[932] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
[1852] | 3 | # mgppbuildproc.pm --
|
---|
[932] | 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
|
---|
[1852] | 27 | # for mgpp to process
|
---|
[932] | 28 |
|
---|
| 29 |
|
---|
| 30 | package mgppbuildproc;
|
---|
| 31 |
|
---|
| 32 | use classify;
|
---|
| 33 | use doc;
|
---|
| 34 | use docproc;
|
---|
| 35 | use util;
|
---|
| 36 |
|
---|
| 37 |
|
---|
| 38 | BEGIN {
|
---|
| 39 | @ISA = ('docproc');
|
---|
| 40 | }
|
---|
| 41 |
|
---|
| 42 |
|
---|
| 43 | sub new {
|
---|
[1694] | 44 | my ($class, $collection, $source_dir, $build_dir,
|
---|
| 45 | $verbosity, $outhandle) = @_;
|
---|
[932] | 46 | my $self = new docproc ();
|
---|
| 47 |
|
---|
[1694] | 48 | # outhandle is where all the debugging info goes
|
---|
| 49 | # output_handle is where the output of the plugins is piped
|
---|
| 50 | # to (i.e. mg, gdbm etc.)
|
---|
| 51 | $outhandle = STDERR unless defined $outhandle;
|
---|
| 52 |
|
---|
[932] | 53 | $self->{'collection'} = $collection;
|
---|
| 54 | $self->{'source_dir'} = $source_dir;
|
---|
| 55 | $self->{'build_dir'} = $build_dir;
|
---|
| 56 | $self->{'verbosity'} = $verbosity;
|
---|
| 57 | $self->{'classifiers'} = [];
|
---|
| 58 | $self->{'mode'} = "text";
|
---|
| 59 | $self->{'assocdir'} = $build_dir;
|
---|
| 60 | $self->{'dontgdbm'} = {};
|
---|
| 61 | $self->{'index'} = "text";
|
---|
| 62 | $self->{'indexexparr'} = [];
|
---|
| 63 | $self->{'output_handle'} = "STDOUT";
|
---|
| 64 | $self->{'num_docs'} = 0;
|
---|
| 65 | $self->{'num_sections'} = 0;
|
---|
| 66 | $self->{'num_bytes'} = 0;
|
---|
[1694] | 67 | $self->{'num_processed_bytes'} = 0;
|
---|
[2771] | 68 | $self->{'store_text'} = 1;
|
---|
[1694] | 69 | $self->{'outhandle'} = $outhandle;
|
---|
[2771] | 70 |
|
---|
| 71 | #used by browse interface
|
---|
| 72 | $self->{'doclist'} = [];
|
---|
| 73 |
|
---|
| 74 | $self->{'indexing_text'} = 0;
|
---|
| 75 |
|
---|
| 76 | #new ones for mgpp
|
---|
[1852] | 77 | $self->{'dontindex'} = {};
|
---|
| 78 | $self->{'indexfieldmap'} = {};
|
---|
[1772] | 79 | $self->{'indexfields'} = {};
|
---|
[1852] | 80 | $self->{'strip_html'}=1;
|
---|
[932] | 81 |
|
---|
[1852] | 82 |
|
---|
[932] | 83 | return bless $self, $class;
|
---|
| 84 | }
|
---|
| 85 |
|
---|
| 86 | sub reset {
|
---|
| 87 | my $self = shift (@_);
|
---|
| 88 |
|
---|
| 89 | $self->{'num_docs'} = 0;
|
---|
| 90 | $self->{'num_sections'} = 0;
|
---|
[1694] | 91 | $self->{'num_processed_bytes'} = 0;
|
---|
[932] | 92 | $self->{'num_bytes'} = 0;
|
---|
| 93 | }
|
---|
| 94 |
|
---|
| 95 | sub get_num_docs {
|
---|
| 96 | my $self = shift (@_);
|
---|
| 97 |
|
---|
| 98 | return $self->{'num_docs'};
|
---|
| 99 | }
|
---|
| 100 |
|
---|
| 101 | sub get_num_sections {
|
---|
| 102 | my $self = shift (@_);
|
---|
| 103 |
|
---|
| 104 | return $self->{'num_sections'};
|
---|
| 105 | }
|
---|
| 106 |
|
---|
[1694] | 107 | # num_bytes is the actual number of bytes in the collection
|
---|
| 108 | # this is normally the same as what's processed during text compression
|
---|
[932] | 109 | sub get_num_bytes {
|
---|
| 110 | my $self = shift (@_);
|
---|
| 111 |
|
