[24626] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
| 3 | # inexport.pm -- useful class to support import.pl and export.pl
|
---|
| 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 |
|
---|
| 30 | no strict 'refs'; # allow filehandles to be variables and vice versa
|
---|
| 31 | no strict 'subs'; # allow barewords (eg STDERR) as function arguments
|
---|
| 32 |
|
---|
| 33 | use arcinfo;
|
---|
| 34 | use colcfg;
|
---|
| 35 | use dbutil;
|
---|
| 36 | use doc;
|
---|
| 37 | use plugin;
|
---|
| 38 | use plugout;
|
---|
| 39 | use manifest;
|
---|
| 40 | use inexport;
|
---|
| 41 | use util;
|
---|
| 42 | use scriptutil;
|
---|
| 43 | use FileHandle;
|
---|
| 44 | use gsprintf 'gsprintf';
|
---|
| 45 | use printusage;
|
---|
| 46 | use parse2;
|
---|
| 47 |
|
---|
| 48 | use File::Basename;
|
---|
| 49 |
|
---|
| 50 | sub new
|
---|
| 51 | {
|
---|
| 52 | my $class = shift (@_);
|
---|
| 53 | my ($mode,$argv,$options,$opt_listall_options) = @_;
|
---|
| 54 |
|
---|
| 55 | my $self = { 'xml' => 0, 'mode' => $mode };
|
---|
| 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 |
|
---|
| 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 |
|
---|
| 86 | if ($self->{'xml'}) {
|
---|
| 87 | &PrintUsage::print_xml_usage($options);
|
---|
| 88 | print "\n";
|
---|
| 89 | return bless $self, $class;
|
---|
| 90 | }
|
---|
| 91 |
|
---|
| 92 | if ($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 |
|
---|
| 101 | if ($intArgLeftinAfterParsing != 1 || (@$argv && $argv->[0] =~ /^\-+h/))
|
---|
| 102 | {
|
---|
| 103 | &PrintUsage::print_txt_usage($options, "{import.params}");
|
---|
| 104 | die "\n";
|
---|
| 105 | }
|
---|
| 106 |
|
---|
| 107 | $self->{'close_out'} = 0;
|
---|
| 108 | my $out = $self->{'out'};
|
---|
| 109 | if ($out !~ /^(STDERR|STDOUT)$/i) {
|
---|
| 110 | open (OUT, ">$out") ||
|
---|
| 111 | (&gsprintf(STDERR, "{common.cannot_open_output_file}: $!\n", $out) && die);
|
---|
| 112 | $out = 'inexport::OUT';
|
---|
| 113 | $self->{'close_out'} = 1;
|
---|
| 114 | }
|
---|
| 115 | $out->autoflush(1);
|
---|
| 116 | $self->{'out'} = $out;
|
---|
| 117 |
|
---|
| 118 | # @ARGV should be only one item, the name of the collection
|
---|
| 119 | $self->{'collection'} = shift @$argv;
|
---|
| 120 |
|
---|
| 121 | if ((defined $self->{'jobs'}) && ($self->{'jobs'}>1)) {
|
---|
| 122 | require ParallelInexport;
|
---|
| 123 | }
|
---|
| 124 |
|
---|
| 125 | return bless $self, $class;
|
---|
| 126 | }
|
---|
| 127 |
|
---|
| 128 | # Simplified version of the contstructor for use with CGI scripts
|
---|
| 129 | sub newCGI
|
---|
| 130 | {
|
---|
| 131 | my $class = shift (@_);
|
---|
| 132 | my ($mode,$collect,$gsdl_cgi,$opt_site) = @_;
|
---|
| 133 |
|
---|
| 134 | my $self = { 'xml' => 0, 'mode' => $mode };
|
---|
| 135 |
|
---|
| 136 | $self->{'out'} = STDERR;
|
---|
| 137 |
|
---|
| 138 | if (defined $gsdl_cgi) {
|
---|
| 139 | $self->{'site'} = $opt_site;
|
---|
| 140 | my $collect_dir = $gsdl_cgi->get_collection_dir($opt_site);
|
---|
| 141 | $self->{'collectdir'} = $collect_dir;
|
---|
| 142 | }
|
---|
| 143 | else {
|
---|
| 144 | $self->{'site'} = "";
|
---|
| 145 | $self->{'collectdir'} = &util::filename_cat($ENV{'GSDLHOME'},"collect");
|
---|
| 146 | }
|
---|
| 147 | $self->{'faillog'} = "";
|
---|
| 148 |
|
---|
| 149 | $self->{'collection'} = $collect;
|
---|
| 150 |
|
---|
| 151 | return bless $self, $class;
|
---|
| 152 | }
|
---|
| 153 | sub get_collection
|
---|
| 154 | {
|
---|
| 155 | my $self = shift @_;
|
---|
| 156 |
|
---|
| 157 | return $self->{'collection'};
|
---|
| 158 | }
|
---|
| 159 |
|
---|
| 160 |
|
---|
| 161 | sub read_collection_cfg
|
---|
| 162 | {
|
---|
| 163 | my $self = shift @_;
|
---|
| 164 | my ($collection,$options) = @_;
|
---|
| 165 |
|
---|
| 166 | my $collectdir = $self->{'collectdir'};
|
---|
| 167 | my $site = $self->{'site'};
|
---|
| 168 | my $out = $self->{'out'};
|
---|
| 169 |
|
---|
| 170 | if (($collection = &colcfg::use_collection($site, $collection, $collectdir)) eq "") {
|
---|
| 171 | &PrintUsage::print_txt_usage($options, "{import.params}");
|
---|
| 172 | die "\n";
|
---|
| 173 | }
|
---|
| 174 |
|
---|
| 175 | # add collection's perllib dir into include path in
|
---|
| 176 | # case we have collection specific modules
|
---|
| 177 | my $collection_perllib_path = $ENV{'GSDLCOLLECTDIR'} . '/perllib';
|
---|
| 178 | my $inc_paths = join(':', @INC);
|
---|
| 179 | if ($inc_paths !~ /$collection_perllib_path/)
|
---|
| 180 | {
|
---|
| 181 | unshift (@INC, $collection_perllib_path); # [jmt12]
|
---|
| 182 | }
|
---|
| 183 |
|
---|
| 184 | # check that we can open the faillog
|
---|
| 185 | my $faillog = $self->{'faillog'};
|
---|
| 186 | if ($faillog eq "") {
|
---|
| 187 | $faillog = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "etc", "fail.log");
|
---|
| 188 | }
|
---|
| 189 | open (FAILLOG, ">$faillog") ||
|
---|
| 190 | (&gsprintf(STDERR, "{import.cannot_open_fail_log}\n", $faillog) && die);
|
---|
| 191 |
|
---|
| 192 |
|
---|
| 193 | my $faillogname = $faillog;
|
---|
| 194 | $faillog = 'inexport::FAILLOG';
|
---|
| 195 | $faillog->autoflush(1);
|
---|
| 196 | $self->{'faillog'} = $faillog;
|
---|
| 197 | $self->{'faillogname'} = $faillogname;
|
---|
| 198 |
|
---|
| 199 | # Read in the collection configuration file.
|
---|
| 200 | my ($config_filename, $gs_mode) = &colcfg::get_collect_cfg_name($out);
|
---|
| 201 | my $collectcfg = &colcfg::read_collection_cfg ($config_filename, $gs_mode);
|
---|
| 202 |
|
---|
| 203 | return ($config_filename,$collectcfg);
|
---|
| 204 | }
|
---|
| 205 |
|
---|
| 206 | sub set_collection_options
|
---|
| 207 | {
|
---|
| 208 | my $self = shift @_;
|
---|
| 209 | my ($collectcfg) = @_;
|
---|
| 210 |
|
---|
| 211 | my $inexport_mode = $self->{'mode'};
|
---|
| 212 |
|
---|
| 213 | my $verbosity = $self->{'verbosity'};
|
---|
| 214 | my $debug = $self->{'debug'};
|
---|
| 215 | my $importdir = $self->{'importdir'};
|
---|
| 216 | my $archivedir = $self->{'archivedir'} || $self->{'exportdir'} || "";
|
---|
| 217 | my $out = $self->{'out'};
|
---|
| 218 |
|
---|
| 219 | # If the infodbtype value wasn't defined in the collect.cfg file, use the default
|
---|
| 220 | if (!defined($collectcfg->{'infodbtype'}))
|
---|
| 221 | {
|
---|
| 222 | $collectcfg->{'infodbtype'} = &dbutil::get_default_infodb_type();
|
---|
| 223 | }
|
---|
| 224 | if ($collectcfg->{'infodbtype'} eq "gdbm-txtgz") {
|
---|
| 225 | # we can't use the text version for archives dbs.
