| 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 File::Basename;
|
|---|
| 31 |
|
|---|
| 32 | use dbutil;
|
|---|
| 33 | use util;
|
|---|
| 34 |
|
|---|
| 35 |
|
|---|
| 36 | sub oid_count_file {
|
|---|
| 37 | my ($archivedir) = @_;
|
|---|
| 38 | return &util::filename_cat ($archivedir, "OIDcount");
|
|---|
| 39 | }
|
|---|
| 40 |
|
|---|
| 41 |
|
|---|
| 42 | sub prime_doc_oid_count
|
|---|
| 43 | {
|
|---|
| 44 | my ($archivedir) = @_;
|
|---|
| 45 | my $oid_count_filename = &oid_count_file($archivedir);
|
|---|
| 46 |
|
|---|
| 47 | if (-e $oid_count_filename) {
|
|---|
| 48 | if (open(OIDIN,"<$oid_count_filename")) {
|
|---|
| 49 | my $OIDcount = <OIDIN>;
|
|---|
| 50 | chomp $OIDcount;
|
|---|
| 51 | close(OIDIN);
|
|---|
| 52 |
|
|---|
| 53 | $doc::OIDcount = $OIDcount;
|
|---|
| 54 | }
|
|---|
| 55 | else {
|
|---|
| 56 |
|
|---|
| 57 | print STDERR "Warning: unable to read document OID count from $oid_count_filename\n";
|
|---|
| 58 | print STDERR "Setting value to 0\n";
|
|---|
| 59 | }
|
|---|
| 60 | }
|
|---|
| 61 |
|
|---|
| 62 | }
|
|---|
| 63 |
|
|---|
| 64 | sub store_doc_oid_count
|
|---|
| 65 | {
|
|---|
| 66 | # Use the file "OIDcount" in the archives directory to record
|
|---|
| 67 | # what value doc.pm got up to
|
|---|
| 68 |
|
|---|
| 69 | my ($archivedir) = @_;
|
|---|
| 70 | my $oid_count_filename = &oid_count_file($archivedir);
|
|---|
| 71 |
|
|---|
| 72 |
|
|---|
| 73 | if (open(OIDOUT,">$oid_count_filename")) {
|
|---|
| 74 | print OIDOUT $doc::OIDcount, "\n";
|
|---|
| 75 |
|
|---|
| 76 | close(OIDOUT);
|
|---|
| 77 | }
|
|---|
| 78 | else {
|
|---|
| 79 | print STDERR "Warning: unable to store document OID count\n";
|
|---|
| 80 | }
|
|---|
| 81 | }
|
|---|
| 82 |
|
|---|
| 83 |
|
|---|
| 84 |
|
|---|
| 85 | sub new_vs_old_import_diff
|
|---|
| 86 | {
|
|---|
| 87 | my ($archive_info,$block_hash,$importdir,$archivedir,$verbosity,$incremental_mode) = @_;
|
|---|
| 88 |
|
|---|
| 89 | # in this method, we want to know if metadata files are modified or not.
