[18457] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
| 3 | # inexport.pm -- useful utilities 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 |
|
---|
[19789] | 30 | use File::Basename;
|
---|
| 31 |
|
---|
[18457] | 32 | use util;
|
---|
| 33 | use GDBMUtils;
|
---|
| 34 |
|
---|
[18528] | 35 |
|
---|
| 36 | sub prime_doc_oid_count
|
---|
| 37 | {
|
---|
| 38 | my ($archivedir) = @_;
|
---|
| 39 | my $oid_count_filename = &util::filename_cat ($archivedir, "OIDcount");
|
---|
| 40 |
|
---|
| 41 | if (-e $oid_count_filename) {
|
---|
| 42 | if (open(OIDIN,"<$oid_count_filename")) {
|
---|
| 43 | my $OIDcount = <OIDIN>;
|
---|
| 44 | chomp $OIDcount;
|
---|
| 45 | close(OIDIN);
|
---|
| 46 |
|
---|
| 47 | $doc::OIDcount = $OIDcount;
|
---|
| 48 | }
|
---|
| 49 | else {
|
---|
| 50 |
|
---|
| 51 | print STDERR "Warning: unable to read document OID count from $oid_count_filename\n";
|
---|
| 52 | print STDERR "Setting value to 0\n";
|
---|
| 53 | }
|
---|
| 54 | }
|
---|
| 55 |
|
---|
| 56 | }
|
---|
| 57 |
|
---|
| 58 | sub store_doc_oid_count
|
---|
| 59 | {
|
---|
| 60 | # Use the file "OIDcount" in the archives directory to record
|
---|
| 61 | # what value doc.pm got up to
|
---|
| 62 |
|
---|
| 63 | my ($archivedir) = @_;
|
---|
| 64 | my $oid_count_filename = &util::filename_cat ($archivedir, "OIDcount");
|
---|
| 65 |
|
---|
| 66 |
|
---|
| 67 | if (open(OIDOUT,">$oid_count_filename")) {
|
---|
| 68 | print OIDOUT $doc::OIDcount, "\n";
|
---|
| 69 |
|
---|
| 70 | close(OIDOUT);
|
---|
| 71 | }
|
---|
| 72 | else {
|
---|
| 73 | print STDERR "Warning: unable to store document OID count\n";
|
---|
| 74 | }
|
---|
| 75 | }
|
---|
| 76 |
|
---|
| 77 |
|
---|
| 78 |
|
---|
[18457] | 79 | sub new_vs_old_import_diff
|
---|
| 80 | {
|
---|
| 81 | my ($archive_info,$block_hash,$importdir) = @_;
|
---|
| 82 |
|
---|
| 83 | # First convert all files to absolute form
|
---|
| 84 | # This is to support the situation where the import folder is not
|
---|
| 85 | # the default
|
---|
| 86 |
|
---|
| 87 | my $prev_all_files = $archive_info->{'prev_import_filelist'};
|
---|
| 88 | my $full_prev_all_files = {};
|
---|
| 89 |
|
---|
| 90 | foreach my $prev_file (keys %$prev_all_files) {
|
---|
| 91 |
|
---|
| 92 | if (!&util::filename_is_absolute($prev_file)) {
|
---|
| 93 | my $full_prev_file = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},$prev_file);
|
---|
| 94 | $full_prev_all_files->{$full_prev_file} = $prev_file;
|
---|
| 95 | }
|
---|
| 96 | else {
|
---|
| 97 | $full_prev_all_files->{$prev_file} = $prev_file;
|
---|
| 98 | }
|
---|
| 99 | }
|
---|
| 100 |
|
---|
[18469] | 101 |
|
---|
[18457] | 102 | # Figure out which are the new files, existing files and so
|
---|
| 103 | # by implication the files from the previous import that are not
|
---|
| 104 | # there any more => mark them for deletion
|
---|
| 105 | foreach my $curr_file (keys %{$block_hash->{'all_files'}}) {
|
---|
| 106 |
|
---|
| 107 | my $full_curr_file = $curr_file;
|
---|
| 108 |
|
---|
| 109 | # entry in 'all_files' is moved to either 'existing_files',
|
---|
| 110 | # 'deleted_files', or 'new_files'
|
---|
| 111 |
|
---|
| 112 | if (!&util::filename_is_absolute($curr_file)) {
|
---|
| 113 | # add in import dir to make absolute
|
---|
| 114 | $full_curr_file = &util::filename_cat($importdir,$curr_file);
|
---|
| 115 | }
|
---|
| 116 |
|
---|
[18469] | 117 | if (defined $block_hash->{'file_blocks'}->{$full_curr_file}) {
