[18457] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
[22413] | 3 | # inexport.pm -- useful class to support import.pl and export.pl
|
---|
[18457] | 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 |
|
---|
| 26 | package inexport;
|
---|
| 27 |
|
---|
| 28 | use strict;
|
---|
| 29 |
|
---|
[22413] | 30 | no strict 'refs'; # allow filehandles to be variables and vice versa
|
---|
| 31 | no strict 'subs'; # allow barewords (eg STDERR) as function arguments
|
---|
[19789] | 32 |
|
---|
[22413] | 33 | use arcinfo;
|
---|
| 34 | use colcfg;
|
---|
[21553] | 35 | use dbutil;
|
---|
[22413] | 36 | use plugin;
|
---|
| 37 | use plugout;
|
---|
| 38 | use manifest;
|
---|
| 39 | use inexport;
|
---|
| 40 | use dbutil;
|
---|
[18457] | 41 | use util;
|
---|
[22413] | 42 | use scriptutil;
|
---|
| 43 | use FileHandle;
|
---|
| 44 | use gsprintf 'gsprintf';
|
---|
| 45 | use printusage;
|
---|
| 46 | use parse2;
|
---|
[18457] | 47 |
|
---|
[22413] | 48 | use File::Basename;
|
---|
[21563] | 49 |
|
---|
[22413] | 50 | sub new
|
---|
| 51 | {
|
---|
| 52 | my $class = shift (@_);
|
---|
[22421] | 53 | my ($mode,$argv,$options,$opt_listall_options) = @_;
|
---|
[22413] | 54 |
|
---|
[22421] | 55 | my $self = { 'xml' => 0, 'mode' => $mode };
|
---|
[22413] | 56 |
|
---|
| 57 | # general options available to all plugins
|
---|
| 58 | my $arguments = $options->{'args'};
|
---|
| 59 | my $intArgLeftinAfterParsing = parse2::parse($argv,$arguments,$self,"allow_extra_options");
|
---|
| 60 | # Parse returns -1 if something has gone wrong
|
---|
| 61 | if ($intArgLeftinAfterParsing == -1)
|
---|
| 62 | {
|
---|
| 63 | &PrintUsage::print_txt_usage($options, "{import.params}");
|
---|
| 64 | die "\n";
|
---|
| 65 | }
|
---|
| 66 |
|
---|
| 67 | my $language = $self->{'language'};
|
---|
| 68 | # If $language has been specified, load the appropriate resource bundle
|
---|
| 69 | # (Otherwise, the default resource bundle will be loaded automatically)
|
---|
| 70 | if ($language && $language =~ /\S/) {
|
---|
| 71 | &gsprintf::load_language_specific_resource_bundle($language);
|
---|
| 72 | }
|
---|
| 73 |
|
---|
[22421] | 74 | if ($self->{'listall'}) {
|
---|
| 75 | if ($self->{'xml'}) {
|
---|
| 76 | &PrintUsage::print_xml_usage($opt_listall_options);
|
---|
| 77 | }
|
---|
| 78 | else
|
---|
| 79 | {
|
---|
| 80 | &PrintUsage::print_txt_usage($opt_listall_options,"{export.params}");
|
---|
| 81 | }
|
---|
| 82 | die "\n";
|
---|
| 83 | }
|
---|
| 84 |
|
---|
| 85 |
|
---|
[22413] | 86 | if ($self->{'xml'}) {
|
---|
| 87 | &PrintUsage::print_xml_usage($options);
|
---|
| 88 | print "\n";
|
---|
| 89 | return;
|
---|
| 90 | }
|
---|
| 91 |
|
---|
| 92 | if ($self->{'gli'}) { # the gli wants strings to be in UTF-8
|
---|
| 93 | &gsprintf::output_strings_in_UTF8;
|
---|
| 94 | }
|
---|
| 95 |
|
---|
| 96 | # now check that we had exactly one leftover arg, which should be
|
---|
| 97 | # the collection name. We don't want to do this earlier, cos
|
---|
| 98 | # -xml arg doesn't need a collection name
|
---|
| 99 | # Or if the user specified -h, then we output the usage also
|
---|
| 100 | if ($intArgLeftinAfterParsing != 1 || (@$argv && $argv->[0] =~ /^\-+h/))
|
---|
| 101 | {
|
---|
| 102 | &PrintUsage::print_txt_usage($options, "{import.params}");
|
---|
| 103 | die "\n";
|
---|
| 104 | }
|
---|
| 105 |
|
---|
| 106 | $self->{'close_out'} = 0;
|
---|
| 107 | my $out = $self->{'out'};
|
---|
| 108 | if ($out !~ /^(STDERR|STDOUT)$/i) {
|
---|
| 109 | open (OUT, ">$out") ||
|
---|
| 110 | (&gsprintf(STDERR, "{common.cannot_open_output_file}: $!\n", $out) && die);
|
---|
| 111 | $out = 'import::OUT';
|
---|
| 112 | $self->{'close_out'} = 1;
|
---|
| 113 | }
|
---|
| 114 | $out->autoflush(1);
|
---|
| 115 | $self->{'out'} = $out;
|
---|
| 116 |
|
---|
| 117 | # @ARGV should be only one item, the name of the collection
|
---|
| 118 | $self->{'collection'} = shift @$argv;
|
---|
| 119 |
|
---|
| 120 | return bless $self, $class;
|
---|
| 121 | }
|
---|
| 122 |
|
---|
| 123 | sub get_collection
|
---|
| 124 | {
|
---|
| 125 | my $self = shift @_;
|
---|
| 126 |
|
---|
| 127 | return $self->{'collection'};
|
---|
| 128 | }
|
---|
| 129 |
|
---|
| 130 |
|
---|
| 131 | sub read_collection_cfg
|
---|
| 132 | {
|
---|
| 133 | my $self = shift @_;
|
---|
| 134 | my ($collection,$options) = @_;
|
---|
| 135 |
|
---|
| 136 | my $collectdir = $self->{'collectdir'};
|
---|
| 137 | my $site = $self->{'site'};
|
---|
| 138 | my $out = $self->{'out'};
|
---|
| 139 |
|
---|
| 140 | if (($collection = &colcfg::use_collection($site, $collection, $collectdir)) eq "") {
|
---|
| 141 | &PrintUsage::print_txt_usage($options, "{import.params}");
|
---|
| 142 | die "\n";
|
---|
| 143 | }
|
---|
| 144 |
|
---|
| 145 | # add collection's perllib dir into include path in
|
---|
| 146 | # case we have collection specific modules
|
---|
| 147 | unshift (@INC, "$ENV{'GSDLCOLLECTDIR'}/perllib");
|
---|
| 148 |
|
---|
| 149 | # check that we can open the faillog
|
---|
| 150 | my $faillog = $self->{'faillog'};
|
---|
| 151 | if ($faillog eq "") {
|
---|
| 152 | $faillog = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "etc", "fail.log");
|
---|
| 153 | }
|
---|
| 154 | open (FAILLOG, ">$faillog") ||
|
---|
| 155 | (&gsprintf(STDERR, "{import.cannot_open_fail_log}\n", $faillog) && die);
|
---|
| 156 |
|
---|
| 157 |
|
---|
| 158 | my $faillogname = $faillog;
|
---|
| 159 | $faillog = 'inexport::FAILLOG';
|
---|
| 160 | $faillog->autoflush(1);
|
---|
| 161 | $self->{'faillog'} = $faillog;
|
---|
| 162 | $self->{'faillogname'} = $faillogname;
|
---|
| 163 |
|
---|
| 164 | # Read in the collection configuration file.
