| 1 | ###########################################################################
|
|---|
| 2 | #
|
|---|
| 3 | # mgppbuildproc.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 mgpp to process
|
|---|
| 28 |
|
|---|
| 29 |
|
|---|
| 30 | package mgppbuildproc;
|
|---|
| 31 |
|
|---|
| 32 | eval {require bytes};
|
|---|
| 33 |
|
|---|
| 34 | use classify;
|
|---|
| 35 | use doc;
|
|---|
| 36 | use docproc;
|
|---|
| 37 | use util;
|
|---|
| 38 |
|
|---|
| 39 |
|
|---|
| 40 | BEGIN {
|
|---|
| 41 | @ISA = ('docproc');
|
|---|
| 42 | }
|
|---|
| 43 |
|
|---|
| 44 |
|
|---|
| 45 | sub new {
|
|---|
| 46 | my ($class, $collection, $source_dir, $build_dir,
|
|---|
| 47 | $verbosity, $outhandle) = @_;
|
|---|
| 48 | my $self = new docproc ();
|
|---|
| 49 |
|
|---|
| 50 | # outhandle is where all the debugging info goes
|
|---|
| 51 | # output_handle is where the output of the plugins is piped
|
|---|
| 52 | # to (i.e. mg, gdbm etc.)
|
|---|
| 53 | $outhandle = STDERR unless defined $outhandle;
|
|---|
| 54 |
|
|---|
| 55 | $self->{'collection'} = $collection;
|
|---|
| 56 | $self->{'source_dir'} = $source_dir;
|
|---|
| 57 | $self->{'build_dir'} = $build_dir;
|
|---|
| 58 | $self->{'verbosity'} = $verbosity;
|
|---|
| 59 | $self->{'classifiers'} = [];
|
|---|
| 60 | $self->{'mode'} = "text";
|
|---|
| 61 | $self->{'assocdir'} = $build_dir;
|
|---|
| 62 | $self->{'dontgdbm'} = {};
|
|---|
| 63 | $self->{'index'} = "text";
|
|---|
| 64 | $self->{'indexexparr'} = [];
|
|---|
| 65 | $self->{'output_handle'} = "STDOUT";
|
|---|
| 66 | $self->{'num_docs'} = 0;
|
|---|
| 67 | $self->{'num_sections'} = 0;
|
|---|
| 68 | $self->{'num_bytes'} = 0;
|
|---|
| 69 | $self->{'num_processed_bytes'} = 0;
|
|---|
| 70 | $self->{'store_text'} = 1;
|
|---|
| 71 | $self->{'outhandle'} = $outhandle;
|
|---|
| 72 |
|
|---|
| 73 | #used by browse interface
|
|---|
| 74 | $self->{'doclist'} = [];
|
|---|
| 75 |
|
|---|
| 76 | $self->{'indexing_text'} = 0;
|
|---|
| 77 |
|
|---|
| 78 | #new ones for mgpp
|
|---|
| 79 | $self->{'dontindex'} = {};
|
|---|
| 80 | $self->{'indexfieldmap'} = {};
|
|---|
| 81 | $self->{'indexfields'} = {};
|
|---|
| 82 | $self->{'strip_html'}=1;
|
|---|
| 83 |
|
|---|
| 84 |
|
|---|
| 85 | return bless $self, $class;
|
|---|
| 86 | }
|
|---|
| 87 |
|
|---|
| 88 | sub reset {
|
|---|
| 89 | my $self = shift (@_);
|
|---|
| 90 |
|
|---|
| 91 | $self->{'num_docs'} = 0;
|
|---|
| 92 | $self->{'num_sections'} = 0;
|
|---|
| 93 | $self->{'num_processed_bytes'} = 0;
|
|---|
| 94 | $self->{'num_bytes'} = 0;
|
|---|
| 95 | }
|
|---|
| 96 |
|
|---|
| 97 | sub get_num_docs {
|
|---|
| 98 | my $self = shift (@_);
|
|---|
| 99 |
|
|---|
| 100 | return $self->{'num_docs'};
|
|---|
| 101 | }
|
|---|
| 102 |
|
|---|
| 103 | sub get_num_sections {
|
|---|
| 104 | my $self = shift (@_);
|
|---|
| 105 |
|
|---|
| 106 | return $self->{'num_sections'};
|
|---|
| 107 | }
|
|---|
| 108 |
|
|---|
| 109 | # num_bytes is the actual number of bytes in the collection
|
|---|
| 110 | # this is normally the same as what's processed during text compression
|
|---|
| 111 | sub get_num_bytes {
|
|---|
| 112 | my $self = shift (@_);
|
|---|
| 113 |
|
|---|
| 114 | return $self->{'num_bytes'};