|
|---|
| 90 | my $arcinfo_doc_filename = &dbutil::get_infodb_file_path("gdbm", "archiveinf-doc", $archivedir);
|
|---|
| 91 |
|
|---|
| 92 | my $archiveinf_timestamp = -M $arcinfo_doc_filename;
|
|---|
| 93 |
|
|---|
| 94 | # First convert all files to absolute form
|
|---|
| 95 | # This is to support the situation where the import folder is not
|
|---|
| 96 | # the default
|
|---|
| 97 |
|
|---|
| 98 | my $prev_all_files = $archive_info->{'prev_import_filelist'};
|
|---|
| 99 | my $full_prev_all_files = {};
|
|---|
| 100 |
|
|---|
| 101 | foreach my $prev_file (keys %$prev_all_files) {
|
|---|
| 102 |
|
|---|
| 103 | if (!&util::filename_is_absolute($prev_file)) {
|
|---|
| 104 | my $full_prev_file = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},$prev_file);
|
|---|
| 105 | $full_prev_all_files->{$full_prev_file} = $prev_file;
|
|---|
| 106 | }
|
|---|
| 107 | else {
|
|---|
| 108 | $full_prev_all_files->{$prev_file} = $prev_file;
|
|---|
| 109 | }
|
|---|
| 110 | }
|
|---|
| 111 |
|
|---|
| 112 |
|
|---|
| 113 | # Figure out which are the new files, existing files and so
|
|---|
| 114 | # by implication the files from the previous import that are not
|
|---|
| 115 | # there any more => mark them for deletion
|
|---|
| 116 | foreach my $curr_file (keys %{$block_hash->{'all_files'}}) {
|
|---|
| 117 |
|
|---|
| 118 | my $full_curr_file = $curr_file;
|
|---|
| 119 |
|
|---|
| 120 | # entry in 'all_files' is moved to either 'existing_files',
|
|---|
| 121 | # 'deleted_files', 'new_files', or 'new_or_modified_metadata_files'
|
|---|
| 122 |
|
|---|
| 123 | if (!&util::filename_is_absolute($curr_file)) {
|
|---|
| 124 | # add in import dir to make absolute
|
|---|
| 125 | $full_curr_file = &util::filename_cat($importdir,$curr_file);
|
|---|
| 126 | }
|
|---|
| 127 |
|
|---|
| 128 | # figure out if new file or not
|
|---|
| 129 | if (defined $full_prev_all_files->{$full_curr_file}) {
|
|---|
| 130 | # delete it so that only files that need deleting are left
|
|---|
| 131 | delete $full_prev_all_files->{$full_curr_file};
|
|---|
| 132 |
|
|---|
| 133 | # had it before. is it a metadata file?
|
|---|
| 134 | if ($block_hash->{'metadata_files'}->{$full_curr_file}) {
|
|---|
| 135 |
|
|---|
| 136 | # is it modified??
|
|---|
| 137 | if (-M $full_curr_file < $archiveinf_timestamp) {
|
|---|
| 138 | print STDERR "*** Detected a modified metadata file: $full_curr_file\n" if $verbosity > 2;
|
|---|
| 139 | # its newer than last build
|
|---|
| 140 | $block_hash->{'new_or_modified_metadata_files'}->{$full_curr_file} = 1;
|
|---|
| 141 | }
|
|---|
| 142 | }
|
|---|
| 143 | else {
|
|---|
| 144 | if ($incremental_mode eq "all") {
|
|---|
| 145 |
|
|---|
| 146 | # had it before
|
|---|
| 147 | $block_hash->{'existing_files'}->{$full_curr_file} = 1;
|
|---|
| 148 |
|
|---|
| 149 | }
|
|---|
| 150 | else {
|
|---|
| 151 | # Warning in "onlyadd" mode, but had it before!
