[1324] | 1 | /**********************************************************************
|
---|
| 2 | *
|
---|
| 3 | * queryfilter.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 |
|
---|
| 27 |
|
---|
| 28 | #include "mgppqueryfilter.h"
|
---|
| 29 | #include "fileutil.h"
|
---|
| 30 | #include <assert.h>
|
---|
| 31 | #include "mgppsearch.h"
|
---|
| 32 |
|
---|
| 33 | /////////////////////////////////
|
---|
| 34 | // functions for queryfilterclass
|
---|
| 35 | /////////////////////////////////
|
---|
| 36 |
|
---|
| 37 |
|
---|
| 38 | mgppqueryfilterclass::mgppqueryfilterclass ()
|
---|
| 39 | : queryfilterclass() {
|
---|
[1835] | 40 |
|
---|
[1324] | 41 |
|
---|
[1835] | 42 | FilterOption_t filtopt;
|
---|
| 43 |
|
---|
[1324] | 44 | // -- onePerTerm Level enumerated
|
---|
[4808] | 45 | // likely to be Doc, Sec, Para, but we dont assume anything now
|
---|
[1324] | 46 | filtopt.clear();
|
---|
| 47 | filtopt.name = "Level";
|
---|
| 48 | filtopt.type = FilterOption_t::enumeratedt;
|
---|
| 49 | filtopt.repeatable = FilterOption_t::onePerTerm;
|
---|
| 50 | filterOptions["Level"] = filtopt;
|
---|
| 51 |
|
---|
[1835] | 52 | // -- IndexField, enumerated, used to list available fields
|
---|
[1324] | 53 | filtopt.clear();
|
---|
[1835] | 54 | filtopt.name = "IndexField";
|
---|
[1324] | 55 | filtopt.type = FilterOption_t::enumeratedt;
|
---|
| 56 | filtopt.repeatable = FilterOption_t::onePerTerm;
|
---|
[4752] | 57 | filtopt.defaultValue = "";
|
---|
[1835] | 58 | filterOptions["IndexField"] = filtopt;
|
---|
[1324] | 59 |
|
---|
| 60 | }
|
---|
| 61 |
|
---|
| 62 | mgppqueryfilterclass::~mgppqueryfilterclass () {
|
---|
| 63 | }
|
---|
| 64 |
|
---|
| 65 |
|
---|
| 66 | //whether a query is a full text browse
|
---|
| 67 | bool mgppqueryfilterclass::full_text_browse (int filterRequestOptions) {
|
---|
| 68 | return (filterRequestOptions & FRfullTextBrowse);
|
---|
| 69 | }
|
---|
| 70 |
|
---|
| 71 | void mgppqueryfilterclass::configure (const text_t &key, const text_tarray &cfgline) {
|
---|
| 72 | queryfilterclass::configure(key, cfgline);
|
---|
| 73 |
|
---|
[1835] | 74 | if (key == "indexfieldmap") {
|
---|
| 75 | indexfieldmap.importmap (cfgline);
|
---|
[1324] | 76 |
|
---|
[1835] | 77 | // update the list of indexes in the filter information
|
---|
| 78 | text_tarray options;
|
---|
| 79 | indexfieldmap.gettoarray (options);
|
---|
[10954] | 80 | filterOptions["IndexField"].validValues = options;
|
---|
[8483] | 81 |
|
---|
[4808] | 82 | } else if (key == "indexlevels") {
|
---|
[1835] | 83 | text_tarray::const_iterator here = cfgline.begin();
|
---|
| 84 | text_tarray::const_iterator end = cfgline.end();
|
---|
[4808] | 85 | bool first=true;
|
---|
[8483] | 86 | filterOptions["Level"].validValues.erase(filterOptions["Level"].validValues.begin(), filterOptions["Level"].validValues.end());
|
---|
[1835] | 87 | while (here != end) {
|
---|
[1909] | 88 | if (!(*here).empty()) {
|
---|