---|
| 112 | return $self->{'num_bytes'};
|
---|
| 113 | }
|
---|
| 114 |
|
---|
[1694] | 115 | # num_processed_bytes is the number of bytes actually passed
|
---|
| 116 | # to mgpp for the current index
|
---|
| 117 | sub get_num_processed_bytes {
|
---|
| 118 | my $self = shift (@_);
|
---|
| 119 |
|
---|
| 120 | return $self->{'num_processed_bytes'};
|
---|
| 121 | }
|
---|
| 122 |
|
---|
[932] | 123 | sub set_output_handle {
|
---|
| 124 | my $self = shift (@_);
|
---|
| 125 | my ($handle) = @_;
|
---|
| 126 |
|
---|
| 127 | $self->{'output_handle'} = $handle;
|
---|
| 128 | }
|
---|
| 129 |
|
---|
| 130 | sub set_mode {
|
---|
| 131 | my $self = shift (@_);
|
---|
| 132 | my ($mode) = @_;
|
---|
| 133 |
|
---|
| 134 | $self->{'mode'} = $mode;
|
---|
| 135 | }
|
---|
| 136 |
|
---|
| 137 | sub set_assocdir {
|
---|
| 138 | my $self = shift (@_);
|
---|
| 139 | my ($assocdir) = @_;
|
---|
| 140 |
|
---|
| 141 | $self->{'assocdir'} = $assocdir;
|
---|
| 142 | }
|
---|
| 143 |
|
---|
| 144 | sub set_dontgdbm {
|
---|
| 145 | my $self = shift (@_);
|
---|
| 146 | my ($dontgdbm) = @_;
|
---|
| 147 |
|
---|
| 148 | $self->{'dontgdbm'} = $dontgdbm;
|
---|
| 149 | }
|
---|
| 150 |
|
---|
| 151 | sub set_index {
|
---|
| 152 | my $self = shift (@_);
|
---|
| 153 | my ($index, $indexexparr) = @_;
|
---|
| 154 |
|
---|
| 155 | $self->{'index'} = $index;
|
---|
| 156 | $self->{'indexexparr'} = $indexexparr if defined $indexexparr;
|
---|
| 157 | }
|
---|
| 158 |
|
---|
[1694] | 159 | sub get_index {
|
---|
| 160 | my $self = shift (@_);
|
---|
| 161 |
|
---|
| 162 | return $self->{'index'};
|
---|
| 163 | }
|
---|
| 164 |
|
---|
[932] | 165 | sub set_classifiers {
|
---|
| 166 | my $self = shift (@_);
|
---|
| 167 | my ($classifiers) = @_;
|
---|
| 168 |
|
---|
| 169 | $self->{'classifiers'} = $classifiers;
|
---|
| 170 | }
|
---|
| 171 |
|
---|
| 172 | sub set_indexing_text {
|
---|
| 173 | my $self = shift (@_);
|
---|
| 174 | my ($indexing_text) = @_;
|
---|
| 175 |
|
---|
| 176 | $self->{'indexing_text'} = $indexing_text;
|
---|
| 177 | }
|
---|
| 178 |
|
---|
[1694] | 179 | sub get_indexing_text {
|
---|
| 180 | my $self = shift (@_);
|
---|
| 181 |
|
---|
| 182 | return $self->{'indexing_text'};
|
---|
| 183 | }
|
---|
| 184 |
|
---|
[2480] | 185 | sub set_store_text {
|
---|
| 186 | my $self = shift (@_);
|
---|
| 187 | my ($store_text) = @_;
|
---|
| 188 |
|
---|
| 189 | $self->{'store_text'} = $store_text;
|
---|
| 190 | }
|
---|
| 191 |
|
---|
[2771] | 192 | sub get_doc_list {
|
---|
| 193 | my $self = shift(@_);
|
---|
| 194 |
|
---|
| 195 | return @{$self->{'doclist'}};
|
---|
| 196 | }
|
---|
| 197 |
|
---|
[1852] | 198 | sub set_indexfieldmap {
|
---|
| 199 | my $self = shift (@_);
|
---|
| 200 | my ($indexmap) = @_;
|
---|
| 201 |
|
---|
| 202 | $self->{'indexfieldmap'} = $indexmap;
|
---|
| 203 | }
|
---|
| 204 |
|
---|
| 205 | sub get_indexfieldmap {
|
---|
| 206 | my $self = shift (@_);
|
---|
| 207 |
|
---|
| 208 | return $self->{'indexfieldmap'};
|
---|
| 209 | }
|
---|
| 210 |
|
---|
| 211 | sub set_levels {
|
---|
| 212 | my $self = shift (@_);
|
---|
| 213 | my ($levels) = @_;
|
---|
| 214 |
|
---|