|
---|
| 226 | $collectcfg->{'infodbtype'} = "gdbm";
|
---|
| 227 | }
|
---|
| 228 | if (defined $collectcfg->{'importdir'} && $importdir eq "") {
|
---|
| 229 | $importdir = $collectcfg->{'importdir'};
|
---|
| 230 | }
|
---|
| 231 | if (defined $collectcfg->{'archivedir'} && $archivedir eq "") {
|
---|
| 232 | $archivedir = $collectcfg->{'archivedir'};
|
---|
| 233 | }
|
---|
| 234 | # fill in the default import and archives directories if none
|
---|
| 235 | # were supplied, turn all \ into / and remove trailing /
|
---|
| 236 | $importdir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "import") if $importdir eq "";
|
---|
| 237 | $importdir =~ s/[\\\/]+/\//g;
|
---|
| 238 | $importdir =~ s/\/$//;
|
---|
| 239 | if (!-e $importdir) {
|
---|
| 240 | &gsprintf($out, "{import.no_import_dir}\n\n", $importdir);
|
---|
| 241 | die "\n";
|
---|
| 242 | }
|
---|
| 243 | $self->{'importdir'} = $importdir;
|
---|
| 244 |
|
---|
| 245 | if ($archivedir eq "") {
|
---|
| 246 | if ($inexport_mode eq "import") {
|
---|
| 247 | $archivedir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "archives");
|
---|
| 248 | }
|
---|
| 249 | elsif ($inexport_mode eq "export") {
|
---|
| 250 | $archivedir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "export");
|
---|
| 251 | }
|
---|
| 252 | else {
|
---|
| 253 | print STDERR "Warning: Unrecognized import/export mode '$inexport_mode'\n";
|
---|
| 254 | print STDERR " Defaulting to 'archives' for file output\n";
|
---|
| 255 | $archivedir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "archives");
|
---|
| 256 | }
|
---|
| 257 | }
|
---|
| 258 |
|
---|
| 259 | $archivedir =~ s/[\\\/]+/\//g;
|
---|
| 260 | $archivedir =~ s/\/$//;
|
---|
| 261 | $self->{'archivedir'} = $archivedir;
|
---|
| 262 |
|
---|
| 263 | if ($verbosity !~ /\d+/) {
|
---|
| 264 | if (defined $collectcfg->{'verbosity'} && $collectcfg->{'verbosity'} =~ /\d+/) {
|
---|
| 265 | $verbosity = $collectcfg->{'verbosity'};
|
---|
| 266 | } else {
|
---|
| 267 | $verbosity = 2; # the default
|
---|
| 268 | }
|
---|
| 269 | }
|
---|
| 270 | $self->{'verbosity'} = $verbosity;
|
---|
| 271 |
|
---|
| 272 | if (defined $collectcfg->{'manifest'} && $self->{'manifest'} eq "") {
|
---|
| 273 | $self->{'manifest'} = $collectcfg->{'manifest'};
|
---|
| 274 | }
|
---|
| 275 |
|
---|
| 276 | if (defined $collectcfg->{'gzip'} && !$self->{'gzip'}) {
|
---|
| 277 | if ($collectcfg->{'gzip'} =~ /^true$/i) {
|
---|
| 278 | $self->{'gzip'} = 1;
|
---|
| 279 | }
|
---|
| 280 | }
|
---|
| 281 |
|
---|
| 282 | if ($self->{'maxdocs'} !~ /\-?\d+/) {
|
---|
| 283 | if (defined $collectcfg->{'maxdocs'} && $collectcfg->{'maxdocs'} =~ /\-?\d+/) {
|
---|
| 284 | $self->{'maxdocs'} = $collectcfg->{'maxdocs'};
|
---|
| 285 | } else {
|
---|
| 286 | $self->{'maxdocs'} = -1; # the default
|
---|
| 287 | }
|
---|
| 288 | }
|
---|
| 289 |
|
---|
| 290 | if ((defined $self->{'groupsize'}) && ($self->{'groupsize'} == 1)) {
|
---|
| 291 | if (defined $collectcfg->{'groupsize'} && $collectcfg->{'groupsize'} =~ /\d+/) {
|
---|
| 292 | $self->{'groupsize'} = $collectcfg->{'groupsize'};
|
---|
| 293 | }
|
---|
| 294 | }
|
---|
| 295 |
|
---|
| 296 | if (!defined $self->{'OIDtype'}
|
---|
| 297 | || ($self->{'OIDtype'} !~ /^(hash|incremental|assigned|dirname)$/ )) {
|
---|
| 298 | if (defined $collectcfg->{'OIDtype'}
|
---|
| 299 | && $collectcfg->{'OIDtype'} =~ /^(hash|incremental|assigned|dirname)$/) {
|
---|
| 300 | $self->{'OIDtype'} = $collectcfg->{'OIDtype'};
|
---|
| 301 | } else {
|
---|
| 302 | $self->{'OIDtype'} = "hash"; # the default
|
---|
| 303 | }
|
---|
| 304 | }
|
---|
| 305 |
|
---|
| 306 | if ((!defined $self->{'OIDmetadata'}) || ($self->{'OIDmetadata'} eq "")) {
|
---|
| 307 | if (defined $collectcfg->{'OIDmetadata'}) {
|
---|
| 308 | $self->{'OIDmetadata'} = $collectcfg->{'OIDmetadata'};
|
---|
| 309 | } else {
|
---|
| 310 | $self->{'OIDmetadata'} = "dc.Identifier"; # the default
|
---|
| 311 | }
|
---|
| 312 | }
|
---|
| 313 |
|
---|
| 314 | my $sortmeta = $self->{'sortmeta'};
|
---|
| 315 | if (defined $collectcfg->{'sortmeta'} && (!defined $sortmeta || $sortmeta eq "")) {
|
---|
| 316 | $sortmeta = $collectcfg->{'sortmeta'};
|
---|
| 317 | }
|
---|
| 318 | # sortmeta cannot be used with group size
|
---|
| 319 | $sortmeta = undef unless defined $sortmeta && $sortmeta =~ /\S/;
|
---|
| 320 | if (defined $sortmeta && $self->{'groupsize'} > 1) {
|
---|
| 321 | &gsprintf($out, "{import.cannot_sort}\n\n");
|
---|
| 322 | $sortmeta = undef;
|
---|
| 323 | }
|
---|
| 324 | $self->{'sortmeta'} = $sortmeta;
|
---|
| 325 |
|
---|
| 326 | if (defined $collectcfg->{'removeprefix'} && $self->{'removeprefix'} eq "") {
|
---|
| 327 | $self->{'removeprefix'} = $collectcfg->{'removeprefix'};
|
---|
| 328 | }
|
---|
| 329 |
|
---|
| 330 | if (defined $collectcfg->{'removesuffix'} && $self->{'removesuffix'} eq "") {
|
---|
| 331 | $self->{'removesuffix'} = $collectcfg->{'removesuffix'};
|
---|
| 332 | }
|
---|
| 333 | if (defined $collectcfg->{'debug'} && $collectcfg->{'debug'} =~ /^true$/i) {
|
---|
| 334 | $self->{'debug'} = 1;
|
---|
| 335 | }
|
---|
| 336 | if (defined $collectcfg->{'gli'} && $collectcfg->{'gli'} =~ /^true$/i) {
|
---|
| 337 | $self->{'gli'} = 1;
|
---|
| 338 | }
|
---|
| 339 | $self->{'gli'} = 0 unless defined $self->{'gli'};
|
---|
| 340 |
|
---|
| 341 | # check keepold and removeold
|
---|
| 342 | my $checkdir = ($inexport_mode eq "import") ? "archives" : "export";
|
---|
| 343 |
|
---|
| 344 | my ($removeold, $keepold, $incremental, $incremental_mode)
|
---|
| 345 | = &scriptutil::check_removeold_and_keepold($self->{'removeold'}, $self->{'keepold'},
|
---|
| 346 | $self->{'incremental'}, $checkdir,
|
---|
| 347 | $collectcfg);
|
---|
| 348 |
|
---|
| 349 | $self->{'removeold'} = $removeold;
|
---|
| 350 | $self->{'keepold'} = $keepold;
|
---|
| 351 | $self->{'incremental'} = $incremental;
|
---|
| 352 | $self->{'incremental_mode'} = $incremental_mode;
|
---|
[24686] | 353 |
|
---|
| 354 | # Since this wasted my morning, let's at least warn a user that manifest
|
---|
| 355 | # files now *only* work if keepold is set. [jmt12]
|
---|
| 356 | if ($self->{'manifest'} && !$self->{'keepold'})
|
---|
| 357 | {
|
---|
| 358 | print STDERR "Warning: -manifest flag should not be specified without also setting -keepold or -incremental. Ignoring.\n";
|
---|
| 359 | }
|
---|
[24626] | 360 | }
|
---|
| 361 |
|
---|
| 362 | sub process_files
|
---|
| 363 | {
|
---|
| 364 | my $self = shift @_;
|
---|
| 365 | my ($config_filename,$collectcfg) = @_;
|
---|
| 366 |
|
---|
| 367 | my $inexport_mode = $self->{'mode'};
|
---|
| 368 |
|
---|
| 369 | my $verbosity = $self->{'verbosity'};
|
---|
| 370 | my $debug = $self->{'debug'};
|
---|
| 371 |
|
---|
| 372 | my $importdir = $self->{'importdir'};
|
---|
| 373 | my $archivedir = $self->{'archivedir'} || $self->{'exportdir'};
|
---|
| 374 |
|
---|
| 375 | my $incremental = $self->{'incremental'};
|
---|
| 376 | my $incremental_mode = $self->{'incremental_mode'};
|
---|
| 377 |
|
---|
| 378 | my $removeold = $self->{'removeold'};
|
---|
| 379 | my $keepold = $self->{'keepold'};
|
---|
| 380 |
|
---|
| 381 | my $saveas = $self->{'saveas'};
|
---|
| 382 | my $OIDtype = $self->{'OIDtype'};
|
---|
