| 1 | ###########################################################################
|
|---|
| 2 | #
|
|---|
| 3 | # mgbuildproc.pm --
|
|---|
| 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 | # This document processor outputs a document
|
|---|
| 27 | # for mg to process
|
|---|
| 28 |
|
|---|
| 29 |
|
|---|
| 30 | package mgbuildproc;
|
|---|
| 31 |
|
|---|
| 32 | eval {require bytes};
|
|---|
| 33 |
|
|---|
| 34 | use classify;
|
|---|
| 35 | use doc;
|
|---|
| 36 | use docproc;
|
|---|
| 37 | use util;
|
|---|
| 38 |
|
|---|
| 39 | BEGIN {
|
|---|
| 40 | @ISA = ('docproc');
|
|---|
| 41 | }
|
|---|
| 42 |
|
|---|
| 43 |
|
|---|
| 44 | sub new {
|
|---|
| 45 | my ($class, $collection, $source_dir, $build_dir,
|
|---|
| 46 | $verbosity, $outhandle) = @_;
|
|---|
| 47 | my $self = new docproc ();
|
|---|
| 48 |
|
|---|
| 49 | # outhandle is where all the debugging info goes
|
|---|
| 50 | # output_handle is where the output of the plugins is piped
|
|---|
| 51 | # to (i.e. mg, gdbm etc.)
|
|---|
| 52 | $outhandle = STDERR unless defined $outhandle;
|
|---|
| 53 |
|
|---|
| 54 | $self->{'collection'} = $collection;
|
|---|
| 55 | $self->{'source_dir'} = $source_dir;
|
|---|
| 56 | $self->{'build_dir'} = $build_dir;
|
|---|
| 57 | $self->{'verbosity'} = $verbosity;
|
|---|
| 58 | $self->{'classifiers'} = [];
|
|---|
| 59 | $self->{'mode'} = "text";
|
|---|
| 60 | $self->{'assocdir'} = $build_dir;
|
|---|
| 61 | $self->{'dontgdbm'} = {};
|
|---|
| 62 | $self->{'index'} = "section:text";
|
|---|
| 63 | $self->{'indexexparr'} = [];
|
|---|
| 64 | $self->{'output_handle'} = "STDOUT";
|
|---|
| 65 | $self->{'num_docs'} = 0;
|
|---|
| 66 | $self->{'num_sections'} = 0;
|
|---|
| 67 | $self->{'num_bytes'} = 0;
|
|---|
| 68 | $self->{'num_processed_bytes'} = 0;
|
|---|
| 69 | $self->{'store_text'} = 1;
|
|---|
| 70 | $self->{'outhandle'} = $outhandle;
|
|---|
| 71 |
|
|---|
| 72 | #used by browse interface
|
|---|
| 73 | $self->{'doclist'} = [];
|
|---|
| 74 |
|
|---|
| 75 | $self->{'indexing_text'} = 0;
|
|---|
| 76 |
|
|---|
| 77 | return bless $self, $class;
|
|---|
| 78 | }
|
|---|
| 79 |
|
|---|
| 80 | sub reset {
|
|---|
| 81 | my $self = shift (@_);
|
|---|
| 82 |
|
|---|
| 83 | $self->{'num_docs'} = 0;
|
|---|
| 84 | $self->{'num_sections'} = 0;
|
|---|
| 85 | $self->{'num_processed_bytes'} = 0;
|
|---|
| 86 | $self->{'num_bytes'} = 0;
|
|---|
| 87 | }
|
|---|
| 88 |
|
|---|
| 89 | sub get_num_docs {
|
|---|
| 90 | my $self = shift (@_);
|
|---|
| 91 |
|
|---|
| 92 | return $self->{'num_docs'};
|
|---|
| 93 | }
|
|---|
| 94 |
|
|---|
| 95 | sub get_num_sections {
|
|---|
| 96 | my $self = shift (@_);
|
|---|
| 97 |
|
|---|
| 98 | return $self->{'num_sections'};
|
|---|
| 99 | }
|
|---|
| 100 |
|
|---|
| 101 | # num_bytes is the actual number of bytes in the collection
|
|---|
| 102 | # this is normally the same as what's processed during text compression
|