| 215 | $self->{'levels'} = $levels;
|
---|
| 216 | }
|
---|
| 217 |
|
---|
| 218 | sub set_strip_html {
|
---|
| 219 | my $self = shift (@_);
|
---|
| 220 | my ($strip) = @_;
|
---|
| 221 | $self->{'strip_html'}=$strip;
|
---|
| 222 | }
|
---|
| 223 |
|
---|
[932] | 224 | sub process {
|
---|
| 225 | my $self = shift (@_);
|
---|
| 226 | my $method = $self->{'mode'};
|
---|
| 227 |
|
---|
| 228 | $self->$method(@_);
|
---|
| 229 | }
|
---|
| 230 |
|
---|
| 231 | # use 'Paged' if document has no more than 2 levels
|
---|
| 232 | # and each section at second level has a number for
|
---|
| 233 | # Title metadata
|
---|
| 234 | sub get_document_type {
|
---|
| 235 | my $self = shift (@_);
|
---|
| 236 | my ($doc_obj) = @_;
|
---|
| 237 |
|
---|
| 238 | my $thistype = "VList";
|
---|
| 239 | my $childtype = "VList";
|
---|
| 240 | my $title;
|
---|
| 241 | my @tmp = ();
|
---|
| 242 |
|
---|
| 243 | my $section = $doc_obj->get_top_section ();
|
---|
| 244 | my $first = 1;
|
---|
| 245 | while (defined $section) {
|
---|
| 246 | @tmp = split /\./, $section;
|
---|
| 247 | if (scalar(@tmp) > 1) {
|
---|
| 248 | return ($thistype, $childtype);
|
---|
| 249 | }
|
---|
| 250 | if (!$first) {
|
---|
| 251 | $title = $doc_obj->get_metadata_element ($section, "Title");
|
---|
| 252 | if (!defined $title || $title !~ /^\d+$/) {
|
---|
| 253 | return ($thistype, $childtype);
|
---|
| 254 | }
|
---|
| 255 | }
|
---|
| 256 | $first = 0;
|
---|
| 257 | $section = $doc_obj->get_next_section($section);
|
---|
| 258 | }
|
---|
| 259 | if ($doc_obj->get_text_length ($doc_obj->get_top_section())) {
|
---|
| 260 | $thistype = "Paged";
|
---|
| 261 | } else {
|
---|
| 262 | $thistype = "Invisible";
|
---|
| 263 | }
|
---|
| 264 | $childtype = "Paged";
|
---|
| 265 | return ($thistype, $childtype);
|
---|
| 266 | }
|
---|
| 267 |
|
---|
| 268 | sub assoc_files {
|
---|
| 269 | my $self = shift (@_);
|
---|
| 270 | my ($doc_obj, $archivedir) = @_;
|
---|
| 271 | my ($afile);
|
---|
| 272 |
|
---|
| 273 | foreach my $assoc_file (@{$doc_obj->get_assoc_files()}) {
|
---|
[3148] | 274 | # if assoc file starts with a slash, we put it relative to the assoc
|
---|
| 275 | # dir, otherwise it is relative to the HASH... directory
|
---|
| 276 | if ($assoc_file->[1] =~ m@^[/\\]@) {
|
---|
| 277 | $afile = &util::filename_cat($self->{'assocdir'},$assoc_file->[1]);
|
---|
[932] | 278 | } else {
|
---|
| 279 | $afile = &util::filename_cat($self->{'assocdir'}, $archivedir, $assoc_file->[1]);
|
---|
| 280 | }
|
---|
| 281 | &util::hard_link ($assoc_file->[0], $afile);
|
---|
| 282 | }
|
---|
| 283 | }
|
---|
| 284 |
|
---|
| 285 | sub infodb {
|
---|
| 286 | my $self = shift (@_);
|
---|
| 287 | my ($doc_obj, $filename) = @_;
|
---|
| 288 | my $handle = $self->{'output_handle'};
|
---|
| 289 |
|
---|
| 290 | my $doctype = $doc_obj->get_doc_type();
|
---|
| 291 |
|
---|
| 292 | # only output this document if it is one to be indexed
|
---|
| 293 | return if ($doctype ne "indexed_doc");
|
---|
| 294 |
|
---|
[1852] | 295 | #if a Section level index is not built, the gdbm file should be at doc
|
---|
| 296 | #level not Section
|
---|
| 297 | my $docs_only = 1;
|
---|
