[8027] | 1 | /**********************************************************************
|
---|
| 2 | *
|
---|
| 3 | * lucenequeryfilter.cpp --
|
---|
| 4 | * Copyright (C) 1999 The New Zealand Digital Library Project
|
---|
| 5 | *
|
---|
| 6 | * A component of the Greenstone digital library software
|
---|
| 7 | * from the New Zealand Digital Library Project at the
|
---|
| 8 | * University of Waikato, New Zealand.
|
---|
| 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 | #include "lucenequeryfilter.h"
|
---|
| 27 | #include "fileutil.h"
|
---|
| 28 | #include "lucenesearch.h"
|
---|
| 29 |
|
---|
| 30 | /////////////////////////////////
|
---|
| 31 | // functions for queryfilterclass
|
---|
| 32 | /////////////////////////////////
|
---|
| 33 |
|
---|
| 34 |
|
---|
| 35 | lucenequeryfilterclass::lucenequeryfilterclass ()
|
---|
| 36 | : queryfilterclass() {
|
---|
| 37 |
|
---|
| 38 |
|
---|
| 39 | FilterOption_t filtopt;
|
---|
| 40 |
|
---|
| 41 | // -- onePerTerm Level enumerated
|
---|
| 42 | // likely to be Doc, Sec, Para, but we dont assume anything now
|
---|
| 43 | filtopt.clear();
|
---|
| 44 | filtopt.name = "Level";
|
---|
| 45 | filtopt.type = FilterOption_t::enumeratedt;
|
---|
| 46 | filtopt.repeatable = FilterOption_t::onePerTerm;
|
---|
| 47 | filterOptions["Level"] = filtopt;
|
---|
| 48 |
|
---|
| 49 | // -- IndexField, enumerated, used to list available fields
|
---|
| 50 | filtopt.clear();
|
---|
| 51 | filtopt.name = "IndexField";
|
---|
| 52 | filtopt.type = FilterOption_t::enumeratedt;
|
---|
| 53 | filtopt.repeatable = FilterOption_t::onePerTerm;
|
---|
| 54 | filtopt.defaultValue = "";
|
---|
| 55 | filterOptions["IndexField"] = filtopt;
|
---|
| 56 |
|
---|
| 57 | }
|
---|
| 58 |
|
---|
| 59 | lucenequeryfilterclass::~lucenequeryfilterclass () {
|
---|
| 60 | }
|
---|
| 61 |
|
---|
| 62 |
|
---|
| 63 | //whether a query is a full text browse
|
---|
| 64 | bool lucenequeryfilterclass::full_text_browse (int filterRequestOptions) {
|
---|
| 65 | return (filterRequestOptions & FRfullTextBrowse);
|
---|
| 66 | }
|
---|
| 67 |
|
---|
| 68 | void lucenequeryfilterclass::configure (const text_t &key, const text_tarray &cfgline) {
|
---|
| 69 | queryfilterclass::configure(key, cfgline);
|
---|
| 70 |
|
---|
| 71 | if (key == "indexfieldmap") {
|
---|
| 72 | indexfieldmap.importmap (cfgline);
|
---|
| 73 |
|
---|
| 74 | // update the list of indexes in the filter information
|
---|
| 75 | text_tarray options;
|
---|
| 76 | indexfieldmap.gettoarray (options);
|
---|
| 77 |
|
---|
| 78 | text_tarray::const_iterator here = options.begin();
|
---|
| 79 | text_tarray::const_iterator end = options.end();
|
---|
| 80 | bool start = true;
|
---|
| 81 | while (here !=end) {
|
---|
| 82 | if (!(*here).empty()) {
|
---|
| 83 | filterOptions["IndexField"].validValues.push_back(*here);
|
---|
| 84 | if (start) {
|
---|
| 85 | filterOptions["IndexField"].defaultValue = *here;
|
---|
| 86 | start = false;
|
---|
| 87 | }
|
---|
| 88 | }
|
---|
[9620] | 89 | ++here;
|
---|
[8027] | 90 | }
|
---|
| 91 | } else if (key == "indexlevels") {
|
---|
| 92 | text_tarray::const_iterator here = cfgline.begin();
|
---|
| 93 | text_tarray::const_iterator end = cfgline.end();
|
---|
| 94 | bool first=true;
|
---|
| 95 | while (here != end) {
|
---|
| 96 | if (!(*here).empty()) {
|
---|
| 97 | if (first) {
|
---|
| 98 | first = false;
|
---|
| 99 | // the default is the first value