|---|
| 103 | sub get_num_bytes {
|
|---|
| 104 | my $self = shift (@_);
|
|---|
| 105 |
|
|---|
| 106 | return $self->{'num_bytes'};
|
|---|
| 107 | }
|
|---|
| 108 |
|
|---|
| 109 | # num_processed_bytes is the number of bytes actually passed
|
|---|
| 110 | # to mg for the current index
|
|---|
| 111 | sub get_num_processed_bytes {
|
|---|
| 112 | my $self = shift (@_);
|
|---|
| 113 |
|
|---|
| 114 | return $self->{'num_processed_bytes'};
|
|---|
| 115 | }
|
|---|
| 116 |
|
|---|
| 117 | sub set_output_handle {
|
|---|
| 118 | my $self = shift (@_);
|
|---|
| 119 | my ($handle) = @_;
|
|---|
| 120 |
|
|---|
| 121 | $self->{'output_handle'} = $handle;
|
|---|
| 122 | }
|
|---|
| 123 |
|
|---|
| 124 | sub set_mode {
|
|---|
| 125 | my $self = shift (@_);
|
|---|
| 126 | my ($mode) = @_;
|
|---|
| 127 |
|
|---|
| 128 | $self->{'mode'} = $mode;
|
|---|
| 129 | }
|
|---|
| 130 |
|
|---|
| 131 | sub set_assocdir {
|
|---|
| 132 | my $self = shift (@_);
|
|---|
| 133 | my ($assocdir) = @_;
|
|---|
| 134 |
|
|---|
| 135 | $self->{'assocdir'} = $assocdir;
|
|---|
| 136 | }
|
|---|
| 137 |
|
|---|
| 138 | sub set_dontgdbm {
|
|---|
| 139 | my $self = shift (@_);
|
|---|
| 140 | my ($dontgdbm) = @_;
|
|---|
| 141 |
|
|---|
| 142 | $self->{'dontgdbm'} = $dontgdbm;
|
|---|
| 143 | }
|
|---|
| 144 |
|
|---|
| 145 | sub set_index {
|
|---|
| 146 | my $self = shift (@_);
|
|---|
| 147 | my ($index, $indexexparr) = @_;
|
|---|
| 148 |
|
|---|
| 149 | $self->{'index'} = $index;
|
|---|
| 150 | $self->{'indexexparr'} = $indexexparr if defined $indexexparr;
|
|---|
| 151 | }
|
|---|
| 152 |
|
|---|
| 153 | sub get_index {
|
|---|
| 154 | my $self = shift (@_);
|
|---|
| 155 |
|
|---|
| 156 | return $self->{'index'};
|
|---|
| 157 | }
|
|---|
| 158 |
|
|---|
| 159 | sub set_classifiers {
|
|---|
| 160 | my $self = shift (@_);
|
|---|
| 161 | my ($classifiers) = @_;
|
|---|
| 162 |
|
|---|
| 163 | $self->{'classifiers'} = $classifiers;
|
|---|
| 164 | }
|
|---|
| 165 |
|
|---|
| 166 | sub set_indexing_text {
|
|---|
| 167 | my $self = shift (@_);
|
|---|
| 168 | my ($indexing_text) = @_;
|
|---|
| 169 |
|
|---|
| 170 | $self->{'indexing_text'} = $indexing_text;
|
|---|
| 171 | }
|
|---|
| 172 |
|
|---|
| 173 | sub get_indexing_text {
|
|---|
| 174 | my $self = shift (@_);
|
|---|
| 175 |
|
|---|
| 176 | return $self->{'indexing_text'};
|
|---|
| 177 | }
|
|---|
| 178 |
|
|---|
| 179 | sub set_store_text {
|
|---|
| 180 | my $self = shift (@_);
|
|---|
| 181 | my ($store_text) = @_;
|
|---|
| 182 |
|
|---|
| 183 | $self->{'store_text'} = $store_text;
|
|---|
| 184 | }
|
|---|
| 185 |
|
|---|
| 186 | sub get_doc_list {
|
|---|
| 187 | my $self = shift(@_);
|
|---|
| 188 |
|
|---|
| 189 | return @{$self->{'doclist'}};
|
|---|
| 190 | }
|
|---|
| 191 |
|
|---|
| 192 |
|
|---|
| 193 | sub process {
|
|---|
| 194 | my $self = shift (@_);
|
|---|
| 195 | my $method = $self->{'mode'};
