| 1 | #! /usr/bin/perl -w
|
|---|
| 2 |
|
|---|
| 3 | ###########################################################################
|
|---|
| 4 | #
|
|---|
| 5 | # phindgen.pl -- produce Phind index files for a GSDL collection.
|
|---|
| 6 | #
|
|---|
| 7 | # Copyright 2000 Gordon W. Paynter
|
|---|
| 8 | # Copyright 2000 The New Zealand Digital Library Project
|
|---|
| 9 | #
|
|---|
| 10 | # A component of the Greenstone digital library software
|
|---|
| 11 | # from the New Zealand Digital Library Project at the
|
|---|
| 12 | # University of Waikato, New Zealand.
|
|---|
| 13 | #
|
|---|
| 14 | # This program is free software; you can redistribute it and/or modify
|
|---|
| 15 | # it under the terms of the GNU General Public License as published by
|
|---|
| 16 | # the Free Software Foundation; either version 2 of the License, or
|
|---|
| 17 | # (at your option) any later version.
|
|---|
| 18 | #
|
|---|
| 19 | # This program is distributed in the hope that it will be useful,
|
|---|
| 20 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|---|
| 21 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|---|
| 22 | # GNU General Public License for more details.
|
|---|
| 23 | #
|
|---|
| 24 | # You should have received a copy of the GNU General Public License
|
|---|
| 25 | # along with this program; if not, write to the Free Software
|
|---|
| 26 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
|---|
| 27 | #
|
|---|
| 28 | ###########################################################################
|
|---|
| 29 |
|
|---|
| 30 |
|
|---|
| 31 | # This program generates phind phrase browsing indexes for a Greenstone
|
|---|
| 32 | # digital library collection.
|
|---|
| 33 | #
|
|---|
| 34 | # The GML files that have been imported to the archives directory are
|
|---|
| 35 | # read, and then the phind indexes are created in the phindex directory.
|
|---|
| 36 | #
|
|---|
| 37 | # This version od Phind uses suffix arrays to extract phrases.
|
|---|
| 38 |
|
|---|
| 39 |
|
|---|
| 40 | BEGIN {
|
|---|
| 41 | die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
|
|---|
| 42 | die "GSDLOS not set\n" unless defined $ENV{'GSDLOS'};
|
|---|
| 43 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib");
|
|---|
| 44 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/plugins");
|
|---|
| 45 | }
|
|---|
| 46 |
|
|---|
| 47 |
|
|---|
| 48 | use strict;
|
|---|
| 49 | use colcfg;
|
|---|
| 50 | use util;
|
|---|
| 51 | use parsargv;
|
|---|
| 52 | use plugin;
|
|---|
| 53 | use phproc;
|
|---|
| 54 |
|
|---|
| 55 |
|
|---|
| 56 | sub print_usage {
|
|---|
| 57 | print STDERR "\n usage: $0 [options] collection-name\n\n";
|
|---|
| 58 | print STDERR " options:\n";
|
|---|
| 59 | print STDERR " -verbosity number 0=none, 3=lots\n";
|
|---|
| 60 | print STDERR " -archivedir directory Where the converted material ends up\n";
|
|---|
| 61 | print STDERR " -phindexdir directory Where to place the phind indexes\n";
|
|---|
| 62 | print STDERR " -languages names The languages to consider in the browser\n";
|
|---|
| 63 | print STDERR " -maxdocs number Maximum number of documents to process\n";
|
|---|
| 64 | print STDERR " -untidy Leave working files in phindex directory\n";
|
|---|
| 65 | print STDERR "\n";
|
|---|
| 66 | }
|
|---|
| 67 |
|
|---|
| 68 |
|
|---|
| 69 | # Define delimiter symbols - this should be abstracted out someplace
|
|---|
| 70 | my $colstart = "COLLECTIONSTART";
|
|---|
| 71 | my $colend = "COLLECTIONEND";
|
|---|
| 72 | my $doclimit = "DOCUMENTLIMIT";
|
|---|
| 73 | my $senlimit = "SENTENCELIMIT";
|
|---|
| 74 | my @delimiters = ($colstart, $colend, $doclimit, $senlimit);
|
|---|
| 75 |
|
|---|
| 76 | &main ();
|
|---|
| 77 |
|
|---|
| 78 |
|
|---|
| 79 | sub main {
|
|---|
| 80 |
|
|---|
| 81 | my ($verbosity, $archivedir, $phindexdir,
|
|---|