|
---|
| 100 | filterOptions["Level"].defaultValue = *here;
|
---|
| 101 | }
|
---|
| 102 | filterOptions["Level"].validValues.push_back(*here);
|
---|
| 103 | }
|
---|
[9620] | 104 | ++here;
|
---|
[8027] | 105 | }
|
---|
| 106 | } else if (key == "textlevel") {
|
---|
[15594] | 107 | ((lucenesearchclass *)textsearchptr)->set_text_level(cfgline[0]);
|
---|
[8027] | 108 | }
|
---|
| 109 |
|
---|
| 110 | }
|
---|
| 111 |
|
---|
| 112 |
|
---|
| 113 | void lucenequeryfilterclass::filter(const FilterRequest_t &request,
|
---|
| 114 | FilterResponse_t &response,
|
---|
| 115 | comerror_t &err, ostream &logout) {
|
---|
| 116 |
|
---|
| 117 | outconvertclass text_t2ascii;
|
---|
| 118 |
|
---|
| 119 | response.clear ();
|
---|
| 120 | err = noError;
|
---|
[15558] | 121 | if (db_ptr == NULL) {
|
---|
[8027] | 122 | // most likely a configuration problem
|
---|
| 123 | logout << text_t2ascii
|
---|
[15558] | 124 | << "configuration error: queryfilter contains a null dbclass\n\n";
|
---|
[8027] | 125 | err = configurationError;
|
---|
| 126 | return;
|
---|
| 127 | }
|
---|
| 128 | if (textsearchptr == NULL) {
|
---|
| 129 | // most likely a configuration problem
|
---|
| 130 | logout << text_t2ascii
|
---|
[9110] | 131 | << "configuration error: queryfilter contains a null textsearchclass (lucene)\n\n";
|
---|
[8027] | 132 | err = configurationError;
|
---|
| 133 | return;
|
---|
| 134 | }
|
---|
| 135 | if (full_text_browse(request.filterResultOptions)) {
|
---|
| 136 | browsefilter(request, response, err, logout);
|
---|
| 137 | return;
|
---|
| 138 | }
|
---|
| 139 | // open the database
|
---|
[15558] | 140 | db_ptr->setlogout(&logout);
|
---|
| 141 | if (!db_ptr->opendatabase (db_filename, DB_READER, 100, false)) {
|
---|
| 142 | // most likely a system problem (we have already checked that the database exists)
|
---|
[8027] | 143 | logout << text_t2ascii
|
---|
[15558] | 144 | << "system problem: open on database \"" << db_filename << "\" failed\n\n";
|
---|
[8027] | 145 | err = systemProblem;
|
---|
| 146 | return;
|
---|
| 147 | }
|
---|
| 148 |
|
---|
| 149 |
|
---|
| 150 | // get the query parameters
|
---|
| 151 | int startresults, endresults;
|
---|
| 152 | text_t phrasematch; // not used here any more
|
---|
| 153 | vector<queryparamclass> queryfilterparams;
|
---|
| 154 | parse_query_params (request, queryfilterparams, startresults,
|
---|
| 155 | endresults, phrasematch, logout);
|
---|
| 156 |
|
---|
| 157 |
|
---|
| 158 | // do query
|
---|
| 159 | queryresultsclass queryresults;
|
---|
| 160 | do_multi_query (request, queryfilterparams, queryresults, err, logout);
|
---|
[12421] | 161 | response.error_message = queryresults.error_message;
|
---|
[8027] | 162 | if (err != noError) return;
|
---|
[12421] | 163 |
|
---|
[8027] | 164 | // assemble document results
|
---|
| 165 | if (need_matching_docs (request.filterResultOptions)) {
|
---|
| 166 |
|
---|
| 167 | int resultnum = 1;
|
---|
| 168 | ResultDocInfo_t resultdoc;
|
---|
| 169 | text_t trans_OID;
|
---|
| 170 | vector<int>::iterator docorder_here = queryresults.docs.docorder.begin();
|
---|
| 171 | vector<int>::iterator docorder_end = queryresults.docs.docorder.end();
|
---|
| 172 |
|
---|
[12655] | 173 | // Now handled by Lucene directly
|
---|
| 174 | //if (endresults == -1) endresults = MAXNUMDOCS;
|
---|
| 175 |
|
---|
| 176 | while (docorder_here != docorder_end)
|
---|
| 177 | {
|
---|