| 383 | my $OIDmetadata = $self->{'OIDmetadata'};
|
---|
| 384 |
|
---|
| 385 | my $out = $self->{'out'};
|
---|
| 386 | my $faillog = $self->{'faillog'};
|
---|
| 387 |
|
---|
| 388 | my $maxdocs = $self->{'maxdocs'};
|
---|
| 389 | my $gzip = $self->{'gzip'};
|
---|
| 390 | my $groupsize = $self->{'groupsize'};
|
---|
| 391 | my $sortmeta = $self->{'sortmeta'};
|
---|
| 392 |
|
---|
| 393 | my $removeprefix = $self->{'removeprefix'};
|
---|
| 394 | my $removesuffix = $self->{'removesuffix'};
|
---|
| 395 |
|
---|
| 396 | my $gli = $self->{'gli'};
|
---|
| 397 |
|
---|
| 398 | my $jobs = $self->{'jobs'};
|
---|
| 399 | my $epoch = $self->{'epoch'};
|
---|
| 400 |
|
---|
| 401 | # related to export
|
---|
| 402 | my $xsltfile = $self->{'xsltfile'};
|
---|
| 403 | my $group_marc = $self->{'group_marc'};
|
---|
| 404 | my $mapping_file = $self->{'mapping_file'};
|
---|
| 405 | my $xslt_mets = $self->{'xslt_mets'};
|
---|
| 406 | my $xslt_txt = $self->{'xslt_txt'};
|
---|
| 407 | my $fedora_namespace = $self->{'fedora_namespace'};
|
---|
| 408 | my $metadata_prefix = $self->{'metadata_prefix'};
|
---|
| 409 |
|
---|
| 410 | if ($inexport_mode eq "import") {
|
---|
| 411 | print STDERR "<Import>\n" if $gli;
|
---|
| 412 | }
|
---|
| 413 | else {
|
---|
| 414 | print STDERR "<export>\n" if $gli;
|
---|
| 415 | }
|
---|
| 416 |
|
---|
| 417 | my $manifest_lookup = new manifest($collectcfg->{'infodbtype'},$archivedir);
|
---|
| 418 | if ($self->{'manifest'} ne "") {
|
---|
| 419 | my $manifest_filename = $self->{'manifest'};
|
---|
| 420 |
|
---|
| 421 | if (!&util::filename_is_absolute($manifest_filename)) {
|
---|
| 422 | $manifest_filename = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, $manifest_filename);
|
---|
| 423 | }
|
---|
| 424 |
|
---|
| 425 | $self->{'manifest'} =~ s/[\\\/]+/\//g;
|
---|
| 426 | $self->{'manifest'} =~ s/\/$//;
|
---|
| 427 |
|
---|
| 428 | $manifest_lookup->parse($manifest_filename);
|
---|
| 429 | }
|
---|
| 430 |
|
---|
| 431 | my $manifest = $self->{'manifest'};
|
---|
| 432 |
|
---|
| 433 | # load all the plugins
|
---|
| 434 | my $plugins = [];
|
---|
| 435 | if (defined $collectcfg->{'plugin'}) {
|
---|
| 436 | $plugins = $collectcfg->{'plugin'};
|
---|
| 437 | }
|
---|
| 438 |
|
---|
| 439 | my $plugin_incr_mode = $incremental_mode;
|
---|
| 440 | if ($manifest ne "") {
|
---|
| 441 | # if we have a manifest file, then we pretend we are fully incremental for plugins
|
---|
| 442 | $plugin_incr_mode = "all";
|
---|
| 443 | }
|
---|
| 444 | #some global options for the plugins
|
---|
| 445 | my @global_opts = ();
|
---|
| 446 |
|
---|
| 447 | my $pluginfo = &plugin::load_plugins ($plugins, $verbosity, $out, $faillog, \@global_opts, $plugin_incr_mode);
|
---|
| 448 | if (scalar(@$pluginfo) == 0) {
|
---|
| 449 | &gsprintf($out, "{import.no_plugins_loaded}\n");
|
---|
| 450 | die "\n";
|
---|
| 451 | }
|
---|
| 452 |
|
---|
| 453 | # remove the old contents of the archives directory (and tmp
|
---|
| 454 | # directory) if needed
|
---|
| 455 |
|
---|
| 456 | if ($removeold) {
|
---|
| 457 | if (-e $archivedir) {
|
---|
| 458 | &gsprintf($out, "{import.removing_archives}\n");
|
---|
| 459 | &util::rm_r ($archivedir);
|
---|
| 460 | }
|
---|
| 461 | my $tmpdir = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "tmp");
|
---|
| 462 | $tmpdir =~ s/[\\\/]+/\//g;
|
---|
| 463 | $tmpdir =~ s/\/$//;
|
---|
| 464 | if (-e $tmpdir) {
|
---|
[25116] | 465 | #&gsprintf($out, "{import.removing_tmpdir}\n");
|
---|
| 466 | #&util::rm_r ($tmpdir);
|
---|
[24626] | 467 | }
|
---|
| 468 | }
|
---|
| 469 |
|
---|
| 470 | # create the archives dir if needed
|
---|
| 471 | &util::mk_all_dir($archivedir);
|
---|
| 472 |
|
---|
| 473 | # read the archive information file
|
---|
| 474 |
|
---|
| 475 | # BACKWARDS COMPATIBILITY: Just in case there are old .ldb/.bdb files (won't do anything for other infodbtypes)
|
---|
| 476 | &util::rename_ldb_or_bdb_file(&util::filename_cat($archivedir, "archiveinf-doc"));
|
---|
| 477 | &util::rename_ldb_or_bdb_file(&util::filename_cat($archivedir, "archiveinf-src"));
|
---|
| 478 |
|
---|
[24686] | 479 | # Warning! Black magic follows. When the following functions are called on
|
---|
| 480 | # the GDBMServer class they will actually prompt the running of the Server
|
---|
| 481 | # and attach themselves as a listener (even though they don't do anything)
|
---|
| 482 | # This is done so that, in parallel importing, the server will persist
|
---|
| 483 | # until the top level import.pl (which will be the first this that calls
|
---|
| 484 | # this function) completes.
|
---|
| 485 | my $arcinfo_doc_filename = &dbutil::get_infodb_file_path($collectcfg->{'infodbtype'}, "archiveinf-doc", $archivedir, 1);
|
---|
| 486 | my $arcinfo_src_filename = &dbutil::get_infodb_file_path($collectcfg->{'infodbtype'}, "archiveinf-src", $archivedir, 1);
|
---|
| 487 |
|
---|
[24626] | 488 | my $archive_info = new arcinfo ($collectcfg->{'infodbtype'});
|
---|
| 489 | $archive_info->load_info ($arcinfo_doc_filename);
|
---|
| 490 |
|
---|
| 491 | if ($manifest eq "") {
|
---|
| 492 | # Load in list of files in import folder from last import (if present)
|
---|
| 493 | $archive_info->load_prev_import_filelist ($arcinfo_src_filename);
|
---|
| 494 | }
|
---|
| 495 |
|
---|
| 496 | ####Use Plugout####
|
---|
| 497 | my $plugout;
|
---|
| 498 |
|
---|
| 499 | if ($inexport_mode eq "import") {
|
---|
| 500 | if (defined $collectcfg->{'plugout'}) {
|
---|
| 501 | # If a plugout was specified in the collect.cfg file, assume it is sensible
|
---|
| 502 | # We can't check the name because it could be anything, if it is a custom plugout
|
---|
| 503 | $plugout = $collectcfg->{'plugout'};
|
---|
| 504 | }
|
---|
| 505 | else{
|
---|
| 506 | if ($saveas !~ /^(GreenstoneXML|GreenstoneMETS)$/) {
|
---|
| 507 | push @$plugout,"GreenstoneXMLPlugout";
|
---|
| 508 | }
|
---|
| 509 | else{
|
---|
| 510 | push @$plugout,$saveas."Plugout";
|
---|
| 511 | }
|
---|
| 512 | }
|
---|
| 513 | }
|
---|
| 514 | else {
|
---|
| 515 | if (defined $collectcfg->{'plugout'} && $collectcfg->{'plugout'} =~ /^(.*METS|DSpace|MARCXML)Plugout/) {
|
---|
| 516 | $plugout = $collectcfg->{'plugout'};
|
---|
| 517 | }
|
---|
| 518 | else{
|
---|
| 519 | if ($saveas !~ /^(GreenstoneMETS|FedoraMETS|DSpace|MARCXML)$/) {
|
---|
| 520 | push @$plugout,"GreenstoneMETSPlugout";
|
---|
| 521 | }
|
---|
| 522 | else{
|
---|
| 523 | push @$plugout,$saveas."Plugout";
|
---|
| 524 | }
|
---|
| 525 | }
|
---|
| 526 | }
|
---|
| 527 |
|
---|
| 528 | my $plugout_name = $plugout->[0];
|
---|
| 529 |
|
---|
| 530 | push @$plugout,("-output_info",$archive_info) if (defined $archive_info);
|
---|
| 531 | push @$plugout,("-verbosity",$verbosity) if (defined $verbosity);
|
---|
| 532 | push @$plugout,("-debug") if ($debug);
|
---|
| 533 | push @$plugout,("-group_size",$groupsize) if (defined $groupsize);
|
---|
| 534 | push @$plugout,("-gzip_output") if ($gzip);
|
---|
| 535 | push @$plugout,("-output_handle",$out) if (defined $out);
|
---|
| 536 |
|
---|
| 537 | push @$plugout,("-xslt_file",$xsltfile) if (defined $xsltfile && $xsltfile ne "");
|
---|
| 538 |
|
---|
| 539 | if ($plugout_name =~ m/^MARCXMLPlugout$/) {
|
---|
| 540 | push @$plugout,("-group") if ($group_marc);