| 82 | $phindcfg, $language, $maxdocs, $untidy,
|
|---|
| 83 | $collection, $configfilename, $collectcfg);
|
|---|
| 84 |
|
|---|
| 85 | # Parse command-line arguments and get the collection name
|
|---|
| 86 | if (!parsargv::parse(\@ARGV,
|
|---|
| 87 | 'verbosity/\d+/1', \$verbosity,
|
|---|
| 88 | 'archivedir/.*/', \$archivedir,
|
|---|
| 89 | 'phindexdir/.*/', \$phindexdir,
|
|---|
| 90 | 'languages/\w+/english', \$language,
|
|---|
| 91 | 'maxdocs/^\-?\d+/-1', \$maxdocs,
|
|---|
| 92 | 'untidy', \$untidy )) {
|
|---|
| 93 | &print_usage();
|
|---|
| 94 | die "\n";
|
|---|
| 95 | }
|
|---|
| 96 | $collection = &util::use_collection(@ARGV);
|
|---|
| 97 | if ($collection eq "") {
|
|---|
| 98 | &print_usage();
|
|---|
| 99 | die "\n";
|
|---|
| 100 | }
|
|---|
| 101 |
|
|---|
| 102 | # Read the collection configuration file
|
|---|
| 103 | $configfilename = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "etc", "collect.cfg");
|
|---|
| 104 | if (-e $configfilename) {
|
|---|
| 105 | $collectcfg = &colcfg::read_collect_cfg ($configfilename);
|
|---|
| 106 | if (defined $collectcfg->{'archivedir'} && $archivedir eq "") {
|
|---|
| 107 | $archivedir = $collectcfg->{'archivedir'};
|
|---|
| 108 | }
|
|---|
| 109 | if (defined $collectcfg->{'phindexdir'} && $phindexdir eq "") {
|
|---|
| 110 | $phindexdir = $collectcfg->{'phindexdir'};
|
|---|
| 111 | }
|
|---|
| 112 | if (defined $collectcfg->{'phind'}) {
|
|---|
| 113 | $phindcfg = $collectcfg->{'phind'};
|
|---|
| 114 | }
|
|---|
| 115 | } else {
|
|---|
| 116 | die "Couldn't find the configuration file $configfilename\n";
|
|---|
| 117 | }
|
|---|
| 118 |
|
|---|
| 119 | # Calculate default archive and phindex directories
|
|---|
| 120 | if ($archivedir eq "") {
|
|---|
| 121 | $archivedir = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "archives");
|
|---|
| 122 | }
|
|---|
| 123 | if ($phindexdir eq "") {
|
|---|
| 124 | $phindexdir = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "phindex");
|
|---|
| 125 | }
|
|---|
| 126 |
|
|---|
| 127 | # Make sure theuser has in fact requested phind indexes
|
|---|
| 128 | if (!defined($phindcfg)) {
|
|---|
| 129 | print "No phind information in $configfilename\n";
|
|---|
| 130 | exit;
|
|---|
| 131 | }
|
|---|
| 132 |
|
|---|
| 133 |
|
|---|
| 134 | # Construct phind indexes
|
|---|
| 135 | my $mode = 1;
|
|---|
| 136 | my $symbol_limit = 40000000;
|
|---|
| 137 | my ($command, $status);
|
|---|
| 138 |
|
|---|
| 139 | # Make a clean phindex directory
|
|---|
| 140 | if (-e "$phindexdir") {
|
|---|
| 141 | &util::rm_r("$phindexdir");
|
|---|
| 142 | }
|
|---|
| 143 | &util::mk_dir("$phindexdir");
|
|---|
| 144 |
|
|---|
| 145 | # Read the archives directory and build the clauses file
|
|---|
| 146 | print "\nReading archive directory\n" if $verbosity;
|
|---|
| 147 | &build_clauses($archivedir, $phindexdir, $language, $phindcfg, $verbosity, $maxdocs);
|
|---|
| 148 |
|
|---|
| 149 | # Generate the vocabulary, symbol statistics, and numbers file
|
|---|
| 150 | # from the clauses file
|
|---|
| 151 | print "\nExtracting vocabulary and statistics\n" if $verbosity;
|
|---|
| 152 | &extract_vocabulary($phindexdir, $language, $verbosity);
|
|---|
| 153 |
|
|---|
| 154 | # Use the suffix program to generate the phindex/phrases file
|
|---|
| 155 | &execute("suffix $phindexdir $symbol_limit $mode", $verbosity);
|
|---|
| 156 |
|
|---|
| 157 | # Create the phrase file and put phrase numbers in phindex/phrases
|
|---|
| 158 | print "\nSorting and Renumbering phrases for input to mgpp\n" if $verbosity;
|
|---|
| 159 | &renumber_phrases("$phindexdir", $verbosity);
|
|---|
| 160 |
|
|---|
| 161 | # Create the mg phrase database
|
|---|
| 162 | my $mgpp = &util::filename_cat($ENV{'GSDLHOME'}, "src", "mgpp");
|
|---|
| 163 | my $mg_passes = &util::filename_cat($mgpp, "text", "mg_passes");