| 178 | // Now handled by Lucene directly
|
---|
| 179 | //if (resultnum > endresults) break;
|
---|
[8027] | 180 |
|
---|
[12655] | 181 | // translate the document number
|
---|
[15558] | 182 | if (!translate(db_ptr, *docorder_here, trans_OID))
|
---|
[12655] | 183 | {
|
---|
| 184 | logout << text_t2ascii
|
---|
| 185 | << "warning: could not translate lucene document number \""
|
---|
| 186 | << *docorder_here << "\" to OID.\n\n";
|
---|
| 187 |
|
---|
| 188 | }
|
---|
| 189 | else
|
---|
| 190 | {
|
---|
| 191 | docresultmap::iterator docset_here = queryresults.docs.docset.find (*docorder_here);
|
---|
[8027] | 192 |
|
---|
[12655] | 193 | // see if there is a result for this number,
|
---|
| 194 | // if it is in the request set (or the request set is empty)
|
---|
| 195 | if (docset_here != queryresults.docs.docset.end() && (request.docSet.empty() || in_set(request.docSet, trans_OID)))
|
---|
| 196 | {
|
---|
| 197 | // Now handled by Lucene directly
|
---|
| 198 | //if (resultnum >= startresults) {
|
---|
[8027] | 199 |
|
---|
[12655] | 200 | // add this document
|
---|
| 201 | resultdoc.OID = trans_OID;
|
---|
| 202 | resultdoc.result_num = resultnum;
|
---|
| 203 | resultdoc.ranking = (int)((*docset_here).second.docweight * 10000.0 + 0.5);
|
---|
| 204 |
|
---|
| 205 | response.docInfo.push_back (resultdoc);
|
---|
| 206 | //}
|
---|
| 207 | ++resultnum;
|
---|
| 208 | }
|
---|
| 209 | } // else
|
---|
| 210 |
|
---|
| 211 | ++docorder_here;
|
---|
| 212 | }
|
---|
[8027] | 213 | } // if need matching docs
|
---|
[12655] | 214 |
|
---|
[8027] | 215 | // assemble the term results
|
---|
| 216 | if (need_term_info(request.filterResultOptions)) {
|
---|
| 217 | // note: the terms have already been sorted and uniqued - ?? have they??
|
---|
| 218 |
|
---|
| 219 | TermInfo_t terminfo;
|
---|
| 220 | bool terms_first = true;
|
---|
| 221 |
|
---|
| 222 | termfreqclassarray::iterator terms_here = queryresults.terms.begin();
|
---|
| 223 | termfreqclassarray::iterator terms_end = queryresults.terms.end();
|
---|
| 224 |
|
---|
| 225 | while (terms_here != terms_end) {
|
---|
| 226 | terminfo.clear();
|
---|
| 227 | terminfo.term = (*terms_here).termstr;
|
---|
| 228 | terminfo.freq = (*terms_here).termfreq;
|
---|
[13063] | 229 | // lucene doesn't return any termvariants at this stage,
|
---|
| 230 | // so make sure the original term is set
|
---|
| 231 | terminfo.matchTerms.push_back(terminfo.term);
|
---|
| 232 |
|
---|
[8027] | 233 | // this bit gets the matchTerms ie the equivalent (stem/casefold) terms
|
---|
| 234 | if (terms_first) {
|
---|
| 235 | text_tset::iterator termvariants_here = queryresults.termvariants.begin();
|
---|
| 236 | text_tset::iterator termvariants_end = queryresults.termvariants.end();
|
---|
| 237 | while (termvariants_here != termvariants_end) {
|
---|
| 238 | terminfo.matchTerms.push_back (*termvariants_here);
|
---|
[9620] | 239 | ++termvariants_here;
|
---|
[8027] | 240 | }
|
---|
| 241 | }
|
---|
| 242 | terms_first = false;
|
---|
| 243 |
|
---|
| 244 | response.termInfo.push_back (terminfo);
|
---|
| 245 |
|
---|
[9620] | 246 | ++terms_here;
|
---|
[8027] | 247 | }
|
---|
[12380] | 248 |
|
---|
| 249 | // add the stop words
|
---|
| 250 | text_tset::iterator stopwords_here = queryresults.stopwords.begin();
|
---|
| 251 | text_tset::iterator stopwords_end = queryresults.stopwords.end();
|
---|
| 252 | while (stopwords_here != stopwords_end) {