|
---|
| 541 | push @$plugout,("-mapping_file",$mapping_file) if (defined $mapping_file && $mapping_file ne "");
|
---|
| 542 | }
|
---|
| 543 | if ($plugout_name =~ m/^.*METSPlugout$/) {
|
---|
| 544 | push @$plugout,("-xslt_mets",$xslt_mets) if (defined $xslt_mets && $xslt_mets ne "");
|
---|
| 545 | push @$plugout,("-xslt_txt",$xslt_txt) if (defined $xslt_txt && $xslt_txt ne "");
|
---|
| 546 | }
|
---|
| 547 |
|
---|
| 548 | if ($plugout_name eq "FedoraMETSPlugout") {
|
---|
| 549 | push @$plugout,("-fedora_namespace",$fedora_namespace) if (defined $fedora_namespace && $fedora_namespace ne "");
|
---|
| 550 | }
|
---|
| 551 |
|
---|
| 552 | if ($plugout_name eq "DSpacePlugout") {
|
---|
| 553 | push @$plugout,("-metadata_prefix",$metadata_prefix) if (defined $metadata_prefix && $metadata_prefix ne "");
|
---|
| 554 | }
|
---|
| 555 |
|
---|
| 556 | my $processor = &plugout::load_plugout($plugout);
|
---|
| 557 | $processor->setoutputdir ($archivedir);
|
---|
| 558 | $processor->set_sortmeta ($sortmeta, $removeprefix, $removesuffix) if defined $sortmeta;
|
---|
| 559 | $processor->set_OIDtype ($OIDtype, $OIDmetadata);
|
---|
| 560 |
|
---|
| 561 | &plugin::begin($pluginfo, $importdir, $processor, $maxdocs, $gli);
|
---|
| 562 |
|
---|
| 563 | if ($removeold) {
|
---|
| 564 | # occasionally, plugins may want to do something on remove
|
---|
| 565 | # old, eg pharos image indexing
|
---|
| 566 | &plugin::remove_all($pluginfo, $importdir, $processor, $maxdocs, $gli);
|
---|
| 567 | }
|
---|
| 568 |
|
---|
| 569 | # process the import directory
|
---|
| 570 | my $block_hash = {};
|
---|
| 571 | $block_hash->{'new_files'} = {};
|
---|
| 572 | $block_hash->{'reindex_files'} = {};
|
---|
[24686] | 573 | # All of these are set somewhere else, so it's kinda nice to define them
|
---|
| 574 | # here. [jmt12]
|
---|
| 575 | $block_hash->{'all_files'} = {};
|
---|
| 576 | $block_hash->{'deleted_files'} = {};
|
---|
| 577 | $block_hash->{'file_blocks'} = {};
|
---|
| 578 | $block_hash->{'metadata_files'} = {};
|
---|
| 579 | $block_hash->{'shared_fileroot'} = '';
|
---|
| 580 | # My new flag so we can tell we had a manifest way down in the plugins
|
---|
| 581 | # [jmt12]
|
---|
| 582 | $block_hash->{'manifest'} = 'false';
|
---|
[24626] | 583 | my $metadata = {};
|
---|
| 584 |
|
---|
| 585 | # global blocking pass may set up some metadata
|
---|
[24686] | 586 | # - when we have a manifest file we don't do this -unless- the collection
|
---|
| 587 | # configuration indicates this collection contains complex (inherited)
|
---|
| 588 | # metadata. [jmt12]
|
---|
| 589 | if ($manifest eq '' || (defined $collectcfg->{'complexmeta'} && $collectcfg->{'complexmeta'} eq 'true'))
|
---|
| 590 | {
|
---|
[25116] | 591 | print "Temporarily skipping global file scan as it is slow as molasses\n";
|
---|
| 592 | #&plugin::file_block_read($pluginfo, $importdir, "", $block_hash, $metadata, $gli);
|
---|
[24686] | 593 | }
|
---|
| 594 | else
|
---|
| 595 | {
|
---|
| 596 | print "Skipping global file scan due to manifest and complexmeta configuration\n";
|
---|
| 597 | }
|
---|
[24626] | 598 |
|
---|
| 599 | if ($manifest ne "") {
|
---|
[24686] | 600 |
|
---|
| 601 | $block_hash->{'manifest'} = 'true';
|
---|
| 602 |
|
---|
[24626] | 603 | #
|
---|
| 604 | # 1. Process delete files first
|
---|
| 605 | #
|
---|
| 606 | my @deleted_files = keys %{$manifest_lookup->{'delete'}};
|
---|
| 607 | my @full_deleted_files = ();
|
---|
| 608 |
|
---|
| 609 | # ensure all filenames are absolute
|
---|
| 610 | foreach my $df (@deleted_files) {
|
---|
| 611 | my $full_df =
|
---|
| 612 | (&util::filename_is_absolute($df))
|
---|
| 613 | ? $df
|
---|
| 614 | : &util::filename_cat($importdir,$df);
|
---|
| 615 |
|
---|
| 616 | if (-d $full_df) {
|
---|
| 617 | &add_dir_contents_to_list($full_df, \@full_deleted_files);
|
---|
| 618 | } else {
|
---|
| 619 | push(@full_deleted_files,$full_df);
|
---|
| 620 | }
|
---|
| 621 | }
|
---|
| 622 |
|
---|
| 623 | &plugin::remove_some($pluginfo, $collectcfg->{'infodbtype'}, $archivedir, \@full_deleted_files);
|
---|
| 624 | mark_docs_for_deletion($archive_info,{},
|
---|
| 625 | \@full_deleted_files,
|
---|
| 626 | $archivedir, $verbosity, "delete");
|
---|
| 627 |
|
---|
| 628 |
|
---|
| 629 | #
|
---|
| 630 | # 2. Now files for reindexing
|
---|
| 631 | #
|
---|
| 632 |
|
---|
| 633 | my @reindex_files = keys %{$manifest_lookup->{'reindex'}};
|
---|
| 634 | my @full_reindex_files = ();
|
---|
| 635 | # ensure all filenames are absolute
|
---|
| 636 | foreach my $rf (@reindex_files) {
|
---|
| 637 | my $full_rf =
|
---|
| 638 | (&util::filename_is_absolute($rf))
|
---|
| 639 | ? $rf
|
---|
| 640 | : &util::filename_cat($importdir,$rf);
|
---|
| 641 |
|
---|
| 642 | if (-d $full_rf) {
|
---|
| 643 | &add_dir_contents_to_list($full_rf, \@full_reindex_files);
|
---|
| 644 | } else {
|
---|
| 645 | push(@full_reindex_files,$full_rf);
|
---|
| 646 | }
|
---|
| 647 | }
|
---|
| 648 |
|
---|
| 649 | &plugin::remove_some($pluginfo, $collectcfg->{'infodbtype'}, $archivedir, \@full_reindex_files);
|
---|
| 650 | mark_docs_for_deletion($archive_info,{},\@full_reindex_files, $archivedir,$verbosity, "reindex");
|
---|
| 651 |
|
---|
| 652 | # And now to ensure the new version of the file processed by
|
---|
| 653 | # appropriate plugin, we need to add it to block_hash reindex list
|
---|
| 654 | foreach my $full_rf (@full_reindex_files) {
|
---|
| 655 | $block_hash->{'reindex_files'}->{$full_rf} = 1;
|
---|
| 656 | }
|
---|
| 657 |
|
---|
| 658 |
|
---|
| 659 | #
|
---|
| 660 | # 3. Now finally any new files - add to block_hash new_files list
|
---|
| 661 | #
|
---|
| 662 |
|
---|
| 663 | my @new_files = keys %{$manifest_lookup->{'index'}};
|
---|
| 664 | my @full_new_files = ();
|
---|
| 665 |
|
---|
| 666 | foreach my $nf (@new_files) {
|
---|
| 667 | # ensure filename is absolute
|
---|
| 668 | my $full_nf =
|
---|
| 669 | (&util::filename_is_absolute($nf))
|
---|
| 670 | ? $nf
|
---|
| 671 | : &util::filename_cat($importdir,$nf);
|
---|
| 672 |
|
---|
| 673 | if (-d $full_nf) {
|
---|
| 674 | &add_dir_contents_to_list($full_nf, \@full_new_files);
|
---|
| 675 | } else {
|
---|
| 676 | push(@full_new_files,$full_nf);
|
---|
| 677 | }
|
---|
| 678 | }
|
---|
| 679 |
|
---|
| 680 | my $arcinfo_src_filename = &dbutil::get_infodb_file_path($collectcfg->{'infodbtype'}, "archiveinf-src", $archivedir);
|
---|
[24686] | 681 | # need to check this file exists before trying to read it. [jmt12]
|
---|
| 682 | if (-e $arcinfo_src_filename)
|
---|
| 683 | {
|
---|
[24626] | 684 | my $arcinfodb_map = {};
|
---|
| 685 | &dbutil::read_infodb_file($collectcfg->{'infodbtype'}, $arcinfo_src_filename, $arcinfodb_map);
|
---|
| 686 | foreach my $f (@full_new_files) {
|
---|
| 687 | # check that we haven't seen it already
|
---|
| 688 | if (defined $arcinfodb_map->{$f}) {
|
---|
| 689 | # TODO make better warning
|
---|
| 690 | print STDERR "Warning: $f already in src archive, \n";
|
---|
| 691 | } else {
|
---|
| 692 | $block_hash->{'new_files'}->{$f} = 1;
|
---|
| 693 | }
|
---|
| 694 | }
|
---|
[24686] | 695 | undef $arcinfodb_map;
|
---|
| 696 | }
|
---|
| 697 | # no existing files - so we can just add all the ones that need adding.