| 298 | if ($self->{'levels'}->{'Section'}) {
|
---|
| 299 | $docs_only = 0;
|
---|
| 300 | }
|
---|
| 301 |
|
---|
[932] | 302 | my ($archivedir) = $filename =~ /^(.*?)(?:\/|\\)[^\/\\]*$/;
|
---|
| 303 | $archivedir = "" unless defined $archivedir;
|
---|
| 304 | $archivedir =~ s/\\/\//g;
|
---|
| 305 | $archivedir =~ s/^\/+//;
|
---|
| 306 | $archivedir =~ s/\/+$//;
|
---|
| 307 |
|
---|
| 308 | $self->assoc_files ($doc_obj, $archivedir);
|
---|
| 309 |
|
---|
| 310 | # this is another document
|
---|
| 311 | $self->{'num_docs'} += 1 unless ($doctype eq "classification");
|
---|
| 312 |
|
---|
| 313 | # is this a paged or a hierarchical document
|
---|
| 314 | my ($thistype, $childtype) = $self->get_document_type ($doc_obj);
|
---|
| 315 |
|
---|
| 316 | my $section = $doc_obj->get_top_section ();
|
---|
| 317 | my $doc_OID = $doc_obj->get_OID();
|
---|
| 318 | my $first = 1;
|
---|
| 319 | my $url = "";
|
---|
| 320 | while (defined $section) {
|
---|
| 321 | # update a few statistics
|
---|
| 322 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
---|
| 323 | $self->{'num_sections'} += 1 unless ($doctype eq "classification");
|
---|
| 324 |
|
---|
| 325 | # output the section name
|
---|
| 326 | if ($section eq "") { print $handle "[$doc_OID]\n"; }
|
---|
| 327 | else { print $handle "[$doc_OID.$section]\n"; }
|
---|
| 328 |
|
---|
[1694] | 329 | # output the fact that this document is a document (unless doctype
|
---|
| 330 | # has been set to something else from within a plugin
|
---|
| 331 | my $dtype = $doc_obj->get_metadata_element ($section, "doctype");
|
---|
| 332 | if (!defined $dtype || $dtype !~ /\w/) {
|
---|
| 333 | print $handle "<doctype>doc\n";
|
---|
| 334 | }
|
---|
[932] | 335 |
|
---|
| 336 | # output whether this node contains text
|
---|
| 337 | if ($doc_obj->get_text_length($section) > 0) {
|
---|
| 338 | print $handle "<hastxt>1\n";
|
---|
| 339 | } else {
|
---|
| 340 | print $handle "<hastxt>0\n";
|
---|
| 341 | }
|
---|
| 342 |
|
---|
| 343 | # output all the section metadata
|
---|
| 344 | my $metadata = $doc_obj->get_all_metadata ($section);
|
---|
[1917] | 345 | foreach my $pair (@$metadata) {
|
---|
[932] | 346 | my ($field, $value) = (@$pair);
|
---|
| 347 |
|
---|
| 348 | if ($field ne "Identifier" && $field !~ /^gsdl/ &&
|
---|
| 349 | defined $value && $value ne "") {
|
---|
| 350 |
|
---|
| 351 | # escape problematic stuff
|
---|
| 352 | $value =~ s/\\/\\\\/g;
|
---|
| 353 | $value =~ s/\n/\\n/g;
|
---|
| 354 | $value =~ s/\r/\\r/g;
|
---|
| 355 |
|
---|
| 356 | # special case for URL metadata
|
---|
| 357 | if ($field =~ /^URL$/i) {
|
---|
| 358 | $url .= "[$value]\n";
|
---|
| 359 | if ($section eq "") {$url .= "<section>$doc_OID\n";}
|
---|
| 360 | else {$url .= "<section>$doc_OID.$section\n";}
|
---|
| 361 | $url .= '-' x 70 . "\n";
|
---|
| 362 | }
|
---|
| 363 |
|
---|
| 364 | if (!defined $self->{'dontgdbm'}->{$field}) {
|
---|
| 365 | print $handle "<$field>$value\n";
|
---|
| 366 | }
|
---|
| 367 | }
|
---|
| 368 | }
|
---|
| 369 |
|
---|
| 370 | # output archivedir if at top level
|
---|
| 371 | if ($section eq $doc_obj->get_top_section()) {
|
---|
