| 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 | use basebuildproc;
|
|---|
| 33 | use cnseg;
|
|---|
| 34 |
|
|---|
| 35 | use strict;
|
|---|
| 36 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
|---|
| 37 |
|
|---|
| 38 |
|
|---|
| 39 | BEGIN {
|
|---|
| 40 | @mgppbuildproc::ISA = ('basebuildproc');
|
|---|
| 41 | }
|
|---|
| 42 |
|
|---|
| 43 | #this must be the same as in mgppbuilder
|
|---|
| 44 | our %level_map = ('document'=>'Doc',
|
|---|
| 45 | 'section'=>'Sec',
|
|---|
| 46 | 'paragraph'=>'Para');
|
|---|
| 47 |
|
|---|
| 48 | # change this so a user can add their own ones in via a file or cfg
|
|---|
| 49 | #add AND, OR, NOT NEAR to this list - these cannot be used as field names
|
|---|
| 50 | #also add the level names (Doc, Sec, Para)
|
|---|
| 51 | our %static_indexfield_map = ('Title'=>'TI',
|
|---|
| 52 | 'TI'=>1,
|
|---|
| 53 | 'Subject'=>'SU',
|
|---|
| 54 | 'SU'=>1,
|
|---|
| 55 | 'Creator'=>'CR',
|
|---|
| 56 | 'CR'=>1,
|
|---|
| 57 | 'Organization'=>'ORG',
|
|---|
| 58 | 'ORG'=>1,
|
|---|
| 59 | 'Source'=>'SO',
|
|---|
| 60 | 'SO'=>1,
|
|---|
| 61 | 'Howto'=>'HT',
|
|---|
| 62 | 'HT'=>1,
|
|---|
| 63 | 'ItemTitle'=>'IT',
|
|---|
| 64 | 'IT'=>1,
|
|---|
| 65 | 'ProgNumber'=>'PN',
|
|---|
| 66 | 'PN'=>1,
|
|---|
| 67 | 'People'=>'PE',
|
|---|
| 68 | 'PE'=>1,
|
|---|
| 69 | 'Coverage'=>'CO',
|
|---|
| 70 | 'CO'=>1,
|
|---|
| 71 | 'allfields'=>'ZZ',
|
|---|
| 72 | 'ZZ'=>1,
|
|---|
| 73 | 'text'=>'TX',
|
|---|
| 74 | 'TX'=>1,
|
|---|
| 75 | 'GPSMapOverlayLabel'=> 'ML',
|
|---|
| 76 | 'ML'=>1,
|
|---|
| 77 | 'Coordinate'=>'CD',
|
|---|
| 78 | 'CD'=>1,
|
|---|
| 79 | 'CoordShort'=>'CS',
|
|---|
| 80 | 'CS'=>1,
|
|---|
| 81 | 'Latitude'=>'LT',
|
|---|
| 82 | 'LT'=>1,
|
|---|
| 83 | 'Longitude'=>'LO',
|
|---|
| 84 | 'LO'=>1,
|
|---|
| 85 | 'LatShort'=>'LA',
|
|---|
| 86 | 'LA'=>1,
|
|---|
| 87 | 'LngShort'=>'LN',
|
|---|
| 88 | 'LN'=>1,
|
|---|
| 89 | 'AND'=>1,
|
|---|
| 90 | 'OR'=>1,
|
|---|
| 91 | 'NOT'=>1,
|
|---|
| 92 | 'NEAR'=>1,
|
|---|
| 93 | 'Doc'=>1,
|
|---|
| 94 | 'Sec'=>1,
|
|---|
| 95 | 'Para'=>1);
|
|---|
| 96 |
|
|---|
| 97 |
|
|---|
| 98 | sub new {
|
|---|
| 99 | my $class = shift @_;
|
|---|
| 100 | my $self = new basebuildproc (@_);
|
|---|
| 101 |
|
|---|
| 102 | # use a different index specification to the default
|
|---|
| 103 | $self->{'index'} = "text";
|
|---|
| 104 |
|
|---|
| 105 | $self->{'dontindex'} = {};
|
|---|
| 106 | $self->{'allindexfields'} = {}; # list of all actually indexed fields
|
|---|
| 107 | $self->{'extraindexfields'} = {}; # indexed fields not specfied in original index list - ie if 'metadata' was specified.