|
---|
| 698 | # [jmt12]
|
---|
| 699 | else
|
---|
| 700 | {
|
---|
| 701 | foreach my $f (@full_new_files)
|
---|
| 702 | {
|
---|
| 703 | $block_hash->{'new_files'}->{$f} = 1;
|
---|
| 704 | }
|
---|
| 705 | }
|
---|
[24626] | 706 |
|
---|
[24686] | 707 | # If we are not using complex inherited metadata (and thus have skipped
|
---|
| 708 | # the global file scan) we need to at least scan the directory of the
|
---|
| 709 | # files being indexed/reindexed. [jmt12]
|
---|
[24697] | 710 | if (!defined $collectcfg->{'complexmeta'} || $collectcfg->{'complexmeta'} ne 'true')
|
---|
[24686] | 711 | {
|
---|
| 712 | my @all_files_to_import = (keys %{$block_hash->{'reindex_files'}}, keys %{$block_hash->{'new_files'}});
|
---|
| 713 | foreach my $file_to_import (@all_files_to_import)
|
---|
| 714 | {
|
---|
| 715 | my $dir_to_import = $file_to_import;
|
---|
| 716 | $dir_to_import =~ s/[^\\\/]*$//;
|
---|
| 717 | # - one day we may need to manually scan this directory for child
|
---|
| 718 | # directories and somehow explicitly block them from being
|
---|
| 719 | # recursed.
|
---|
| 720 | if (-d $dir_to_import)
|
---|
| 721 | {
|
---|
| 722 | &plugin::file_block_read($pluginfo, $dir_to_import, '', $block_hash, $metadata, $gli);
|
---|
| 723 | }
|
---|
| 724 | }
|
---|
| 725 | }
|
---|
[24626] | 726 | }
|
---|
| 727 | else {
|
---|
| 728 | # if incremental, we read through the import folder to see whats changed.
|
---|
| 729 |
|
---|
| 730 | if ($incremental || $incremental_mode eq "onlyadd") {
|
---|
| 731 | prime_doc_oid_count($archivedir);
|
---|
| 732 |
|
---|
| 733 | # Can now work out which files were new, already existed, and have
|
---|
| 734 | # been deleted
|
---|
| 735 |
|
---|
| 736 | new_vs_old_import_diff($archive_info,$block_hash,$importdir,
|
---|
| 737 | $archivedir,$verbosity,$incremental_mode);
|
---|
| 738 |
|
---|
| 739 | my @new_files = sort keys %{$block_hash->{'new_files'}};
|
---|
| 740 | if (scalar(@new_files>0)) {
|
---|
| 741 | print STDERR "New files and modified metadata files since last import:\n ";
|
---|
| 742 | print STDERR join("\n ",@new_files), "\n";
|
---|
| 743 | }
|
---|
| 744 |
|
---|
| 745 | if ($incremental) {
|
---|
| 746 | # only look for deletions if we are truely incremental
|
---|
| 747 | my @deleted_files = sort keys %{$block_hash->{'deleted_files'}};
|
---|
| 748 | # Filter out any in gsdl/tmp area
|
---|
| 749 | my @filtered_deleted_files = ();
|
---|
| 750 | my $gsdl_tmp_area = &util::filename_cat($ENV{'GSDLHOME'}, "tmp");
|
---|
| 751 | my $collect_tmp_area = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "tmp");
|
---|
| 752 | $gsdl_tmp_area = &util::filename_to_regex($gsdl_tmp_area);
|
---|
| 753 | $collect_tmp_area = &util::filename_to_regex($collect_tmp_area);
|
---|
| 754 |
|
---|
| 755 | foreach my $df (@deleted_files) {
|
---|
| 756 | next if ($df =~ m/^$gsdl_tmp_area/);
|
---|
| 757 | next if ($df =~ m/^$collect_tmp_area/);
|
---|
| 758 |
|
---|
| 759 | push(@filtered_deleted_files,$df);
|
---|
| 760 | }
|
---|
| 761 |
|
---|
| 762 |
|
---|
| 763 | @deleted_files = @filtered_deleted_files;
|
---|
| 764 |
|
---|
| 765 | if (scalar(@deleted_files)>0) {
|
---|
| 766 | print STDERR "Files deleted since last import:\n ";
|
---|
| 767 | print STDERR join("\n ",@deleted_files), "\n";
|
---|
| 768 |
|
---|
| 769 |
|
---|
| 770 | &plugin::remove_some($pluginfo, $collectcfg->{'infodbtype'}, $archivedir, \@deleted_files);
|
---|
| 771 |
|
---|
| 772 | mark_docs_for_deletion($archive_info,$block_hash,\@deleted_files, $archivedir,$verbosity, "delete");
|
---|
| 773 | }
|
---|
| 774 |
|
---|
| 775 | my @reindex_files = sort keys %{$block_hash->{'reindex_files'}};
|
---|
| 776 |
|
---|
| 777 | if (scalar(@reindex_files)>0) {
|
---|
| 778 | print STDERR "Files to reindex since last import:\n ";
|
---|
| 779 | print STDERR join("\n ",@reindex_files), "\n";
|
---|
| 780 | &plugin::remove_some($pluginfo, $collectcfg->{'infodbtype'}, $archivedir, \@reindex_files);
|
---|
| 781 | mark_docs_for_deletion($archive_info,$block_hash,\@reindex_files, $archivedir,$verbosity, "reindex");
|
---|
| 782 | }
|
---|
| 783 |
|
---|
| 784 | }
|
---|
| 785 | }
|
---|
| 786 | }
|
---|
| 787 |
|
---|
| 788 | # Check for existence of the file that's to contain earliestDateStamp in archivesdir
|
---|
| 789 | # Do nothing if the file already exists (file exists on incremental build).