|
|---|
| 115 | }
|
|---|
| 116 |
|
|---|
| 117 | # num_processed_bytes is the number of bytes actually passed
|
|---|
| 118 | # to mgpp for the current index
|
|---|
| 119 | sub get_num_processed_bytes {
|
|---|
| 120 | my $self = shift (@_);
|
|---|
| 121 |
|
|---|
| 122 | return $self->{'num_processed_bytes'};
|
|---|
| 123 | }
|
|---|
| 124 |
|
|---|
| 125 | sub set_output_handle {
|
|---|
| 126 | my $self = shift (@_);
|
|---|
| 127 | my ($handle) = @_;
|
|---|
| 128 |
|
|---|
| 129 | $self->{'output_handle'} = $handle;
|
|---|
| 130 | }
|
|---|
| 131 |
|
|---|
| 132 | sub set_mode {
|
|---|
| 133 | my $self = shift (@_);
|
|---|
| 134 | my ($mode) = @_;
|
|---|
| 135 |
|
|---|
| 136 | $self->{'mode'} = $mode;
|
|---|
| 137 | }
|
|---|
| 138 |
|
|---|
| 139 | sub set_assocdir {
|
|---|
| 140 | my $self = shift (@_);
|
|---|
| 141 | my ($assocdir) = @_;
|
|---|
| 142 |
|
|---|
| 143 | $self->{'assocdir'} = $assocdir;
|
|---|
| 144 | }
|
|---|
| 145 |
|
|---|
| 146 | sub set_dontgdbm {
|
|---|
| 147 | my $self = shift (@_);
|
|---|
| 148 | my ($dontgdbm) = @_;
|
|---|
| 149 |
|
|---|
| 150 | $self->{'dontgdbm'} = $dontgdbm;
|
|---|
| 151 | }
|
|---|
| 152 |
|
|---|
| 153 | sub set_index {
|
|---|
| 154 | my $self = shift (@_);
|
|---|
| 155 | my ($index, $indexexparr) = @_;
|
|---|
| 156 |
|
|---|
| 157 | $self->{'index'} = $index;
|
|---|
| 158 | $self->{'indexexparr'} = $indexexparr if defined $indexexparr;
|
|---|
| 159 | }
|
|---|
| 160 |
|
|---|
| 161 | sub get_index {
|
|---|
| 162 | my $self = shift (@_);
|
|---|
| 163 |
|
|---|
| 164 | return $self->{'index'};
|
|---|
| 165 | }
|
|---|
| 166 |
|
|---|
| 167 | sub set_classifiers {
|
|---|
| 168 | my $self = shift (@_);
|
|---|
| 169 | my ($classifiers) = @_;
|
|---|
| 170 |
|
|---|
| 171 | $self->{'classifiers'} = $classifiers;
|
|---|
| 172 | }
|
|---|
| 173 |
|
|---|
| 174 | sub set_indexing_text {
|
|---|
| 175 | my $self = shift (@_);
|
|---|
| 176 | my ($indexing_text) = @_;
|
|---|
| 177 |
|
|---|
| 178 | $self->{'indexing_text'} = $indexing_text;
|
|---|
| 179 | }
|
|---|
| 180 |
|
|---|
| 181 | sub get_indexing_text {
|
|---|
| 182 | my $self = shift (@_);
|
|---|
| 183 |
|
|---|
| 184 | return $self->{'indexing_text'};
|
|---|
| 185 | }
|
|---|
| 186 |
|
|---|
| 187 | sub set_store_text {
|
|---|
| 188 | my $self = shift (@_);
|
|---|
| 189 | my ($store_text) = @_;
|
|---|
| 190 |
|
|---|
| 191 | $self->{'store_text'} = $store_text;
|
|---|
| 192 | }
|
|---|
| 193 |
|
|---|
| 194 | sub get_doc_list {
|
|---|
| 195 | my $self = shift(@_);
|
|---|
| 196 |
|
|---|
| 197 | return @{$self->{'doclist'}};
|
|---|
| 198 | }
|
|---|
| 199 |
|
|---|
| 200 | sub set_indexfieldmap {
|
|---|
| 201 | my $self = shift (@_);
|
|---|
| 202 | my ($indexmap) = @_;
|
|---|
| 203 |
|
|---|
| 204 | $self->{'indexfieldmap'} = $indexmap;
|
|---|
| 205 | }
|
|---|
| 206 |
|
|---|
| 207 | sub get_indexfieldmap {
|
|---|
| 208 | my $self = shift (@_);
|
|---|
| 209 |
|
|---|
| 210 | return $self->{'indexfieldmap'};
|
|---|
| 211 | }
|
|---|
| 212 |
|
|---|
| 213 | sub set_levels {
|
|---|
| 214 | my $self = shift (@_);
|
|---|
| 215 | my ($levels) = @_;
|
|---|
| 216 |
|
|---|