[4808] | 89 | if (first) {
|
---|
| 90 | first = false;
|
---|
| 91 | // the default is the first value
|
---|
| 92 | filterOptions["Level"].defaultValue = *here;
|
---|
[1909] | 93 | }
|
---|
[4808] | 94 | filterOptions["Level"].validValues.push_back(*here);
|
---|
[1909] | 95 | }
|
---|
[9620] | 96 | ++here;
|
---|
[1835] | 97 | }
|
---|
[4808] | 98 | } else if (key == "textlevel") {
|
---|
[8025] | 99 | ((mgppsearchclass *)textsearchptr)->set_gdbm_level( cfgline[0]);
|
---|
[9937] | 100 | } else if (key == "indexstem") {
|
---|
| 101 | ((mgppsearchclass *)textsearchptr)->set_indexstem (cfgline[0]);
|
---|
[10954] | 102 | } else if (key == "defaultindex") { // used for fields in mgpp
|
---|
| 103 | indexfieldmap.from2to (cfgline[0], filterOptions["IndexField"].defaultValue);
|
---|
[1835] | 104 | }
|
---|
[4808] | 105 |
|
---|
[1324] | 106 | }
|
---|
| 107 |
|
---|
[10954] | 108 | bool mgppqueryfilterclass::init (ostream &logout) {
|
---|
| 109 |
|
---|
| 110 | if (!queryfilterclass::init(logout)) {
|
---|
| 111 | return false;
|
---|
| 112 | }
|
---|
| 113 |
|
---|
| 114 | if (filterOptions["IndexField"].defaultValue.empty()) {
|
---|
| 115 | // use first index in map as default if no default is set explicitly
|
---|
| 116 | text_tarray fromarray;
|
---|
| 117 | indexfieldmap.getfromarray(fromarray);
|
---|
| 118 | if (fromarray.size()) {
|
---|
| 119 | filterOptions["IndexField"].defaultValue = fromarray[0];
|
---|
| 120 | }
|
---|
| 121 | }
|
---|
| 122 | return true;
|
---|
| 123 | }
|
---|
[1324] | 124 |
|
---|
| 125 | void mgppqueryfilterclass::filter(const FilterRequest_t &request,
|
---|
| 126 | FilterResponse_t &response,
|
---|
| 127 | comerror_t &err, ostream &logout) {
|
---|
| 128 |
|
---|
| 129 |
|
---|
| 130 | outconvertclass text_t2ascii;
|
---|
| 131 |
|
---|
| 132 | response.clear ();
|
---|
| 133 | err = noError;
|
---|
| 134 | if (gdbmptr == NULL) {
|
---|
| 135 | // most likely a configuration problem
|
---|
| 136 | logout << text_t2ascii
|
---|
| 137 | << "configuration error: queryfilter contains a null gdbmclass\n\n";
|
---|
| 138 | err = configurationError;
|
---|
| 139 | return;
|
---|
| 140 | }
|
---|
[8025] | 141 | if (textsearchptr == NULL) {
|
---|
[1324] | 142 | // most likely a configuration problem
|
---|
| 143 | logout << text_t2ascii
|
---|
[8025] | 144 | << "configuration error: queryfilter contains a null textsearchclass for mgpp\n\n";
|
---|
[1324] | 145 | err = configurationError;
|
---|
| 146 | return;
|
---|
| 147 | }
|
---|
| 148 | if (full_text_browse(request.filterResultOptions)) {
|
---|
| 149 | browsefilter(request, response, err, logout);
|
---|
| 150 | return;
|
---|
| 151 | }
|
---|
| 152 | // open the database
|
---|
| 153 | gdbmptr->setlogout(&logout);
|
---|
| 154 | if (!gdbmptr->opendatabase (gdbm_filename, GDBM_READER, 100, false)) {
|
---|
| 155 | // most likely a system problem (we have already checked that the
|
---|
| 156 | // gdbm database exists)