|
---|
| 790 | # If the file doesn't exist, as happens on full build, create it and write out the current datestamp into it
|
---|
| 791 | # In buildcol, read the file's contents and set the earliestdateStamp in GS2's build.cfg / GS3's buildconfig.xml
|
---|
| 792 | # In doc.pm have set_oaiLastModified similar to set_lastmodified, and create the doc fields
|
---|
| 793 | # oailastmodified and oailastmodifieddate
|
---|
| 794 | my $earliestDatestampFile = &util::filename_cat($archivedir, "earliestDatestamp");
|
---|
| 795 | if (!-f $earliestDatestampFile && -d $archivedir) {
|
---|
| 796 | my $current_time_in_seconds = time; # in seconds
|
---|
| 797 |
|
---|
| 798 | if(open(FOUT, ">$earliestDatestampFile")) {
|
---|
| 799 | # || (&gsprintf(STDERR, "{common.cannot_open}: $!\n", $earliestDatestampFile) && die);
|
---|
| 800 | print FOUT $current_time_in_seconds;
|
---|
| 801 | close(FOUT);
|
---|
| 802 | }
|
---|
| 803 | else {
|
---|
| 804 | &gsprintf(STDERR, "{import.cannot_write_earliestdatestamp}\n", $earliestDatestampFile);
|
---|
| 805 | }
|
---|
| 806 |
|
---|
| 807 | }
|
---|
| 808 |
|
---|
| 809 | # now, whichever mode we are in, we can process the entire import folder
|
---|
| 810 | if ((defined $jobs) && ($jobs > 1))
|
---|
| 811 | {
|
---|
| 812 | # if jobs are set to >1, run in parallel using MPI helper
|
---|
| 813 | # [hs, 1 july 2010]
|
---|
| 814 | &ParallelInexport::farm_out_processes($jobs, $epoch, $importdir, $block_hash,
|
---|
| 815 | $self->{'collection'}, $self->{'site'});
|
---|
| 816 | }
|
---|
| 817 | else
|
---|
| 818 | {
|
---|
| 819 | &plugin::read ($pluginfo, $importdir, "", $block_hash, $metadata, $processor, $maxdocs, 0, $gli);
|
---|
| 820 | }
|
---|
| 821 |
|
---|
| 822 |
|
---|
| 823 | if ($saveas eq "FedoraMETS") {
|
---|
| 824 | # create collection "doc obj" for Fedora that contains
|
---|
| 825 | # collection-level metadata
|
---|
| 826 |
|
---|
| 827 | my $doc_obj = new doc($config_filename,"nonindexed_doc","none");
|
---|
| 828 | $doc_obj->set_OID("collection");
|
---|
| 829 |
|
---|
| 830 | my $col_name = undef;
|
---|
| 831 | my $col_meta = $collectcfg->{'collectionmeta'};
|
---|
| 832 |
|
---|
| 833 | if (defined $col_meta) {
|
---|
| 834 | store_collectionmeta($col_meta,"collectionname",$doc_obj); # in GS3 this is a collection's name
|
---|
| 835 | store_collectionmeta($col_meta,"collectionextra",$doc_obj); # in GS3 this is a collection's description
|
---|
| 836 | }
|
---|
| 837 | $processor->process($doc_obj);
|
---|
| 838 | }
|
---|
| 839 |
|
---|
| 840 | &plugin::end($pluginfo, $processor);
|
---|
| 841 |
|
---|
| 842 | &plugin::deinit($pluginfo, $processor);
|
---|
| 843 |
|
---|
| 844 | # Store the value of OIDCount (used in doc.pm) so it can be
|
---|
| 845 | # restored correctly to this value on an incremental build
|
---|
| 846 | store_doc_oid_count($archivedir);
|
---|
| 847 |
|
---|
| 848 | # write out the archive information file
|
---|
| 849 | $processor->close_file_output() if (defined $groupsize) && ($groupsize > 1);
|
---|
| 850 | $processor->close_group_output() if $processor->is_group();
|
---|
| 851 |
|
---|
| 852 | # for backwards compatability with archvies.inf file
|
---|
| 853 | if ($arcinfo_doc_filename =~ m/(contents)|(\.inf)$/) {
|
---|
| 854 | $archive_info->save_info($arcinfo_doc_filename);
|
---|
| 855 | }
|
---|
| 856 | else {
|
---|
| 857 | $archive_info->save_revinfo_db($arcinfo_src_filename);
|
---|
| 858 | }
|
---|
| 859 |
|
---|
| 860 | return $pluginfo;
|
---|
| 861 | }
|
---|
| 862 |
|
---|
| 863 |
|
---|
| 864 | sub generate_statistics
|
---|
| 865 | {
|
---|
| 866 | my $self = shift @_;
|
---|
| 867 | my ($pluginfo) = @_;
|
---|
| 868 |
|
---|
| 869 | my $inexport_mode = $self->{'mode'};
|
---|
| 870 |
|
---|
| 871 | my $statsfile = $self->{'statsfile'};
|
---|
| 872 | my $out = $self->{'out'};
|
---|
| 873 | my $faillogname = $self->{'faillogname'};
|
---|
| 874 | my $gli = $self->{'gli'};
|
---|
| 875 | my $jobs = $self->{'jobs'};
|
---|
| 876 |
|
---|
| 877 | # write out import stats
|
---|
| 878 |
|
---|
| 879 | if ((!defined $jobs) || ($jobs == 1))
|
---|
| 880 | {
|
---|
| 881 | # only output statistics if there are multiple jobs
|
---|
| 882 | # [hs, 1 july 2010]
|
---|
| 883 |
|
---|
| 884 | my $close_stats = 0;
|
---|
| 885 | if ($statsfile !~ /^(STDERR|STDOUT)$/i) {
|
---|
| 886 | if (open (STATS, ">$statsfile")) {
|
---|
| 887 | $statsfile = 'inexport::STATS';
|
---|
| 888 | $close_stats = 1;
|
---|
| 889 | } else {
|
---|
| 890 | &gsprintf($out, "{import.cannot_open_stats_file}", $statsfile);
|
---|
| 891 | &gsprintf($out, "{import.stats_backup}\n");
|
---|
| 892 | $statsfile = 'STDERR';
|
---|
| 893 | }
|
---|
| 894 | }
|
---|
| 895 |
|
---|
| 896 | &gsprintf($out, "\n");
|
---|
| 897 | &gsprintf($out, "*********************************************\n");
|
---|
| 898 | &gsprintf($out, "{$inexport_mode.complete}\n");
|
---|
| 899 | &gsprintf($out, "*********************************************\n");
|
---|
| 900 |
|
---|
| 901 | &plugin::write_stats($pluginfo, $statsfile, $faillogname, $gli);
|
---|
| 902 | if ($close_stats) {
|
---|
| 903 | close STATS;
|
---|
| 904 | }
|
---|
| 905 | }
|
---|
| 906 |
|
---|
| 907 | close OUT if $self->{'close_out'};
|
---|
| 908 | close FAILLOG;
|
---|
| 909 | }
|
---|
| 910 |
|
---|
| 911 |
|
---|
| 912 | sub store_collectionmeta
|
---|
| 913 | {
|
---|
| 914 | my ($collectionmeta,$field,$doc_obj) = @_;
|
---|
| 915 |
|
---|
| 916 | my $section = $doc_obj->get_top_section();
|
---|
| 917 |
|
---|
| 918 | my $field_hash = $collectionmeta->{$field};
|
---|
| 919 |
|
---|
| 920 | foreach my $k (keys %$field_hash)
|
---|
| 921 | {
|
---|
| 922 | my $val = $field_hash->{$k};
|
---|
| 923 |
|
---|
| 924 | ### print STDERR "*** $k = $field_hash->{$k}\n";
|
---|
| 925 |
|
---|
| 926 | my $md_label = "ex.$field";
|
---|
| 927 |
|
---|
| 928 |
|
---|
| 929 | if ($k =~ m/^\[l=(.*?)\]$/)
|
---|
| 930 | {
|
---|
| 931 |
|
---|
| 932 | my $md_suffix = $1;
|
---|
| 933 | $md_label .= "^$md_suffix";
|
---|
| 934 | }
|
---|
| 935 |
|
---|
| 936 |
|
---|
| 937 | $doc_obj->add_utf8_metadata($section,$md_label, $val);
|
---|
| 938 |
|
---|
| 939 | # see collConfigxml.pm: GS2's "collectionextra" is called "description" in GS3,
|
---|
| 940 | # while "collectionname" in GS2 is called "name" in GS3.
|
---|
| 941 | # Variable $nameMap variable in collConfigxml.pm maps between GS2 and GS3
|
---|
| 942 | if (($md_label eq "ex.collectionname^en") || ($md_label eq "ex.collectionname"))
|
---|
| 943 | {
|
---|
| 944 | $doc_obj->add_utf8_metadata($section,"dc.Title", $val);
|
---|
| 945 | }
|
---|
| 946 |
|
---|
| 947 | }
|
---|
| 948 | }
|
---|
| 949 |
|
---|
| 950 |
|
---|
| 951 | sub oid_count_file {
|
---|
| 952 | my ($archivedir) = @_;
|
---|
| 953 | return &util::filename_cat ($archivedir, "OIDcount");
|
---|
| 954 | }
|
---|
| 955 |
|
---|
| 956 |
|
---|
| 957 | sub prime_doc_oid_count
|
---|
| 958 | {
|
---|
| 959 | my ($archivedir) = @_;
|
---|
| 960 | my $oid_count_filename = &oid_count_file($archivedir);
|
---|
| 961 |
|
---|
| 962 | if (-e $oid_count_filename) {
|
---|
| 963 | if (open(OIDIN,"<$oid_count_filename")) {
|
---|
| 964 | my $OIDcount = <OIDIN>;
|
---|
| 965 | chomp $OIDcount;
|
---|
| 966 | close(OIDIN);
|
---|
| 967 |
|
---|
| 968 | $doc::OIDcount = $OIDcount;
|
---|
| 969 | }
|
---|
| 970 | else {
|
---|
| 971 | &gsprintf(STDERR, "{import.cannot_read_OIDcount}\n", $oid_count_filename);
|
---|
| 972 | }
|
---|
| 973 | }
|
---|
| 974 |
|
---|
| 975 | }
|
---|
| 976 |
|
---|
| 977 | sub store_doc_oid_count
|
---|
| 978 | {
|
---|
| 979 | # Use the file "OIDcount" in the archives directory to record
|
---|
| 980 | # what value doc.pm got up to
|
---|
| 981 |
|
---|
| 982 | my ($archivedir) = @_;
|
---|
| 983 | my $oid_count_filename = &oid_count_file($archivedir);
|
---|
| 984 |
|
---|
| 985 |
|
---|
| 986 | if (open(OIDOUT,">$oid_count_filename")) {
|
---|
| 987 | print OIDOUT $doc::OIDcount, "\n";
|
---|
| 988 |
|
---|
| 989 | close(OIDOUT);
|
---|
| 990 | }
|
---|
| 991 | else {
|
---|
| 992 | &gsprintf(STDERR, "{import.cannot_write_OIDcount}\n", $oid_count_filename);
|
---|
| 993 | }
|
---|
| 994 | }
|
---|
| 995 |
|
---|
| 996 |
|
---|
| 997 |
|
---|
| 998 | sub new_vs_old_import_diff
|
---|
| 999 | {
|
---|
| 1000 | my ($archive_info,$block_hash,$importdir,$archivedir,$verbosity,$incremental_mode) = @_;
|
---|
| 1001 |
|
---|
| 1002 | # Get the infodbtype value for this collection from the arcinfo object
|
---|
| 1003 | my $infodbtype = $archive_info->{'infodbtype'};
|
---|
| 1004 |
|
---|
| 1005 | # in this method, we want to know if metadata files are modified or not.
