[537] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
| 3 | # ArcPlug.pm --
|
---|
| 4 | # A component of the Greenstone digital library software
|
---|
| 5 | # from the New Zealand Digital Library Project at the
|
---|
| 6 | # University of Waikato, New Zealand.
|
---|
| 7 | #
|
---|
| 8 | # Copyright (C) 1999 New Zealand Digital Library Project
|
---|
| 9 | #
|
---|
| 10 | # This program is free software; you can redistribute it and/or modify
|
---|
| 11 | # it under the terms of the GNU General Public License as published by
|
---|
| 12 | # the Free Software Foundation; either version 2 of the License, or
|
---|
| 13 | # (at your option) any later version.
|
---|
| 14 | #
|
---|
| 15 | # This program is distributed in the hope that it will be useful,
|
---|
| 16 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 17 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 18 | # GNU General Public License for more details.
|
---|
| 19 | #
|
---|
| 20 | # You should have received a copy of the GNU General Public License
|
---|
| 21 | # along with this program; if not, write to the Free Software
|
---|
| 22 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 23 | #
|
---|
| 24 | ###########################################################################
|
---|
| 25 |
|
---|
[317] | 26 | # plugin which recurses through an archives.inf file
|
---|
| 27 | # (i.e. the file generated in the archives directory
|
---|
| 28 | # when an import is done), processing each file it finds
|
---|
[4] | 29 |
|
---|
[3540] | 30 | # 12-05-02 Added usage datastructure - John Thompson
|
---|
| 31 |
|
---|
[4] | 32 | package ArcPlug;
|
---|
| 33 |
|
---|
| 34 | use util;
|
---|
| 35 | use BasPlug;
|
---|
| 36 | use plugin;
|
---|
| 37 | use arcinfo;
|
---|
[5680] | 38 | use gsprintf;
|
---|
[4] | 39 |
|
---|
| 40 | BEGIN {
|
---|
[8716] | 41 | @ArcPlug::ISA = ('BasPlug');
|
---|
[4] | 42 | }
|
---|
| 43 |
|
---|
[4744] | 44 | my $options = { 'name' => "ArcPlug",
|
---|
[5680] | 45 | 'desc' => "{ArcPlug.desc}",
|
---|
[6408] | 46 | 'abstract' => "no",
|
---|
| 47 | 'inherits' => "yes" };
|
---|
[3540] | 48 |
|
---|
[5680] | 49 | sub gsprintf
|
---|
| 50 | {
|
---|
| 51 | return &gsprintf::gsprintf(@_);
|
---|
| 52 | }
|
---|
| 53 |
|
---|
[4] | 54 | sub new {
|
---|
[10218] | 55 | my ($class) = shift (@_);
|
---|
| 56 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
---|
| 57 | push(@$pluginlist, $class);
|
---|
[4] | 58 |
|
---|
[10218] | 59 | if(defined $arguments){print "SETSEGE\n"; push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
|
---|
| 60 | if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
|
---|
| 61 |
|
---|
| 62 | my $self = (defined $hashArgOptLists)? new BasPlug($pluginlist,$inputargs,$hashArgOptLists): new BasPlug($pluginlist,$inputargs);
|
---|
| 63 |
|
---|
[4] | 64 | return bless $self, $class;
|
---|
| 65 | }
|
---|
| 66 |
|
---|
[10156] | 67 | sub deinit {
|
---|
| 68 | my ($self) = @_;
|
---|
| 69 |
|
---|
| 70 | my $archive_info = $self->{'archive_info'};
|
---|
| 71 |
|
---|
| 72 | if (defined $archive_info) {
|
---|
| 73 | my $archive_info_filename = $self->{'archive_info_filename'};
|
---|
| 74 |
|
---|
| 75 | my $file_list = $archive_info->get_file_list();
|
---|