| 372 | print $handle "<archivedir>$archivedir\n";
|
---|
| 373 | }
|
---|
| 374 |
|
---|
| 375 | # output document display type
|
---|
| 376 | if ($first) {
|
---|
| 377 | print $handle "<thistype>$thistype\n";
|
---|
| 378 | }
|
---|
| 379 |
|
---|
[1852] | 380 | if (!$docs_only) {
|
---|
| 381 | # output a list of children
|
---|
| 382 | my $children = $doc_obj->get_children ($section);
|
---|
| 383 | if (scalar(@$children) > 0) {
|
---|
| 384 | print $handle "<childtype>$childtype\n";
|
---|
| 385 | print $handle "<contains>";
|
---|
| 386 | my $firstchild = 1;
|
---|
| 387 | foreach $child (@$children) {
|
---|
| 388 | print $handle ";" unless $firstchild;
|
---|
| 389 | $firstchild = 0;
|
---|
| 390 | if ($child =~ /^.*?\.(\d+)$/) {
|
---|
| 391 | print $handle "\".$1";
|
---|
| 392 | } else {
|
---|
| 393 | print $handle "\".$child";
|
---|
| 394 | }
|
---|
[932] | 395 | # if ($child eq "") { print $handle "$doc_OID"; }
|
---|
| 396 | # elsif ($section eq "") { print $handle "$doc_OID.$child"; }
|
---|
| 397 | # else { print $handle "$doc_OID.$section.$child"; }
|
---|
[1852] | 398 | }
|
---|
| 399 | print $handle "\n";
|
---|
[932] | 400 | }
|
---|
[1852] | 401 | #output the matching doc number
|
---|
| 402 | print $handle "<docnum>$self->{'num_sections'}\n";
|
---|
| 403 |
|
---|
| 404 | } # if (!$docs_only)
|
---|
| 405 | else { #docs only, doc num is num_docs not num_sections
|
---|
| 406 | # output the matching document number
|
---|
| 407 | print $handle "<docnum>$self->{'num_docs'}\n";
|
---|
[932] | 408 | }
|
---|
[1852] | 409 |
|
---|
[932] | 410 | print $handle '-' x 70, "\n";
|
---|
| 411 |
|
---|
| 412 |
|
---|
| 413 | # output a database entry for the document number
|
---|
[1852] | 414 | if ($docs_only) {
|
---|
| 415 | print $handle "[$self->{'num_docs'}]\n";
|
---|
| 416 | print $handle "<section>$doc_OID\n";
|
---|
| 417 | }
|
---|
| 418 | else {
|
---|
| 419 | print $handle "[$self->{'num_sections'}]\n";
|
---|
| 420 | if ($section eq "") { print $handle "<section>$doc_OID\n"; }
|
---|
| 421 | else { print $handle "<section>$doc_OID.$section\n"; }
|
---|
| 422 | }
|
---|
[932] | 423 | print $handle '-' x 70, "\n";
|
---|
| 424 |
|
---|
| 425 | # output entry for url
|
---|
| 426 | if ($url ne "") {
|
---|
| 427 | print $handle $url;
|
---|
| 428 | }
|
---|
| 429 |
|
---|
| 430 | $first = 0;
|
---|
| 431 | $section = $doc_obj->get_next_section($section);
|
---|
[1852] | 432 | last if ($docs_only); # if no sections wanted, only gdbm the docs
|
---|
[932] | 433 | }
|
---|
| 434 |
|
---|
[2771] | 435 | #add this document to the browse structure
|
---|
| 436 | push(@{$self->{'doclist'}},$doc_obj->get_OID())
|
---|
| 437 | unless ($doctype eq "classification");
|
---|
| 438 |
|
---|
[932] | 439 | # classify this document
|
---|
| 440 | &classify::classify_doc ($self->{'classifiers'}, $doc_obj);
|
---|
| 441 |
|
---|
| 442 | }
|
---|
| 443 |
|
---|
| 444 | sub find_paragraphs {
|
---|
| 445 | $_[1] =~ s/(<p\b)/<Paragraph>$1/gi;
|
---|
| 446 | }
|
---|
| 447 |
|
---|
[1852] | 448 | #this function strips the html tags from the doc if ($strip_html) and
|
---|
| 449 | # if ($para) replaces <p> with <Paragraph> tags.