|
|---|
| 152 | print STDERR "Warning: File $full_curr_file previously imported.\n";
|
|---|
| 153 | print STDERR " Treating as new file\n";
|
|---|
| 154 |
|
|---|
| 155 | $block_hash->{'new_files'}->{$full_curr_file} = 1;
|
|---|
| 156 |
|
|---|
| 157 | }
|
|---|
| 158 | }
|
|---|
| 159 | }
|
|---|
| 160 | else {
|
|---|
| 161 | if ($block_hash->{'metadata_files'}->{$full_curr_file}) {
|
|---|
| 162 | # the new file is the special sort of file greenstone uses
|
|---|
| 163 | # to attach metadata to src documents
|
|---|
| 164 | # i.e metadata.xml
|
|---|
| 165 | # (but note, the filename used is not constrained in
|
|---|
| 166 | # Greenstone to always be this)
|
|---|
| 167 |
|
|---|
| 168 | print STDERR "***** Detected new metadata file: $full_curr_file\n" if $verbosity > 2;
|
|---|
| 169 | $block_hash->{'new_or_modified_metadata_files'}->{$full_curr_file} = 1;
|
|---|
| 170 | }
|
|---|
| 171 | else {
|
|---|
| 172 | $block_hash->{'new_files'}->{$full_curr_file} = 1;
|
|---|
| 173 | }
|
|---|
| 174 | }
|
|---|
| 175 |
|
|---|
| 176 |
|
|---|
| 177 | delete $block_hash->{'all_files'}->{$curr_file};
|
|---|
| 178 | }
|
|---|
| 179 |
|
|---|
| 180 |
|
|---|
| 181 |
|
|---|
| 182 |
|
|---|
| 183 | # Deal with complication of new or modified metadata files by forcing
|
|---|
| 184 | # everything from this point down in the file hierarchy to
|
|---|
| 185 | # be freshly imported.
|
|---|
| 186 | #
|
|---|
| 187 | # This may mean files that have not changed are reindexed, but does
|
|---|
| 188 | # guarantee by the end of processing all new metadata is correctly
|
|---|
| 189 | # associated with the relevant document(s).
|
|---|
| 190 |
|
|---|
| 191 | foreach my $new_mdf (keys %{$block_hash->{'new_or_modified_metadata_files'}}) {
|
|---|
| 192 | my ($fileroot,$situated_dir,$ext) = fileparse($new_mdf, "\\.[^\\.]+\$");
|
|---|
| 193 |
|
|---|
| 194 | $situated_dir =~ s/[\\\/]+$//; # remove tailing slashes
|
|---|
| 195 | $situated_dir =~ s/\\/\\\\/g; # need to protect windows slash \ in regular expression
|
|---|
| 196 |
|
|---|
| 197 | # Go through existing_files, and mark anything that is contained
|
|---|
| 198 | # within 'situated_dir' to be reindexed (in case some of the metadata
|
|---|
| 199 | # attaches to one of these files)
|
|---|
| 200 |
|
|---|
| 201 | my $reindex_files = [];
|
|---|
| 202 |
|
|---|
| 203 | foreach my $existing_f (keys %{$block_hash->{'existing_files'}}) {
|
|---|
| 204 |
|
|---|
| 205 | if ($existing_f =~ m/^$situated_dir/) {
|
|---|
| 206 | push(@$reindex_files,$existing_f);
|
|---|
| 207 | $block_hash->{'reindex_files'}->{$existing_f} = 1;
|
|---|
| 208 | delete $block_hash->{'existing_files'}->{$existing_f};
|
|---|
| 209 |
|
|---|
| 210 | }
|
|---|
| 211 | }
|
|---|
| 212 |
|
|---|
| 213 | # metadata file needs to be in new_files list so parsed by MetadataXMLPlug
|
|---|
| 214 | # (or equivalent)
|
|---|
| 215 | $block_hash->{'new_files'}->{$new_mdf} = 1;
|
|---|
| 216 |
|
|---|
| 217 | }
|
|---|
| 218 |
|
|---|
| 219 | # go through remaining existing files and work out what has changed and needs to be reindexed.