|
|---|
| 108 | $self->{'fieldnamemap'} = {'allfields'=>'ZZ',
|
|---|
| 109 | 'ZZ'=>1,
|
|---|
| 110 | 'text'=>'TX',
|
|---|
| 111 | 'TX'=>1}; # mapping between index full names and short names. Once we have decided on a mapping it goes in here, whether we have indexed something or not.
|
|---|
| 112 | $self->{'strip_html'}=1;
|
|---|
| 113 |
|
|---|
| 114 | return bless $self, $class;
|
|---|
| 115 | }
|
|---|
| 116 |
|
|---|
| 117 | sub set_levels {
|
|---|
| 118 | my $self = shift (@_);
|
|---|
| 119 | my ($levels) = @_;
|
|---|
| 120 |
|
|---|
| 121 | $self->{'levels'} = $levels;
|
|---|
| 122 | }
|
|---|
| 123 |
|
|---|
| 124 | sub set_strip_html {
|
|---|
| 125 | my $self = shift (@_);
|
|---|
| 126 | my ($strip) = @_;
|
|---|
| 127 | $self->{'strip_html'}=$strip;
|
|---|
| 128 | }
|
|---|
| 129 |
|
|---|
| 130 | #sub find_paragraphs {
|
|---|
| 131 | # $_[1] =~ s/(<p\b)/<Paragraph>$1/gi;
|
|---|
| 132 | #}
|
|---|
| 133 |
|
|---|
| 134 | sub remove_gtlt {
|
|---|
| 135 | my $self =shift(@_);
|
|---|
| 136 | my ($text, $para) = @_;
|
|---|
| 137 | $text =~s/[<>]//g;
|
|---|
| 138 | return "$para$text$para";
|
|---|
| 139 | }
|
|---|
| 140 |
|
|---|
| 141 | sub process_tags {
|
|---|
| 142 | my $self = shift(@_);
|
|---|
| 143 | my ($text, $para) = @_;
|
|---|
| 144 | if ($text =~ /^p\b/i) {
|
|---|
| 145 | return $para;
|
|---|
| 146 | }
|
|---|
| 147 | return "";
|
|---|
| 148 | }
|
|---|
| 149 |
|
|---|
| 150 | sub preprocess_text {
|
|---|
| 151 | my $self = shift (@_);
|
|---|
| 152 | my ($text, $strip_html, $para) = @_;
|
|---|
| 153 | # at this stage, we do not do paragraph tags unless have strip_html -
|
|---|
| 154 | # it will result in a huge mess of non-xml
|
|---|
| 155 | return unless $strip_html;
|
|---|
| 156 |
|
|---|
| 157 | my $new_text = $text;
|
|---|
| 158 |
|
|---|
| 159 | # if we have <pre> tags, we can have < > inside them, need to delete
|
|---|
| 160 | # the <> before stripping tags
|
|---|
| 161 | $new_text =~ s/<pre>(.*?)<\/pre>/$self->remove_gtlt($1,$para)/gse;
|
|---|
| 162 |
|
|---|
| 163 | if ($para eq "") {
|
|---|
| 164 | # just remove all tags
|
|---|
| 165 | $new_text =~ s/<[^>]*>//gs;
|
|---|
| 166 | } else {
|
|---|
| 167 | # strip all tags except <p> tags which get turned into $para
|
|---|
| 168 | $new_text =~ s/<([^>]*)>/$self->process_tags($1, $para)/gse;
|
|---|
| 169 |
|
|---|
| 170 | }
|
|---|
| 171 | return $new_text;
|
|---|
| 172 | }
|
|---|
| 173 | #this function strips the html tags from the doc if ($strip_html) and
|
|---|
| 174 | # if ($para) replaces <p> with <Paragraph> tags.