|
|---|
| 164 | my $mg_compression_dict = &util::filename_cat($mgpp, "text", "mg_compression_dict");
|
|---|
| 165 |
|
|---|
| 166 | my $mg_perf_hash_build = &util::filename_cat($mgpp, "text", "mg_perf_hash_build");
|
|---|
| 167 | my $mg_weights_build = &util::filename_cat($mgpp, "text", "mg_weights_build");
|
|---|
| 168 | my $mg_invf_dict = &util::filename_cat($mgpp, "text", "mg_invf_dict");
|
|---|
| 169 | my $mg_stem_idx = &util::filename_cat($mgpp, "text", "mg_stem_idx");
|
|---|
| 170 |
|
|---|
| 171 | print "\nCreating phrase databases\n";
|
|---|
| 172 | my $mg_input = &util::filename_cat($phindexdir, "pdata.txt");
|
|---|
| 173 | my $mg_stem = "pdata";
|
|---|
| 174 |
|
|---|
| 175 | &execute("$mg_passes -d $phindexdir -f $mg_stem -T1 $mg_input", $verbosity);
|
|---|
| 176 | &execute("$mg_compression_dict -d $phindexdir -f $mg_stem", $verbosity);
|
|---|
| 177 | &execute("$mg_passes -d $phindexdir -f $mg_stem -T2 $mg_input", $verbosity);
|
|---|
| 178 |
|
|---|
| 179 | # create the mg index of words
|
|---|
| 180 | print "\nCreating word-level search indexes\n";
|
|---|
| 181 | $mg_input = &util::filename_cat($phindexdir, "pword.txt");
|
|---|
| 182 | $mg_stem = "pword";
|
|---|
| 183 |
|
|---|
| 184 | &execute("$mg_passes -d $phindexdir -f $mg_stem -T1 -I1 $mg_input", $verbosity);
|
|---|
| 185 | &execute("$mg_compression_dict -d $phindexdir -f $mg_stem", $verbosity);
|
|---|
| 186 | &execute("$mg_perf_hash_build -d $phindexdir -f $mg_stem", $verbosity);
|
|---|
| 187 | &execute("$mg_passes -d $phindexdir -f $mg_stem -T2 -I2 $mg_input", $verbosity);
|
|---|
| 188 | &execute("$mg_weights_build -d $phindexdir -f $mg_stem", $verbosity);
|
|---|
| 189 | &execute("$mg_invf_dict -d $phindexdir -f $mg_stem", $verbosity);
|
|---|
| 190 |
|
|---|
| 191 | &execute("$mg_stem_idx -d $phindexdir -f $mg_stem -s 1", $verbosity);
|
|---|
| 192 | &execute("$mg_stem_idx -d $phindexdir -f $mg_stem -s 2", $verbosity);
|
|---|
| 193 | &execute("$mg_stem_idx -d $phindexdir -f $mg_stem -s 3", $verbosity);
|
|---|
| 194 |
|
|---|
| 195 | # create the mg document information database
|
|---|
| 196 | print "\nCreating document information databases\n";
|
|---|
| 197 | $mg_input = &util::filename_cat($phindexdir, "docs.txt");
|
|---|
| 198 | $mg_stem = "docs";
|
|---|
| 199 |
|
|---|
| 200 | &execute("$mg_passes -d $phindexdir -f $mg_stem -T1 $mg_input", $verbosity);
|
|---|
| 201 | &execute("$mg_compression_dict -d $phindexdir -f $mg_stem", $verbosity);
|
|---|
| 202 | &execute("$mg_passes -d $phindexdir -f $mg_stem -T2 $mg_input", $verbosity);
|
|---|
| 203 |
|
|---|
| 204 |
|
|---|
| 205 | # Tidy up stray files
|
|---|
| 206 | if (!$untidy) {
|
|---|
| 207 | print "\nCleaning up\n" if $verbosity;
|
|---|
| 208 | &util::rm("$phindexdir/clauses", "$phindexdir/clauses.numbers",
|
|---|
| 209 | "$phindexdir/clauses.vocab", "$phindexdir/clauses.stats",
|
|---|
| 210 | "$phindexdir/phrases", "$phindexdir/docs.txt",
|
|---|
| 211 | "$phindexdir/pdata.txt", "$phindexdir/pword.txt");
|
|---|
| 212 | my $outfile = 1;
|
|---|
| 213 | while (-e "$phindexdir/outPhrase.$outfile") {
|
|---|
| 214 | &util::rm("$phindexdir/outPhrase.$outfile");
|
|---|
| 215 | $outfile++;
|
|---|
| 216 | }
|
|---|
| 217 | }
|
|---|
| 218 | }
|
|---|
| 219 |
|
|---|
| 220 | # Execute a system command
|
|---|
| 221 |
|
|---|
| 222 | sub execute {
|
|---|
| 223 | my ($command, $verbosity) = @_;
|
|---|
| 224 | print "Executing: $command\n" if $verbosity;
|
|---|
| 225 | my $status = system($command);
|
|---|
| 226 | if ($status != 0) {
|
|---|
| 227 | print STDERR "phindgen.pl - Error executing $command: $!\n";
|
|---|
| 228 | exit($status);
|
|---|
| 229 | }
|
|---|
| 230 | }
|
|---|
| 231 |
|
|---|
| 232 |
|
|---|
| 233 |
|
|---|
| 234 |
|
|---|
| 235 |
|
|---|
| 236 | # Get the text for the collection.