|
---|
| 118 | # If in block list, we want to ignore it
|
---|
| 119 | delete $block_hash->{'all_files'}->{$curr_file};
|
---|
| 120 |
|
---|
| 121 | if (defined $full_prev_all_files->{$full_curr_file}) {
|
---|
| 122 | # also make sure it is gone from 'previous' list so
|
---|
| 123 | # not mistaken for a file that needs to be deleted
|
---|
| 124 | delete $full_prev_all_files->{$full_curr_file};
|
---|
| 125 | }
|
---|
| 126 | next;
|
---|
| 127 | }
|
---|
| 128 |
|
---|
[19498] | 129 | # figure out if new file or not
|
---|
[18457] | 130 | if (defined $full_prev_all_files->{$full_curr_file}) {
|
---|
[18469] | 131 |
|
---|
[18457] | 132 | # had it before
|
---|
[18469] | 133 | $block_hash->{'existing_files'}->{$full_curr_file} = 1;
|
---|
| 134 |
|
---|
[18457] | 135 | # Now remove it, so by end of loop only the files
|
---|
| 136 | # that need deleting are left
|
---|
| 137 |
|
---|
| 138 | delete $full_prev_all_files->{$full_curr_file};
|
---|
| 139 | }
|
---|
| 140 | else {
|
---|
[18469] | 141 | $block_hash->{'new_files'}->{$full_curr_file} = 1;
|
---|
[18457] | 142 | }
|
---|
| 143 |
|
---|
| 144 | delete $block_hash->{'all_files'}->{$curr_file};
|
---|
| 145 | }
|
---|
| 146 |
|
---|
[18469] | 147 | # By this point full_prev_all_files contains the files
|
---|
| 148 | # mentioned in archiveinf-src.db but are not in the 'import'
|
---|
| 149 | # folder (or whatever was specified through -importdir ...)
|
---|
| 150 |
|
---|
| 151 | # This list can contain files that were created in the 'tmp' or
|
---|
| 152 | # 'cache' areas (such as screen-size and thumbnail images).
|
---|
[18457] | 153 | #
|
---|
[18469] | 154 | # In building the final list of files to delete, we test to see if
|
---|
| 155 | # it exists on the filesystem and if it does (unusual for a file
|
---|
[19498] | 156 | # that's allegedly deleted!), supress it from going into the final
|
---|
[18469] | 157 | # list
|
---|
| 158 |
|
---|
| 159 | my $collectdir = $ENV{'GSDLCOLLECTDIR'};
|
---|
| 160 |
|
---|
[18457] | 161 | my @deleted_files = values %$full_prev_all_files;
|
---|
[18469] | 162 | map { my $curr_file = $_;
|
---|
| 163 | my $full_curr_file = $curr_file;
|
---|
| 164 |
|
---|
| 165 | if (!&util::filename_is_absolute($curr_file)) {
|
---|
| 166 | # add in import dir to make absolute
|
---|
| 167 |
|
---|
| 168 | $full_curr_file = &util::filename_cat($collectdir,$curr_file);
|
---|
| 169 | }
|
---|
| 170 |
|
---|
| 171 |
|
---|
| 172 | if (!-e $full_curr_file) {
|
---|
| 173 | $block_hash->{'deleted_files'}->{$curr_file} = 1;
|
---|
| 174 | }
|
---|
| 175 | } @deleted_files;
|
---|
[18457] | 176 | }
|
---|
| 177 |
|
---|
[18554] | 178 |
|
---|
[19498] | 179 | sub is_assoc_file
|
---|
| 180 | {
|
---|
| 181 | my ($file,$doc_rec) = @_;
|
---|
| 182 |
|
---|
[19789] | 183 | my ($file_root,$dirname,$suffix) = fileparse($file, "\\.[^\\.]+\$");
|
---|
| 184 |
|
---|
[19498] | 185 | foreach my $af (@{$doc_rec->{'assoc-file'}}) {
|
---|
[19789] | 186 | my $full_af = &util::filename_cat($dirname,$af);
|
---|
| 187 |
|
---|
| 188 | return 1 if ($full_af eq $file);
|
---|
[19498] | 189 | }
|
---|
| 190 |
|
---|
| 191 | return 0;
|
---|
| 192 | }
|
---|
| 193 |
|
---|
| 194 |
|
---|
[18554] | 195 | sub _mark_docs_for_deletion
|
---|
[18457] | 196 | {
|
---|
[19498] | 197 | my ($archive_info,$block_hash,$deleted_files,$archivedir,$verbosity,$mode_text) = @_;