|
---|
| 1006 | my $arcinfo_doc_filename = &dbutil::get_infodb_file_path($infodbtype, "archiveinf-doc", $archivedir);
|
---|
| 1007 |
|
---|
| 1008 | my $archiveinf_timestamp = -M $arcinfo_doc_filename;
|
---|
| 1009 |
|
---|
| 1010 | # First convert all files to absolute form
|
---|
| 1011 | # This is to support the situation where the import folder is not
|
---|
| 1012 | # the default
|
---|
| 1013 |
|
---|
| 1014 | my $prev_all_files = $archive_info->{'prev_import_filelist'};
|
---|
| 1015 | my $full_prev_all_files = {};
|
---|
| 1016 |
|
---|
| 1017 | foreach my $prev_file (keys %$prev_all_files) {
|
---|
| 1018 |
|
---|
| 1019 | if (!&util::filename_is_absolute($prev_file)) {
|
---|
| 1020 | my $full_prev_file = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},$prev_file);
|
---|
| 1021 | $full_prev_all_files->{$full_prev_file} = $prev_file;
|
---|
| 1022 | }
|
---|
| 1023 | else {
|
---|
| 1024 | $full_prev_all_files->{$prev_file} = $prev_file;
|
---|
| 1025 | }
|
---|
| 1026 | }
|
---|
| 1027 |
|
---|
| 1028 |
|
---|
| 1029 | # Figure out which are the new files, existing files and so
|
---|
| 1030 | # by implication the files from the previous import that are not
|
---|
| 1031 | # there any more => mark them for deletion
|
---|
| 1032 | foreach my $curr_file (keys %{$block_hash->{'all_files'}}) {
|
---|
| 1033 |
|
---|
| 1034 | my $full_curr_file = $curr_file;
|
---|
| 1035 |
|
---|
| 1036 | # entry in 'all_files' is moved to either 'existing_files',
|
---|
| 1037 | # 'deleted_files', 'new_files', or 'new_or_modified_metadata_files'
|
---|
| 1038 |
|
---|
| 1039 | if (!&util::filename_is_absolute($curr_file)) {
|
---|
| 1040 | # add in import dir to make absolute
|
---|
| 1041 | $full_curr_file = &util::filename_cat($importdir,$curr_file);
|
---|
| 1042 | }
|
---|
| 1043 |
|
---|
| 1044 | # figure out if new file or not
|
---|
| 1045 | if (defined $full_prev_all_files->{$full_curr_file}) {
|
---|
| 1046 | # delete it so that only files that need deleting are left
|
---|
| 1047 | delete $full_prev_all_files->{$full_curr_file};
|
---|
| 1048 |
|
---|
| 1049 | # had it before. is it a metadata file?
|
---|
| 1050 | if ($block_hash->{'metadata_files'}->{$full_curr_file}) {
|
---|
| 1051 |
|
---|
| 1052 | # is it modified??
|
---|
| 1053 | if (-M $full_curr_file < $archiveinf_timestamp) {
|
---|
| 1054 | print STDERR "*** Detected a *modified metadata* file: $full_curr_file\n" if $verbosity >= 2;
|
---|
| 1055 | # its newer than last build
|
---|
| 1056 | $block_hash->{'new_or_modified_metadata_files'}->{$full_curr_file} = 1;
|
---|
| 1057 | }
|
---|
| 1058 | }
|
---|
| 1059 | else {
|
---|
| 1060 | if ($incremental_mode eq "all") {
|
---|
| 1061 |
|
---|
| 1062 | # had it before
|
---|
| 1063 | $block_hash->{'existing_files'}->{$full_curr_file} = 1;
|
---|
| 1064 |
|
---|
| 1065 | }
|
---|
| 1066 | else {
|
---|
| 1067 | # Warning in "onlyadd" mode, but had it before!
|
---|
| 1068 | print STDERR "Warning: File $full_curr_file previously imported.\n";
|
---|
| 1069 | print STDERR " Treating as new file\n";
|
---|
| 1070 |
|
---|
| 1071 | $block_hash->{'new_files'}->{$full_curr_file} = 1;
|
---|
| 1072 |
|
---|
| 1073 | }
|
---|
| 1074 | }
|
---|
| 1075 | }
|
---|
| 1076 | else {
|
---|
| 1077 | if ($block_hash->{'metadata_files'}->{$full_curr_file}) {
|
---|
| 1078 | # the new file is the special sort of file greenstone uses
|
---|
| 1079 | # to attach metadata to src documents
|
---|
| 1080 | # i.e metadata.xml
|
---|
| 1081 | # (but note, the filename used is not constrained in
|
---|
| 1082 | # Greenstone to always be this)
|
---|
| 1083 |
|
---|
| 1084 | print STDERR "*** Detected *new* metadata file: $full_curr_file\n" if $verbosity >= 2;
|
---|
| 1085 | $block_hash->{'new_or_modified_metadata_files'}->{$full_curr_file} = 1;
|
---|
| 1086 | }
|
---|
| 1087 | else {
|
---|
| 1088 | $block_hash->{'new_files'}->{$full_curr_file} = 1;
|
---|
| 1089 | }
|
---|
| 1090 | }
|
---|
| 1091 |
|
---|
| 1092 |
|
---|
| 1093 | delete $block_hash->{'all_files'}->{$curr_file};
|
---|
| 1094 | }
|
---|
| 1095 |
|
---|
| 1096 |
|
---|
| 1097 |
|
---|
| 1098 |
|
---|
| 1099 | # Deal with complication of new or modified metadata files by forcing
|
---|
| 1100 | # everything from this point down in the file hierarchy to
|
---|
| 1101 | # be freshly imported.
|
---|
| 1102 | #
|
---|
| 1103 | # This may mean files that have not changed are reindexed, but does
|
---|
| 1104 | # guarantee by the end of processing all new metadata is correctly
|
---|
| 1105 | # associated with the relevant document(s).
|
---|
| 1106 |
|
---|
| 1107 | foreach my $new_mdf (keys %{$block_hash->{'new_or_modified_metadata_files'}}) {
|
---|
| 1108 | my ($fileroot,$situated_dir,$ext) = fileparse($new_mdf, "\\.[^\\.]+\$");
|
---|
| 1109 |
|
---|
| 1110 | $situated_dir =~ s/[\\\/]+$//; # remove tailing slashes
|
---|
| 1111 | $situated_dir =~ s/\\/\\\\/g; # need to protect windows slash \ in regular expression
|
---|
| 1112 |
|
---|
| 1113 | # Go through existing_files, and mark anything that is contained
|
---|
| 1114 | # within 'situated_dir' to be reindexed (in case some of the metadata
|
---|
| 1115 | # attaches to one of these files)
|
---|
| 1116 |
|
---|
| 1117 | my $reindex_files = [];
|
---|
| 1118 |
|
---|
| 1119 | foreach my $existing_f (keys %{$block_hash->{'existing_files'}}) {
|
---|
| 1120 |
|
---|
| 1121 | if ($existing_f =~ m/^$situated_dir/) {
|
---|
| 1122 |
|
---|
| 1123 | print STDERR "**** Existing file $existing_f\nis located within\n$situated_dir\n";
|
---|
| 1124 |
|
---|
| 1125 | push(@$reindex_files,$existing_f);
|
---|
| 1126 | $block_hash->{'reindex_files'}->{$existing_f} = 1;
|
---|
| 1127 | delete $block_hash->{'existing_files'}->{$existing_f};
|
---|
| 1128 |
|
---|
| 1129 | }
|
---|
| 1130 | }
|
---|
| 1131 |
|
---|
| 1132 | # metadata file needs to be in new_files list so parsed by MetadataXMLPlug
|
---|
| 1133 | # (or equivalent)
|
---|
| 1134 | $block_hash->{'new_files'}->{$new_mdf} = 1;
|
---|
| 1135 |
|
---|
| 1136 | }
|
---|
| 1137 |
|
---|
| 1138 | # go through remaining existing files and work out what has changed and needs to be reindexed.
|
---|
| 1139 | my @existing_files = sort keys %{$block_hash->{'existing_files'}};
|
---|
| 1140 |
|
---|
| 1141 | my $reindex_files = [];
|
---|
| 1142 |
|
---|
| 1143 | foreach my $existing_filename (@existing_files) {
|
---|
| 1144 | if (-M $existing_filename < $archiveinf_timestamp) {
|
---|
| 1145 | # file is newer than last build
|
---|
| 1146 |
|
---|
| 1147 | my $existing_file = $existing_filename;
|
---|
| 1148 | #my $collectdir = &util::filename_cat($ENV{'GSDLCOLLECTDIR'});
|
---|
| 1149 |
|
---|
| 1150 | #my $collectdir_resafe = &util::filename_to_regex($collectdir);
|
---|
| 1151 | #$existing_file =~ s/^$collectdir_resafe(\\|\/)?//;
|
---|
| 1152 |
|
---|
| 1153 | print STDERR "**** Reindexing existing file: $existing_file\n";
|
---|
| 1154 |
|
---|
| 1155 | push(@$reindex_files,$existing_file);
|
---|
| 1156 | $block_hash->{'reindex_files'}->{$existing_filename} = 1;
|
---|
| 1157 | }
|
---|
| 1158 |
|
---|
| 1159 | }
|
---|
| 1160 |
|
---|
| 1161 |
|
---|
| 1162 | # By this point full_prev_all_files contains the files
|
---|
| 1163 | # mentioned in archiveinf-src.db but are not in the 'import'
|
---|
| 1164 | # folder (or whatever was specified through -importdir ...)
