#!/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 Getopt::Long;
use Equinox::Migration::MapDrivenMARCXMLProc 1.005;
use Equinox::Migration::MARCXMLSampler;

my $VERSION = '1.001';

=pod

TODO

  * Have detail mode report on number of subfields per datafield

=cut

my $c = initialize();
$| = 1;

# run samples if we've been asked for them
run_samples($c) if ($c->{sample} or $c->{samplemap} or $c->{samplestr});
extract_holdings($c) if ($c->{map});
print "\n";

#--------------------------

sub extract_holdings {
    my ($c) = @_;
    print "Parsing records for extraction:\n";
    my $m = Equinox::Migration::MapDrivenMARCXMLProc->new( marcfile => $c->{marcfile},
                                                           mapfile  => $c->{map},
                                                           verbose  => 1,
                                                         );

    print "Writing holdings to output file(s)...\n";
    # open main holdings file
    open HOLDINGS, '>', ($c->{prefix} . "-HOLDINGS.pg");
    # create multi files
    my $multis = $m->get_multis;
    my %MULTIFILE = ();
    for my $t ( keys %{$multis} ) {
        for my $s ( keys %{$multis->{$t}}) 
          { open my $fh, ">", ($c->{prefix} . "-HOLDINGS-MULT-$t$s.pg"); $MULTIFILE{"$t$s"} = $fh }
    }

    my $parallel_fields = $m->get_parallel_fields;

    my $i = 0; # record counter
    my $j = 0; # holdings counter

    while (  $m->{data}{recs}[$i] ) {
        print HOLDINGS "BEGIN;\n\negid, hseq, " unless $j;
        my $rec = $m->{data}{recs}[$i];
        my $k = 0; # holding-within-record pointer

        for my $holdidx ( @{ $rec->{tmap}{ $c->{holdings} } } ) {
            # for each holdings tag in the record...
            my $tagid = $rec->{tags}[$holdidx]{tag};
            $k++;

            my @out = ();            # clear the output buffer
            push @out, $rec->{egid}; # slug in the egid first thing
            push @out, $j;           # holding seq goes next

            # grab the unary mappings and slug 'em in
            for my $sub ( sort keys %{$rec->{tags}[$holdidx]{uni}} ) {
                push @out, $rec->{tags}[$holdidx]{uni}{$sub};
                print HOLDINGS "l_", $m->name($tagid, $sub),", " unless $j;
            }

            # handle holdings multis
            for my $sub ( sort keys %{$multis->{$tagid}} ) {
                for my $value ( @{$rec->{tags}[$holdidx]{multi}{$sub}} ) {
                  my $fh = $MULTIFILE{"$tagid$sub"};
                  print $fh join("\t", $rec->{egid}, $j, $value), "\n";
              }
            }


            # now get everything else in the mapping
            for my $othertag ( sort keys %{$rec->{tmap}} ) {
                next if $othertag eq $c->{holdings};  # ignoring the holdings, o'course
                my $test_idx = $rec->{tmap}{$othertag}[0]; # get index into tags struct
                unless (defined $test_idx) {
                    push @out, '';
                    next;
                }

                # handle parallel fields
                if (exists($parallel_fields->{$othertag})) {
                    my $num_fields = $#{ $rec->{tmap}{$othertag} };
                    my $tag_idx;
                    if ($holdidx > $num_fields) {
                        $tag_idx = -1;
                    } else {
                        $tag_idx = $rec->{tmap}{$othertag}[$holdidx];
                    }
                    for my $sub ( sort keys %{ $parallel_fields->{$othertag } } ) {
                        push @out, $tag_idx > -1 ? $rec->{tags}[$tag_idx]{parallel}{$sub}->[0] : '';
                        print HOLDINGS "l_", $m->name($rec->{tags}[$tag_idx]{tag}, $sub), ", " unless $j;
                    }
                }

                # handle only first other tag unless it is known to be multi
                my $limit = 0;
                if (exists($multis->{$othertag})) {
                    $limit = $#{ $rec->{tmap}{$othertag} };
                }
                foreach my $idx (0..$limit) {
                    my $tag_idx = $rec->{tmap}{$othertag}[$idx];
                    for my $sub ( sort keys %{$rec->{tags}[$tag_idx]{uni}} ) {
                        if ($m->first_only($rec->{tags}[$tag_idx]{tag}, $sub)) {
                            push @out, ($k == 1) ? $rec->{tags}[$tag_idx]{uni}{$sub} : '';
                        } else {
                            push @out, $rec->{tags}[$tag_idx]{uni}{$sub};
                        }
                        print HOLDINGS "l_", $m->name($rec->{tags}[$tag_idx]{tag}, $sub), ", " unless $j;
                    }
                    next unless exists($multis->{$othertag});
                    for my $sub ( sort keys %{$multis->{$othertag}} ) {
                        next if $m->first_only($rec->{tags}[$tag_idx]{tag}, $sub) and ($k > 1);
                        for my $value ( @{$rec->{tags}[$tag_idx]{multi}{$sub}} ) {
                            my $fh = $MULTIFILE{"$othertag$sub"};
                            print $fh normalize_output(join("\t", $rec->{egid}, $j, $value)), "\n";
                        }
                    }
                }
            }

            # and dump it
            print HOLDINGS "\n" unless $j;
            print HOLDINGS normalize_output(join("\t", @out));
            print HOLDINGS "\n";
            $j++;
        }
        $i++;
        print "\r$i $j";
    }
    print "\n";
}