|
|---|
| 220 | my @existing_files = sort keys %{$block_hash->{'existing_files'}};
|
|---|
| 221 |
|
|---|
| 222 | my $reindex_files = [];
|
|---|
| 223 |
|
|---|
| 224 | foreach my $existing_filename (@existing_files) {
|
|---|
| 225 | if (-M $existing_filename < $archiveinf_timestamp) {
|
|---|
| 226 | # file is newer than last build
|
|---|
| 227 |
|
|---|
| 228 | my $existing_file = $existing_filename;
|
|---|
| 229 | #my $collectdir = &util::filename_cat($ENV{'GSDLCOLLECTDIR'});
|
|---|
| 230 |
|
|---|
| 231 | #my $collectdir_resafe = &util::filename_to_regex($collectdir);
|
|---|
| 232 | #$existing_file =~ s/^$collectdir_resafe(\\|\/)?//;
|
|---|
| 233 |
|
|---|
| 234 | print STDERR "**** Reindexing existing file: $existing_file\n";
|
|---|
| 235 |
|
|---|
| 236 | push(@$reindex_files,$existing_file);
|
|---|
| 237 | $block_hash->{'reindex_files'}->{$existing_filename} = 1;
|
|---|
| 238 | }
|
|---|
| 239 |
|
|---|
| 240 | }
|
|---|
| 241 |
|
|---|
| 242 |
|
|---|
| 243 | # By this point full_prev_all_files contains the files
|
|---|
| 244 | # mentioned in archiveinf-src.db but are not in the 'import'
|
|---|
| 245 | # folder (or whatever was specified through -importdir ...)
|
|---|
| 246 |
|
|---|
| 247 | # This list can contain files that were created in the 'tmp' or
|
|---|
| 248 | # 'cache' areas (such as screen-size and thumbnail images).
|
|---|
| 249 | #
|
|---|
| 250 | # In building the final list of files to delete, we test to see if
|
|---|
| 251 | # it exists on the filesystem and if it does (unusual for a "normal"
|
|---|
| 252 | # file in import, but possible in the case of 'tmp' files),
|
|---|
| 253 | # supress it from going into the final list
|
|---|
| 254 |
|
|---|
| 255 | my $collectdir = $ENV{'GSDLCOLLECTDIR'};
|
|---|
| 256 |
|
|---|
| 257 | my @deleted_files = values %$full_prev_all_files;
|
|---|
| 258 | map { my $curr_file = $_;
|
|---|
| 259 | my $full_curr_file = $curr_file;
|
|---|
| 260 |
|
|---|
| 261 | if (!&util::filename_is_absolute($curr_file)) {
|
|---|
| 262 | # add in import dir to make absolute
|
|---|
| 263 |
|
|---|
| 264 | $full_curr_file = &util::filename_cat($collectdir,$curr_file);
|
|---|
| 265 | }
|
|---|
| 266 |
|
|---|
| 267 |
|
|---|
| 268 | if (!-e $full_curr_file) {
|
|---|
| 269 | $block_hash->{'deleted_files'}->{$curr_file} = 1;
|
|---|
| 270 | }
|
|---|
| 271 | } @deleted_files;
|
|---|
| 272 |
|
|---|
| 273 |
|
|---|
| 274 |
|
|---|
| 275 | }
|
|---|
| 276 |
|
|---|
| 277 |
|
|---|
| 278 | # this is used to delete "deleted" docs, and to remove old versions of "changed" docs
|
|---|
| 279 | # $mode is 'delete' or 'reindex'
|
|---|
| 280 | sub mark_docs_for_deletion
|
|---|
| 281 | {
|
|---|
| 282 | my ($archive_info,$block_hash,$deleted_files,$archivedir,$verbosity,$mode) = @_;
|
|---|
| 283 |
|
|---|
| 284 | my $mode_text = "deleted from index";
|
|---|
| 285 | if ($mode eq "reindex") {
|
|---|
| 286 | $mode_text = "reindexed";
|
|---|
| 287 | }
|
|---|
| 288 | my $arcinfo_doc_filename = &dbutil::get_infodb_file_path("gdbm", "archiveinf-doc", $archivedir);
|
|---|