|
---|
[22421] | 165 | my ($config_filename, $gs_mode) = &colcfg::get_collect_cfg_name($out);
|
---|
| 166 | my $collectcfg = &colcfg::read_collection_cfg ($config_filename, $gs_mode);
|
---|
[22413] | 167 |
|
---|
[22421] | 168 | return ($config_filename,$collectcfg);
|
---|
[22413] | 169 | }
|
---|
| 170 |
|
---|
| 171 | sub set_collection_options
|
---|
| 172 | {
|
---|
| 173 | my $self = shift @_;
|
---|
[22421] | 174 | my ($collectcfg) = @_;
|
---|
[22413] | 175 |
|
---|
[22421] | 176 | my $inexport_mode = $self->{'mode'};
|
---|
| 177 |
|
---|
[22413] | 178 | my $verbosity = $self->{'verbosity'};
|
---|
| 179 | my $debug = $self->{'debug'};
|
---|
| 180 | my $importdir = $self->{'importdir'};
|
---|
| 181 | my $archivedir = $self->{'archivedir'};
|
---|
| 182 | my $out = $self->{'out'};
|
---|
| 183 |
|
---|
| 184 | # If the infodbtype value wasn't defined in the collect.cfg file, use the default
|
---|
| 185 | if (!defined($collectcfg->{'infodbtype'}))
|
---|
| 186 | {
|
---|
| 187 | $collectcfg->{'infodbtype'} = &dbutil::get_default_infodb_type();
|
---|
| 188 | }
|
---|
| 189 |
|
---|
| 190 | if (defined $collectcfg->{'importdir'} && $importdir eq "") {
|
---|
| 191 | $importdir = $collectcfg->{'importdir'};
|
---|
| 192 | }
|
---|
| 193 | if (defined $collectcfg->{'archivedir'} && $archivedir eq "") {
|
---|
| 194 | $archivedir = $collectcfg->{'archivedir'};
|
---|
| 195 | }
|
---|
| 196 | # fill in the default import and archives directories if none
|
---|
| 197 | # were supplied, turn all \ into / and remove trailing /
|
---|
| 198 | $importdir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "import") if $importdir eq "";
|
---|
| 199 | $importdir =~ s/[\\\/]+/\//g;
|
---|
| 200 | $importdir =~ s/\/$//;
|
---|
| 201 | if (!-e $importdir) {
|
---|
| 202 | &gsprintf($out, "{import.no_import_dir}\n\n", $importdir);
|
---|
| 203 | die "\n";
|
---|
| 204 | }
|
---|
| 205 | $self->{'importdir'} = $importdir;
|
---|
| 206 |
|
---|
| 207 | if ($archivedir eq "") {
|
---|
| 208 | if ($inexport_mode eq "import") {
|
---|
| 209 | $archivedir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "archives");
|
---|
| 210 | }
|
---|
| 211 | elsif ($inexport_mode eq "export") {
|
---|
| 212 | $archivedir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "export");
|
---|
| 213 | }
|
---|
| 214 | else {
|
---|
| 215 | print STDERR "Warning: Unrecognized import/export mode '$inexport_mode'\n";
|
---|
| 216 | print STDERR " Defaulting to 'archives' for file output\n";
|
---|
| 217 | $archivedir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "archives");
|
---|
| 218 | }
|
---|
| 219 | }
|
---|
| 220 |
|
---|
| 221 | $archivedir =~ s/[\\\/]+/\//g;
|
---|
| 222 | $archivedir =~ s/\/$//;
|
---|
| 223 | $self->{'archivedir'} = $archivedir;
|
---|
| 224 |
|
---|
| 225 | if ($verbosity !~ /\d+/) {
|
---|
| 226 | if (defined $collectcfg->{'verbosity'} && $collectcfg->{'verbosity'} =~ /\d+/) {
|
---|
| 227 | $verbosity = $collectcfg->{'verbosity'};
|
---|
| 228 | } else {
|
---|
| 229 | $verbosity = 2; # the default
|
---|
| 230 | }
|
---|
| 231 | }
|
---|
[22421] | 232 | $self->{'verbosity'} = $verbosity;
|
---|
| 233 |
|
---|
[22413] | 234 | if (defined $collectcfg->{'manifest'} && $self->{'manifest'} eq "") {
|
---|
| 235 | $self->{'manifest'} = $collectcfg->{'manifest'};
|
---|
| 236 | }
|
---|
| 237 |
|
---|
| 238 | if (defined $collectcfg->{'gzip'} && !$self->{'gzip'}) {
|
---|
| 239 | if ($collectcfg->{'gzip'} =~ /^true$/i) {
|
---|
| 240 | $self->{'gzip'} = 1;
|
---|
| 241 | }
|
---|
| 242 | }
|
---|
| 243 |
|
---|
| 244 | if ($self->{'maxdocs'} !~ /\-?\d+/) {
|
---|
| 245 | if (defined $collectcfg->{'maxdocs'} && $collectcfg->{'maxdocs'} =~ /\-?\d+/) {
|
---|
| 246 | $self->{'maxdocs'} = $collectcfg->{'maxdocs'};
|
---|
| 247 | } else {
|
---|
| 248 | $self->{'maxdocs'} = -1; # the default
|
---|
| 249 | }
|
---|
| 250 | }
|
---|
[22421] | 251 |
|
---|
| 252 | if ((defined $self->{'groupsize'}) && ($self->{'groupsize'} == 1)) {
|
---|
[22413] | 253 | if (defined $collectcfg->{'groupsize'} && $collectcfg->{'groupsize'} =~ /\d+/) {
|
---|
| 254 | $self->{'groupsize'} = $collectcfg->{'groupsize'};
|
---|
| 255 | }
|
---|
| 256 | }
|
---|
| 257 |
|
---|
| 258 | if (!defined $self->{'OIDtype'}
|
---|
| 259 | || ($self->{'OIDtype'} !~ /^(hash|incremental|assigned|dirname)$/ )) {
|
---|
| 260 | if (defined $collectcfg->{'OIDtype'}
|
---|
| 261 | && $collectcfg->{'OIDtype'} =~ /^(hash|incremental|assigned|dirname)$/) {
|
---|
| 262 | $self->{'OIDtype'} = $collectcfg->{'OIDtype'};
|
---|
| 263 | } else {
|
---|
| 264 | $self->{'OIDtype'} = "hash"; # the default
|
---|
| 265 | }
|
---|
| 266 | }
|
---|
| 267 |
|
---|
| 268 | if ((!defined $self->{'OIDmetadata'}) || ($self->{'OIDmetadata'} eq "")) {
|
---|
| 269 | if (defined $collectcfg->{'OIDmetadata'}) {
|
---|
| 270 | $self->{'OIDmetadata'} = $collectcfg->{'OIDmetadata'};
|
---|
| 271 | } else {
|
---|
| 272 | $self->{'OIDmetadata'} = "dc.Identifier"; # the default
|
---|
| 273 | }
|
---|
| 274 | }
|
---|
| 275 |
|
---|
| 276 | my $sortmeta = $self->{'sortmeta'};
|
---|
| 277 | if (defined $collectcfg->{'sortmeta'} && (!defined $sortmeta || $sortmeta eq "")) {
|
---|
| 278 | $sortmeta = $collectcfg->{'sortmeta'};
|
---|
| 279 | }
|
---|
| 280 | # sortmeta cannot be used with group size
|
---|
| 281 | $sortmeta = undef unless defined $sortmeta && $sortmeta =~ /\S/;
|
---|
| 282 | if (defined $sortmeta && $self->{'groupsize'} > 1) {
|
---|
| 283 | &gsprintf($out, "{import.cannot_sort}\n\n");
|
---|
| 284 | $sortmeta = undef;
|
---|
| 285 | }
|
---|
| 286 | $self->{'sortmeta'} = $sortmeta;
|
---|
| 287 |
|
---|
| 288 | if (defined $collectcfg->{'removeprefix'} && $self->{'removeprefix'} eq "") {
|
---|
| 289 | $self->{'removeprefix'} = $collectcfg->{'removeprefix'};
|
---|
| 290 | }
|
---|
| 291 |
|
---|