|
|---|
| 175 | # if both are false, the original text is returned
|
|---|
| 176 | #assumes that <pre> and </pre> have no spaces, and removes all < and > inside
|
|---|
| 177 | #these tags
|
|---|
| 178 | sub preprocess_text_old_and_slow {
|
|---|
| 179 | my $self = shift (@_);
|
|---|
| 180 | my ($text, $strip_html, $para) = @_;
|
|---|
| 181 | my ($outtext) = "";
|
|---|
| 182 | if ($strip_html) {
|
|---|
| 183 | while ($text =~ /<([^>]*)>/ && $text ne "") {
|
|---|
| 184 |
|
|---|
| 185 | my $tag = $1;
|
|---|
| 186 | $outtext .= $`." "; #add everything before the matched tag
|
|---|
| 187 | $text = $'; #'everything after the matched tag
|
|---|
| 188 | if ($para && $tag =~ /^\s*p\s/i) {
|
|---|
| 189 | $outtext .= $para;
|
|---|
| 190 | }
|
|---|
| 191 | elsif ($tag =~ /^pre$/) { # a pre tag
|
|---|
| 192 | $text =~ /<\/pre>/; # find the closing pre tag
|
|---|
| 193 | my $tmp_text = $`; #everything before the closing pre tag
|
|---|
| 194 | $text = $'; #'everything after the </pre>
|
|---|
| 195 | $tmp_text =~ s/[<>]//g; # remove all < and >
|
|---|
| 196 | $outtext.= $tmp_text . " ";
|
|---|
| 197 | }
|
|---|
| 198 | }
|
|---|
| 199 |
|
|---|
| 200 | $outtext .= $text; # add any remaining text
|
|---|
| 201 | return $outtext;
|
|---|
| 202 | } #if strip_html
|
|---|
| 203 |
|
|---|
| 204 | #if ($para) {
|
|---|
| 205 | #$text =~ s/(<p\b)/$para$1/gi;
|
|---|
| 206 | #return $text;
|
|---|
| 207 | # }
|
|---|
| 208 | return $text;
|
|---|
| 209 | }
|
|---|
| 210 |
|
|---|
| 211 | sub text {
|
|---|
| 212 | my $self = shift (@_);
|
|---|
| 213 | my ($doc_obj) = @_;
|
|---|
| 214 | my $handle = $self->{'output_handle'};
|
|---|
| 215 | my $outhandle = $self->{'outhandle'};
|
|---|
| 216 |
|
|---|
| 217 | # only output this document if it is one to be indexed
|
|---|
| 218 | return if ($doc_obj->get_doc_type() ne "indexed_doc");
|
|---|
| 219 |
|
|---|
| 220 | my $indexed_doc = $self->is_subcollection_doc($doc_obj);
|
|---|
| 221 |
|
|---|
| 222 | # this is another document
|
|---|
| 223 | $self->{'num_docs'} += 1;
|
|---|
| 224 |
|
|---|
| 225 | # get the parameters for the output
|
|---|
| 226 | # split on : just in case there is subcoll and lang stuff
|
|---|
| 227 | my ($fields) = split (/:/, $self->{'index'});
|
|---|
| 228 |
|
|---|
| 229 | # we always do text and index on Doc and Sec levels
|
|---|
| 230 | my ($documenttag) = "\n<". $level_map{'document'} . ">\n";
|
|---|
| 231 | my ($documentendtag) = "\n</". $level_map{'document'} . ">\n";
|
|---|
| 232 | my ($sectiontag) = "\n<". $level_map{'section'} . ">\n";
|
|---|
| 233 | my ($sectionendtag) = "\n</". $level_map{'section'} . ">\n";
|
|---|
| 234 |
|
|---|
| 235 | my ($paratag) = "";
|
|---|
| 236 |
|
|---|
| 237 | # paragraph tags will only be used for indexing (can't retrieve
|
|---|
| 238 | # paragraphs), and can ony be used if we are stripping HTML tags
|
|---|
| 239 | if ($self->{'indexing_text'} && $self->{'levels'}->{'paragraph'}) {
|
|---|
| 240 | if ($self->{'strip_html'}) {
|
|---|
| 241 | $paratag = "<". $level_map{'paragraph'} . ">";
|
|---|
| 242 | } else {
|
|---|
| 243 | print $outhandle "Paragraph level can not be used with no_strip_html!. Not indexing Paragraphs.\n";
|
|---|
| 244 | }
|
|---|
| 245 | }
|
|---|
| 246 |
|
|---|
| 247 | my $doc_section = 0; # just for this document
|
|---|
| 248 |
|
|---|
| 249 | my $text = $documenttag;
|
|---|
| 250 |
|
|---|
| 251 | # get the text for this document
|
|---|
| 252 | my $section = $doc_obj->get_top_section();
|
|---|
| 253 |
|
|---|
| 254 | while (defined $section) {
|
|---|
| 255 | # update a few statistics
|
|---|
| 256 | $doc_section++;
|
|---|
| 257 | $self->{'num_sections'} += 1;
|
|---|
| 258 | $text .= "$sectiontag";
|
|---|
| 259 |
|
|---|
| 260 | my $indexed_section = $doc_obj->get_metadata_element($section, "gsdldoctype") || "indexed_section";
|
|---|
| 261 | if (($indexed_doc == 0) || ($indexed_section ne "indexed_section" && $indexed_section ne "indexed_doc")) {
|
|---|
| 262 | # we are not actually indexing anything for this document,
|
|---|
| 263 | # but we want to keep the section numbers the same, so we just
|
|---|
| 264 | # output section tags for each section (which is done above)
|
|---|
| 265 | $text .= "$sectionendtag";
|
|---|
| 266 | $section = $doc_obj->get_next_section($section);
|
|---|
| 267 | next;
|
|---|
| 268 | }
|
|---|
| 269 |
|
|---|
| 270 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
|---|
| 271 |
|
|---|
| 272 | # has the user added a 'metadata' index?