|
|---|
| 196 |
|
|---|
| 197 | $self->$method(@_);
|
|---|
| 198 | }
|
|---|
| 199 |
|
|---|
| 200 | # use 'Paged' if document has no more than 2 levels
|
|---|
| 201 | # and each section at second level has a number for
|
|---|
| 202 | # Title metadata
|
|---|
| 203 | #also use Paged if gsdlthistype metadata is set to Paged
|
|---|
| 204 | sub get_document_type {
|
|---|
| 205 | my $self = shift (@_);
|
|---|
| 206 | my ($doc_obj) = @_;
|
|---|
| 207 |
|
|---|
| 208 | my $thistype = "VList";
|
|---|
| 209 | my $childtype = "VList";
|
|---|
| 210 | my $title;
|
|---|
| 211 | my @tmp = ();
|
|---|
| 212 |
|
|---|
| 213 | my $section = $doc_obj->get_top_section ();
|
|---|
| 214 |
|
|---|
| 215 | my $gsdlthistype = $doc_obj->get_metadata_element ($section, "gsdlthistype");
|
|---|
| 216 | if (defined $gsdlthistype) {
|
|---|
| 217 | if ($gsdlthistype eq "Paged") {
|
|---|
| 218 | $thistype = "Paged";
|
|---|
| 219 | $childtype = "Paged";
|
|---|
| 220 | return ($thistype, $childtype);
|
|---|
| 221 | } elsif ($gsdlthistype eq "Hierarchy") {
|
|---|
| 222 | return ($thistype, $childtype); # use VList, VList
|
|---|
| 223 | }
|
|---|
| 224 | }
|
|---|
| 225 | my $first = 1;
|
|---|
| 226 | while (defined $section) {
|
|---|
| 227 | @tmp = split /\./, $section;
|
|---|
| 228 | if (scalar(@tmp) > 1) {
|
|---|
| 229 | return ($thistype, $childtype);
|
|---|
| 230 | }
|
|---|
| 231 | if (!$first) {
|
|---|
| 232 | $title = $doc_obj->get_metadata_element ($section, "Title");
|
|---|
| 233 | if (!defined $title || $title !~ /^\d+$/) {
|
|---|
| 234 | return ($thistype, $childtype);
|
|---|
| 235 | }
|
|---|
| 236 | }
|
|---|
| 237 | $first = 0;
|
|---|
| 238 | $section = $doc_obj->get_next_section($section);
|
|---|
| 239 | }
|
|---|
| 240 | if ($doc_obj->get_text_length ($doc_obj->get_top_section())) {
|
|---|
| 241 | $thistype = "Paged";
|
|---|
| 242 | } else {
|
|---|
| 243 | $thistype = "Invisible";
|
|---|
| 244 | }
|
|---|
| 245 | $childtype = "Paged";
|
|---|
| 246 | return ($thistype, $childtype);
|
|---|
| 247 | }
|
|---|
| 248 |
|
|---|
| 249 | sub assoc_files {
|
|---|
| 250 | my $self = shift (@_);
|
|---|
| 251 | my ($doc_obj, $archivedir) = @_;
|
|---|
| 252 | my ($afile);
|
|---|
| 253 |
|
|---|
| 254 | foreach my $assoc_file (@{$doc_obj->get_assoc_files()}) {
|
|---|
| 255 | # if assoc file starts with a slash, we put it relative to the assoc
|
|---|
| 256 | # dir, otherwise it is relative to the HASH... directory
|
|---|
| 257 | if ($assoc_file->[1] =~ m@^[/\\]@) {
|
|---|
| 258 | $afile = &util::filename_cat($self->{'assocdir'},$assoc_file->[1]);
|
|---|
| 259 | } else {
|
|---|
| 260 | $afile = &util::filename_cat($self->{'assocdir'}, $archivedir, $assoc_file->[1]);
|
|---|
| 261 | }
|
|---|
| 262 | &util::hard_link ($assoc_file->[0], $afile);
|
|---|
| 263 | }
|
|---|
| 264 | }
|
|---|
| 265 |
|
|---|
| 266 | sub infodb {
|
|---|
| 267 | my $self = shift (@_);