| 292 | if (defined $collectcfg->{'removesuffix'} && $self->{'removesuffix'} eq "") {
|
---|
| 293 | $self->{'removesuffix'} = $collectcfg->{'removesuffix'};
|
---|
| 294 | }
|
---|
| 295 | if (defined $collectcfg->{'debug'} && $collectcfg->{'debug'} =~ /^true$/i) {
|
---|
| 296 | $self->{'debug'} = 1;
|
---|
| 297 | }
|
---|
| 298 | if (defined $collectcfg->{'gli'} && $collectcfg->{'gli'} =~ /^true$/i) {
|
---|
| 299 | $self->{'gli'} = 1;
|
---|
| 300 | }
|
---|
| 301 | $self->{'gli'} = 0 unless defined $self->{'gli'};
|
---|
| 302 |
|
---|
| 303 | # check keepold and removeold
|
---|
[22421] | 304 | my $checkdir = ($inexport_mode eq "import") ? "archives" : "export";
|
---|
| 305 |
|
---|
[22413] | 306 | my ($removeold, $keepold, $incremental, $incremental_mode)
|
---|
| 307 | = &scriptutil::check_removeold_and_keepold($self->{'removeold'}, $self->{'keepold'},
|
---|
[22421] | 308 | $self->{'incremental'}, $checkdir,
|
---|
[22413] | 309 | $collectcfg);
|
---|
| 310 |
|
---|
| 311 | $self->{'removeold'} = $removeold;
|
---|
| 312 | $self->{'keepold'} = $keepold;
|
---|
| 313 | $self->{'incremental'} = $incremental;
|
---|
| 314 | $self->{'incremental_mode'} = $incremental_mode;
|
---|
| 315 | }
|
---|
| 316 |
|
---|
| 317 | sub process_files
|
---|
| 318 | {
|
---|
| 319 | my $self = shift @_;
|
---|
[22421] | 320 | my ($config_filename,$collectcfg) = @_;
|
---|
[22413] | 321 |
|
---|
[22421] | 322 | my $inexport_mode = $self->{'mode'};
|
---|
| 323 |
|
---|
[22413] | 324 | my $verbosity = $self->{'verbosity'};
|
---|
| 325 | my $debug = $self->{'debug'};
|
---|
| 326 |
|
---|
| 327 | my $importdir = $self->{'importdir'};
|
---|
| 328 | my $archivedir = $self->{'archivedir'};
|
---|
| 329 |
|
---|
| 330 | my $incremental = $self->{'incremental'};
|
---|
| 331 | my $incremental_mode = $self->{'incremental_mode'};
|
---|
| 332 |
|
---|
| 333 | my $removeold = $self->{'removeold'};
|
---|
| 334 | my $keepold = $self->{'keepold'};
|
---|
| 335 |
|
---|
| 336 | my $saveas = $self->{'saveas'};
|
---|
| 337 | my $OIDtype = $self->{'OIDtype'};
|
---|
| 338 | my $OIDmetadata = $self->{'OIDmetadata'};
|
---|
| 339 |
|
---|
| 340 | my $out = $self->{'out'};
|
---|
| 341 | my $faillog = $self->{'faillog'};
|
---|
| 342 |
|
---|
| 343 | my $maxdocs = $self->{'maxdocs'};
|
---|
| 344 | my $gzip = $self->{'gzip'};
|
---|
| 345 | my $groupsize = $self->{'groupsize'};
|
---|
| 346 | my $sortmeta = $self->{'sortmeta'};
|
---|
| 347 |
|
---|
| 348 | my $removeprefix = $self->{'removeprefix'};
|
---|
| 349 | my $removesuffix = $self->{'removesuffix'};
|
---|
| 350 |
|
---|
[22421] | 351 | my $gli = $self->{'gli'};
|
---|
[22413] | 352 |
|
---|
[22421] | 353 | # related to export
|
---|
| 354 | my $xsltfile = $self->{'xsltfile'};
|
---|
| 355 | my $group_marc = $self->{'group_marc'};
|
---|
| 356 | my $mapping_file = $self->{'mapping_file'};
|
---|
| 357 | my $xslt_mets = $self->{'xslt_mets'};
|
---|
| 358 | my $xslt_txt = $self->{'xslt_txt'};
|
---|
| 359 | my $fedora_namespace = $self->{'fedora_namespace'};
|
---|
| 360 |
|
---|
| 361 | if ($inexport_mode eq "import") {
|
---|
| 362 | print STDERR "<Import>\n" if $gli;
|
---|
| 363 | }
|
---|
| 364 | else {
|
---|
| 365 | print STDERR "<export>\n" if $gli;
|
---|
| 366 | }
|
---|
[22413] | 367 |
|
---|
| 368 | my $manifest_lookup = new manifest($collectcfg->{'infodbtype'},$archivedir);
|
---|
| 369 | if ($self->{'manifest'} ne "") {
|
---|
| 370 | my $manifest_filename = $self->{'manifest'};
|
---|
| 371 |
|
---|
| 372 | if (!&util::filename_is_absolute($manifest_filename)) {
|
---|
| 373 | $manifest_filename = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, $manifest_filename);
|
---|
| 374 | }
|
---|
| 375 |
|
---|
| 376 | $self->{'manifest'} =~ s/[\\\/]+/\//g;
|
---|
| 377 | $self->{'manifest'} =~ s/\/$//;
|
---|
| 378 |
|
---|
| 379 | $manifest_lookup->parse($manifest_filename);
|
---|
| 380 | }
|
---|
| 381 |
|
---|
| 382 | my $manifest = $self->{'manifest'};
|
---|
| 383 |
|
---|
| 384 | # load all the plugins
|
---|
| 385 | my $plugins = [];
|
---|
| 386 | if (defined $collectcfg->{'plugin'}) {
|
---|
| 387 | $plugins = $collectcfg->{'plugin'};
|
---|
| 388 | }
|
---|
| 389 |
|
---|
| 390 | #some global options for the plugins
|
---|
| 391 | my @global_opts = ();
|
---|
| 392 |
|
---|
| 393 | my $pluginfo = &plugin::load_plugins ($plugins, $verbosity, $out, $faillog, \@global_opts, $incremental_mode);
|
---|
| 394 | if (scalar(@$pluginfo) == 0) {
|
---|
| 395 | &gsprintf($out, "{import.no_plugins_loaded}\n");
|
---|
| 396 | die "\n";
|
---|
| 397 | }
|
---|
| 398 |
|
---|
| 399 | # remove the old contents of the archives directory (and tmp directory) if needed
|
---|
| 400 | if ($removeold) {
|
---|
| 401 | if (-e $archivedir) {
|
---|
| 402 | &gsprintf($out, "{import.removing_archives}\n");
|
---|
| 403 | &util::rm_r ($archivedir);
|
---|
| 404 | }
|
---|
| 405 | my $tmpdir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "tmp");
|
---|
| 406 | $tmpdir =~ s/[\\\/]+/\//g;
|
---|
| 407 | $tmpdir =~ s/\/$//;
|
---|
| 408 | if (-e $tmpdir) {
|
---|
| 409 | &gsprintf($out, "{import.removing_tmpdir}\n");
|
---|
| 410 | &util::rm_r ($tmpdir);
|
---|
| 411 | }
|
---|
| 412 | }
|
---|
| 413 |
|
---|
| 414 | # create the archives dir if needed
|
---|
| 415 | &util::mk_all_dir($archivedir);
|
---|
| 416 |
|
---|
| 417 | # read the archive information file
|
---|
| 418 | ## my $arcinfo_doc_filename = &util::filename_cat ($archivedir, "archives.inf");
|
---|
| 419 |
|
---|
| 420 | # BACKWARDS COMPATIBILITY: Just in case there are old .ldb/.bdb files (won't do anything for other infodbtypes)
|
---|
| 421 | &util::rename_ldb_or_bdb_file(&util::filename_cat($archivedir, "archiveinf-doc"));
|
---|
| 422 | &util::rename_ldb_or_bdb_file(&util::filename_cat($archivedir, "archiveinf-src"));
|
---|
| 423 |
|
---|