|
|---|
| 273 | my $all_metadata_specified = 0;
|
|---|
| 274 | # which fields have already been indexed? (same as fields, but in a map)
|
|---|
| 275 | my $specified_fields = {};
|
|---|
| 276 | foreach my $field (split (/;/, $fields)) {
|
|---|
| 277 | # only deal with this field if it doesn't start with top or
|
|---|
| 278 | # this is the first section
|
|---|
| 279 | my $real_field = $field;
|
|---|
| 280 | next if (($real_field =~ s/^top//) && ($doc_section != 1));
|
|---|
| 281 |
|
|---|
| 282 | my $new_text = "";
|
|---|
| 283 |
|
|---|
| 284 | # we get allfields by default
|
|---|
| 285 | next if ($real_field eq "allfields");
|
|---|
| 286 |
|
|---|
| 287 | # metadata - output all metadata we know about except gsdl stuff
|
|---|
| 288 | # each metadata is in a separate index field
|
|---|
| 289 | if ($real_field eq "metadata") {
|
|---|
| 290 | # we will process this later, so we are not reindexing metadata already indexed
|
|---|
| 291 | $all_metadata_specified = 1;
|
|---|
| 292 | next;
|
|---|
| 293 | }
|
|---|
| 294 |
|
|---|
| 295 | #individual metadata and or text specified - could be
|
|---|
| 296 | # a comma separated list
|
|---|
| 297 | $specified_fields->{$real_field} = 1;
|
|---|
| 298 | my $shortname="";
|
|---|
| 299 |
|
|---|
| 300 | if (defined $self->{'fieldnamemap'}->{$real_field}) {
|
|---|
| 301 | $shortname = $self->{'fieldnamemap'}->{$real_field};
|
|---|
| 302 | } else {
|
|---|
| 303 | $shortname = $self->create_shortname($real_field);
|
|---|
| 304 | $self->{'fieldnamemap'}->{$real_field} = $shortname;
|
|---|
| 305 | $self->{'fieldnamemap'}->{$shortname} = 1;
|
|---|
| 306 | }
|
|---|
| 307 |
|
|---|
| 308 | my @metadata_list = (); # put any meta values in here
|
|---|
| 309 | my $section_text = ""; # put any text in here
|
|---|
| 310 | foreach my $submeta (split /,/, $real_field) {
|
|---|
| 311 | if ($submeta eq "text") {
|
|---|
| 312 | # no point in indexing text more than once
|
|---|
| 313 | if ($section_text eq "") {
|
|---|
| 314 | $section_text = $doc_obj->get_text($section);
|
|---|
| 315 | if ($self->{'indexing_text'}) {
|
|---|
| 316 | if ($paratag ne "") {
|
|---|
| 317 | # we fiddle around with splitting text into paragraphs
|
|---|
| 318 | $section_text = $self->preprocess_text($section_text, $self->{'strip_html'}, "</$shortname>$paratag<$shortname>");
|
|---|
| 319 | }
|
|---|
| 320 | else {
|
|---|
| 321 | $section_text = $self->preprocess_text($section_text, $self->{'strip_html'}, "");
|
|---|
| 322 | }
|
|---|
| 323 | }
|
|---|
| 324 | }
|
|---|
| 325 | }
|
|---|
| 326 | else {
|
|---|