|
---|
| 450 | # if both are false, the original text is returned
|
---|
| 451 | #assumes that <pre> and </pre> have no spaces, and removes all < and > inside
|
---|
| 452 | #these tags
|
---|
| 453 | sub preprocess_text {
|
---|
| 454 | my $self = shift (@_);
|
---|
| 455 | my ($text, $strip_html, $para) = @_;
|
---|
| 456 |
|
---|
| 457 | my ($outtext) = "";
|
---|
| 458 | if ($strip_html) {
|
---|
| 459 | while ($text =~ /<([^>]*)>/ && $text ne "") {
|
---|
| 460 |
|
---|
| 461 | $tag = $1;
|
---|
| 462 | $outtext .= $`." "; #add everything before the matched tag
|
---|
| 463 | $text = $'; #everything after the matched tag
|
---|
| 464 | if ($para && $tag =~ /^\s*p\s/) {
|
---|
| 465 | $outtext .= "<Paragraph> ";
|
---|
| 466 | }
|
---|
| 467 | elsif ($tag =~ /^pre$/) { # a pre tag
|
---|
| 468 | $text =~ /<\/pre>/; # find the closing pre tag
|
---|
| 469 | my $tmp_text = $`; #everything before the closing pre tag
|
---|
| 470 | $text = $'; #everything after the </pre>
|
---|
| 471 | $tmp_text =~ s/[<>]//g; # remove all < and >
|
---|
| 472 | $outtext.= $tmp_text . " ";
|
---|
| 473 | }
|
---|
| 474 | }
|
---|
| 475 |
|
---|
| 476 | $outtext .= $text; # add any remaining text
|
---|
| 477 | return $outtext;
|
---|
| 478 | } #if strip_html
|
---|
| 479 |
|
---|
| 480 | if ($para) {
|
---|
| 481 | $text =~ s/(<p\b)/<Paragraph>$1/gi;
|
---|
| 482 | return $text;
|
---|
| 483 | }
|
---|
| 484 | return $text;
|
---|
| 485 | }
|
---|
| 486 |
|
---|
| 487 |
|
---|
| 488 |
|
---|
[932] | 489 | sub filter_text {
|
---|
| 490 | # $self->filter_text ($field, $new_text);
|
---|
| 491 | # don't want to do anything for this version, however,
|
---|
| 492 | # in a particular collection you might want to override
|
---|
| 493 | # this method to post-process certain fields depending on
|
---|
| 494 | # the field, or whether we are outputting it for indexing
|
---|
| 495 | }
|
---|
| 496 |
|
---|
| 497 | sub text {
|
---|
| 498 | my $self = shift (@_);
|
---|
| 499 | my ($doc_obj) = @_;
|
---|
| 500 | my $handle = $self->{'output_handle'};
|
---|
| 501 | my $indexed_doc = 1;
|
---|
| 502 |
|
---|
| 503 | # only output this document if it is one to be indexed
|
---|
| 504 | return if ($doc_obj->get_doc_type() ne "indexed_doc");
|
---|
| 505 |
|
---|
| 506 | # see if this document belongs to this subcollection
|
---|
[1917] | 507 | foreach my $indexexp (@{$self->{'indexexparr'}}) {
|
---|
[932] | 508 | $indexed_doc = 0;
|
---|
| 509 | my ($field, $exp, $options) = split /\//, $indexexp;
|
---|
| 510 | if (defined ($field) && defined ($exp)) {
|
---|
| 511 | my ($bool) = $field =~ /^(.)/;
|
---|
| 512 | $field =~ s/^.// if $bool eq '!';
|
---|
| 513 | if ($field =~ /^filename$/i) {
|
---|
| 514 | $field = $doc_obj->get_source_filename();
|
---|
| 515 | } else {
|
---|
| 516 | $field = $doc_obj->get_metadata_element($doc_obj->get_top_section(), $field);
|
---|
| 517 | }
|
---|
| 518 | next unless defined $field;
|
---|
| 519 | if ($bool eq '!') {
|
---|
| 520 | if ($options =~ /^i$/i) {
|
---|
| 521 | if ($field !~ /$exp/i) {$indexed_doc = 1; last;}
|
---|
| 522 | } else {
|
---|
| 523 | if ($field !~ /$exp/) {$indexed_doc = 1; last;}
|
---|
| 524 | }
|
---|
| 525 | } else {