| 424 | my $arcinfo_doc_filename = &dbutil::get_infodb_file_path($collectcfg->{'infodbtype'}, "archiveinf-doc", $archivedir);
|
---|
| 425 | my $arcinfo_src_filename = &dbutil::get_infodb_file_path($collectcfg->{'infodbtype'}, "archiveinf-src", $archivedir);
|
---|
| 426 |
|
---|
| 427 | my $archive_info = new arcinfo ($collectcfg->{'infodbtype'});
|
---|
| 428 | $archive_info->load_info ($arcinfo_doc_filename);
|
---|
| 429 |
|
---|
| 430 | if ($manifest eq "") {
|
---|
| 431 | # Load in list of files in import folder from last import (if present)
|
---|
| 432 | $archive_info->load_prev_import_filelist ($arcinfo_src_filename);
|
---|
| 433 | }
|
---|
| 434 |
|
---|
| 435 | ####Use Plugout####
|
---|
[22421] | 436 | my $plugout;
|
---|
| 437 |
|
---|
| 438 | if ($inexport_mode eq "import") {
|
---|
| 439 | if (defined $collectcfg->{'plugout'}) {
|
---|
| 440 | # If a plugout was specified in the collect.cfg file, assume it is sensible
|
---|
| 441 | # We can't check the name because it could be anything, if it is a custom plugout
|
---|
| 442 | $plugout = $collectcfg->{'plugout'};
|
---|
| 443 | }
|
---|
| 444 | else{
|
---|
| 445 | if ($saveas !~ /^(GreenstoneXML|GreenstoneMETS)$/) {
|
---|
| 446 | push @$plugout,"GreenstoneXMLPlugout";
|
---|
| 447 | }
|
---|
| 448 | else{
|
---|
| 449 | push @$plugout,$saveas."Plugout";
|
---|
| 450 | }
|
---|
| 451 | }
|
---|
[22413] | 452 | }
|
---|
[22421] | 453 | else {
|
---|
| 454 | if (defined $collectcfg->{'plugout'} && $collectcfg->{'plugout'} =~ /^(.*METS|DSpace|MARCXML)Plugout/) {
|
---|
| 455 | $plugout = $collectcfg->{'plugout'};
|
---|
[22413] | 456 | }
|
---|
| 457 | else{
|
---|
[22421] | 458 | if ($saveas !~ /^(GreenstoneMETS|FedoraMETS|DSpace|MARCXML)$/) {
|
---|
| 459 | push @$plugout,"GreenstoneMETSPlugout";
|
---|
| 460 | }
|
---|
| 461 | else{
|
---|
| 462 | push @$plugout,$saveas."Plugout";
|
---|
| 463 | }
|
---|
[22413] | 464 | }
|
---|
| 465 | }
|
---|
[22421] | 466 |
|
---|
| 467 | my $plugout_name = $plugout->[0];
|
---|
[22413] | 468 |
|
---|
[22421] | 469 | push @$plugout,("-output_info",$archive_info) if (defined $archive_info);
|
---|
| 470 | push @$plugout,("-verbosity",$verbosity) if (defined $verbosity);
|
---|
| 471 | push @$plugout,("-debug") if ($debug);
|
---|
| 472 | push @$plugout,("-group_size",$groupsize) if (defined $groupsize);
|
---|
| 473 | push @$plugout,("-gzip_output") if ($gzip);
|
---|
| 474 | push @$plugout,("-output_handle",$out) if (defined $out);
|
---|
| 475 |
|
---|
| 476 | push @$plugout,("-xslt_file",$xsltfile) if (defined $xsltfile && $xsltfile ne "");
|
---|
| 477 |
|
---|
| 478 | if ($plugout_name =~ m/^MARCXMLPlugout$/) {
|
---|
| 479 | push @$plugout,("-group") if ($group_marc);
|
---|
| 480 | push @$plugout,("-mapping_file",$mapping_file) if (defined $mapping_file && $mapping_file ne "");
|
---|
| 481 | }
|
---|
| 482 | if ($plugout_name =~ m/^.*METSPlugout$/) {
|
---|
| 483 | push @$plugout,("-xslt_mets",$xslt_mets) if (defined $xslt_mets && $xslt_mets ne "");
|
---|
| 484 | push @$plugout,("-xslt_txt",$xslt_txt) if (defined $xslt_txt && $xslt_txt ne "");
|
---|
| 485 | }
|
---|
| 486 |
|
---|
| 487 | if ($plugout_name eq "FedoraMETSPlugout") {
|
---|
| 488 | push @$plugout,("-fedora_namespace",$fedora_namespace) if (defined $fedora_namespace && $fedora_namespace ne "");
|
---|
| 489 | }
|
---|
| 490 |
|
---|
| 491 |
|
---|
[22413] | 492 | my $processor = &plugout::load_plugout($plugout);
|
---|
| 493 | $processor->setoutputdir ($archivedir);
|
---|
| 494 | $processor->set_sortmeta ($sortmeta, $removeprefix, $removesuffix) if defined $sortmeta;
|
---|
| 495 | $processor->set_OIDtype ($OIDtype, $OIDmetadata);
|
---|
| 496 |
|
---|
| 497 | &plugin::begin($pluginfo, $importdir, $processor, $maxdocs, $gli);
|
---|
| 498 |
|
---|
| 499 | if ($removeold) {
|
---|
| 500 | # occasionally, plugins may want to do something on remove old, eg pharos image indexing
|
---|
| 501 | &plugin::remove_all($pluginfo, $importdir, $processor, $maxdocs, $gli);
|
---|
| 502 | }
|
---|
| 503 | if ($manifest eq "") {
|
---|
| 504 | # process the import directory
|
---|
| 505 | my $block_hash = {};
|
---|
| 506 | my $metadata = {};
|
---|
| 507 | # gobal blocking pass may set up some metadata
|
---|
| 508 | &plugin::file_block_read($pluginfo, $importdir, "", $block_hash, $metadata, $gli);
|
---|
| 509 |
|
---|
| 510 | if ($incremental || $incremental_mode eq "onlyadd") {
|
---|
| 511 |
|
---|
| 512 | prime_doc_oid_count($archivedir);
|
---|
| 513 |
|
---|
| 514 | # Can now work out which files were new, already existed, and have
|
---|
| 515 | # been deleted
|
---|
| 516 |
|
---|
| 517 | new_vs_old_import_diff($archive_info,$block_hash,$importdir,
|
---|
| 518 | $archivedir,$verbosity,$incremental_mode);
|
---|
| 519 |
|
---|
| 520 | my @new_files = sort keys %{$block_hash->{'new_files'}};
|
---|
| 521 | if (scalar(@new_files>0)) {
|
---|
| 522 | print STDERR "New files and modified metadata files since last import:\n ";
|
---|
| 523 | print STDERR join("\n ",@new_files), "\n";
|
---|
| 524 | }
|
---|
| 525 |
|
---|
| 526 | if ($incremental) {
|
---|
| 527 | # only look for deletions if we are truely incremental
|
---|
| 528 | my @deleted_files = sort keys %{$block_hash->{'deleted_files'}};
|
---|
| 529 | # Filter out any in gsdl/tmp area
|
---|
| 530 | my @filtered_deleted_files = ();
|
---|
| 531 | my $gsdl_tmp_area = &util::filename_cat($ENV{'GSDLHOME'}, "tmp");
|
---|
| 532 | my $collect_tmp_area = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "tmp");
|
---|
| 533 | $gsdl_tmp_area = &util::filename_to_regex($gsdl_tmp_area);
|
---|
| 534 | $collect_tmp_area = &util::filename_to_regex($collect_tmp_area);
|
---|
| 535 |
|
---|
| 536 | foreach my $df (@deleted_files) {
|
---|
| 537 | next if ($df =~ m/^$gsdl_tmp_area/);
|
---|
| 538 | next if ($df =~ m/^$collect_tmp_area/);
|
---|
| 539 |
|
---|