| 327 | $submeta =~ s/^ex\.([^.]+)$/$1/; #strip off ex. iff it's the only metadata set prefix (will leave ex.dc.* intact)
|
|---|
| 328 | # its a metadata element
|
|---|
| 329 | my @section_metadata = @{$doc_obj->get_metadata ($section, $submeta)};
|
|---|
| 330 | if ($section ne $doc_obj->get_top_section() && $self->{'indexing_text'} && defined ($self->{'sections_index_document_metadata'})) {
|
|---|
| 331 | if ($self->{'sections_index_document_metadata'} eq "always" || ( scalar(@section_metadata) == 0 && $self->{'sections_index_document_metadata'} eq "unless_section_metadata_exists")) {
|
|---|
| 332 | push (@section_metadata, @{$doc_obj->get_metadata ($doc_obj->get_top_section(), $submeta)});
|
|---|
| 333 | }
|
|---|
| 334 | }
|
|---|
| 335 | push (@metadata_list, @section_metadata);
|
|---|
| 336 | }
|
|---|
| 337 | } # for each field in index
|
|---|
| 338 |
|
|---|
| 339 |
|
|---|
| 340 | # now we add the text and/or the metadata into new_text
|
|---|
| 341 | if ($section_text ne "" || scalar(@metadata_list)) {
|
|---|
| 342 | if ($self->{'indexing_text'}) {
|
|---|
| 343 | # only add tags in if indexing
|
|---|
| 344 | $new_text .= "$paratag<$shortname>";
|
|---|
| 345 | }
|
|---|
| 346 | if ($section_text ne "") {
|
|---|
| 347 | $new_text .= "$section_text ";
|
|---|
| 348 | if ($self->{'indexing_text'} && $paratag ne "" && scalar(@metadata_list)) {
|
|---|
| 349 | $new_text .= "</$shortname>$paratag<$shortname>";
|
|---|
| 350 | }
|
|---|
| 351 | }
|
|---|
| 352 | foreach my $item (@metadata_list) {
|
|---|
| 353 | $new_text .= "$item ";
|
|---|
| 354 | }
|
|---|
| 355 | if ($self->{'indexing_text'}) {
|
|---|
| 356 | # only add tags in if indexing
|
|---|
| 357 | $new_text .= "</$shortname>";
|
|---|
| 358 | $self->{'allindexfields'}->{$real_field} = 1;
|
|---|
| 359 | }
|
|---|
| 360 | }
|
|---|
| 361 |
|
|---|
| 362 | # filter the text
|
|---|
| 363 | $new_text = $self->filter_text ($field, $new_text);
|
|---|
| 364 |
|
|---|
| 365 | $self->{'num_processed_bytes'} += length ($new_text);
|
|---|
| 366 | $text .= "$new_text";
|
|---|
| 367 | } # foreach field
|
|---|
| 368 |
|
|---|
| 369 | if ($all_metadata_specified) {
|
|---|
| 370 | my $new_text = "";
|
|---|
| 371 | my $shortname = "";
|
|---|
| 372 | my $metadata = $doc_obj->get_all_metadata ($section);
|
|---|
| 373 | foreach my $pair (@$metadata) {
|
|---|
| 374 | my ($mfield, $mvalue) = (@$pair);
|
|---|
| 375 | # no value
|
|---|
| 376 | next unless defined $mvalue && $mvalue ne "";
|
|---|
| 377 | # we have already indexed this
|
|---|
| 378 | next if defined ($specified_fields->{$mfield});
|
|---|
| 379 | # check fields here, maybe others dont want - change to use dontindex!!