|
---|
| 526 | if ($options =~ /^i$/i) {
|
---|
| 527 | if ($field =~ /$exp/i) {$indexed_doc = 1; last;}
|
---|
| 528 | } else {
|
---|
| 529 | if ($field =~ /$exp/) {$indexed_doc = 1; last;}
|
---|
| 530 | }
|
---|
| 531 | }
|
---|
| 532 | }
|
---|
| 533 | }
|
---|
| 534 |
|
---|
| 535 | # this is another document
|
---|
| 536 | $self->{'num_docs'} += 1;
|
---|
| 537 |
|
---|
| 538 | # get the parameters for the output
|
---|
| 539 | my ($fields) = $self->{'index'};
|
---|
| 540 |
|
---|
[1852] | 541 | my ($sectiontag) = "";
|
---|
| 542 | if ($self->{'levels'}->{'Section'}) {
|
---|
| 543 | $sectiontag = "\n<Section>\n";
|
---|
| 544 | }
|
---|
| 545 | my ($paratag) = "";
|
---|
| 546 | if ($self->{'levels'}->{'Paragraph'}) {
|
---|
| 547 | $paratag = "<Paragraph>";
|
---|
| 548 | }
|
---|
[932] | 549 | my $doc_section = 0; # just for this document
|
---|
[1917] | 550 | my $text = "<Document>\n";
|
---|
| 551 |
|
---|
[932] | 552 | # get the text for this document
|
---|
| 553 | my $section = $doc_obj->get_top_section();
|
---|
| 554 | while (defined $section) {
|
---|
| 555 | # update a few statistics
|
---|
| 556 | $doc_section++;
|
---|
| 557 | $self->{'num_sections'} += 1;
|
---|
[1852] | 558 | $text .= $sectiontag;
|
---|
| 559 |
|
---|
[932] | 560 | if ($indexed_doc) {
|
---|
| 561 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
---|
[1917] | 562 | foreach my $field (split (/,/, $fields)) {
|
---|
[932] | 563 | # only deal with this field if it doesn't start with top or
|
---|
| 564 | # this is the first section
|
---|
| 565 | my $real_field = $field;
|
---|
| 566 | if (!($real_field =~ s/^top//) || ($doc_section == 1)) {
|
---|
| 567 | my $new_text = "";
|
---|
[1852] | 568 | my $tmp_text = "";
|
---|
[932] | 569 | if ($real_field eq "text") {
|
---|
[1852] | 570 | if ($self->{'indexing_text'}) { #tag the text with <Text>...</Text>, add the <Paragraph> tags and strip out html if needed
|
---|
| 571 | $new_text .= "<TX>\n";
|
---|
[2480] | 572 | $tmp_text .= $doc_obj->get_text ($section) if $self->{'store_text'};
|
---|
[1852] | 573 | $tmp_text = $self->preprocess_text($tmp_text, $self->{'strip_html'}, $self->{'levels'}->{'Paragraph'});
|
---|
| 574 |
|
---|
| 575 | $new_text .= "$tmp_text</TX>\n";
|
---|
| 576 | if (!defined $self->{'indexfields'}->{'TextOnly'}) {
|
---|
| 577 | $self->{'indexfields'}->{'TextOnly'} = 1;
|
---|
| 578 | }
|
---|
| 579 | }
|
---|
| 580 | else { # leave html stuff in, and dont add Paragraph tags - never retrieve paras at the moment
|
---|
[2480] | 581 | $new_text .= $doc_obj->get_text ($section) if $self->{'store_text'};
|
---|
[1852] | 582 | #if ($self->{'levels'}->{'Paragraph'}) {
|
---|
| 583 | #$self->find_paragraphs($new_text);
|
---|
| 584 | #}
|
---|
| 585 | }
|
---|
[932] | 586 | } else { # metadata field
|
---|
| 587 | if ($real_field eq "metadata") { # insert all metadata
|
---|
[1852] | 588 | #except gsdl stuff
|
---|
| 589 | my $shortname = "";
|
---|
[932] | 590 | my $metadata = $doc_obj->get_all_metadata ($section);
|
---|
| 591 | foreach $pair (@$metadata) {
|
---|
| 592 | my ($mfield, $mvalue) = (@$pair);
|
---|
[1852] | 593 | # check fields here, maybe others dont want - change to use dontindex!!