| 540 | push(@filtered_deleted_files,$df);
|
---|
| 541 | }
|
---|
| 542 |
|
---|
| 543 |
|
---|
| 544 | @deleted_files = @filtered_deleted_files;
|
---|
| 545 |
|
---|
| 546 | if (scalar(@deleted_files)>0) {
|
---|
| 547 | print STDERR "Files deleted since last import:\n ";
|
---|
| 548 | print STDERR join("\n ",@deleted_files), "\n";
|
---|
| 549 |
|
---|
| 550 |
|
---|
| 551 | &plugin::remove_some($pluginfo, $collectcfg->{'infodbtype'}, $archivedir, \@deleted_files);
|
---|
| 552 |
|
---|
| 553 | mark_docs_for_deletion($archive_info,$block_hash,\@deleted_files, $archivedir,$verbosity, "delete");
|
---|
| 554 | }
|
---|
| 555 |
|
---|
| 556 | my @reindex_files = sort keys %{$block_hash->{'reindex_files'}};
|
---|
| 557 |
|
---|
| 558 | if (scalar(@reindex_files)>0) {
|
---|
| 559 | print STDERR "Files to reindex since last import:\n ";
|
---|
| 560 | print STDERR join("\n ",@reindex_files), "\n";
|
---|
| 561 | &plugin::remove_some($pluginfo, $collectcfg->{'infodbtype'}, $archivedir, \@reindex_files);
|
---|
| 562 | mark_docs_for_deletion($archive_info,$block_hash,\@reindex_files, $archivedir,$verbosity, "reindex");
|
---|
| 563 | }
|
---|
| 564 |
|
---|
| 565 | }
|
---|
| 566 |
|
---|
| 567 | # Play it safe, and run through the entire folder, only processing new or edited files
|
---|
| 568 | &plugin::read ($pluginfo, $importdir, "", $block_hash, $metadata, $processor, $maxdocs, 0, $gli);
|
---|
| 569 |
|
---|
| 570 | }
|
---|
| 571 | else {
|
---|
| 572 | &plugin::read ($pluginfo, $importdir, "", $block_hash, $metadata, $processor, $maxdocs, 0, $gli);
|
---|
| 573 | }
|
---|
| 574 |
|
---|
| 575 | }
|
---|
| 576 | else
|
---|
| 577 | {
|
---|
| 578 | #
|
---|
| 579 | # 1. Process delete files first
|
---|
| 580 | #
|
---|
| 581 |
|
---|
| 582 | my @deleted_files = keys %{$manifest_lookup->{'delete'}};
|
---|
| 583 | my @full_deleted_files = ();
|
---|
| 584 |
|
---|
| 585 | # ensure all filenames are absolute
|
---|
| 586 | foreach my $df (@deleted_files) {
|
---|
| 587 | my $full_df =
|
---|
| 588 | (&util::filename_is_absolute($df))
|
---|
| 589 | ? $df
|
---|
| 590 | : &util::filename_cat($importdir,$df);
|
---|
| 591 |
|
---|
| 592 | push(@full_deleted_files,$full_df);
|
---|
| 593 | }
|
---|
| 594 |
|
---|
| 595 | &plugin::remove_some($pluginfo, $collectcfg->{'infodbtype'}, $archivedir, \@full_deleted_files);
|
---|
| 596 | mark_docs_for_deletion($archive_info,{},
|
---|
[22421] | 597 | \@full_deleted_files,
|
---|
| 598 | $archivedir, $verbosity, "delete");
|
---|
[22413] | 599 |
|
---|
| 600 |
|
---|
| 601 | #
|
---|
| 602 | # 2. Now files for reindexing
|
---|
| 603 | #
|
---|
| 604 |
|
---|
| 605 | my @reindex_files = keys %{$manifest_lookup->{'reindex'}};
|
---|
| 606 | my @full_reindex_files = ();
|
---|
| 607 |
|
---|
| 608 | # ensure all filenames are absolute
|
---|
| 609 | foreach my $rf (@reindex_files) {
|
---|
| 610 | my $full_rf =
|
---|
| 611 | (&util::filename_is_absolute($rf))
|
---|
| 612 | ? $rf
|
---|
| 613 | : &util::filename_cat($importdir,$rf);
|
---|
| 614 |
|
---|
| 615 | push(@full_reindex_files,$full_rf);
|
---|
| 616 | }
|
---|
| 617 |
|
---|
| 618 | &plugin::remove_some($pluginfo, $collectcfg->{'infodbtype'}, $archivedir, \@full_reindex_files);
|
---|
| 619 | mark_docs_for_deletion($archive_info,{},\@full_reindex_files, $archivedir,$verbosity, "reindex");
|
---|
| 620 |
|
---|
| 621 | # And now ensure the new version of the file processed by appropriate
|
---|
| 622 | # plugin
|
---|
| 623 | foreach my $full_rf (@full_reindex_files) {
|
---|
| 624 | &plugin::read ($pluginfo, "", $full_rf, {}, {}, $processor, $maxdocs, 0, $gli);
|
---|
| 625 | }
|
---|
| 626 |
|
---|
| 627 |
|
---|
| 628 | #
|
---|
| 629 | # 3. Now finally any new files
|
---|
| 630 | #
|
---|
| 631 |
|
---|
| 632 | foreach my $file (keys %{$manifest_lookup->{'index'}}) {
|
---|
| 633 | &plugin::read ($pluginfo, $importdir, $file, {}, {}, $processor, $maxdocs, 0, $gli);
|
---|
| 634 | }
|
---|
| 635 |
|
---|
| 636 |
|
---|
| 637 | }
|
---|
| 638 |
|
---|
[22421] | 639 | if ($saveas eq "FedoraMETS") {
|
---|
| 640 | # create collection "doc obj" for Fedora that contains
|
---|
| 641 | # collection-level metadata
|
---|
| 642 |
|
---|
| 643 | my $doc_obj = new doc($config_filename,"nonindexed_doc","none");
|
---|
| 644 | $doc_obj->set_OID("collection");
|
---|
| 645 |
|
---|
| 646 | my $col_name = undef;
|
---|
| 647 | my $col_meta = $collectcfg->{'collectionmeta'};
|
---|
| 648 |
|
---|
| 649 | if (defined $col_meta) {
|
---|
| 650 | store_collectionmeta($col_meta,"collectionname",$doc_obj); # in GS3 this is a collection's name
|
---|
| 651 | store_collectionmeta($col_meta,"collectionextra",$doc_obj); # in GS3 this is a collection's description
|
---|
| 652 | }
|
---|
| 653 | $processor->process($doc_obj);
|
---|
| 654 | }
|
---|
| 655 |
|
---|
[22413] | 656 | &plugin::end($pluginfo, $processor);
|
---|
| 657 |
|
---|
| 658 | &plugin::deinit($pluginfo, $processor);
|
---|
| 659 |
|
---|
| 660 | # Store the value of OIDCount (used in doc.pm) so it can be
|
---|
| 661 | # restored correctly to this value on an incremental build
|
---|
| 662 | store_doc_oid_count($archivedir);
|
---|
| 663 |
|
---|
| 664 | # write out the archive information file
|
---|
| 665 | $processor->close_file_output() if $groupsize > 1;
|
---|
| 666 | $processor->close_group_output() if $processor->is_group();
|
---|
| 667 |
|
---|
| 668 | # for backwards compatability with archvies.inf file
|
---|
| 669 | if ($arcinfo_doc_filename =~ m/(contents)|(\.inf)$/) {
|
---|
| 670 | $archive_info->save_info($arcinfo_doc_filename);
|
---|
| 671 | }
|
---|
| 672 | else {
|
---|
| 673 | $archive_info->save_revinfo_db($arcinfo_src_filename);
|
---|
| 674 | }
|
---|
| 675 |
|
---|
| 676 | return $pluginfo;
|
---|
| 677 | }
|
---|
| 678 |
|
---|
| 679 |
|
---|
| 680 | sub generate_statistics
|
---|
| 681 | {
|