|
|---|
| 268 | my ($doc_obj, $filename) = @_;
|
|---|
| 269 | my $handle = $self->{'output_handle'};
|
|---|
| 270 | # $handle = "main::STDOUT";
|
|---|
| 271 |
|
|---|
| 272 | my $doctype = $doc_obj->get_doc_type();
|
|---|
| 273 |
|
|---|
| 274 | # only output this document if it is one to be indexed
|
|---|
| 275 | return if ($doctype ne "indexed_doc");
|
|---|
| 276 |
|
|---|
| 277 | my ($archivedir) = $filename =~ /^(.*?)(?:\/|\\)[^\/\\]*$/;
|
|---|
| 278 | $archivedir = "" unless defined $archivedir;
|
|---|
| 279 | $archivedir =~ s/\\/\//g;
|
|---|
| 280 | $archivedir =~ s/^\/+//;
|
|---|
| 281 | $archivedir =~ s/\/+$//;
|
|---|
| 282 |
|
|---|
| 283 | # resolve the final filenames of the files associated with this document
|
|---|
| 284 | $self->assoc_files ($doc_obj, $archivedir);
|
|---|
| 285 |
|
|---|
| 286 | #GRB: moved 1/06/2004 from GRB01062004
|
|---|
| 287 | #add this document to the browse structure
|
|---|
| 288 | push(@{$self->{'doclist'}},$doc_obj->get_OID())
|
|---|
| 289 | unless ($doctype eq "classification");
|
|---|
| 290 |
|
|---|
| 291 | # classify this document
|
|---|
| 292 | &classify::classify_doc ($self->{'classifiers'}, $doc_obj);
|
|---|
| 293 | #GRB: end of moved block
|
|---|
| 294 |
|
|---|
| 295 | # this is another document
|
|---|
| 296 | $self->{'num_docs'} += 1 unless ($doctype eq "classification");
|
|---|
| 297 |
|
|---|
| 298 | # is this a paged or a hierarchical document
|
|---|
| 299 | my ($thistype, $childtype) = $self->get_document_type ($doc_obj);
|
|---|
| 300 |
|
|---|
| 301 | my $section = $doc_obj->get_top_section ();
|
|---|
| 302 | my $doc_OID = $doc_obj->get_OID();
|
|---|
| 303 | my $first = 1;
|
|---|
| 304 | my $url = "";
|
|---|
| 305 | while (defined $section) {
|
|---|
| 306 | # update a few statistics
|
|---|
| 307 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
|---|
| 308 | $self->{'num_sections'} += 1 unless ($doctype eq "classification");
|
|---|
| 309 |
|
|---|
| 310 | # output the section name
|
|---|
| 311 | if ($section eq "") { print $handle "[$doc_OID]\n"; }
|
|---|
| 312 | else { print $handle "[$doc_OID.$section]\n"; }
|
|---|
| 313 |
|
|---|
| 314 | # output the fact that this document is a document (unless doctype
|
|---|
| 315 | # has been set to something else from within a plugin
|
|---|
| 316 | my $dtype = $doc_obj->get_metadata_element ($section, "doctype");
|
|---|
| 317 | if (!defined $dtype || $dtype !~ /\w/) {
|
|---|
| 318 | print $handle "<doctype>doc\n";
|
|---|
| 319 | }
|
|---|
| 320 |
|
|---|
| 321 | # output whether this node contains text
|
|---|
| 322 | if ($doc_obj->get_text_length($section) > 0) {
|
|---|
| 323 | print $handle "<hastxt>1\n";
|
|---|
| 324 | } else {
|
|---|
| 325 | print $handle "<hastxt>0\n";
|
|---|
| 326 | }
|
|---|
| 327 |
|
|---|
| 328 | # output all the section metadata
|
|---|
| 329 | my $metadata = $doc_obj->get_all_metadata ($section);