| 217 | $self->{'levels'} = $levels;
|
|---|
| 218 | }
|
|---|
| 219 |
|
|---|
| 220 | sub set_strip_html {
|
|---|
| 221 | my $self = shift (@_);
|
|---|
| 222 | my ($strip) = @_;
|
|---|
| 223 | $self->{'strip_html'}=$strip;
|
|---|
| 224 | }
|
|---|
| 225 |
|
|---|
| 226 | sub process {
|
|---|
| 227 | my $self = shift (@_);
|
|---|
| 228 | my $method = $self->{'mode'};
|
|---|
| 229 |
|
|---|
| 230 | $self->$method(@_);
|
|---|
| 231 | }
|
|---|
| 232 |
|
|---|
| 233 | # use 'Paged' if document has no more than 2 levels
|
|---|
| 234 | # and each section at second level has a number for
|
|---|
| 235 | # Title metadata
|
|---|
| 236 | sub get_document_type {
|
|---|
| 237 | my $self = shift (@_);
|
|---|
| 238 | my ($doc_obj) = @_;
|
|---|
| 239 |
|
|---|
| 240 | my $thistype = "VList";
|
|---|
| 241 | my $childtype = "VList";
|
|---|
| 242 | my $title;
|
|---|
| 243 | my @tmp = ();
|
|---|
| 244 |
|
|---|
| 245 | my $section = $doc_obj->get_top_section ();
|
|---|
| 246 | my $first = 1;
|
|---|
| 247 | while (defined $section) {
|
|---|
| 248 | @tmp = split /\./, $section;
|
|---|
| 249 | if (scalar(@tmp) > 1) {
|
|---|
| 250 | return ($thistype, $childtype);
|
|---|
| 251 | }
|
|---|
| 252 | if (!$first) {
|
|---|
| 253 | $title = $doc_obj->get_metadata_element ($section, "Title");
|
|---|
| 254 | if (!defined $title || $title !~ /^\d+$/) {
|
|---|
| 255 | return ($thistype, $childtype);
|
|---|
| 256 | }
|
|---|
| 257 | }
|
|---|
| 258 | $first = 0;
|
|---|
| 259 | $section = $doc_obj->get_next_section($section);
|
|---|
| 260 | }
|
|---|
| 261 | if ($doc_obj->get_text_length ($doc_obj->get_top_section())) {
|
|---|
| 262 | $thistype = "Paged";
|
|---|
| 263 | } else {
|
|---|
| 264 | $thistype = "Invisible";
|
|---|
| 265 | }
|
|---|
| 266 | $childtype = "Paged";
|
|---|
| 267 | return ($thistype, $childtype);
|
|---|
| 268 | }
|
|---|
| 269 |
|
|---|
| 270 | sub assoc_files {
|
|---|
| 271 | my $self = shift (@_);
|
|---|
| 272 | my ($doc_obj, $archivedir) = @_;
|
|---|
| 273 | my ($afile);
|
|---|
| 274 |
|
|---|
| 275 | foreach my $assoc_file (@{$doc_obj->get_assoc_files()}) {
|
|---|
| 276 | # if assoc file starts with a slash, we put it relative to the assoc
|
|---|
| 277 | # dir, otherwise it is relative to the HASH... directory
|
|---|
| 278 | if ($assoc_file->[1] =~ m@^[/\\]@) {
|
|---|
| 279 | $afile = &util::filename_cat($self->{'assocdir'},$assoc_file->[1]);
|
|---|
| 280 | } else {
|
|---|
| 281 | $afile = &util::filename_cat($self->{'assocdir'}, $archivedir, $assoc_file->[1]);
|
|---|
| 282 | }
|
|---|
| 283 | &util::hard_link ($assoc_file->[0], $afile);
|
|---|
| 284 | }
|
|---|
| 285 | }
|
|---|
| 286 |
|
|---|
| 287 | sub infodb {
|
|---|
| 288 | my $self = shift (@_);
|
|---|
| 289 | my ($doc_obj, $filename) = @_;
|
|---|
| 290 | my $handle = $self->{'output_handle'};
|
|---|
| 291 |
|
|---|
| 292 | my $doctype = $doc_obj->get_doc_type();
|
|---|
| 293 |
|
|---|
| 294 | # only output this document if it is one to be indexed
|
|---|
| 295 | return if ($doctype ne "indexed_doc");
|
|---|
| 296 |
|
|---|
| 297 | #if a Section level index is not built, the gdbm file should be at doc
|
|---|
| 298 | #level not Section
|
|---|