---|
| 682 | my $self = shift @_;
|
---|
[22421] | 683 | my ($pluginfo) = @_;
|
---|
[22413] | 684 |
|
---|
[22421] | 685 | my $inexport_mode = $self->{'mode'};
|
---|
| 686 |
|
---|
[22413] | 687 | my $statsfile = $self->{'statsfile'};
|
---|
| 688 | my $out = $self->{'out'};
|
---|
| 689 | my $faillogname = $self->{'faillogname'};
|
---|
| 690 | my $gli = $self->{'gli'};
|
---|
| 691 |
|
---|
| 692 | # write out import stats
|
---|
| 693 | my $close_stats = 0;
|
---|
| 694 | if ($statsfile !~ /^(STDERR|STDOUT)$/i) {
|
---|
| 695 | if (open (STATS, ">$statsfile")) {
|
---|
| 696 | $statsfile = 'import::STATS';
|
---|
| 697 | $close_stats = 1;
|
---|
| 698 | } else {
|
---|
| 699 | &gsprintf($out, "{import.cannot_open_stats_file}", $statsfile);
|
---|
| 700 | &gsprintf($out, "{import.stats_backup}\n");
|
---|
| 701 | $statsfile = 'STDERR';
|
---|
| 702 | }
|
---|
| 703 | }
|
---|
| 704 |
|
---|
| 705 | &gsprintf($out, "\n");
|
---|
| 706 | &gsprintf($out, "*********************************************\n");
|
---|
[22421] | 707 | &gsprintf($out, "{$inexport_mode.complete}\n");
|
---|
[22413] | 708 | &gsprintf($out, "*********************************************\n");
|
---|
| 709 |
|
---|
| 710 | &plugin::write_stats($pluginfo, $statsfile, $faillogname, $gli);
|
---|
| 711 | if ($close_stats) {
|
---|
| 712 | close STATS;
|
---|
| 713 | }
|
---|
| 714 |
|
---|
| 715 | close OUT if $self->{'close_out'};
|
---|
| 716 | close FAILLOG;
|
---|
| 717 | }
|
---|
| 718 |
|
---|
| 719 |
|
---|
| 720 |
|
---|
| 721 |
|
---|
| 722 |
|
---|
| 723 |
|
---|
| 724 |
|
---|
[21306] | 725 | sub oid_count_file {
|
---|
| 726 | my ($archivedir) = @_;
|
---|
| 727 | return &util::filename_cat ($archivedir, "OIDcount");
|
---|
| 728 | }
|
---|
| 729 |
|
---|
| 730 |
|
---|
[18528] | 731 | sub prime_doc_oid_count
|
---|
| 732 | {
|
---|
| 733 | my ($archivedir) = @_;
|
---|
[21306] | 734 | my $oid_count_filename = &oid_count_file($archivedir);
|
---|
[18528] | 735 |
|
---|
| 736 | if (-e $oid_count_filename) {
|
---|
| 737 | if (open(OIDIN,"<$oid_count_filename")) {
|
---|
| 738 | my $OIDcount = <OIDIN>;
|
---|
| 739 | chomp $OIDcount;
|
---|
| 740 | close(OIDIN);
|
---|
| 741 |
|
---|
| 742 | $doc::OIDcount = $OIDcount;
|
---|
| 743 | }
|
---|
| 744 | else {
|
---|
| 745 |
|
---|
| 746 | print STDERR "Warning: unable to read document OID count from $oid_count_filename\n";
|
---|
| 747 | print STDERR "Setting value to 0\n";
|
---|
| 748 | }
|
---|
| 749 | }
|
---|
| 750 |
|
---|
| 751 | }
|
---|
| 752 |
|
---|
| 753 | sub store_doc_oid_count
|
---|
| 754 | {
|
---|
| 755 | # Use the file "OIDcount" in the archives directory to record
|
---|
| 756 | # what value doc.pm got up to
|
---|
| 757 |
|
---|
| 758 | my ($archivedir) = @_;
|
---|
[21306] | 759 | my $oid_count_filename = &oid_count_file($archivedir);
|
---|
[18528] | 760 |
|
---|
| 761 |
|
---|
| 762 | if (open(OIDOUT,">$oid_count_filename")) {
|
---|
| 763 | print OIDOUT $doc::OIDcount, "\n";
|
---|
| 764 |
|
---|
| 765 | close(OIDOUT);
|
---|
| 766 | }
|
---|
| 767 | else {
|
---|
| 768 | print STDERR "Warning: unable to store document OID count\n";
|
---|
| 769 | }
|
---|
| 770 | }
|
---|
| 771 |
|
---|
| 772 |
|
---|
| 773 |
|
---|
[18457] | 774 | sub new_vs_old_import_diff
|
---|
| 775 | {
|
---|
[20578] | 776 | my ($archive_info,$block_hash,$importdir,$archivedir,$verbosity,$incremental_mode) = @_;
|
---|
[18457] | 777 |
|
---|
[21620] | 778 | # Get the infodbtype value for this collection from the arcinfo object
|
---|
| 779 | my $infodbtype = $archive_info->{'infodbtype'};
|
---|
| 780 |
|
---|
[20776] | 781 | # in this method, we want to know if metadata files are modified or not.
|
---|
[21620] | 782 | my $arcinfo_doc_filename = &dbutil::get_infodb_file_path($infodbtype, "archiveinf-doc", $archivedir);
|
---|
[20776] | 783 |
|
---|
| 784 | my $archiveinf_timestamp = -M $arcinfo_doc_filename;
|
---|
| 785 |
|
---|
[18457] | 786 | # First convert all files to absolute form
|
---|
| 787 | # This is to support the situation where the import folder is not
|
---|
| 788 | # the default
|
---|
| 789 |
|
---|
| 790 | my $prev_all_files = $archive_info->{'prev_import_filelist'};
|
---|
| 791 | my $full_prev_all_files = {};
|
---|
| 792 |
|
---|
| 793 | foreach my $prev_file (keys %$prev_all_files) {
|
---|
| 794 |
|
---|
| 795 | if (!&util::filename_is_absolute($prev_file)) {
|
---|
| 796 | my $full_prev_file = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},$prev_file);
|
---|
| 797 | $full_prev_all_files->{$full_prev_file} = $prev_file;
|
---|
| 798 | }
|
---|
| 799 | else {
|
---|
| 800 | $full_prev_all_files->{$prev_file} = $prev_file;
|
---|
| 801 | }
|
---|
| 802 | }
|
---|
| 803 |
|
---|
[18469] | 804 |
|
---|
[18457] | 805 | # Figure out which are the new files, existing files and so
|
---|
| 806 | # by implication the files from the previous import that are not
|
---|
| 807 | # there any more => mark them for deletion
|
---|
| 808 | foreach my $curr_file (keys %{$block_hash->{'all_files'}}) {
|
---|
| 809 |
|
---|
| 810 | my $full_curr_file = $curr_file;
|
---|
| 811 |
|
---|
| 812 | # entry in 'all_files' is moved to either 'existing_files',
|
---|
[20776] | 813 | # 'deleted_files', 'new_files', or 'new_or_modified_metadata_files'
|
---|
[18457] | 814 |
|
---|
| 815 | if (!&util::filename_is_absolute($curr_file)) {
|
---|
| 816 | # add in import dir to make absolute
|
---|
| 817 | $full_curr_file = &util::filename_cat($importdir,$curr_file);
|
---|
| 818 | }
|
---|
| 819 |
|
---|
[19498] | 820 | # figure out if new file or not
|
---|
[18457] | 821 | if (defined $full_prev_all_files->{$full_curr_file}) {
|
---|
[20776] | 822 | # delete it so that only files that need deleting are left
|
---|
| 823 | delete $full_prev_all_files->{$full_curr_file};
|
---|
| 824 |
|
---|
| 825 | # had it before. is it a metadata file?