|
---|
[18457] | 198 |
|
---|
[18658] | 199 | my $doc_db = "archiveinf-doc.gdb";
|
---|
| 200 | my $src_db = "archiveinf-src.gdb";
|
---|
[18457] | 201 | my $arcinfo_doc_filename = &util::filename_cat ($archivedir, $doc_db);
|
---|
| 202 | my $arcinfo_src_filename = &util::filename_cat ($archivedir, $src_db);
|
---|
| 203 |
|
---|
| 204 |
|
---|
| 205 | # record files marked for deletion in arcinfo
|
---|
[19498] | 206 | foreach my $file (@$deleted_files) {
|
---|
[18457] | 207 | # use 'archiveinf-src' GDBM file to look up all the OIDs
|
---|
[19789] | 208 | # that this file is used in (note in most cases, it's just one OID)
|
---|
[18457] | 209 |
|
---|
| 210 | my $src_rec = GDBMUtils::gdbmRecordToHash($arcinfo_src_filename,$file);
|
---|
| 211 | my $oids = $src_rec->{'oid'};
|
---|
| 212 | foreach my $oid (@$oids) {
|
---|
| 213 |
|
---|
[19498] | 214 | # Find out if it's a main doc, assoc file, or metadata
|
---|
[18457] | 215 |
|
---|
| 216 | my $doc_rec = GDBMUtils::gdbmRecordToHash($arcinfo_doc_filename,$oid);
|
---|
| 217 |
|
---|
[19498] | 218 | if (is_assoc_file($file,$doc_rec)) {
|
---|
[18457] | 219 | # assoc file => mark it for re-indexing (safest thing to do)
|
---|
| 220 | my $curr_status = $archive_info->get_status_info($oid);
|
---|
| 221 |
|
---|
| 222 |
|
---|
| 223 | if (defined($curr_status) && (($curr_status ne "D") && ($curr_status ne "R"))) {
|
---|
| 224 | if ($verbosity > 1) {
|
---|
| 225 | print STDERR "$oid marked for (potential) reindexing\n";
|
---|
| 226 | print STDERR " (because associated file $file deleted)\n";
|
---|
| 227 | }
|
---|
| 228 | $archive_info->set_status_info($oid,"R");
|
---|
| 229 |
|
---|
| 230 | my $val = &GDBMUtils::gdbmDatabaseGet($arcinfo_doc_filename,$oid);
|
---|
| 231 | $val =~ s/^<index-status>(.*)$/<index-status>R/m;
|
---|
| 232 | &GDBMUtils::gdbmDatabaseSet($arcinfo_doc_filename,$oid,$val);
|
---|
| 233 | }
|
---|
[19498] | 234 | GDBMUtils::gdbmDatabaseRemove($arcinfo_src_filename,$file);
|
---|
[18457] | 235 | }
|
---|
[19498] | 236 | else {
|
---|
| 237 | # either src-file or metadata.xml file linking to src-file
|
---|
[18469] | 238 |
|
---|
[19498] | 239 | my $src_file;
|
---|
| 240 |
|
---|
| 241 | if ($doc_rec->{'src-file'}->[0] ne $file) {
|
---|
| 242 | # it's a metadata file attached to this OID
|
---|
| 243 | # => workout the src-file it matches to
|
---|
| 244 |
|
---|
| 245 | $src_file = $doc_rec->{'src-file'}->[0];
|
---|
| 246 |
|
---|
| 247 | my $src_filename = $src_file;
|
---|
| 248 | if (!&util::filename_is_absolute($src_file)) {
|
---|
| 249 | $src_filename = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},$src_file);
|
---|
| 250 | }
|
---|
| 251 |
|
---|
| 252 | $block_hash->{'reindex_files'}->{$src_filename} = 1;
|
---|
[19789] | 253 |
|
---|
| 254 | # remove the metadata file from the src-database
|
---|
| 255 | GDBMUtils::gdbmDatabaseRemove($arcinfo_src_filename,$file);
|
---|
[19498] | 256 | }
|
---|
| 257 | else {
|
---|
| 258 | # It's the main doc
|
---|
| 259 | # => make it the target and mark it for deletion
|
---|
| 260 | $src_file = $file;
|
---|
| 261 | }
|
---|
| 262 |
|
---|
| 263 | # Whether the main file directly or indirectly, mark for deletion/reindex
|
---|
| 264 |
|
---|
| 265 | my $val = &GDBMUtils::gdbmDatabaseGet($arcinfo_doc_filename,$oid);
|
---|
| 266 | my ($index_status) = ($val =~ m/^<index-status>(.*)$/m);
|
---|
[19789] | 267 |
|
---|
| 268 |