|
|---|
| 237 | # Save the document names and identifiers as we go.
|
|---|
| 238 |
|
|---|
| 239 | sub build_clauses {
|
|---|
| 240 | my ($archive_dir, $phindex_dir, $language, $phindcfg, $verbosity, $maxdocs) = @_;
|
|---|
| 241 |
|
|---|
| 242 | # create a "pluginfo" for ArcPlug and RecPlug
|
|---|
| 243 | my $plugins = [["GMLPlug"], ["ArcPlug"]];
|
|---|
| 244 | my $pluginfo = &plugin::load_plugins ($plugins, $verbosity);
|
|---|
| 245 | die "No plugins were loaded." if (scalar(@$pluginfo) == 0);
|
|---|
| 246 |
|
|---|
| 247 |
|
|---|
| 248 | # create a phind document processor object to process the documents
|
|---|
| 249 | my $processor = new phproc ($archive_dir, $phindex_dir, $phindcfg, $language,
|
|---|
| 250 | $doclimit, $verbosity, "STDOUT");
|
|---|
| 251 |
|
|---|
| 252 | # process the archives directory
|
|---|
| 253 | &plugin::read ($pluginfo, $archive_dir, "", {}, $processor, $maxdocs);
|
|---|
| 254 | }
|
|---|
| 255 |
|
|---|
| 256 |
|
|---|
| 257 |
|
|---|
| 258 | # Generate the vocabulary, symbol statistics, and numbers file from the
|
|---|
| 259 | # clauses file. This is legacy code, so is a bit messy and probably wont
|
|---|
| 260 | # run under windows.
|
|---|
| 261 |
|
|---|
| 262 | sub extract_vocabulary {
|
|---|
| 263 | my ($phindex_dir, $language, $verbosity) = @_;
|
|---|
| 264 |
|
|---|
| 265 | my ($w, $l, $line, $word);
|
|---|
| 266 |
|
|---|
| 267 | my ($first_delimiter, $last_delimiter,
|
|---|
| 268 | $first_stopword, $last_stopword,
|
|---|
| 269 | $first_extractword, $last_extractword,
|
|---|
| 270 | $first_contentword, $last_contentword,
|
|---|
| 271 | $phrasedelimiter);
|
|---|
| 272 |
|
|---|
| 273 | my ($use_thesaurus, %thesaurus, $first_thesaurusword, $last_thesaurusword);
|
|---|
| 274 |
|
|---|
| 275 |
|
|---|
| 276 | my %symbol;
|
|---|
| 277 | my (%freq);
|
|---|
| 278 |
|
|---|
| 279 | print "Calculating vocabulary\n" if ($verbosity > 1);
|
|---|
| 280 |
|
|---|
| 281 | # Read and store the stopwords
|
|---|
| 282 | my $words = `find $ENV{'GSDLHOME'}/etc/phind/$language -name "*.sw" | xargs cat`;
|
|---|
| 283 | my %stopwords;
|
|---|
| 284 | foreach my $w (split(/\s+/, $words)) {
|
|---|
| 285 | $l = lc($w);
|
|---|
| 286 | $stopwords{$l} = $w;
|
|---|
| 287 | }
|
|---|
| 288 |
|
|---|
| 289 | # Read and store the thesaurus terms
|
|---|
| 290 | $use_thesaurus = 0;
|
|---|
| 291 | my $lex_file = &util::filename_cat("$ENV{'GSDLHOME'}", "etc", "phind",
|
|---|
| 292 | "$language", "agrovoc.lex");
|
|---|
| 293 | if (-e "$lex_file") {
|
|---|
| 294 | open(TH, "<$lex_file");
|
|---|
| 295 | while(<TH>) {
|
|---|
| 296 | s/^\d+ //;
|
|---|
| 297 | s/\(.*\)//;
|
|---|
| 298 | foreach my $w (split(/\s+/, $_)) {
|
|---|
| 299 | $thesaurus{lc($w)} = $w;
|
|---|
| 300 | }
|
|---|
| 301 | }
|
|---|
| 302 | close TH;
|
|---|
| 303 | $use_thesaurus = 1;
|
|---|
| 304 | }
|
|---|
| 305 |
|
|---|
| 306 | # Read words in the text and count occurences
|
|---|
| 307 | open(TXT, "<$phindex_dir/clauses");
|
|---|
| 308 | my @words;
|
|---|
| 309 |
|
|---|
| 310 | while(<TXT>) {
|
|---|
| 311 | $line = $_;
|
|---|
| 312 | next unless ($line =~ /./);
|
|---|
| 313 |
|
|---|
| 314 | @words = split(/\s+/, $line);
|
|---|
| 315 | foreach $w (@words) {
|
|---|
| 316 | $l = lc($w);
|
|---|