| 299 | my $docs_only = 1;
|
|---|
| 300 | if ($self->{'levels'}->{'Section'}) {
|
|---|
| 301 | $docs_only = 0;
|
|---|
| 302 | }
|
|---|
| 303 |
|
|---|
| 304 | my ($archivedir) = $filename =~ /^(.*?)(?:\/|\\)[^\/\\]*$/;
|
|---|
| 305 | $archivedir = "" unless defined $archivedir;
|
|---|
| 306 | $archivedir =~ s/\\/\//g;
|
|---|
| 307 | $archivedir =~ s/^\/+//;
|
|---|
| 308 | $archivedir =~ s/\/+$//;
|
|---|
| 309 |
|
|---|
| 310 | $self->assoc_files ($doc_obj, $archivedir);
|
|---|
| 311 |
|
|---|
| 312 | # this is another document
|
|---|
| 313 | $self->{'num_docs'} += 1 unless ($doctype eq "classification");
|
|---|
| 314 |
|
|---|
| 315 | # is this a paged or a hierarchical document
|
|---|
| 316 | my ($thistype, $childtype) = $self->get_document_type ($doc_obj);
|
|---|
| 317 |
|
|---|
| 318 | my $section = $doc_obj->get_top_section ();
|
|---|
| 319 | my $doc_OID = $doc_obj->get_OID();
|
|---|
| 320 | my $first = 1;
|
|---|
| 321 | my $url = "";
|
|---|
| 322 | while (defined $section) {
|
|---|
| 323 | # update a few statistics
|
|---|
| 324 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
|---|
| 325 | $self->{'num_sections'} += 1 unless ($doctype eq "classification");
|
|---|
| 326 |
|
|---|
| 327 | # output the section name
|
|---|
| 328 | if ($section eq "") { print $handle "[$doc_OID]\n"; }
|
|---|
| 329 | else { print $handle "[$doc_OID.$section]\n"; }
|
|---|
| 330 |
|
|---|
| 331 | # output the fact that this document is a document (unless doctype
|
|---|
| 332 | # has been set to something else from within a plugin
|
|---|
| 333 | my $dtype = $doc_obj->get_metadata_element ($section, "doctype");
|
|---|
| 334 | if (!defined $dtype || $dtype !~ /\w/) {
|
|---|
| 335 | print $handle "<doctype>doc\n";
|
|---|
| 336 | }
|
|---|
| 337 |
|
|---|
| 338 | # output whether this node contains text
|
|---|
| 339 | if ($doc_obj->get_text_length($section) > 0) {
|
|---|
| 340 | print $handle "<hastxt>1\n";
|
|---|
| 341 | } else {
|
|---|
| 342 | print $handle "<hastxt>0\n";
|
|---|
| 343 | }
|
|---|
| 344 |
|
|---|
| 345 | # output all the section metadata
|
|---|
| 346 | my $metadata = $doc_obj->get_all_metadata ($section);
|
|---|
| 347 | foreach my $pair (@$metadata) {
|
|---|
| 348 | my ($field, $value) = (@$pair);
|
|---|
| 349 |
|
|---|
| 350 | if ($field !~ /^(?:Identifier|Encoding|Language|gsdl)/
|
|---|
| 351 | && defined $value && $value ne "") {
|
|---|
| 352 |
|
|---|
| 353 | # escape problematic stuff
|
|---|
| 354 | $value =~ s/\\/\\\\/g;
|
|---|
| 355 | $value =~ s/\n/\\n/g;
|
|---|
| 356 | $value =~ s/\r/\\r/g;
|
|---|
| 357 |
|
|---|
| 358 | # special case for URL metadata
|
|---|
| 359 | if ($field =~ /^URL$/i) {
|
|---|
| 360 | $url .= "[$value]\n";
|
|---|
| 361 | if ($section eq "") {$url .= "<section>$doc_OID\n";}
|
|---|
| 362 | else {$url .= "<section>$doc_OID.$section\n";}
|
|---|
| 363 | $url .= '-' x 70 . "\n";
|
|---|
| 364 | }
|
|---|
| 365 |
|
|---|
| 366 | if (!defined $self->{'dontgdbm'}->{$field}) {
|
|---|
| 367 | print $handle "<$field>$value\n";
|
|---|
| 368 | }
|
|---|
| 369 | }
|
|---|
| 370 | }
|
|---|
| 371 |
|
|---|
| 372 | # output archivedir if at top level
|
|---|
| 373 | if ($section eq $doc_obj->get_top_section()) {