|
---|
| 157 | logout << text_t2ascii
|
---|
| 158 | << "system problem: open on gdbm database \""
|
---|
| 159 | << gdbm_filename << "\" failed\n\n";
|
---|
| 160 | err = systemProblem;
|
---|
| 161 | return;
|
---|
| 162 | }
|
---|
| 163 |
|
---|
| 164 |
|
---|
| 165 | // get the query parameters
|
---|
| 166 | int startresults, endresults;
|
---|
| 167 | text_t phrasematch; // not used here any more
|
---|
| 168 | vector<queryparamclass> queryfilterparams;
|
---|
| 169 | parse_query_params (request, queryfilterparams, startresults,
|
---|
| 170 | endresults, phrasematch, logout);
|
---|
| 171 |
|
---|
| 172 |
|
---|
| 173 | // do query
|
---|
| 174 | queryresultsclass queryresults;
|
---|
| 175 | do_multi_query (request, queryfilterparams, queryresults, err, logout);
|
---|
| 176 | if (err != noError) return;
|
---|
| 177 | // assemble document results
|
---|
| 178 | if (need_matching_docs (request.filterResultOptions)) {
|
---|
| 179 |
|
---|
| 180 | int resultnum = 1;
|
---|
| 181 | ResultDocInfo_t resultdoc;
|
---|
| 182 | text_t trans_OID;
|
---|
| 183 | vector<int>::iterator docorder_here = queryresults.docs.docorder.begin();
|
---|
| 184 | vector<int>::iterator docorder_end = queryresults.docs.docorder.end();
|
---|
| 185 |
|
---|
| 186 | if (endresults == -1) endresults = MAXNUMDOCS;
|
---|
| 187 | while (docorder_here != docorder_end) {
|
---|
| 188 | if (resultnum > endresults) break;
|
---|
| 189 |
|
---|
| 190 | // translate the document number
|
---|
| 191 | if (!translate(gdbmptr, *docorder_here, trans_OID)) {
|
---|
| 192 | logout << text_t2ascii
|
---|
| 193 | << "warning: could not translate mgpp document number \""
|
---|
| 194 | << *docorder_here << "\"to OID.\n\n";
|
---|
| 195 |
|
---|
| 196 | } else {
|
---|
| 197 | docresultmap::iterator docset_here = queryresults.docs.docset.find (*docorder_here);
|
---|
| 198 |
|
---|
| 199 | // see if there is a result for this number,
|
---|
| 200 | // if it is in the request set (or the request set is empty)
|
---|
| 201 | if (docset_here != queryresults.docs.docset.end() &&
|
---|
| 202 | (request.docSet.empty() || in_set(request.docSet, trans_OID))) {
|
---|
| 203 | if (resultnum >= startresults) {
|
---|
| 204 | // add this document
|
---|
| 205 | resultdoc.OID = trans_OID;
|
---|
| 206 | resultdoc.result_num = resultnum;
|
---|
| 207 | resultdoc.ranking = (int)((*docset_here).second.docweight * 10000.0 + 0.5);
|
---|
| 208 |
|
---|
| 209 | response.docInfo.push_back (resultdoc);
|
---|
| 210 | }
|
---|
| 211 |
|
---|
[9620] | 212 | ++resultnum;
|
---|
[1324] | 213 | }
|
---|
| 214 | } // else
|
---|
| 215 |
|
---|
[9620] | 216 | ++docorder_here;
|
---|
[1324] | 217 | }
|
---|
| 218 | } // if need matching docs
|
---|
| 219 |
|
---|
| 220 | // assemble the term results
|
---|
| 221 | if (need_term_info(request.filterResultOptions)) {
|
---|
[1835] | 222 | // note: the terms have already been sorted and uniqued - ?? have they??