| 76 |
|
---|
| 77 | # change each file to "Been Indexed"
|
---|
| 78 |
|
---|
| 79 | foreach my $subfile (@$file_list) {
|
---|
| 80 | my $doc_oid = $subfile->[1];
|
---|
| 81 | $index_status = $archive_info->get_status_info($doc_oid);
|
---|
| 82 | $archive_info->set_status_info($doc_oid,"B");
|
---|
| 83 | }
|
---|
| 84 |
|
---|
| 85 | $archive_info->save_info($archive_info_filename);
|
---|
| 86 | }
|
---|
| 87 | }
|
---|
| 88 |
|
---|
[4] | 89 | # return 1 if this class might recurse using $pluginfo
|
---|
| 90 | sub is_recursive {
|
---|
| 91 | my $self = shift (@_);
|
---|
| 92 |
|
---|
| 93 | return 1;
|
---|
| 94 | }
|
---|
| 95 |
|
---|
[10156] | 96 |
|
---|
| 97 |
|
---|
| 98 |
|
---|
[317] | 99 | # return number of files processed, undef if can't process
|
---|
[4] | 100 | # Note that $base_dir might be "" and that $file might
|
---|
| 101 | # include directories
|
---|
| 102 | sub read {
|
---|
| 103 | my $self = shift (@_);
|
---|
[9853] | 104 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs,$total_count, $gli) = @_;
|
---|
[1424] | 105 | my $outhandle = $self->{'outhandle'};
|
---|
[4] | 106 |
|
---|
[317] | 107 | my $count = 0;
|
---|
| 108 |
|
---|
[4] | 109 | # see if this has a archives information file within it
|
---|
[1244] | 110 | my $archive_info_filename = &util::filename_cat($base_dir,$file,"archives.inf");
|
---|
[10156] | 111 | $self->{'archive_info_filename'} = $archive_info_filename;
|
---|
[4] | 112 |
|
---|
| 113 | if (-e $archive_info_filename) {
|
---|
| 114 |
|
---|
[317] | 115 | # found an archives.inf file
|
---|
[6332] | 116 | &gsprintf($outhandle, "ArcPlug: {common.processing} $archive_info_filename\n") if $self->{'verbosity'} > 1;
|
---|
[317] | 117 |
|
---|
[4] | 118 | # read in the archives information file
|
---|
| 119 | my $archive_info = new arcinfo ();
|
---|
[10156] | 120 | $self->{'archive_info'} = $archive_info;
|
---|
| 121 |
|
---|
[4] | 122 | $archive_info->load_info ($archive_info_filename);
|
---|
| 123 |
|
---|
[230] | 124 | my $file_list = $archive_info->get_file_list();
|
---|
[4] | 125 |
|
---|
| 126 | # process each file
|
---|
[1244] | 127 | foreach my $subfile (@$file_list) {
|
---|
[9853] | 128 | last if ($maxdocs != -1 && ($total_count + $count) >= $maxdocs);
|
---|
[317] | 129 |
|
---|
[4] | 130 | my $tmp = &util::filename_cat ($file, $subfile->[0]);
|
---|
| 131 | next if $tmp eq $file;
|
---|
[10156] | 132 |
|
---|
| 133 | my $doc_oid = $subfile->[1];
|
---|
| 134 |
|
---|
| 135 | $index_status = $archive_info->get_status_info($doc_oid);
|
---|
| 136 |
|
---|
| 137 | my $processor_mode = $processor->get_mode();
|
---|
| 138 |
|
---|
| 139 | if (($processor_mode eq "infodb") || ($index_status eq "I")) {
|
---|
| 140 |
|
---|
| 141 | # note: metadata is not carried on to the next level
|
---|
| 142 | $count += &plugin::read ($pluginfo, $base_dir, $tmp, {}, $processor, $maxdocs, ($total_count+$count), $gli);
|
---|
| 143 | }
|
---|
| 144 |
|
---|
[4] | 145 | }
|
---|
| 146 |
|
---|
[317] | 147 | return $count;
|
---|
[4] | 148 | }
|
---|
| 149 |
|
---|
| 150 | # wasn't an archives directory, someone else will have to process it
|
---|
[317] | 151 | return undef;
|
---|
[4] | 152 | }
|
---|
| 153 |
|
---|
| 154 | 1;
|
---|