|
|---|
| 374 | print $handle "<archivedir>$archivedir\n";
|
|---|
| 375 | }
|
|---|
| 376 |
|
|---|
| 377 | # output document display type
|
|---|
| 378 | if ($first) {
|
|---|
| 379 | print $handle "<thistype>$thistype\n";
|
|---|
| 380 | }
|
|---|
| 381 |
|
|---|
| 382 | if (!$docs_only) {
|
|---|
| 383 | # output a list of children
|
|---|
| 384 | my $children = $doc_obj->get_children ($section);
|
|---|
| 385 | if (scalar(@$children) > 0) {
|
|---|
| 386 | print $handle "<childtype>$childtype\n";
|
|---|
| 387 | print $handle "<contains>";
|
|---|
| 388 | my $firstchild = 1;
|
|---|
| 389 | foreach $child (@$children) {
|
|---|
| 390 | print $handle ";" unless $firstchild;
|
|---|
| 391 | $firstchild = 0;
|
|---|
| 392 | if ($child =~ /^.*?\.(\d+)$/) {
|
|---|
| 393 | print $handle "\".$1";
|
|---|
| 394 | } else {
|
|---|
| 395 | print $handle "\".$child";
|
|---|
| 396 | }
|
|---|
| 397 | # if ($child eq "") { print $handle "$doc_OID"; }
|
|---|
| 398 | # elsif ($section eq "") { print $handle "$doc_OID.$child"; }
|
|---|
| 399 | # else { print $handle "$doc_OID.$section.$child"; }
|
|---|
| 400 | }
|
|---|
| 401 | print $handle "\n";
|
|---|
| 402 | }
|
|---|
| 403 | #output the matching doc number
|
|---|
| 404 | print $handle "<docnum>$self->{'num_sections'}\n";
|
|---|
| 405 |
|
|---|
| 406 | } # if (!$docs_only)
|
|---|
| 407 | else { #docs only, doc num is num_docs not num_sections
|
|---|
| 408 | # output the matching document number
|
|---|
| 409 | print $handle "<docnum>$self->{'num_docs'}\n";
|
|---|
| 410 | }
|
|---|
| 411 |
|
|---|
| 412 | print $handle '-' x 70, "\n";
|
|---|
| 413 |
|
|---|
| 414 |
|
|---|
| 415 | # output a database entry for the document number
|
|---|
| 416 | if ($docs_only) {
|
|---|
| 417 | print $handle "[$self->{'num_docs'}]\n";
|
|---|
| 418 | print $handle "<section>$doc_OID\n";
|
|---|
| 419 | }
|
|---|
| 420 | else {
|
|---|
| 421 | print $handle "[$self->{'num_sections'}]\n";
|
|---|
| 422 | if ($section eq "") { print $handle "<section>$doc_OID\n"; }
|
|---|
| 423 | else { print $handle "<section>$doc_OID.$section\n"; }
|
|---|
| 424 | }
|
|---|
| 425 | print $handle '-' x 70, "\n";
|
|---|
| 426 |
|
|---|
| 427 | # output entry for url
|
|---|
| 428 | if ($url ne "") {
|
|---|
| 429 | print $handle $url;
|
|---|
| 430 | }
|
|---|
| 431 |
|
|---|
| 432 | $first = 0;
|
|---|
| 433 | $section = $doc_obj->get_next_section($section);
|
|---|
| 434 | last if ($docs_only); # if no sections wanted, only gdbm the docs
|
|---|
| 435 | }
|
|---|
| 436 |
|
|---|
| 437 | #add this document to the browse structure
|
|---|
| 438 | push(@{$self->{'doclist'}},$doc_obj->get_OID())
|
|---|
| 439 | unless ($doctype eq "classification");
|
|---|
| 440 |
|
|---|
| 441 | # classify this document
|
|---|
| 442 | &classify::classify_doc ($self->{'classifiers'}, $doc_obj);
|
|---|
| 443 |
|
|---|
| 444 | }
|
|---|
| 445 |
|
|---|
| 446 | sub find_paragraphs {
|
|---|
| 447 | $_[1] =~ s/(<p\b)/<Paragraph>$1/gi;
|
|---|
| 448 | }
|
|---|
| 449 |
|
|---|
| 450 | #this function strips the html tags from the doc if ($strip_html) and
|
|---|
| 451 | # if ($para) replaces <p> with <Paragraph> tags.