| 317 | $w = $l if ((defined $stopwords{$l}) || (defined $thesaurus{$l}));
|
|---|
| 318 | $freq{$w}++;
|
|---|
| 319 | }
|
|---|
| 320 | $freq{$senlimit}++;
|
|---|
| 321 | }
|
|---|
| 322 |
|
|---|
| 323 | # Calculate the "best" form of each word
|
|---|
| 324 | my (%bestform, %totalfreq, %bestfreq);
|
|---|
| 325 |
|
|---|
| 326 | foreach $w (sort (keys %freq)) {
|
|---|
| 327 | $l = lc($w);
|
|---|
| 328 |
|
|---|
| 329 | # totalfreq is the number of times a term appears in any form
|
|---|
| 330 | $totalfreq{$l} += $freq{$w};
|
|---|
| 331 |
|
|---|
| 332 | if (defined $stopwords{$l}) {
|
|---|
| 333 | $bestform{$l} = $stopwords{$l};
|
|---|
| 334 |
|
|---|
| 335 | } elsif (defined $thesaurus{$l}) {
|
|---|
| 336 | $bestform{$l} = $thesaurus{$l};
|
|---|
| 337 |
|
|---|
| 338 | } elsif (!$bestform{$l} || ($freq{$w} > $bestfreq{$l})) {
|
|---|
| 339 | $bestfreq{$l} = $freq{$w};
|
|---|
| 340 | $bestform{$l} = $w;
|
|---|
| 341 | }
|
|---|
| 342 | }
|
|---|
| 343 |
|
|---|
| 344 | undef %freq;
|
|---|
| 345 | undef %bestfreq;
|
|---|
| 346 |
|
|---|
| 347 |
|
|---|
| 348 | # Assign symbol numbers to tokens
|
|---|
| 349 | my $nextsymbol = 1;
|
|---|
| 350 | my (@vocab);
|
|---|
| 351 |
|
|---|
| 352 | # Delimiters
|
|---|
| 353 | $first_delimiter = 1;
|
|---|
| 354 |
|
|---|
| 355 | foreach $word (@delimiters) {
|
|---|
| 356 |
|
|---|
| 357 | $word = lc($word);
|
|---|
| 358 | $bestform{$word} = uc($word);
|
|---|
| 359 | $vocab[$nextsymbol] = $word;
|
|---|
| 360 | $symbol{$word} = $nextsymbol;
|
|---|
| 361 | $nextsymbol++;
|
|---|
| 362 | }
|
|---|
| 363 | $last_delimiter = $nextsymbol - 1;
|
|---|
| 364 |
|
|---|
| 365 | # Stopwords
|
|---|
| 366 | $first_stopword = $nextsymbol;
|
|---|
| 367 |
|
|---|
| 368 | foreach my $word (sort keys %stopwords) {
|
|---|
| 369 |
|
|---|
| 370 | # don't incluse stopword unless it occurs in the text
|
|---|
| 371 | $word = lc($word);
|
|---|
| 372 | next unless ($totalfreq{$word});
|
|---|
| 373 | next if ($symbol{$word});
|
|---|
| 374 |
|
|---|
| 375 | $vocab[$nextsymbol] = $word;
|
|---|
| 376 | $symbol{$word} = $nextsymbol;
|
|---|
| 377 | $nextsymbol++;
|
|---|
| 378 | }
|
|---|
| 379 | $last_stopword = $nextsymbol - 1;
|
|---|
| 380 | $first_contentword = $nextsymbol;
|
|---|
| 381 |
|
|---|
| 382 | # Thesaurus terms
|
|---|
| 383 | if ($use_thesaurus) {
|
|---|
| 384 | $first_thesaurusword = $nextsymbol;
|
|---|
| 385 |
|
|---|
| 386 | foreach my $word (sort keys %thesaurus) {
|
|---|
| 387 |
|
|---|
| 388 | $word = lc($word);
|
|---|
| 389 | next if ($symbol{$word});
|
|---|
| 390 | $bestform{$word} = $thesaurus{$word};
|
|---|
| 391 |
|
|---|
| 392 | $vocab[$nextsymbol] = $word;
|
|---|
| 393 | $symbol{$word} = $nextsymbol;
|
|---|
| 394 | $nextsymbol++;
|
|---|
| 395 |
|
|---|
| 396 | }
|
|---|
| 397 | $last_thesaurusword = $nextsymbol - 1;
|
|---|
| 398 | }
|
|---|
| 399 |
|
|---|
| 400 | # Other content words
|
|---|
| 401 | $first_extractword = $nextsymbol;
|
|---|
| 402 |
|
|---|
| 403 | foreach my $word (sort (keys %bestform)) {
|
|---|
| 404 |
|
|---|
| 405 | next if ($symbol{$word});
|
|---|
| 406 |
|
|---|
| 407 | $vocab[$nextsymbol] = $word;
|
|---|
| 408 | $symbol{$word} = $nextsymbol;
|
|---|
| 409 | $nextsymbol++;
|
|---|
| 410 | }
|
|---|
| 411 | $last_extractword = $nextsymbol - 1;