|
---|
[19498] | 269 | if ($index_status ne "D") {
|
---|
[19789] | 270 | if ($verbosity>1) {
|
---|
| 271 | print STDERR "$oid ($src_file) marked to be $mode_text on next buildcol.pl\n";
|
---|
| 272 | }
|
---|
| 273 | $archive_info->set_status_info($oid,"D");
|
---|
| 274 |
|
---|
[19498] | 275 | $val =~ s/^<index-status>(.*)$/<index-status>D/m;
|
---|
| 276 | &GDBMUtils::gdbmDatabaseSet($arcinfo_doc_filename,$oid,$val);
|
---|
[19789] | 277 |
|
---|
| 278 | GDBMUtils::gdbmDatabaseRemove($arcinfo_src_filename,$src_file);
|
---|
[19498] | 279 | }
|
---|
| 280 |
|
---|
[19789] | 281 |
|
---|
[19498] | 282 | }
|
---|
| 283 |
|
---|
[18457] | 284 | }
|
---|
| 285 | }
|
---|
| 286 | }
|
---|
| 287 |
|
---|
| 288 |
|
---|
[18554] | 289 | sub mark_docs_for_deletion
|
---|
| 290 | {
|
---|
| 291 | _mark_docs_for_deletion(@_,"deleted from index");
|
---|
| 292 | }
|
---|
[18457] | 293 |
|
---|
[18554] | 294 |
|
---|
[18469] | 295 | sub mark_docs_for_reindex
|
---|
| 296 | {
|
---|
[19498] | 297 | my ($archive_info,$block_hash,$archivedir,$verbosity) = @_;
|
---|
[18469] | 298 |
|
---|
| 299 | # Reindexing is accomplished by deleting the previously indexed
|
---|
| 300 | # version of the document, and then allowing the new version to
|
---|
| 301 | # be indexed (as would a new document be indexed).
|
---|
| 302 | #
|
---|
| 303 | # The first step (marking for deletion) is implemented by this routine.
|
---|
| 304 | #
|
---|
| 305 | # By default in Greenstone a new version of an index will hash to
|
---|
| 306 | # a new unique OID, and the above strategy of reindex=delete+add
|
---|
| 307 | # works fine. A special case arises when a persistent OID is
|
---|
| 308 | # allocated to a document (for instance through a metadata field),
|
---|
| 309 | # and the second step to reindexing (see XXXX) detects this and
|
---|
| 310 | # deals with it appropriately.
|
---|
| 311 |
|
---|
[19498] | 312 | my @existing_files = sort keys %{$block_hash->{'existing_files'}};
|
---|
| 313 |
|
---|
[18658] | 314 | my $doc_db = "archiveinf-doc.gdb";
|
---|
[18469] | 315 | my $arcinfo_doc_filename = &util::filename_cat ($archivedir, $doc_db);
|
---|
| 316 |
|
---|
| 317 | my $archiveinf_timestamp = -M $arcinfo_doc_filename;
|
---|
| 318 |
|
---|
[19498] | 319 | my $reindex_files = [];
|
---|
[18469] | 320 |
|
---|
[19498] | 321 | foreach my $existing_filename (@existing_files) {
|
---|
[18469] | 322 |
|
---|
| 323 | if (-M $existing_filename < $archiveinf_timestamp) {
|
---|
| 324 | # file is newer than last build
|
---|
| 325 |
|
---|
| 326 | my $existing_file = $existing_filename;
|
---|
| 327 | my $collectdir = &util::filename_cat($ENV{'GSDLCOLLECTDIR'});
|
---|
| 328 |
|
---|
[19266] | 329 | my $collectdir_resafe = &util::filename_to_regex($collectdir);
|
---|
| 330 | $existing_file =~ s/^$collectdir_resafe(\\|\/)?//;
|
---|
[18469] | 331 |
|
---|
[18554] | 332 | ### print STDERR "**** Deleting existing file: $existing_file\n";
|
---|
[18469] | 333 |
|
---|
[19498] | 334 | push(@$reindex_files,$existing_file);
|
---|
| 335 | $block_hash->{'reindex_files'}->{$existing_filename} = 1;
|
---|
[18469] | 336 | }
|
---|
| 337 |
|
---|
| 338 | }
|
---|
| 339 |
|
---|
[19498] | 340 | _mark_docs_for_deletion($archive_info,$block_hash,$reindex_files,$archivedir,$verbosity, "reindex");
|
---|
[18469] | 341 |
|
---|
| 342 | }
|
---|
| 343 |
|
---|
| 344 |
|
---|
| 345 |
|
---|
[18457] | 346 | 1;
|
---|