|
---|
| 826 | if ($block_hash->{'metadata_files'}->{$full_curr_file}) {
|
---|
[20578] | 827 |
|
---|
[20776] | 828 | # is it modified??
|
---|
| 829 | if (-M $full_curr_file < $archiveinf_timestamp) {
|
---|
| 830 | print STDERR "*** Detected a modified metadata file: $full_curr_file\n" if $verbosity > 2;
|
---|
| 831 | # its newer than last build
|
---|
| 832 | $block_hash->{'new_or_modified_metadata_files'}->{$full_curr_file} = 1;
|
---|
| 833 | }
|
---|
[20578] | 834 | }
|
---|
| 835 | else {
|
---|
[20776] | 836 | if ($incremental_mode eq "all") {
|
---|
| 837 |
|
---|
| 838 | # had it before
|
---|
| 839 | $block_hash->{'existing_files'}->{$full_curr_file} = 1;
|
---|
| 840 |
|
---|
| 841 | }
|
---|
| 842 | else {
|
---|
| 843 | # Warning in "onlyadd" mode, but had it before!
|
---|
| 844 | print STDERR "Warning: File $full_curr_file previously imported.\n";
|
---|
| 845 | print STDERR " Treating as new file\n";
|
---|
| 846 |
|
---|
| 847 | $block_hash->{'new_files'}->{$full_curr_file} = 1;
|
---|
| 848 |
|
---|
| 849 | }
|
---|
[20578] | 850 | }
|
---|
| 851 | }
|
---|
| 852 | else {
|
---|
| 853 | if ($block_hash->{'metadata_files'}->{$full_curr_file}) {
|
---|
| 854 | # the new file is the special sort of file greenstone uses
|
---|
| 855 | # to attach metadata to src documents
|
---|
| 856 | # i.e metadata.xml
|
---|
| 857 | # (but note, the filename used is not constrained in
|
---|
| 858 | # Greenstone to always be this)
|
---|
[18457] | 859 |
|
---|
[20776] | 860 | print STDERR "***** Detected new metadata file: $full_curr_file\n" if $verbosity > 2;
|
---|
| 861 | $block_hash->{'new_or_modified_metadata_files'}->{$full_curr_file} = 1;
|
---|
[20578] | 862 | }
|
---|
| 863 | else {
|
---|
| 864 | $block_hash->{'new_files'}->{$full_curr_file} = 1;
|
---|
| 865 | }
|
---|
[18457] | 866 | }
|
---|
[20578] | 867 |
|
---|
[18457] | 868 |
|
---|
| 869 | delete $block_hash->{'all_files'}->{$curr_file};
|
---|
| 870 | }
|
---|
| 871 |
|
---|
[20578] | 872 |
|
---|
[21306] | 873 |
|
---|
| 874 |
|
---|
[20776] | 875 | # Deal with complication of new or modified metadata files by forcing
|
---|
[20578] | 876 | # everything from this point down in the file hierarchy to
|
---|
| 877 | # be freshly imported.
|
---|
| 878 | #
|
---|
| 879 | # This may mean files that have not changed are reindexed, but does
|
---|
| 880 | # guarantee by the end of processing all new metadata is correctly
|
---|
| 881 | # associated with the relevant document(s).
|
---|
| 882 |
|
---|
[20776] | 883 | foreach my $new_mdf (keys %{$block_hash->{'new_or_modified_metadata_files'}}) {
|
---|
[20578] | 884 | my ($fileroot,$situated_dir,$ext) = fileparse($new_mdf, "\\.[^\\.]+\$");
|
---|
| 885 |
|
---|
| 886 | $situated_dir =~ s/[\\\/]+$//; # remove tailing slashes
|
---|
[20769] | 887 | $situated_dir =~ s/\\/\\\\/g; # need to protect windows slash \ in regular expression
|
---|
| 888 |
|
---|
[20578] | 889 | # Go through existing_files, and mark anything that is contained
|
---|
| 890 | # within 'situated_dir' to be reindexed (in case some of the metadata
|
---|
| 891 | # attaches to one of these files)
|
---|
| 892 |
|
---|
| 893 | my $reindex_files = [];
|
---|
| 894 |
|
---|
| 895 | foreach my $existing_f (keys %{$block_hash->{'existing_files'}}) {
|
---|
[20769] | 896 |
|
---|
[20578] | 897 | if ($existing_f =~ m/^$situated_dir/) {
|
---|
| 898 | push(@$reindex_files,$existing_f);
|
---|
| 899 | $block_hash->{'reindex_files'}->{$existing_f} = 1;
|
---|
[21306] | 900 | delete $block_hash->{'existing_files'}->{$existing_f};
|
---|
[20578] | 901 |
|
---|
| 902 | }
|
---|
| 903 | }
|
---|
| 904 |
|
---|
| 905 | # metadata file needs to be in new_files list so parsed by MetadataXMLPlug
|
---|
| 906 | # (or equivalent)
|
---|
| 907 | $block_hash->{'new_files'}->{$new_mdf} = 1;
|
---|
| 908 |
|
---|
| 909 | }
|
---|
| 910 |
|
---|
[21306] | 911 | # go through remaining existing files and work out what has changed and needs to be reindexed.
|
---|
| 912 | my @existing_files = sort keys %{$block_hash->{'existing_files'}};
|
---|
| 913 |
|
---|
| 914 | my $reindex_files = [];
|
---|
| 915 |
|
---|
| 916 | foreach my $existing_filename (@existing_files) {
|
---|
| 917 | if (-M $existing_filename < $archiveinf_timestamp) {
|
---|
| 918 | # file is newer than last build
|
---|
| 919 |
|
---|
| 920 | my $existing_file = $existing_filename;
|
---|
| 921 | #my $collectdir = &util::filename_cat($ENV{'GSDLCOLLECTDIR'});
|
---|
| 922 |
|
---|
| 923 | #my $collectdir_resafe = &util::filename_to_regex($collectdir);
|
---|
| 924 | #$existing_file =~ s/^$collectdir_resafe(\\|\/)?//;
|
---|
| 925 |
|
---|
| 926 | print STDERR "**** Reindexing existing file: $existing_file\n";
|
---|
| 927 |
|
---|
| 928 | push(@$reindex_files,$existing_file);
|
---|
| 929 | $block_hash->{'reindex_files'}->{$existing_filename} = 1;
|
---|
| 930 | }
|
---|
| 931 |
|
---|
| 932 | }
|
---|
| 933 |
|
---|
[20578] | 934 |
|
---|
[18469] | 935 | # By this point full_prev_all_files contains the files
|
---|
| 936 | # mentioned in archiveinf-src.db but are not in the 'import'
|
---|
| 937 | # folder (or whatever was specified through -importdir ...)
|
---|
| 938 |
|
---|
| 939 | # This list can contain files that were created in the 'tmp' or
|
---|
| 940 | # 'cache' areas (such as screen-size and thumbnail images).