|
|---|
| 412 | $last_contentword = $nextsymbol - 1;
|
|---|
| 413 |
|
|---|
| 414 |
|
|---|
| 415 | # Outut the words
|
|---|
| 416 | print "Saving vocabulary in $phindex_dir/clauses.vocab\n" if ($verbosity > 1);
|
|---|
| 417 | open(VOC, ">$phindex_dir/clauses.vocab");
|
|---|
| 418 |
|
|---|
| 419 | for (my $i = 1; $i < $nextsymbol; $i++) {
|
|---|
| 420 | $w = $vocab[$i];
|
|---|
| 421 |
|
|---|
| 422 | print VOC "$bestform{$w}\n";
|
|---|
| 423 | $totalfreq{$w} = 0 unless ($totalfreq{$w});
|
|---|
| 424 | }
|
|---|
| 425 | close VOC;
|
|---|
| 426 |
|
|---|
| 427 |
|
|---|
| 428 | # Output statistics about the vocablary
|
|---|
| 429 | print "Saving statistics in $phindex_dir/clauses.stats\n" if ($verbosity > 1);
|
|---|
| 430 | &util::rm("$phindex_dir/clauses.stats") if (-e "$phindex_dir/clauses.stats");
|
|---|
| 431 | open(STAT, ">$phindex_dir/clauses.stats")
|
|---|
| 432 | || die "Cannot open $phindex_dir/clauses.stats: $!";
|
|---|
| 433 |
|
|---|
| 434 | print STAT "first_delimiter $first_delimiter\n";
|
|---|
| 435 | print STAT "last_delimiter $last_delimiter\n";
|
|---|
| 436 | print STAT "first_stopword $first_stopword\n";
|
|---|
| 437 | print STAT "last_stopword $last_stopword\n";
|
|---|
| 438 | if ($use_thesaurus) {
|
|---|
| 439 | print STAT "first_thesaurusword $first_thesaurusword\n";
|
|---|
| 440 | print STAT "last_thesaurusword $last_thesaurusword\n";
|
|---|
| 441 | }
|
|---|
| 442 | print STAT "first_extractword $first_extractword\n";
|
|---|
| 443 | print STAT "last_extractword $last_extractword\n";
|
|---|
| 444 | print STAT "first_contentword $first_contentword\n";
|
|---|
| 445 | print STAT "last_contentword $last_contentword\n";
|
|---|
| 446 | print STAT "first_symbol $first_delimiter\n";
|
|---|
| 447 | print STAT "last_symbol $last_contentword\n";
|
|---|
| 448 | print STAT "first_word $first_stopword\n";
|
|---|
| 449 | print STAT "last_word $last_contentword\n";
|
|---|
| 450 | close STAT;
|
|---|
| 451 |
|
|---|
| 452 | undef @vocab;
|
|---|
| 453 |
|
|---|
| 454 |
|
|---|
| 455 | # Save text as symbol numbers
|
|---|
| 456 | print "Saving text as numbers in $phindex_dir/clauses.numbers\n" if ($verbosity > 1);
|
|---|
| 457 |
|
|---|
| 458 | open(TXT, "<$phindex_dir/clauses");
|
|---|
| 459 | open(NUM, ">$phindex_dir/clauses.numbers");
|
|---|
| 460 |
|
|---|
| 461 | $phrasedelimiter = $symbol{lc($senlimit)};
|
|---|
| 462 | print NUM "$symbol{lc($colstart)}\n";
|
|---|
| 463 |
|
|---|
| 464 | # set up the special symbols that delimit documents and sentences
|
|---|
| 465 | while(<TXT>) {
|
|---|
| 466 |
|
|---|
| 467 | # split sentence into a list of tokens
|
|---|
| 468 | $line = $_;
|
|---|
| 469 | next unless ($line =~ /./);
|
|---|
| 470 | @words = split(/\s+/, $line);
|
|---|
| 471 |
|
|---|
| 472 | # output one token at a time
|
|---|
| 473 | foreach $word (@words) {
|
|---|
| 474 | $word = lc($word);
|
|---|
| 475 | print NUM "$symbol{$word}\n";
|
|---|
| 476 | }
|
|---|
| 477 |
|
|---|
| 478 | # output phrase delimiter
|
|---|
| 479 | print NUM "$phrasedelimiter\n";
|
|---|
| 480 | }
|
|---|
| 481 |
|
|---|
| 482 | print NUM "$symbol{lc($colend)}\n";
|
|---|
| 483 |
|
|---|
| 484 | }
|
|---|
| 485 |
|
|---|
| 486 |
|
|---|
| 487 | # Prepare the phrases file to be input to mgpp.
