X-Git-Url: http://git.equinoxoli.org/?p=migration-tools.git;a=blobdiff_plain;f=fingerprinter;h=c0a87413dce1236b4ab0791bad3c3fe838f6f574;hp=7b20e3e43297d93d06594acdf3b4dbffa0b8cc2a;hb=81f067cea40ef8e347202188fbf4f68c69220698;hpb=d478b47d3998baa905f245e8e05faf83bc314069 diff --git a/fingerprinter b/fingerprinter index 7b20e3e..c0a8741 100755 --- a/fingerprinter +++ b/fingerprinter @@ -74,29 +74,30 @@ 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; + $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 @@ -126,9 +127,11 @@ sub populate_marc { if (defined $oclc and $oclc =~ /\(OCoLC\)/ and $oclc =~/([0-9]+)/); } - # "Accompanying material" (300e) - $marc{accomp} = $record->field('300')->subfield('e') - if $record->field('300'); + # "Accompanying material" and check for "copy" (300) + if ($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 $record->field('010'); @@ -208,69 +211,70 @@ sub marc_isvalid { Assign a score to the record based on various criteria. -Score is constructed by pushing elements onto a list. At the end of -the routine, the list is flattened into a string via join(); +Score is constructed by pushing elements onto a list, via a dispatch +table. This allows order of fingerprints in the output file to be +varied. =cut sub score_marc { my ($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, ( $chunk =~ /^o/i ? 1 : 0 ); - } else { - push @score, 0; - } - } - - # does 040a contain "dlc"? - if ($conf->{scores}{dlc}) { - if ($record->field('040') and $record->field('040')->subfield('a')) { - $chunk = $record->field('040')->subfield('a'); - push @score, ( $chunk =~ /dlc/i ? 1 : 0 ); - } else { - push @score, 0; - } - } - - # number of 650 datafields - # zero-padded to 4 digits with printf - if ($conf->{scores}{num_650}) { - if ($record->field('650')) { - my @tags = $record->field('650'); - push @score, scalar @tags; - } else { - push @score, 0; - } - } - - # number of tags in total - # zero-padded to 4 digits with printf - if ($conf->{scores}{num_tags}) { - my @tags = $record->fields; - push @score, scalar @tags; - } - - # encoding level - if ($conf->{scores}{enc_lvl}) { - my $enc = substr($record->leader, 17, 1); + my $json = '{'; + + #---------------------------------- + # static criteria scoring + #---------------------------------- + $marc->{misc_score} = 999; + # -1 if 008 has been padded, -2 if it doesn't exist + if ($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); + + #---------------------------------- + # 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->{misc_score}, @score); + $marc->{score} = "$compact\t$json"; } =head2 dump_fingerprints @@ -389,12 +393,10 @@ 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', @@ -421,6 +423,7 @@ 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 +434,21 @@ 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'); } my @keys = keys %{$c}; @@ -484,20 +483,24 @@ 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 --fingerprints=LIST Fingerprints to generate, comma separated Default: oclc,isbn,edition,issn,lccn,accomp,authpub Others: baseline + --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 + + --marctype=TYPE Defaults to 'XML' HELP exit 1; }