#--------------------------

sub run_samples {
    my ($c) = @_;
    my $s;
    print "Parsing records for sampling... ";
    if ($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} );
    }

    dump_sample_overview($c, $s) if $c->{sample};
    dump_sample_detail($c, $s) if ($c->{samplemap} or $c->{samplestr});
}

sub dump_sample_detail {
    my ($c, $s) = @_;
    my $tags = $s->{data}{samp};
    my $count = $s->{data}{rcnt};
    my $scnt  = $s->{data}{scnt};

    open DETAIL, '>', ($c->{prefix} . "-HOLDINGS-DETAIL.txt");
    select DETAIL;
    for my $tag (sort keys %{ $tags }) {
        print ">>>>> TAG $tag\n\n";
        for my $subkey (sort keys %{ $tags->{$tag} }) {
            my $sub = $tags->{$tag}{$subkey};
            print "|| $subkey | ", $sub->{value}, " | ", 
              $sub->{count}, "/", $sub->{tcnt}, " | ", ($sub->{count} > $sub->{tcnt}) ? "MULTI" : "", " ||\n";
        }
        print "\n";
    }
    close DETAIL;
    open SCOUNT, '>', ($c->{prefix} . "-HOLDINGS-SUBCOUNTS.txt");
    select SCOUNT;
    for my $tag (sort keys %{ $scnt }) {
        print ">>>>> TAG $tag\n\n";
        for my $len (sort keys %{ $scnt->{$tag} }) 
          { print "|| $len | ", $scnt->{$tag}{$len}, " ||\n" }
        print "\n";
    }
    select STDOUT;
    print "Saved results as ", ($c->{prefix} . "-HOLDINGS-DETAIL.txt"), " and ",
        ($c->{prefix} . "-HOLDINGS-SUBCOUNTS.txt"), "\n";
}

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 "SAMPLE REPORT FOR ", $c->{prefix},": $count records\n\n";
    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 " (", sprintf("%03d", int($tags->{ $tagsbyname[$i] } / $count * 100)), "%)";
        print "      ";
        print $tagsbycount[$i], (" " x (16 - length $tags->{ $tagsbycount[$i] })),
          $tags->{ $tagsbycount[$i] };
        print " (", sprintf("%03d", int($tags->{ $tagsbycount[$i] } / $count * 100)), "%)\n";
    }
    select STDOUT;
    print "Saved results as ", ($c->{prefix} . "-HOLDINGS-OVERVIEW.txt"), "\n";
    close SAMPLE;
}

#--------------------------

sub initialize {
    my $c = {};
    my @missing = ();

    # set mode on existing filehandles
    binmode(STDIN, ':utf8');

    my $rc = GetOptions( $c,
                         'sample|s',
                         'samplemap|sm=s',
                         'samplestr|ss=s',
                         'marcfile=s',
                         'map|m=s',
                         'holdings|h=i',
                         'copyid|c=s',
                         'prefix|p=s',
                         'disable-pg-normalization',
                         'version|v',
                         'help',
                       );
    show_help() unless $rc;
    show_help() if ($c->{help});
    show_help("Nothing to do!")
      unless ($c->{map} or $c->{sample} or $c->{samplemap} or $c->{samplestr});
    show_help("map, holdings, and copyid must be specified together!")
      if ($c->{map} and !($c->{holdings} and $c->{copyid}));
    show_version() if $c->{version};

    if ($c->{prefix} and !$c->{marcfile}) {
        $c->{marcfile} = $c->{prefix} . ".clean.marc.xml";
    }

    my @keys = keys %{$c};
    for my $key ('prefix', 'marcfile')
      { push @missing, $key unless $c->{$key} }
    if (@missing) {
        print "Required option: ", join(', ', @missing), " missing!\n";
        show_help();
    }

    return $c;
}

sub normalize_output {
    my $str = shift;
    $str =~ s!\\!\\\\!g unless $c->{'disable-pg-normalization'};
    return $str;
}

sub show_help {
    my ($msg) = @_;
    print "\nERROR - $msg\n" if $msg;
    print <<HELP;

Usage is: extract_holdings --prefix PREFIX --marcfile MARCFILE [ARGUMENTS]

REQUIRED ARGUMENTS
  --prefix   -p  Prefix string for output filenames
  --marcfile     MARCXML to use as source data
                 Defaults to 'PREFIX.clean.marc.xml'

SAMPLING ARGUMENTS
  --sample    -s   Generate a report of all tags in the MARC data
  --samplemap -sm  Specify a Equinox::Migration::SimpleTagList map file which
                   will be used to generate subfield breakdown reports about
                   specific tags in the MARC data
  --samplestr -ss  As above, but with a one-liner map specified on the command
                   line as a string (e.g. '-ss "852 999"')

  If --samplemap and --samplestr are both specified, --samplemap wins.

HOLDINGS EXTRACTION ARGUMENTS
  --map      -m  Equinox::Migration::SubfieldMapper map file which will be
                 used to extract holdings data from the input MARC file
  --holdings -h  Specifies actual holdings tag
  --copyid   -c  Specifies subfield of holdings with unique copy identifier
  --disable-pg-normalization  By default, output is normalized so that a Postgres
                              copy or \\copy can import the data without choking on
                              backslashes; use this command-line option if
                              output is not meant to be consumed by psql.

  All three of these must be given together.
HELP
    exit;
}

sub show_version { print "extract_holdings v$VERSION\n"; exit }