|
---|
| 1165 |
|
---|
| 1166 | # This list can contain files that were created in the 'tmp' or
|
---|
| 1167 | # 'cache' areas (such as screen-size and thumbnail images).
|
---|
| 1168 | #
|
---|
| 1169 | # In building the final list of files to delete, we test to see if
|
---|
| 1170 | # it exists on the filesystem and if it does (unusual for a "normal"
|
---|
| 1171 | # file in import, but possible in the case of 'tmp' files),
|
---|
| 1172 | # supress it from going into the final list
|
---|
| 1173 |
|
---|
| 1174 | my $collectdir = $ENV{'GSDLCOLLECTDIR'};
|
---|
| 1175 |
|
---|
| 1176 | my @deleted_files = values %$full_prev_all_files;
|
---|
| 1177 | map { my $curr_file = $_;
|
---|
| 1178 | my $full_curr_file = $curr_file;
|
---|
| 1179 |
|
---|
| 1180 | if (!&util::filename_is_absolute($curr_file)) {
|
---|
| 1181 | # add in import dir to make absolute
|
---|
| 1182 |
|
---|
| 1183 | $full_curr_file = &util::filename_cat($collectdir,$curr_file);
|
---|
| 1184 | }
|
---|
| 1185 |
|
---|
| 1186 |
|
---|
| 1187 | if (!-e $full_curr_file) {
|
---|
| 1188 | $block_hash->{'deleted_files'}->{$curr_file} = 1;
|
---|
| 1189 | }
|
---|
| 1190 | } @deleted_files;
|
---|
| 1191 |
|
---|
| 1192 |
|
---|
| 1193 |
|
---|
| 1194 | }
|
---|
| 1195 |
|
---|
| 1196 |
|
---|
| 1197 | # this is used to delete "deleted" docs, and to remove old versions of "changed" docs
|
---|
| 1198 | # $mode is 'delete' or 'reindex'
|
---|
| 1199 | sub mark_docs_for_deletion
|
---|
| 1200 | {
|
---|
| 1201 | my ($archive_info,$block_hash,$deleted_files,$archivedir,$verbosity,$mode) = @_;
|
---|
| 1202 |
|
---|
| 1203 | my $mode_text = "deleted from index";
|
---|
| 1204 | if ($mode eq "reindex") {
|
---|
| 1205 | $mode_text = "reindexed";
|
---|
| 1206 | }
|
---|
| 1207 |
|
---|
| 1208 | # Get the infodbtype value for this collection from the arcinfo object
|
---|
| 1209 | my $infodbtype = $archive_info->{'infodbtype'};
|
---|
| 1210 |
|
---|
| 1211 | my $arcinfo_doc_filename = &dbutil::get_infodb_file_path($infodbtype, "archiveinf-doc", $archivedir);
|
---|
| 1212 | my $arcinfo_src_filename = &dbutil::get_infodb_file_path($infodbtype, "archiveinf-src", $archivedir);
|
---|
| 1213 |
|
---|
| 1214 |
|
---|
| 1215 | # record files marked for deletion in arcinfo
|
---|
| 1216 | foreach my $file (@$deleted_files) {
|
---|
| 1217 | # use 'archiveinf-src' info database file to look up all the OIDs
|
---|
| 1218 | # that this file is used in (note in most cases, it's just one OID)
|
---|
| 1219 |
|
---|
| 1220 | my $src_rec = &dbutil::read_infodb_entry($infodbtype, $arcinfo_src_filename, $file);
|
---|
| 1221 | my $oids = $src_rec->{'oid'};
|
---|
| 1222 | my $file_record_deleted = 0;
|
---|
| 1223 |
|
---|
| 1224 | # delete the src record
|
---|
| 1225 | my $src_infodb_file_handle = &dbutil::open_infodb_write_handle($infodbtype, $arcinfo_src_filename, "append");
|
---|
| 1226 | &dbutil::delete_infodb_entry($infodbtype, $src_infodb_file_handle, $file);
|
---|
| 1227 | &dbutil::close_infodb_write_handle($infodbtype, $src_infodb_file_handle);
|
---|
| 1228 |
|
---|
| 1229 |
|
---|
| 1230 | foreach my $oid (@$oids) {
|
---|
| 1231 |
|
---|
| 1232 | # find the source doc (the primary file that becomes this oid)
|
---|
| 1233 | my $doc_rec = &dbutil::read_infodb_entry($infodbtype, $arcinfo_doc_filename, $oid);
|
---|
| 1234 | my $doc_source_file = $doc_rec->{'src-file'}->[0];
|
---|
| 1235 | if (!&util::filename_is_absolute($doc_source_file)) {
|
---|
| 1236 | $doc_source_file = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},$doc_source_file);
|
---|
| 1237 | }
|
---|
| 1238 |
|
---|
| 1239 | if ($doc_source_file ne $file) {
|
---|
| 1240 | # its an associated or metadata file
|
---|
| 1241 |
|
---|
| 1242 | # mark source doc for reimport as one of its assoc files has changed or deleted
|
---|
| 1243 | $block_hash->{'reindex_files'}->{$doc_source_file} = 1;
|
---|
| 1244 |
|
---|
| 1245 | }
|
---|
| 1246 | my $curr_status = $archive_info->get_status_info($oid);
|
---|
| 1247 | if (defined($curr_status) && (($curr_status ne "D"))) {
|
---|
| 1248 | if ($verbosity>1) {
|
---|
| 1249 | print STDERR "$oid ($doc_source_file) marked to be $mode_text on next buildcol.pl\n";
|
---|
| 1250 | }
|
---|
| 1251 | # mark oid for deletion (it will be deleted or reimported)
|
---|
| 1252 | $archive_info->set_status_info($oid,"D");
|
---|
| 1253 | my $val = &dbutil::read_infodb_rawentry($infodbtype, $arcinfo_doc_filename, $oid);
|
---|
| 1254 | $val =~ s/^<index-status>(.*)$/<index-status>D/m;
|
---|
| 1255 |
|
---|
| 1256 | my $val_rec = &dbutil::convert_infodb_string_to_hash($val);
|
---|
| 1257 | my $doc_infodb_file_handle = &dbutil::open_infodb_write_handle($infodbtype, $arcinfo_doc_filename, "append");
|
---|
| 1258 |
|
---|
| 1259 | &dbutil::write_infodb_entry($infodbtype, $doc_infodb_file_handle, $oid, $val_rec);
|
---|
| 1260 | &dbutil::close_infodb_write_handle($infodbtype, $doc_infodb_file_handle);
|
---|
| 1261 | }
|
---|
| 1262 | }
|
---|
| 1263 |
|
---|
| 1264 | }
|
---|
| 1265 |
|
---|
| 1266 | # now go through and check that we haven't marked any primary
|
---|
| 1267 | # files for reindex (because their associated files have
|
---|
| 1268 | # changed/deleted) when they have been deleted themselves. only in
|
---|
| 1269 | # delete mode.
|
---|
| 1270 |
|
---|
| 1271 | if ($mode eq "delete") {
|
---|
| 1272 | foreach my $file (@$deleted_files) {
|
---|
| 1273 | if (defined $block_hash->{'reindex_files'}->{$file}) {
|
---|
| 1274 | delete $block_hash->{'reindex_files'}->{$file};
|
---|
| 1275 | }
|
---|
| 1276 | }
|
---|
| 1277 | }
|
---|
| 1278 |
|
---|
| 1279 |
|
---|
| 1280 | }
|
---|
| 1281 |
|
---|
| 1282 | sub add_dir_contents_to_list {
|
---|
| 1283 |
|
---|
| 1284 | my ($dirname, $list) = @_;
|
---|
| 1285 |
|
---|
| 1286 | # Recur over directory contents.
|
---|
| 1287 | my (@dir, $subfile);
|
---|
| 1288 |
|
---|
| 1289 | # find all the files in the directory
|
---|
| 1290 | if (!opendir (DIR, $dirname)) {
|
---|
| 1291 | print STDERR "inexport: WARNING - couldn't read directory $dirname\n";
|
---|
| 1292 | return -1; # error in processing
|
---|
| 1293 | }
|
---|
| 1294 | @dir = readdir (DIR);
|
---|
| 1295 | closedir (DIR);
|
---|
| 1296 |
|
---|
| 1297 | for (my $i = 0; $i < scalar(@dir); $i++) {
|
---|
| 1298 | my $subfile = $dir[$i];
|
---|
| 1299 | next if ($subfile =~ m/^\.\.?$/);
|
---|
| 1300 | next if ($subfile =~ /^\.svn$/);
|
---|
| 1301 | my $full_file = &util::filename_cat($dirname, $subfile);
|
---|
| 1302 | if (-d $full_file) {
|
---|
| 1303 | &add_dir_contents_to_list($full_file, $list);
|
---|
| 1304 | } else {
|
---|
| 1305 | push (@$list, $full_file);
|
---|
| 1306 | }
|
---|
| 1307 | }
|
---|
| 1308 |
|
---|
| 1309 | }
|
---|
| 1310 |
|
---|
| 1311 |
|
---|
| 1312 | 1;
|
---|