|
---|
[2480] | 594 | if ($mfield ne "Identifier" && $mfield ne "classifytype" && $mfield ne "assocfilepath" &&
|
---|
[932] | 595 | $mfield !~ /^gsdl/ && defined $mvalue && $mvalue ne "") {
|
---|
[1852] | 596 |
|
---|
| 597 | if (defined $self->{'indexfieldmap'}->{$mfield}) {
|
---|
| 598 | $shortname = $self->{'indexfieldmap'}->{$mfield};
|
---|
| 599 | }
|
---|
| 600 | else {
|
---|
| 601 | $shortname = $self->create_shortname($mfield);
|
---|
| 602 | $self->{'indexfieldmap'}->{$mfield} = $shortname;
|
---|
| 603 | $self->{'indexfieldmap'}->{$shortname} = 1;
|
---|
| 604 | }
|
---|
| 605 | $new_text .= "$paratag<$shortname>$mvalue</$shortname>\n";
|
---|
| 606 | if (!defined $self->{'indexfields'}->{$mfield}) {
|
---|
| 607 | $self->{'indexfields'}->{$mfield} = 1;
|
---|
| 608 | }
|
---|
[932] | 609 | }
|
---|
| 610 | }
|
---|
[1852] | 611 |
|
---|
[932] | 612 | }
|
---|
| 613 | else { #individual metadata specified
|
---|
[1852] | 614 | my $shortname="";
|
---|
[1772] | 615 | if (!defined $self->{'indexfields'}->{$real_field}) {
|
---|
| 616 | $self->{'indexfields'}->{$real_field} = 1;
|
---|
[1852] | 617 | }
|
---|
| 618 | if (defined $self->{'indexfieldmap'}->{$real_field}) {
|
---|
| 619 | $shortname = $self->{'indexfieldmap'}->{$real_field};
|
---|
| 620 | }
|
---|
| 621 | else {
|
---|
| 622 | $shortname = $self->create_shortname($real_field);
|
---|
| 623 | $self->{'indexfieldmap'}->{$real_field} = $shortname;
|
---|
| 624 | $self->{'indexfieldmap'}->{$shortname} = 1;
|
---|
| 625 | }
|
---|
[932] | 626 | foreach $item (@{$doc_obj->get_metadata ($section, $real_field)}) {
|
---|
[1852] | 627 | $new_text .= "$paratag<$shortname>$item</$shortname>\n";
|
---|
[932] | 628 | }
|
---|
| 629 | }
|
---|
| 630 |
|
---|
| 631 | }
|
---|
| 632 |
|
---|
| 633 | # filter the text
|
---|
| 634 | $self->filter_text ($field, $new_text);
|
---|
| 635 |
|
---|
[1852] | 636 | $self->{'num_processed_bytes'} += length ($new_text);
|
---|
[932] | 637 | $text .= "$new_text";
|
---|
| 638 | }
|
---|
| 639 | }
|
---|
| 640 | } # if (indexed_doc)
|
---|
| 641 |
|
---|
| 642 | $section = $doc_obj->get_next_section($section);
|
---|
| 643 | } #while defined section
|
---|
[1917] | 644 | print $handle "$text\n</Document>\n";
|
---|
[932] | 645 | }
|
---|
| 646 |
|
---|
[1852] | 647 | sub create_shortname {
|
---|
| 648 | $self = shift(@_);
|
---|
| 649 |
|
---|
| 650 | my ($realname) = @_;
|
---|
| 651 | #take the first two chars
|
---|
| 652 | my ($shortname) = $realname =~ /^(\w\w)/;
|
---|
| 653 | $shortname =~ tr/a-z/A-Z/;
|
---|
| 654 |
|
---|
| 655 | #if already used, take the first and third letters and so on
|
---|
| 656 | $count = 1;
|
---|
| 657 | while (defined $self->{'indexfieldmap'}->{$shortname}) {
|
---|
| 658 | if ($realname =~ /^(\w).{$count}(\w)/) {
|
---|
| 659 | $shortname = "$1$2";
|
---|
| 660 | $count++;
|
---|
| 661 | $shortname =~ tr/a-z/A-Z/;
|
---|
| 662 |
|
---|
| 663 | }
|
---|
| 664 | else {
|
---|
| 665 | $realname =~ s/^.//;
|
---|
| 666 | $count = 0;
|
---|
| 667 | }
|
---|
| 668 | }
|
---|
| 669 |
|
---|
| 670 | return $shortname;
|
---|
| 671 | }
|
---|
| 672 |
|
---|
[932] | 673 | 1;
|
---|
| 674 |
|
---|