|
|---|
| 452 | # if both are false, the original text is returned
|
|---|
| 453 | #assumes that <pre> and </pre> have no spaces, and removes all < and > inside
|
|---|
| 454 | #these tags
|
|---|
| 455 | sub preprocess_text {
|
|---|
| 456 | my $self = shift (@_);
|
|---|
| 457 | my ($text, $strip_html, $para) = @_;
|
|---|
| 458 |
|
|---|
| 459 | my ($outtext) = "";
|
|---|
| 460 | if ($strip_html) {
|
|---|
| 461 | while ($text =~ /<([^>]*)>/ && $text ne "") {
|
|---|
| 462 |
|
|---|
| 463 | $tag = $1;
|
|---|
| 464 | $outtext .= $`." "; #add everything before the matched tag
|
|---|
| 465 | $text = $'; #everything after the matched tag
|
|---|
| 466 | if ($para && $tag =~ /^\s*p\s/) {
|
|---|
| 467 | $outtext .= "<Paragraph> ";
|
|---|
| 468 | }
|
|---|
| 469 | elsif ($tag =~ /^pre$/) { # a pre tag
|
|---|
| 470 | $text =~ /<\/pre>/; # find the closing pre tag
|
|---|
| 471 | my $tmp_text = $`; #everything before the closing pre tag
|
|---|
| 472 | $text = $'; #everything after the </pre>
|
|---|
| 473 | $tmp_text =~ s/[<>]//g; # remove all < and >
|
|---|
| 474 | $outtext.= $tmp_text . " ";
|
|---|
| 475 | }
|
|---|
| 476 | }
|
|---|
| 477 |
|
|---|
| 478 | $outtext .= $text; # add any remaining text
|
|---|
| 479 | return $outtext;
|
|---|
| 480 | } #if strip_html
|
|---|
| 481 |
|
|---|
| 482 | if ($para) {
|
|---|
| 483 | $text =~ s/(<p\b)/<Paragraph>$1/gi;
|
|---|
| 484 | return $text;
|
|---|
| 485 | }
|
|---|
| 486 | return $text;
|
|---|
| 487 | }
|
|---|
| 488 |
|
|---|
| 489 |
|
|---|
| 490 |
|
|---|
| 491 | sub filter_text {
|
|---|
| 492 | # $self->filter_text ($field, $new_text);
|
|---|
| 493 | # don't want to do anything for this version, however,
|
|---|
| 494 | # in a particular collection you might want to override
|
|---|
| 495 | # this method to post-process certain fields depending on
|
|---|
| 496 | # the field, or whether we are outputting it for indexing
|
|---|
| 497 | }
|
|---|
| 498 |
|
|---|
| 499 | sub text {
|
|---|
| 500 | my $self = shift (@_);
|
|---|
| 501 | my ($doc_obj) = @_;
|
|---|
| 502 | my $handle = $self->{'output_handle'};
|
|---|
| 503 | my $indexed_doc = 1;
|
|---|
| 504 |
|
|---|
| 505 | # only output this document if it is one to be indexed
|
|---|
| 506 | return if ($doc_obj->get_doc_type() ne "indexed_doc");
|
|---|
| 507 |
|
|---|
| 508 | # see if this document belongs to this subcollection
|
|---|
| 509 | foreach my $indexexp (@{$self->{'indexexparr'}}) {
|
|---|
| 510 | $indexed_doc = 0;
|
|---|
| 511 | my ($field, $exp, $options) = split /\//, $indexexp;
|
|---|
| 512 | if (defined ($field) && defined ($exp)) {
|
|---|
| 513 | my ($bool) = $field =~ /^(.)/;
|
|---|
| 514 | $field =~ s/^.// if $bool eq '!';
|
|---|
| 515 | if ($field =~ /^filename$/i) {
|
|---|
| 516 | $field = $doc_obj->get_source_filename();
|
|---|
| 517 | } else {
|
|---|
| 518 | $field = $doc_obj->get_metadata_element($doc_obj->get_top_section(), $field);
|
|---|
| 519 | }
|
|---|
| 520 | next unless defined $field;
|
|---|
| 521 | if ($bool eq '!') {
|
|---|
| 522 | if ($options =~ /^i$/i) {
|
|---|
| 523 | if ($field !~ /$exp/i) {$indexed_doc = 1; last;}
|
|---|
| 524 | } else {
|
|---|
| 525 | if ($field !~ /$exp/) {$indexed_doc = 1; last;}