| 289 | my $arcinfo_src_filename = &dbutil::get_infodb_file_path("gdbm", "archiveinf-src", $archivedir);
|
|---|
| 290 | my $doc_infodb_file_handle = &dbutil::open_infodb_write_handle("gdbm", $arcinfo_doc_filename, "append");
|
|---|
| 291 | my $src_infodb_file_handle = &dbutil::open_infodb_write_handle("gdbm", $arcinfo_src_filename, "append");
|
|---|
| 292 |
|
|---|
| 293 |
|
|---|
| 294 | # record files marked for deletion in arcinfo
|
|---|
| 295 | foreach my $file (@$deleted_files) {
|
|---|
| 296 | # use 'archiveinf-src' info database file to look up all the OIDs
|
|---|
| 297 | # that this file is used in (note in most cases, it's just one OID)
|
|---|
| 298 |
|
|---|
| 299 | my $src_rec_string = &dbutil::read_infodb_entry("gdbm", $arcinfo_src_filename, $file);
|
|---|
| 300 | my $src_rec = &dbutil::convert_infodb_string_to_hash($src_rec_string);
|
|---|
| 301 | my $oids = $src_rec->{'oid'};
|
|---|
| 302 | my $file_record_deleted = 0;
|
|---|
| 303 |
|
|---|
| 304 | # delete the src record
|
|---|
| 305 | &dbutil::delete_infodb_entry("gdbm", $src_infodb_file_handle, $file);
|
|---|
| 306 |
|
|---|
| 307 | foreach my $oid (@$oids) {
|
|---|
| 308 |
|
|---|
| 309 | # find the source doc (the primary file that becomes this oid)
|
|---|
| 310 | my $doc_rec_string = &dbutil::read_infodb_entry("gdbm", $arcinfo_doc_filename, $oid);
|
|---|
| 311 | my $doc_rec = &dbutil::convert_infodb_string_to_hash($doc_rec_string);
|
|---|
| 312 | my $doc_source_file = $doc_rec->{'src-file'}->[0];
|
|---|
| 313 | if (!&util::filename_is_absolute($doc_source_file)) {
|
|---|
| 314 | $doc_source_file = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},$doc_source_file);
|
|---|
| 315 | }
|
|---|
| 316 |
|
|---|
| 317 | if ($doc_source_file ne $file) {
|
|---|
| 318 | # its an associated or metadata file
|
|---|
| 319 |
|
|---|
| 320 | # mark source doc for reimport as one of its assoc files has changed or deleted
|
|---|
| 321 | $block_hash->{'reindex_files'}->{$doc_source_file} = 1;
|
|---|
| 322 |
|
|---|
| 323 | }
|
|---|
| 324 | my $curr_status = $archive_info->get_status_info($oid);
|
|---|
| 325 | if (defined($curr_status) && (($curr_status ne "D"))) {
|
|---|
| 326 | if ($verbosity>1) {
|
|---|
| 327 | print STDERR "$oid ($doc_source_file) marked to be $mode_text on next buildcol.pl\n";
|
|---|
| 328 | }
|
|---|
| 329 | # mark oid for deletion (it will be deleted or reimported)
|
|---|
| 330 | $archive_info->set_status_info($oid,"D");
|
|---|
| 331 | my $val = &dbutil::read_infodb_entry("gdbm", $arcinfo_doc_filename, $oid);
|
|---|
| 332 | $val =~ s/^<index-status>(.*)$/<index-status>D/m;
|
|---|
| 333 |
|
|---|
| 334 | my $val_rec = &dbutil::convert_infodb_string_to_hash($val);
|
|---|
| 335 | &dbutil::write_infodb_entry("gdbm", $doc_infodb_file_handle, $oid, $val_rec);
|
|---|
| 336 | }
|
|---|
| 337 | }
|
|---|
| 338 | }
|
|---|
| 339 |
|
|---|
| 340 | &dbutil::close_infodb_write_handle("gdbm", $doc_infodb_file_handle);
|
|---|
| 341 | &dbutil::close_infodb_write_handle("gdbm", $src_infodb_file_handle);
|
|---|
| 342 | }
|
|---|
| 343 |
|
|---|
| 344 |
|
|---|
| 345 |
|
|---|
| 346 | 1;
|
|---|