|
|---|
| 488 | # This means renumbering the phrases in order of decreasing frequency.
|
|---|
| 489 |
|
|---|
| 490 |
|
|---|
| 491 | # This is legacy code, and a little ugly, and may be unix-specific
|
|---|
| 492 | # (particularly the sort command).
|
|---|
| 493 |
|
|---|
| 494 | sub renumber_phrases {
|
|---|
| 495 | my ($phindex_dir, $verbosity) = @_;
|
|---|
| 496 |
|
|---|
| 497 | # Sort the phrases into order of increasing frequency
|
|---|
| 498 | # This means the expansions will be sorted correctly later on.
|
|---|
| 499 | print "Sorting phrases into freq order\n" if ($verbosity);
|
|---|
| 500 | system("sort -rnt ':' +2 -o $phindex_dir/phrases $phindex_dir/phrases");
|
|---|
| 501 |
|
|---|
| 502 | my @symbol;
|
|---|
| 503 |
|
|---|
| 504 | # Read the vocabulary
|
|---|
| 505 | print "Reading the vocabulary\n" if ($verbosity);
|
|---|
| 506 | open(V, "<$phindex_dir/clauses.vocab")
|
|---|
| 507 | || die "Cannot open $phindex_dir/clauses.vocab: $!";
|
|---|
| 508 |
|
|---|
| 509 | my $i = 1;
|
|---|
| 510 | while(<V>) {
|
|---|
| 511 | chomp;
|
|---|
| 512 | $symbol[$i++] = $_;
|
|---|
| 513 | }
|
|---|
| 514 |
|
|---|
| 515 | # Create file for phrase data
|
|---|
| 516 | #
|
|---|
| 517 | # The phrases file looks something like this
|
|---|
| 518 | # 159396-1:s5175:4:1:116149-2:3:d2240,2;d2253;d2254
|
|---|
| 519 | # 159409-1:s5263:6:1:159410-2:6:d2122;d2128;d2129;d2130;d2215;d2380
|
|---|
| 520 | # 159415-1:s5267:9:1:159418-2:8:d3,2;d632;d633;d668;d1934;d2010;d2281;d2374
|
|---|
| 521 | # 159426-1:s5273:5:2:159429-2,115168-17:5:d252;d815;d938;d939;d2361
|
|---|
| 522 |
|
|---|
| 523 | # The first field on each line is a unique phrase identifier.
|
|---|
| 524 | # We need to calculate phrase numbers for each phrase
|
|---|
| 525 | print "Calculate phrase numbers\n" if ($verbosity);
|
|---|
| 526 |
|
|---|
| 527 | my %phrasenumber;
|
|---|
| 528 | my $nextphrase = 1;
|
|---|
| 529 | my ($line);
|
|---|
| 530 |
|
|---|
| 531 | open(IN, "<$phindex_dir/phrases");
|
|---|
| 532 | while(<IN>) {
|
|---|
| 533 |
|
|---|
| 534 | # read the line
|
|---|
| 535 | chomp;
|
|---|
| 536 | $line = $_;
|
|---|
| 537 |
|
|---|
| 538 | # we're only interested in the first field
|
|---|
| 539 | $line =~ s/:.*//;
|
|---|
| 540 |
|
|---|
| 541 | # get a phrase number for this line
|
|---|
| 542 | $phrasenumber{$line} = $nextphrase;
|
|---|
| 543 | $nextphrase++;
|
|---|
| 544 | }
|
|---|
| 545 |
|
|---|
| 546 |
|
|---|
| 547 | # Now we create a new phrase file using phrase numbers, not the old IDs.