|
|---|
| 526 | }
|
|---|
| 527 | } else {
|
|---|
| 528 | if ($options =~ /^i$/i) {
|
|---|
| 529 | if ($field =~ /$exp/i) {$indexed_doc = 1; last;}
|
|---|
| 530 | } else {
|
|---|
| 531 | if ($field =~ /$exp/) {$indexed_doc = 1; last;}
|
|---|
| 532 | }
|
|---|
| 533 | }
|
|---|
| 534 | }
|
|---|
| 535 | }
|
|---|
| 536 |
|
|---|
| 537 | # this is another document
|
|---|
| 538 | $self->{'num_docs'} += 1;
|
|---|
| 539 |
|
|---|
| 540 | # get the parameters for the output
|
|---|
| 541 | my ($fields) = $self->{'index'};
|
|---|
| 542 |
|
|---|
| 543 | my ($sectiontag) = "";
|
|---|
| 544 | if ($self->{'levels'}->{'Section'}) {
|
|---|
| 545 | $sectiontag = "\n<Section>\n";
|
|---|
| 546 | }
|
|---|
| 547 | my ($paratag) = "";
|
|---|
| 548 | if ($self->{'levels'}->{'Paragraph'}) {
|
|---|
| 549 | $paratag = "<Paragraph>";
|
|---|
| 550 | }
|
|---|
| 551 | my $doc_section = 0; # just for this document
|
|---|
| 552 | my $text = "<Document>\n";
|
|---|
| 553 |
|
|---|
| 554 | # get the text for this document
|
|---|
| 555 | my $section = $doc_obj->get_top_section();
|
|---|
| 556 | while (defined $section) {
|
|---|
| 557 | # update a few statistics
|
|---|
| 558 | $doc_section++;
|
|---|
| 559 | $self->{'num_sections'} += 1;
|
|---|
| 560 | $text .= $sectiontag;
|
|---|
| 561 |
|
|---|
| 562 | if ($indexed_doc) {
|
|---|
| 563 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
|---|
| 564 | foreach my $field (split (/,/, $fields)) {
|
|---|
| 565 | # only deal with this field if it doesn't start with top or
|
|---|
| 566 | # this is the first section
|
|---|
| 567 | my $real_field = $field;
|
|---|
| 568 | if (!($real_field =~ s/^top//) || ($doc_section == 1)) {
|
|---|
| 569 | my $new_text = "";
|
|---|
| 570 | my $tmp_text = "";
|
|---|
| 571 | if ($real_field eq "text") {
|
|---|
| 572 | if ($self->{'indexing_text'}) { #tag the text with <Text>...</Text>, add the <Paragraph> tags and strip out html if needed
|
|---|
| 573 | $new_text .= "<TX>\n";
|
|---|
| 574 | $tmp_text .= $doc_obj->get_text ($section) if $self->{'store_text'};
|
|---|
| 575 | $tmp_text = $self->preprocess_text($tmp_text, $self->{'strip_html'}, $self->{'levels'}->{'Paragraph'});
|
|---|
| 576 |
|
|---|
| 577 | $new_text .= "$tmp_text</TX>\n";
|
|---|
| 578 | if (!defined $self->{'indexfields'}->{'TextOnly'}) {
|
|---|
| 579 | $self->{'indexfields'}->{'TextOnly'} = 1;
|
|---|
| 580 | }
|
|---|
| 581 | }
|
|---|
| 582 | else { # leave html stuff in, and dont add Paragraph tags - never retrieve paras at the moment
|
|---|
| 583 | $new_text .= $doc_obj->get_text ($section) if $self->{'store_text'};
|
|---|
| 584 | #if ($self->{'levels'}->{'Paragraph'}) {
|
|---|
| 585 | #$self->find_paragraphs($new_text);
|
|---|
| 586 | #}
|
|---|
| 587 | }
|
|---|
| 588 | } else { # metadata field
|
|---|
| 589 | if ($real_field eq "metadata") { # insert all metadata
|
|---|
| 590 | #except gsdl stuff
|
|---|
| 591 | my $shortname = "";
|
|---|
| 592 | my $metadata = $doc_obj->get_all_metadata ($section);
|
|---|
| 593 | foreach $pair (@$metadata) {
|
|---|
| 594 | my ($mfield, $mvalue) = (@$pair);
|
|---|
| 595 | # check fields here, maybe others dont want - change to use dontindex!!
