504,19 @@ sub initialize { binmode(STDIN, ':utf8'); my $rc = GetOptions( $c, - 'incoming', - 'incumbent', 'exception|x=s', - 'marctype|m=s', 'output|o=s', - 'runtype|r=s', + 'prefix|p=s', + 'marctype|m=s', 'subfield|s=s', 'tag|t=s', 'fingerprints=s', 'scores=s', + 'arbitrarily-lose-above=i', + 'arbitrarily-lose-below=i', + 'newwins', + 'excludelist=s', + 'ignoresubtitle|i', 'quiet|q', 'help|h', ); @@ -411,6 +529,7 @@ sub initialize { my %valid_fps = ( oclc => 1, isbn => 1, issn => 1, lccn => 1, edition => 1, accomp => 1, authpub => 1, baseline => 1, crap => 1, + koha_bib_id => 1, ); for (split /,/, $c->{fingerprints}) { die "Invalid fingerprint '$_'\n" unless $valid_fps{$_}; @@ -421,6 +540,7 @@ sub initialize { $c->{fingerprints} = {oclc => 1, isbn => 1, edition => 1, issn => 1, lccn => 1, accomp => 1, authpub => 1}; } + # check scores list for validity if ($c->{scores}) { my %scores = (); @@ -431,25 +551,27 @@ sub initialize { die "Invalid score mode '$_'\n" unless $valid_scores{$_}; $scores{$_} = 1; } + $c->{dyn_scores} = [split /,/, $c->{scores}]; $c->{scores} = \%scores; } else { $c->{scores} = {oclc => 1, dlc => 1, num_650 => 1, num_tags => 1, enc_lvl => 1}; + $c->{dyn_scores} = [ qw/oclc dlc num_650 num_tags enc_lvl/ ]; } - # set defaults if told to do so - if ($c->{incoming}) { - $c->{tag} = 903 unless defined $c->{tag}; - $c->{subfield} = 'a' unless defined $c->{subfield}; - $c->{marctype} = 'XML' unless defined $c->{marctype}; - $c->{output} = 'incoming.fp' unless defined $c->{output}; - $c->{exception} = 'incoming.ex' unless defined $c->{exception}; - } elsif ($c->{incumbent}) { - $c->{tag} = 901 unless defined $c->{tag}; - $c->{subfield} = 'c' unless defined $c->{subfield}; - $c->{marctype} = 'XML' unless defined $c->{marctype}; - $c->{output} = 'incumbent.fp' unless defined $c->{output}; - $c->{exception} = 'incumbent.ex' unless defined $c->{exception}; + # set defaults + $c->{tag} = 903 unless defined $c->{tag}; + $c->{subfield} = 'a' unless defined $c->{subfield}; + $c->{marctype} = 'XML' unless defined $c->{marctype}; + if ($c->{prefix}) { + $c->{output} = join('.',$c->{prefix},'fp'); + $c->{exception} = join('.',$c->{prefix},'fp','ex'); + } + + # get SFM object if excludelist was specified + if ($c->{excludelist}) { + $c->{excludelist} = + Equinox::Migration::SubfieldMapper->new( file => $c->{excludelist} ); } my @keys = keys %{$c}; @@ -484,20 +606,32 @@ sub show_help { print < Req'd Arguments - --tag=N -t Which tag to use - --subfield=X -s Which subfield to use - --output= -o Output filename - --exceptions= -x Exception report filename + --output= -o Output filename + --exceptions= -x Exception report filename + or + --prefix=> -p Shared prefix for output/exception files. Will + produce PREFIX.fp and PREFIX.fp.ex Options - --incoming '-t 903 -s a -o incoming.fp -x incoming.ex' - --incumbent '-t 901 -s c -o incumbent.fp -x incumbent.ex' + --tag=N -t Which tag to use (default 903) + --subfield=X -s Which subfield to use (default 'a') + --quiet -q Don't write status messages to STDOUT + --ignoresubtitle -i Ignore 245\$b and construct the title from 245\$a alone. --fingerprints=LIST Fingerprints to generate, comma separated Default: oclc,isbn,edition,issn,lccn,accomp,authpub - Others: baseline + Others: baseline,koha_bib_id + --excludelist=FILE Name of fingerprints exclusions file + --scores=LIST Scores to calculate, comma separated Default: oclc,dlc,num_650,num_tags,enc_level - --quiet -q Don't write status messages to STDOUT + --newwins New record IDs score higher (default is old wins) + --arbitrarily-lose-above + --arbitrarily-lose-below + --arbitrarily-decrease-score-by + Modify fingerprint scoring of records whose EG id is above or below a + given value, inclusive (so 5 is <= 5 or >= 5) such that they lose. + + --marctype=TYPE Defaults to 'XML' HELP exit 1; }