|
|---|
| 330 | foreach my $pair (@$metadata) {
|
|---|
| 331 | my ($field, $value) = (@$pair);
|
|---|
| 332 |
|
|---|
| 333 | if ($field ne "Identifier" && $field !~ /^gsdl/ &&
|
|---|
| 334 | defined $value && $value ne "") {
|
|---|
| 335 |
|
|---|
| 336 | # escape problematic stuff
|
|---|
| 337 | $value =~ s/\\/\\\\/g;
|
|---|
| 338 | $value =~ s/\n/\\n/g;
|
|---|
| 339 | $value =~ s/\r/\\r/g;
|
|---|
| 340 | if ($value =~ /-{70,}/) {
|
|---|
| 341 | # if value contains 70 or more hyphens in a row we need
|
|---|
| 342 | # to escape them to prevent txt2db from treating them
|
|---|
| 343 | # as a separator
|
|---|
| 344 | $value =~ s/-/&\#045;/gi;
|
|---|
| 345 | }
|
|---|
| 346 |
|
|---|
| 347 | # special case for URL metadata
|
|---|
| 348 | if ($field =~ /^URL$/i) {
|
|---|
| 349 | $url .= "[$value]\n";
|
|---|
| 350 | if ($section eq "") {$url .= "<section>$doc_OID\n";}
|
|---|
| 351 | else {$url .= "<section>$doc_OID.$section\n";}
|
|---|
| 352 | $url .= '-' x 70 . "\n";
|
|---|
| 353 | }
|
|---|
| 354 |
|
|---|
| 355 | if (!defined $self->{'dontgdbm'}->{$field}) {
|
|---|
| 356 | print $handle "<$field>$value\n";
|
|---|
| 357 | }
|
|---|
| 358 | }
|
|---|
| 359 | }
|
|---|
| 360 |
|
|---|
| 361 | # output archivedir if at top level
|
|---|
| 362 | if ($section eq $doc_obj->get_top_section()) {
|
|---|
| 363 | print $handle "<archivedir>$archivedir\n";
|
|---|
| 364 | }
|
|---|
| 365 |
|
|---|
| 366 | # output document display type
|
|---|
| 367 | if ($first) {
|
|---|
| 368 | print $handle "<thistype>$thistype\n";
|
|---|
| 369 | }
|
|---|
| 370 |
|
|---|
| 371 | # output a list of children
|
|---|
| 372 | my $children = $doc_obj->get_children ($section);
|
|---|
| 373 | if (scalar(@$children) > 0) {
|
|---|
| 374 | print $handle "<childtype>$childtype\n";
|
|---|
| 375 | print $handle "<contains>";
|
|---|
| 376 | my $firstchild = 1;
|
|---|
| 377 | foreach my $child (@$children) {
|
|---|
| 378 | print $handle ";" unless $firstchild;
|
|---|
| 379 | $firstchild = 0;
|
|---|
| 380 | if ($child =~ /^.*?\.(\d+)$/) {
|
|---|
| 381 | print $handle "\".$1";
|
|---|
| 382 | } else {
|
|---|
| 383 | print $handle "\".$child";
|
|---|
| 384 | }
|
|---|
| 385 | # if ($child eq "") { print $handle "$doc_OID"; }
|
|---|
| 386 | # elsif ($section eq "") { print $handle "$doc_OID.$child"; }
|
|---|
| 387 | # else { print $handle "$doc_OID.$section.$child"; }
|
|---|
| 388 | }
|
|---|
| 389 | print $handle "\n";
|
|---|
| 390 | }
|
|---|
| 391 |
|
|---|
| 392 | # output the matching document number
|
|---|
| 393 | print $handle "<docnum>$self->{'num_sections'}\n";
|
|---|
| 394 |
|
|---|
| 395 | print $handle '-' x 70, "\n";
|
|---|
| 396 |
|
|---|
| 397 |
|
|---|
| 398 | # output a database entry for the document number
|
|---|
| 399 | print $handle "[$self->{'num_sections'}]\n";
|
|---|
| 400 | if ($section eq "") { print $handle "<section>$doc_OID\n"; }