|
---|
[18457] | 941 | #
|
---|
[18469] | 942 | # In building the final list of files to delete, we test to see if
|
---|
[20578] | 943 | # it exists on the filesystem and if it does (unusual for a "normal"
|
---|
| 944 | # file in import, but possible in the case of 'tmp' files),
|
---|
| 945 | # supress it from going into the final list
|
---|
[18469] | 946 |
|
---|
| 947 | my $collectdir = $ENV{'GSDLCOLLECTDIR'};
|
---|
| 948 |
|
---|
[18457] | 949 | my @deleted_files = values %$full_prev_all_files;
|
---|
[18469] | 950 | map { my $curr_file = $_;
|
---|
| 951 | my $full_curr_file = $curr_file;
|
---|
| 952 |
|
---|
| 953 | if (!&util::filename_is_absolute($curr_file)) {
|
---|
| 954 | # add in import dir to make absolute
|
---|
| 955 |
|
---|
| 956 | $full_curr_file = &util::filename_cat($collectdir,$curr_file);
|
---|
| 957 | }
|
---|
| 958 |
|
---|
| 959 |
|
---|
| 960 | if (!-e $full_curr_file) {
|
---|
| 961 | $block_hash->{'deleted_files'}->{$curr_file} = 1;
|
---|
| 962 | }
|
---|
| 963 | } @deleted_files;
|
---|
[20578] | 964 |
|
---|
| 965 |
|
---|
| 966 |
|
---|
[18457] | 967 | }
|
---|
| 968 |
|
---|
[19498] | 969 |
|
---|
[20788] | 970 | # this is used to delete "deleted" docs, and to remove old versions of "changed" docs
|
---|
[21306] | 971 | # $mode is 'delete' or 'reindex'
|
---|
| 972 | sub mark_docs_for_deletion
|
---|
[18457] | 973 | {
|
---|
[21306] | 974 | my ($archive_info,$block_hash,$deleted_files,$archivedir,$verbosity,$mode) = @_;
|
---|
[18457] | 975 |
|
---|
[21306] | 976 | my $mode_text = "deleted from index";
|
---|
| 977 | if ($mode eq "reindex") {
|
---|
| 978 | $mode_text = "reindexed";
|
---|
| 979 | }
|
---|
[18457] | 980 |
|
---|
[21620] | 981 | # Get the infodbtype value for this collection from the arcinfo object
|
---|
| 982 | my $infodbtype = $archive_info->{'infodbtype'};
|
---|
[18457] | 983 |
|
---|
[21620] | 984 | my $arcinfo_doc_filename = &dbutil::get_infodb_file_path($infodbtype, "archiveinf-doc", $archivedir);
|
---|
| 985 | my $arcinfo_src_filename = &dbutil::get_infodb_file_path($infodbtype, "archiveinf-src", $archivedir);
|
---|
| 986 |
|
---|
[22010] | 987 |
|
---|
[18457] | 988 | # record files marked for deletion in arcinfo
|
---|
[19498] | 989 | foreach my $file (@$deleted_files) {
|
---|
[21564] | 990 | # use 'archiveinf-src' info database file to look up all the OIDs
|
---|
[19789] | 991 | # that this file is used in (note in most cases, it's just one OID)
|
---|
[18457] | 992 |
|
---|
[21620] | 993 | my $src_rec_string = &dbutil::read_infodb_entry($infodbtype, $arcinfo_src_filename, $file);
|
---|
[21554] | 994 | my $src_rec = &dbutil::convert_infodb_string_to_hash($src_rec_string);
|
---|
[18457] | 995 | my $oids = $src_rec->{'oid'};
|
---|
[20776] | 996 | my $file_record_deleted = 0;
|
---|
[20788] | 997 |
|
---|
| 998 | # delete the src record
|
---|
[22010] | 999 | my $src_infodb_file_handle = &dbutil::open_infodb_write_handle($infodbtype, $arcinfo_src_filename, "append");
|
---|
[21620] | 1000 | &dbutil::delete_infodb_entry($infodbtype, $src_infodb_file_handle, $file);
|
---|
[22010] | 1001 | &dbutil::close_infodb_write_handle($infodbtype, $src_infodb_file_handle);
|
---|
| 1002 |
|
---|
| 1003 |
|
---|
[18457] | 1004 | foreach my $oid (@$oids) {
|
---|
| 1005 |
|
---|
[20788] | 1006 | # find the source doc (the primary file that becomes this oid)
|
---|
[21620] | 1007 | my $doc_rec_string = &dbutil::read_infodb_entry($infodbtype, $arcinfo_doc_filename, $oid);
|
---|
[21554] | 1008 | my $doc_rec = &dbutil::convert_infodb_string_to_hash($doc_rec_string);
|
---|
[20776] | 1009 | my $doc_source_file = $doc_rec->{'src-file'}->[0];
|
---|
| 1010 | if (!&util::filename_is_absolute($doc_source_file)) {
|
---|
| 1011 | $doc_source_file = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},$doc_source_file);
|
---|
| 1012 | }
|
---|
[18457] | 1013 |
|
---|
[20788] | 1014 | if ($doc_source_file ne $file) {
|
---|
| 1015 | # its an associated or metadata file
|
---|
| 1016 |
|
---|
[20776] | 1017 | # mark source doc for reimport as one of its assoc files has changed or deleted
|
---|
| 1018 | $block_hash->{'reindex_files'}->{$doc_source_file} = 1;
|
---|
[20788] | 1019 |
|
---|
[18457] | 1020 | }
|
---|
[20788] | 1021 | my $curr_status = $archive_info->get_status_info($oid);
|
---|
| 1022 | if (defined($curr_status) && (($curr_status ne "D"))) {
|
---|
| 1023 | if ($verbosity>1) {
|
---|
| 1024 | print STDERR "$oid ($doc_source_file) marked to be $mode_text on next buildcol.pl\n";
|
---|
[19498] | 1025 | }
|
---|
[20788] | 1026 | # mark oid for deletion (it will be deleted or reimported)
|
---|
| 1027 | $archive_info->set_status_info($oid,"D");
|
---|
[21620] | 1028 | my $val = &dbutil::read_infodb_entry($infodbtype, $arcinfo_doc_filename, $oid);
|
---|
[20788] | 1029 | $val =~ s/^<index-status>(.*)$/<index-status>D/m;
|
---|
[21557] | 1030 |
|
---|
| 1031 | my $val_rec = &dbutil::convert_infodb_string_to_hash($val);
|
---|
[22010] | 1032 | my $doc_infodb_file_handle = &dbutil::open_infodb_write_handle($infodbtype, $arcinfo_doc_filename, "append");
|
---|
| 1033 |
|
---|
[21620] | 1034 | &dbutil::write_infodb_entry($infodbtype, $doc_infodb_file_handle, $oid, $val_rec);
|
---|
[22010] | 1035 | &dbutil::close_infodb_write_handle($infodbtype, $doc_infodb_file_handle);
|
---|
[19498] | 1036 | }
|
---|
[18457] | 1037 | }
|
---|
[22327] | 1038 |
|
---|
[18457] | 1039 | }
|
---|
[22327] | 1040 | # now go through and check that we haven't marked any primary files for reindex (because their associated files have changed/deleted) when they have been deleted themselves.
|
---|
| 1041 | foreach my $file (@$deleted_files) {
|
---|
| 1042 | if (defined $block_hash->{'reindex_files'}->{$file}) {
|
---|
| 1043 | delete $block_hash->{'reindex_files'}->{$file};
|
---|
| 1044 | }
|
---|
| 1045 | }
|
---|
[21560] | 1046 |
|
---|
[22010] | 1047 |
|
---|
[18457] | 1048 | }
|
---|
| 1049 |
|
---|
| 1050 |
|
---|
[18554] | 1051 |
|
---|
[18457] | 1052 | 1;
|
---|