|
---|
[1324] | 223 |
|
---|
| 224 | TermInfo_t terminfo;
|
---|
| 225 | bool terms_first = true;
|
---|
| 226 |
|
---|
| 227 | termfreqclassarray::iterator terms_here = queryresults.terms.begin();
|
---|
| 228 | termfreqclassarray::iterator terms_end = queryresults.terms.end();
|
---|
| 229 |
|
---|
| 230 | while (terms_here != terms_end) {
|
---|
| 231 | terminfo.clear();
|
---|
| 232 | terminfo.term = (*terms_here).termstr;
|
---|
| 233 | terminfo.freq = (*terms_here).termfreq;
|
---|
[1788] | 234 |
|
---|
| 235 | // this bit gets the matchTerms ie the equivalent (stem/casefold) terms
|
---|
[1324] | 236 | if (terms_first) {
|
---|
| 237 | text_tset::iterator termvariants_here = queryresults.termvariants.begin();
|
---|
| 238 | text_tset::iterator termvariants_end = queryresults.termvariants.end();
|
---|
| 239 | while (termvariants_here != termvariants_end) {
|
---|
| 240 | terminfo.matchTerms.push_back (*termvariants_here);
|
---|
[9620] | 241 | ++termvariants_here;
|
---|
[1324] | 242 | }
|
---|
| 243 | }
|
---|
| 244 | terms_first = false;
|
---|
[1835] | 245 |
|
---|
[1324] | 246 | response.termInfo.push_back (terminfo);
|
---|
| 247 |
|
---|
[9620] | 248 | ++terms_here;
|
---|
[1324] | 249 | }
|
---|
| 250 | }
|
---|
| 251 |
|
---|
| 252 | response.numDocs = queryresults.docs_matched;
|
---|
| 253 | response.isApprox = queryresults.is_approx;
|
---|
| 254 | }
|
---|
| 255 |
|
---|
| 256 | void mgppqueryfilterclass::browsefilter(const FilterRequest_t &request,
|
---|
| 257 | FilterResponse_t &response,
|
---|
| 258 | comerror_t &err, ostream &logout) {
|
---|
| 259 |
|
---|
| 260 | outconvertclass text_t2ascii;
|
---|
| 261 |
|
---|
| 262 | // get the query parameters
|
---|
| 263 | int startresults, endresults;
|
---|
| 264 | text_t phrasematch; // not used here any more, just have it so can use
|
---|
| 265 | // parse_query_params function
|
---|
| 266 |
|
---|
| 267 | vector<queryparamclass> queryfilterparams;
|
---|
| 268 | parse_query_params (request, queryfilterparams, startresults,
|
---|
| 269 | endresults, phrasematch, logout);
|
---|
[1909] | 270 |
|
---|
| 271 | vector<queryparamclass>::const_iterator query_here = queryfilterparams.begin();
|
---|
[1324] | 272 |
|
---|
| 273 | // do query
|
---|
| 274 | queryresultsclass queryresults;
|
---|
| 275 | queryresults.clear();
|
---|
| 276 |
|
---|
| 277 | int numDocs = endresults-startresults;
|
---|
[8025] | 278 | textsearchptr->setcollectdir (collectdir);
|
---|
[1324] | 279 |
|
---|
[8025] | 280 | if (!((mgppsearchclass*)textsearchptr)->browse_search((*query_here), startresults, numDocs, queryresults)) {
|
---|
[1324] | 281 | // most likely a system problem
|
---|
| 282 | logout << text_t2ascii
|
---|
| 283 | << "system problem: could not do full text browse with mgpp for index \""
|
---|
| 284 | << (*query_here).index << (*query_here).subcollection
|
---|
| 285 | << (*query_here).language << "\".\n\n";
|
---|
| 286 | err = systemProblem;
|
---|
| 287 | return;
|
---|
| 288 | }
|
---|
| 289 |
|
---|
| 290 | // assemble the term results
|
---|
| 291 | TermInfo_t terminfo;