|
|---|
| 401 | else { print $handle "<section>$doc_OID.$section\n"; }
|
|---|
| 402 | print $handle '-' x 70, "\n";
|
|---|
| 403 |
|
|---|
| 404 | # output entry for url
|
|---|
| 405 | if ($url ne "") {
|
|---|
| 406 | print $handle $url;
|
|---|
| 407 | }
|
|---|
| 408 |
|
|---|
| 409 | $first = 0;
|
|---|
| 410 | $section = $doc_obj->get_next_section($section);
|
|---|
| 411 | }
|
|---|
| 412 |
|
|---|
| 413 | #GRB01062004: see code above moved from here
|
|---|
| 414 | }
|
|---|
| 415 |
|
|---|
| 416 | sub find_paragraphs {
|
|---|
| 417 | $_[1] =~ s/(<p\b)/\cC$1/gi;
|
|---|
| 418 | }
|
|---|
| 419 |
|
|---|
| 420 | sub filter_text {
|
|---|
| 421 | # $self->filter_text ($field, $new_text);
|
|---|
| 422 | # don't want to do anything for this version, however,
|
|---|
| 423 | # in a particular collection you might want to override
|
|---|
| 424 | # this method to post-process certain fields depending on
|
|---|
| 425 | # the field, or whether we are outputting it for indexing
|
|---|
| 426 | }
|
|---|
| 427 |
|
|---|
| 428 | sub text {
|
|---|
| 429 | my $self = shift (@_);
|
|---|
| 430 | my ($doc_obj) = @_;
|
|---|
| 431 | my $handle = $self->{'output_handle'};
|
|---|
| 432 | my $indexed_doc = 1;
|
|---|
| 433 |
|
|---|
| 434 | # only output this document if it is one to be indexed
|
|---|
| 435 | return if ($doc_obj->get_doc_type() ne "indexed_doc");
|
|---|
| 436 |
|
|---|
| 437 | # see if this document belongs to this subcollection
|
|---|
| 438 | foreach my $indexexp (@{$self->{'indexexparr'}}) {
|
|---|
| 439 | $indexed_doc = 0;
|
|---|
| 440 | my ($field, $exp, $options) = split /\//, $indexexp;
|
|---|
| 441 | if (defined ($field) && defined ($exp)) {
|
|---|
| 442 | my ($bool) = $field =~ /^(.)/;
|
|---|
| 443 | $field =~ s/^.// if $bool eq '!';
|
|---|
| 444 | if ($field =~ /^filename$/i) {
|
|---|
| 445 | $field = $doc_obj->get_source_filename();
|
|---|
| 446 | } else {
|
|---|
| 447 | $field = $doc_obj->get_metadata_element($doc_obj->get_top_section(), $field);
|
|---|
| 448 | }
|
|---|
| 449 | next unless defined $field;
|
|---|
| 450 | if ($bool eq '!') {
|
|---|
| 451 | if ($options =~ /^i$/i) {
|
|---|
| 452 | if ($field !~ /$exp/i) {$indexed_doc = 1; last;}
|
|---|
| 453 | } else {
|
|---|
| 454 | if ($field !~ /$exp/) {$indexed_doc = 1; last;}
|
|---|
| 455 | }
|
|---|
| 456 | } else {
|
|---|
| 457 | if ($options =~ /^i$/i) {
|
|---|
| 458 | if ($field =~ /$exp/i) {$indexed_doc = 1; last;}
|
|---|
| 459 | } else {
|
|---|
| 460 | if ($field =~ /$exp/) {$indexed_doc = 1; last;}
|
|---|
| 461 | }
|
|---|
| 462 | }
|
|---|
| 463 | }
|
|---|
| 464 | }
|
|---|
| 465 |
|
|---|
| 466 | # this is another document
|
|---|
| 467 | $self->{'num_docs'} += 1;
|
|---|
| 468 |
|
|---|
| 469 | # get the parameters for the output
|
|---|
| 470 | my ($level, $fields) = split (/:/, $self->{'index'});
|
|---|
| 471 | $fields =~ s/\ball\b/Title,Creator,text/;