|
|---|
| 548 | print "Format phrase data for MGPP\n" if ($verbosity);
|
|---|
| 549 |
|
|---|
| 550 | open(IN, "<$phindex_dir/phrases");
|
|---|
| 551 | open(DATA, ">$phindex_dir/pdata.txt");
|
|---|
| 552 | open(IDX, ">$phindex_dir/pword.txt");
|
|---|
| 553 |
|
|---|
| 554 | my ($key, $tf, $num, $countexp, $expansions, $countdocs, $documents, $text, $word);
|
|---|
| 555 | my @fields;
|
|---|
| 556 | my @documents;
|
|---|
| 557 | my (@newexp, $k, $n);
|
|---|
| 558 |
|
|---|
| 559 | my $linenumber = 0;
|
|---|
| 560 |
|
|---|
| 561 | while(<IN>) {
|
|---|
| 562 |
|
|---|
| 563 | # read the line
|
|---|
| 564 | chomp;
|
|---|
| 565 | $line = $_;
|
|---|
| 566 | @fields = split(/:/, $line);
|
|---|
| 567 |
|
|---|
| 568 | # get a phrase number for this line
|
|---|
| 569 | $key = shift @fields;
|
|---|
| 570 | die unless (defined($phrasenumber{$key}));
|
|---|
| 571 | $num = $phrasenumber{$key};
|
|---|
| 572 |
|
|---|
| 573 | # get the text of the phrase
|
|---|
| 574 | $text = shift @fields;
|
|---|
| 575 | $text =~ s/s(\d+)/$symbol[$1]/g;
|
|---|
| 576 | if ($text =~ / /) {
|
|---|
| 577 | $word = "";
|
|---|
| 578 | } else {
|
|---|
| 579 | $word = $text;
|
|---|
| 580 | }
|
|---|
| 581 |
|
|---|
| 582 | $linenumber++;
|
|---|
| 583 | if ($linenumber % 1000 == 0) {
|
|---|
| 584 | print "line $linenumber:\t$num\t$key\t($text)\n" if ($verbosity > 2);
|
|---|
| 585 | }
|
|---|
| 586 | print "$num: $key\t($text)\n" if ($verbosity > 3);
|
|---|
| 587 |
|
|---|
| 588 | # get the phrase frequency
|
|---|
| 589 | $tf = shift @fields;
|
|---|
| 590 |
|
|---|
| 591 | # get the number of expansions
|
|---|
| 592 | $countexp = shift @fields;
|
|---|
| 593 |
|
|---|
| 594 | # get the expansions and convert them into phrase numbers
|
|---|
| 595 | $expansions = shift @fields;
|
|---|
| 596 | @newexp = ();
|
|---|
| 597 | foreach $k (split(/,/, $expansions)) {
|
|---|
| 598 | die "ERROR - no phrase number for: $k" unless (defined($phrasenumber{$k}));
|
|---|
| 599 | $n = $phrasenumber{$k};
|
|---|
| 600 | push @newexp, $n;
|
|---|
| 601 | }
|
|---|
| 602 | @newexp = sort numerically @newexp;
|
|---|
| 603 |
|
|---|
| 604 | # get the number of documents
|
|---|
| 605 | $countdocs = shift @fields;
|
|---|
| 606 |
|
|---|
| 607 | # get the documents
|
|---|
| 608 | $documents = shift @fields;
|
|---|
| 609 | $documents =~ s/d//g;
|
|---|
| 610 | @documents = split(/;/, $documents);
|
|---|
| 611 | @documents = sort by_frequency @documents;
|
|---|
| 612 |
|
|---|
| 613 | # output the phrase data
|
|---|
| 614 | print DATA "<Document>";
|
|---|
| 615 | print DATA "$num:$text:$tf:$countexp:$countdocs:";
|
|---|
| 616 | print DATA join(",", @newexp), ":", join(";", @documents), "\n";
|
|---|
| 617 |
|
|---|
| 618 | # output the word index search data
|
|---|
| 619 | print IDX "<Document>$word\n";
|
|---|
| 620 |
|
|---|
| 621 |
|
|---|
| 622 | }
|
|---|
| 623 | }
|
|---|
| 624 |
|
|---|
| 625 | # sort routines used to renumber phrases
|
|---|
| 626 |
|
|---|
| 627 | sub numerically { $a <=> $b }
|
|---|
| 628 |
|
|---|
| 629 | sub by_frequency {
|
|---|
| 630 | my $fa = 1;
|
|---|
| 631 | if ($a =~ /,/) {
|
|---|
| 632 | $fa = $a;
|
|---|
| 633 | $fa =~ s/\d+,//;
|
|---|
| 634 | }
|
|---|
| 635 | my $fb = 1;
|
|---|
| 636 | if ($b =~ /,/) {
|
|---|
| 637 | $fb = $b;
|
|---|
| 638 | $fb =~ s/\d+,//;
|
|---|
| 639 | }
|
|---|
| 640 |
|
|---|
| 641 | return ($fb <=> $fa);
|
|---|
| 642 | }
|
|---|
| 643 |
|
|---|