|
|---|
| 596 | if ($mfield !~ /^(?:Identifier|Encoding|Language|gsdl)/
|
|---|
| 597 | && $mfield ne "classifytype"
|
|---|
| 598 | && $mfield ne "assocfilepath"
|
|---|
| 599 | && defined $mvalue && $mvalue ne "") {
|
|---|
| 600 |
|
|---|
| 601 | if (defined $self->{'indexfieldmap'}->{$mfield}) {
|
|---|
| 602 | $shortname = $self->{'indexfieldmap'}->{$mfield};
|
|---|
| 603 | }
|
|---|
| 604 | else {
|
|---|
| 605 | $shortname = $self->create_shortname($mfield);
|
|---|
| 606 | $self->{'indexfieldmap'}->{$mfield} = $shortname;
|
|---|
| 607 | $self->{'indexfieldmap'}->{$shortname} = 1;
|
|---|
| 608 | }
|
|---|
| 609 | $new_text .= "$paratag<$shortname>$mvalue</$shortname>\n";
|
|---|
| 610 | if (!defined $self->{'indexfields'}->{$mfield}) {
|
|---|
| 611 | $self->{'indexfields'}->{$mfield} = 1;
|
|---|
| 612 | }
|
|---|
| 613 | }
|
|---|
| 614 | }
|
|---|
| 615 |
|
|---|
| 616 | }
|
|---|
| 617 | else { #individual metadata specified
|
|---|
| 618 | my $shortname="";
|
|---|
| 619 | if (!defined $self->{'indexfields'}->{$real_field}) {
|
|---|
| 620 | $self->{'indexfields'}->{$real_field} = 1;
|
|---|
| 621 | }
|
|---|
| 622 | if (defined $self->{'indexfieldmap'}->{$real_field}) {
|
|---|
| 623 | $shortname = $self->{'indexfieldmap'}->{$real_field};
|
|---|
| 624 | }
|
|---|
| 625 | else {
|
|---|
| 626 | $shortname = $self->create_shortname($real_field);
|
|---|
| 627 | $self->{'indexfieldmap'}->{$real_field} = $shortname;
|
|---|
| 628 | $self->{'indexfieldmap'}->{$shortname} = 1;
|
|---|
| 629 | }
|
|---|
| 630 | foreach $item (@{$doc_obj->get_metadata ($section, $real_field)}) {
|
|---|
| 631 | $new_text .= "$paratag<$shortname>$item</$shortname>\n";
|
|---|
| 632 | }
|
|---|
| 633 | }
|
|---|
| 634 |
|
|---|
| 635 | }
|
|---|
| 636 |
|
|---|
| 637 | # filter the text
|
|---|
| 638 | $self->filter_text ($field, $new_text);
|
|---|
| 639 |
|
|---|
| 640 | $self->{'num_processed_bytes'} += length ($new_text);
|
|---|
| 641 | $text .= "$new_text";
|
|---|
| 642 | }
|
|---|
| 643 | }
|
|---|
| 644 | } # if (indexed_doc)
|
|---|
| 645 |
|
|---|
| 646 | $section = $doc_obj->get_next_section($section);
|
|---|
| 647 | } #while defined section
|
|---|
| 648 | print $handle "$text\n</Document>\n";
|
|---|
| 649 | }
|
|---|
| 650 |
|
|---|
| 651 | #chooses the first two letters or digits for the shortname
|
|---|
| 652 | #now ignores non-letdig characters
|
|---|
| 653 | sub create_shortname {
|
|---|
| 654 | $self = shift(@_);
|
|---|
| 655 |
|
|---|
| 656 | my ($realname) = @_;
|
|---|
| 657 | #take the first two chars
|
|---|
| 658 | my $shortname;
|
|---|
| 659 | if ($realname =~ /^[^\w]*(\w)[^\w]*(\w)/) {
|
|---|
| 660 | $shortname = "$1$2";
|
|---|
| 661 | } else {
|
|---|
| 662 | # there aren't two letdig's in the field - try arbitrary combinations
|
|---|
| 663 | $realname = "ABCDEFGHIJKLMNOPQRSTUVWXYZ";
|
|---|
| 664 | $shortname = "AB";
|
|---|
| 665 | }
|
|---|
| 666 | $shortname =~ tr/a-z/A-Z/;
|
|---|
| 667 |
|
|---|
| 668 | #if already used, take the first and third letdigs and so on
|
|---|
| 669 | $count = 1;
|
|---|
| 670 | while (defined $self->{'indexfieldmap'}->{$shortname}) {
|
|---|
| 671 | if ($realname =~ /^[^\w]*(\w)([^\w]*\w){$count}[^\w]*(\w)/) {
|
|---|
| 672 | $shortname = "$1$3";
|
|---|
| 673 | $count++;
|
|---|
| 674 | $shortname =~ tr/a-z/A-Z/;
|
|---|
| 675 |
|
|---|
| 676 | }
|
|---|
| 677 | else {
|
|---|
| 678 | #remove up to and incl the first letdig
|
|---|
| 679 | $realname =~ s/^[^\w]*\w//;
|
|---|
| 680 | $count = 0;
|
|---|
| 681 | }
|
|---|
| 682 | }
|
|---|
| 683 |
|
|---|
| 684 | return $shortname;
|
|---|
| 685 | }
|
|---|
| 686 |
|
|---|
| 687 | 1;
|
|---|
| 688 |
|
|---|