|
|---|
| 472 | $fields =~ s/\btopall\b/topTitle,topCreator,toptext/;
|
|---|
| 473 |
|
|---|
| 474 | my $doc_section = 0; # just for this document
|
|---|
| 475 | my $text = "";
|
|---|
| 476 | my $text_extra = "";
|
|---|
| 477 |
|
|---|
| 478 | # get the text for this document
|
|---|
| 479 | my $section = $doc_obj->get_top_section();
|
|---|
| 480 | while (defined $section) {
|
|---|
| 481 | # update a few statistics
|
|---|
| 482 | $doc_section++;
|
|---|
| 483 | $self->{'num_sections'} += 1;
|
|---|
| 484 |
|
|---|
| 485 | if ($indexed_doc) {
|
|---|
| 486 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
|---|
| 487 | foreach my $field (split (/,/, $fields)) {
|
|---|
| 488 | # only deal with this field if it doesn't start with top or
|
|---|
| 489 | # this is the first section
|
|---|
| 490 | my $real_field = $field;
|
|---|
| 491 | if (!($real_field =~ s/^top//) || ($doc_section == 1)) {
|
|---|
| 492 | my $new_text = "";
|
|---|
| 493 | if ($level eq "dummy") {
|
|---|
| 494 | # a dummy index is a special case used when no
|
|---|
| 495 | # indexes are specified (since there must always be
|
|---|
| 496 | # at least one index or we can't retrieve the
|
|---|
| 497 | # compressed text) - we add a small amount of text
|
|---|
| 498 | # to these dummy indexes which will never be seen
|
|---|
| 499 | # but will overcome mg's problems with building
|
|---|
| 500 | # empty indexes
|
|---|
| 501 | $new_text = "this is dummy text to stop mg barfing";
|
|---|
| 502 | $self->{'num_processed_bytes'} += length ($new_text);
|
|---|
| 503 |
|
|---|
| 504 | } elsif ($real_field eq "text") {
|
|---|
| 505 | $new_text = $doc_obj->get_text ($section) if $self->{'store_text'};
|
|---|
| 506 | $self->{'num_processed_bytes'} += length ($new_text);
|
|---|
| 507 | $new_text =~ s/[\cB\cC]//g;
|
|---|
| 508 | $self->find_paragraphs($new_text);
|
|---|
| 509 |
|
|---|
| 510 | } else {
|
|---|
| 511 | my $first = 1;
|
|---|
| 512 | foreach $meta (@{$doc_obj->get_metadata ($section, $real_field)}) {
|
|---|
| 513 | $meta =~ s/[\cB\cC]//g;
|
|---|
| 514 | $self->{'num_processed_bytes'} += length ($meta);
|
|---|
| 515 | $new_text .= "\cC" unless $first;
|
|---|
| 516 | $new_text .= $meta if $self->{'store_text'};
|
|---|
| 517 | $first = 0;
|
|---|
| 518 | }
|
|---|
| 519 | }
|
|---|
| 520 |
|
|---|
| 521 | # filter the text
|
|---|
| 522 | $self->filter_text ($field, $new_text);
|
|---|
| 523 |
|
|---|
| 524 | $text .= "$new_text\cC";
|
|---|
| 525 | }
|
|---|
| 526 | }
|
|---|
| 527 | }
|
|---|
| 528 |
|
|---|
| 529 | if ($level eq "document") { $text_extra .= "\cB"; }
|
|---|
| 530 | else { $text .= "\cB"; }
|
|---|
| 531 |
|
|---|
| 532 | $section = $doc_obj->get_next_section($section);
|
|---|
| 533 | }
|
|---|
| 534 |
|
|---|
| 535 | print $handle "$text$text_extra";
|
|---|
| 536 | }
|
|---|
| 537 |
|
|---|
| 538 | 1;
|
|---|
| 539 |
|
|---|
