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