|
|---|
| 380 | next if ($mfield eq "Identifier" || $mfield eq "classifytype" || $mfield eq "assocfilepath");
|
|---|
| 381 | next if ($mfield =~ /^gsdl/);
|
|---|
| 382 |
|
|---|
| 383 | if (defined $self->{'fieldnamemap'}->{$mfield}) {
|
|---|
| 384 | $shortname = $self->{'fieldnamemap'}->{$mfield};
|
|---|
| 385 | } else {
|
|---|
| 386 | $shortname = $self->create_shortname($mfield);
|
|---|
| 387 | $self->{'fieldnamemap'}->{$mfield} = $shortname;
|
|---|
| 388 | $self->{'fieldnamemap'}->{$shortname} = 1;
|
|---|
| 389 | }
|
|---|
| 390 | $self->{'allindexfields'}->{$mfield} = 1;
|
|---|
| 391 | $new_text .= "$paratag<$shortname>$mvalue</$shortname>\n";
|
|---|
| 392 | if (!defined $self->{'extraindexfields'}->{$mfield}) {
|
|---|
| 393 | $self->{'extraindexfields'}->{$mfield} = 1;
|
|---|
| 394 | }
|
|---|
| 395 |
|
|---|
| 396 | }
|
|---|
| 397 | # filter the text
|
|---|
| 398 | $new_text = $self->filter_text ("metadata", $new_text);
|
|---|
| 399 |
|
|---|
| 400 | $self->{'num_processed_bytes'} += length ($new_text);
|
|---|
| 401 | $text .= "$new_text";
|
|---|
| 402 |
|
|---|
| 403 |
|
|---|
| 404 | }
|
|---|
| 405 |
|
|---|
| 406 | $text .= "$sectionendtag";
|
|---|
| 407 | $section = $doc_obj->get_next_section($section);
|
|---|
| 408 | } # while defined section
|
|---|
| 409 | print $handle "$text\n$documentendtag";
|
|---|
| 410 | #print STDERR "***********\n$text\n***************\n";
|
|---|
| 411 |
|
|---|
| 412 | }
|
|---|
| 413 |
|
|---|
| 414 | #chooses the first two letters or digits for the shortname
|
|---|
| 415 | #now ignores non-letdig characters
|
|---|
| 416 | sub create_shortname {
|
|---|
| 417 | my $self = shift(@_);
|
|---|
| 418 |
|
|---|
| 419 | my ($realname) = @_;
|
|---|
| 420 | my @realnamelist = split(",", $realname);
|
|---|
| 421 | map {$_=~ s/^[a-zA-Z]+\.//;} @realnamelist; #remove namespaces
|
|---|
| 422 | my ($singlename) = $realnamelist[0];
|
|---|
| 423 |
|
|---|
| 424 | # try our predefined static mapping
|
|---|
| 425 | my $name;
|
|---|
| 426 | if (defined ($name = $static_indexfield_map{$singlename})) {
|
|---|
| 427 | if (! defined $self->{'fieldnamemap'}->{$name}) {
|
|---|
| 428 | # has this shortname already been used??
|
|---|
| 429 | return $static_indexfield_map{$singlename};
|
|---|
| 430 | }
|
|---|
| 431 | }
|
|---|
| 432 | # we can't use the quick map, so join all fields back together (without namespaces), and try sets of two characters.
|
|---|
| 433 | $realname = join ("", @realnamelist);
|
|---|
| 434 | #try the first two chars
|
|---|
| 435 | my $shortname;
|
|---|
| 436 | if ($realname =~ /^[^\w]*(\w)[^\w]*(\w)/) {
|
|---|
| 437 | $shortname = "$1$2";
|
|---|
| 438 | } else {
|
|---|
| 439 | # there aren't two letdig's in the field - try arbitrary combinations
|
|---|
| 440 | $realname = "ABCDEFGHIJKLMNOPQRSTUVWXYZ";
|
|---|
| 441 | $shortname = "AB";
|
|---|
| 442 | }
|
|---|
| 443 | $shortname =~ tr/a-z/A-Z/;
|
|---|
| 444 |
|
|---|
| 445 | #if already used, take the first and third letdigs and so on
|
|---|
| 446 | my $count = 1;
|
|---|
| 447 | while (defined $self->{'fieldnamemap'}->{$shortname} || defined $static_indexfield_map{$shortname}) {
|
|---|
| 448 | if ($realname =~ /^[^\w]*(\w)([^\w]*\w){$count}[^\w]*(\w)/) {
|
|---|
| 449 | $shortname = "$1$3";
|
|---|
| 450 | $count++;
|
|---|
| 451 | $shortname =~ tr/a-z/A-Z/;
|
|---|
| 452 |
|
|---|
| 453 | }
|
|---|
| 454 | else {
|
|---|
| 455 | #remove up to and incl the first letdig
|
|---|
| 456 | $realname =~ s/^[^\w]*\w//;
|
|---|
| 457 | $count = 0;
|
|---|
| 458 | }
|
|---|
| 459 | }
|
|---|
| 460 |
|
|---|
| 461 | return $shortname;
|
|---|
| 462 | }
|
|---|
| 463 |
|
|---|
| 464 | 1;
|
|---|
| 465 |
|
|---|