|
---|
| 253 | response.stopwords.insert(*stopwords_here);
|
---|
| 254 | ++stopwords_here;
|
---|
| 255 | }
|
---|
[8027] | 256 | }
|
---|
| 257 |
|
---|
[15558] | 258 | db_ptr->closedatabase(); // Important that local library doesn't leave any files open
|
---|
[8027] | 259 | response.numDocs = queryresults.docs_matched;
|
---|
| 260 | response.isApprox = queryresults.is_approx;
|
---|
| 261 | }
|
---|
| 262 |
|
---|
| 263 | void lucenequeryfilterclass::browsefilter(const FilterRequest_t &request,
|
---|
| 264 | FilterResponse_t &response,
|
---|
| 265 | comerror_t &err, ostream &logout) {
|
---|
| 266 |
|
---|
| 267 | outconvertclass text_t2ascii;
|
---|
| 268 |
|
---|
| 269 | // get the query parameters
|
---|
| 270 | int startresults, endresults;
|
---|
| 271 | text_t phrasematch; // not used here any more, just have it so can use
|
---|
| 272 | // parse_query_params function
|
---|
| 273 |
|
---|
| 274 | vector<queryparamclass> queryfilterparams;
|
---|
| 275 | parse_query_params (request, queryfilterparams, startresults,
|
---|
| 276 | endresults, phrasematch, logout);
|
---|
| 277 |
|
---|
| 278 | vector<queryparamclass>::const_iterator query_here = queryfilterparams.begin();
|
---|
| 279 |
|
---|
| 280 | // do query
|
---|
| 281 | queryresultsclass queryresults;
|
---|
| 282 | queryresults.clear();
|
---|
| 283 |
|
---|
| 284 | int numDocs = endresults-startresults;
|
---|
| 285 | textsearchptr->setcollectdir (collectdir);
|
---|
| 286 |
|
---|
| 287 | if (!((lucenesearchclass*)textsearchptr)->browse_search((*query_here), startresults, numDocs, queryresults)) {
|
---|
| 288 | // most likely a system problem
|
---|
| 289 | logout << text_t2ascii
|
---|
| 290 | << "system problem: could not do full text browse with lucene for index \""
|
---|
| 291 | << (*query_here).index << (*query_here).subcollection
|
---|
| 292 | << (*query_here).language << "\".\n\n";
|
---|
| 293 | err = systemProblem;
|
---|
| 294 | return;
|
---|
| 295 | }
|
---|
| 296 |
|
---|
| 297 | // assemble the term results
|
---|
| 298 | TermInfo_t terminfo;
|
---|
| 299 |
|
---|
| 300 | termfreqclassarray::iterator terms_here = queryresults.terms.begin();
|
---|
| 301 | termfreqclassarray::iterator terms_end = queryresults.terms.end();
|
---|
| 302 |
|
---|
| 303 | while (terms_here != terms_end) {
|
---|
| 304 | terminfo.clear();
|
---|
| 305 | terminfo.term = (*terms_here).termstr;
|
---|
| 306 | terminfo.freq = (*terms_here).termfreq;
|
---|
| 307 |
|
---|
| 308 | response.termInfo.push_back (terminfo);
|
---|
| 309 |
|
---|
[9620] | 310 | ++terms_here;
|
---|
[8027] | 311 | }
|
---|
| 312 |
|
---|
| 313 |
|
---|
| 314 | }
|
---|
| 315 |
|
---|
[15558] | 316 | // lucenesearchptr and db_ptr are assumed to be valid
|
---|
[8027] | 317 | void lucenequeryfilterclass::do_multi_query (const FilterRequest_t &request,
|
---|
| 318 | const vector<queryparamclass> &query_params,
|
---|
| 319 | queryresultsclass &multiresults,
|
---|
| 320 | comerror_t &err, ostream &logout) {
|
---|
| 321 | outconvertclass text_t2ascii;
|
---|
| 322 |
|
---|
| 323 | err = noError;
|
---|
| 324 | textsearchptr->setcollectdir (collectdir);
|
---|
| 325 | multiresults.clear();
|
---|
| 326 |
|
---|
| 327 | vector<queryparamclass>::const_iterator query_here = query_params.begin();
|
---|
| 328 | vector<queryparamclass>::const_iterator query_end = query_params.end();
|
---|
| 329 | while (query_here != query_end) {
|
---|
| 330 | queryresultsclass thisqueryresults;
|
---|
| 331 | if (!textsearchptr->search((*query_here), thisqueryresults)) {
|
---|
| 332 | // most likely a system problem
|
---|
| 333 | logout << text_t2ascii
|
---|
| 334 | << "system problem: could not do search with lucene for index \""
|
---|
[9090] | 335 | << (*query_here).index << (*query_here).level
|
---|
| 336 | << (*query_here).subcollection
|
---|
[8027] | 337 | << (*query_here).language << "\".\n\n";
|
---|
| 338 | err = systemProblem;
|
---|
| 339 | return;
|
---|
| 340 | }
|
---|
| 341 |
|
---|
| 342 | // check for syntax error
|
---|
| 343 | if (thisqueryresults.syntax_error==true) {
|
---|
| 344 | logout << text_t2ascii
|
---|
| 345 | << "syntax problem: invalid query string \""
|
---|
| 346 | << (*query_here).querystring<<"\".\n";
|
---|
| 347 | err = syntaxError;
|
---|
| 348 | return;
|
---|
| 349 | }
|
---|
| 350 | // combine the results
|
---|
| 351 | if (need_matching_docs (request.filterResultOptions)) {
|
---|
| 352 |
|
---|
| 353 | if (query_params.size() == 1) {
|
---|
[12421] | 354 | multiresults.error_message = thisqueryresults.error_message;
|
---|
[8027] | 355 | multiresults.docs = thisqueryresults.docs; // just one set of results
|
---|
| 356 | multiresults.docs_matched = thisqueryresults.docs_matched;
|
---|
| 357 | multiresults.is_approx = thisqueryresults.is_approx;
|
---|
| 358 |
|
---|
| 359 | } else {
|
---|
| 360 | if ((*query_here).combinequery == "and") {
|
---|
| 361 | multiresults.docs.combine_and (thisqueryresults.docs);
|
---|
| 362 | } else if ((*query_here).combinequery == "or") {
|
---|
| 363 | multiresults.docs.combine_or (thisqueryresults.docs);
|
---|
| 364 | } else if ((*query_here).combinequery == "not") {
|
---|
| 365 | multiresults.docs.combine_not (thisqueryresults.docs);
|
---|
| 366 | }
|
---|
| 367 | multiresults.docs_matched = multiresults.docs.docset.size();
|
---|
| 368 | multiresults.is_approx = Exact;
|
---|
| 369 | }
|
---|
| 370 | }
|
---|
| 371 |
|
---|
| 372 | // combine the term information
|
---|
| 373 | if (need_term_info (request.filterResultOptions)) {
|
---|
| 374 | // append the terms
|
---|
| 375 | multiresults.orgterms.insert(multiresults.orgterms.end(),
|
---|
| 376 | thisqueryresults.orgterms.begin(),
|
---|
| 377 | thisqueryresults.orgterms.end());
|
---|
| 378 |
|
---|
| 379 |
|
---|
| 380 | // add the term variants -
|
---|
| 381 | text_tset::iterator termvar_here = thisqueryresults.termvariants.begin();
|
---|
| 382 | text_tset::iterator termvar_end = thisqueryresults.termvariants.end();
|
---|
| 383 | while (termvar_here != termvar_end) {
|
---|
| 384 | multiresults.termvariants.insert(*termvar_here);
|
---|
[9620] | 385 | ++termvar_here;
|
---|
[8027] | 386 | }
|
---|
[12380] | 387 |
|
---|
| 388 | // add the stop words
|
---|
| 389 | text_tset::iterator stopwords_here = thisqueryresults.stopwords.begin();
|
---|
| 390 | text_tset::iterator stopwords_end = thisqueryresults.stopwords.end();
|
---|
| 391 | while (stopwords_here != stopwords_end) {
|
---|
| 392 | multiresults.stopwords.insert(*stopwords_here);
|
---|
| 393 | ++stopwords_here;
|
---|
| 394 | }
|
---|
[8027] | 395 | }
|
---|
| 396 |
|
---|
[9620] | 397 | ++query_here;
|
---|
[8027] | 398 | }
|
---|
| 399 |
|
---|
| 400 | // sort and unique the query terms
|
---|
| 401 | multiresults.sortuniqqueryterms ();
|
---|
| 402 | }
|
---|
| 403 |
|
---|
| 404 |
|
---|
| 405 |
|
---|