|
---|
| 292 |
|
---|
| 293 | termfreqclassarray::iterator terms_here = queryresults.terms.begin();
|
---|
| 294 | termfreqclassarray::iterator terms_end = queryresults.terms.end();
|
---|
| 295 |
|
---|
| 296 | while (terms_here != terms_end) {
|
---|
| 297 | terminfo.clear();
|
---|
| 298 | terminfo.term = (*terms_here).termstr;
|
---|
| 299 | terminfo.freq = (*terms_here).termfreq;
|
---|
| 300 |
|
---|
| 301 | response.termInfo.push_back (terminfo);
|
---|
| 302 |
|
---|
[9620] | 303 | ++terms_here;
|
---|
[1324] | 304 | }
|
---|
| 305 |
|
---|
| 306 |
|
---|
| 307 | }
|
---|
| 308 |
|
---|
| 309 | // mgppsearchptr and gdbmptr are assumed to be valid
|
---|
| 310 | void mgppqueryfilterclass::do_multi_query (const FilterRequest_t &request,
|
---|
| 311 | const vector<queryparamclass> &query_params,
|
---|
| 312 | queryresultsclass &multiresults,
|
---|
| 313 | comerror_t &err, ostream &logout) {
|
---|
| 314 | outconvertclass text_t2ascii;
|
---|
| 315 |
|
---|
| 316 | err = noError;
|
---|
[8025] | 317 | textsearchptr->setcollectdir (collectdir);
|
---|
[1324] | 318 | multiresults.clear();
|
---|
| 319 |
|
---|
| 320 | vector<queryparamclass>::const_iterator query_here = query_params.begin();
|
---|
| 321 | vector<queryparamclass>::const_iterator query_end = query_params.end();
|
---|
| 322 | while (query_here != query_end) {
|
---|
| 323 | queryresultsclass thisqueryresults;
|
---|
| 324 | text_t indx((*query_here).index);
|
---|
[8025] | 325 | if (!textsearchptr->search((*query_here), thisqueryresults)) {
|
---|
[1324] | 326 | // most likely a system problem
|
---|
| 327 | logout << text_t2ascii
|
---|
| 328 | << "system problem: could not do search with mgpp for index \""
|
---|
| 329 | << (*query_here).index << (*query_here).subcollection
|
---|
| 330 | << (*query_here).language << "\".\n\n";
|
---|
| 331 | err = systemProblem;
|
---|
| 332 | return;
|
---|
| 333 | }
|
---|
| 334 |
|
---|
[4218] | 335 | // check for syntax error
|
---|
| 336 | if (thisqueryresults.syntax_error==true) {
|
---|
| 337 | logout << text_t2ascii
|
---|
| 338 | << "syntax problem: invalid query string \""
|
---|
| 339 | << (*query_here).querystring<<"\".\n";
|
---|
| 340 | err = syntaxError;
|
---|
| 341 | return;
|
---|
| 342 | }
|
---|
[1324] | 343 | // combine the results
|
---|
| 344 | if (need_matching_docs (request.filterResultOptions)) {
|
---|
| 345 |
|
---|
| 346 | if (query_params.size() == 1) {
|
---|
| 347 | multiresults.docs = thisqueryresults.docs; // just one set of results
|
---|
| 348 | multiresults.docs_matched = thisqueryresults.docs_matched;
|
---|
| 349 | multiresults.is_approx = thisqueryresults.is_approx;
|
---|
| 350 |
|
---|
| 351 | } else {
|
---|
| 352 | if ((*query_here).combinequery == "and") {
|
---|
| 353 | multiresults.docs.combine_and (thisqueryresults.docs);
|
---|
| 354 | } else if ((*query_here).combinequery == "or") {
|
---|
| 355 | multiresults.docs.combine_or (thisqueryresults.docs);
|
---|
| 356 | } else if ((*query_here).combinequery == "not") {
|
---|
| 357 | multiresults.docs.combine_not (thisqueryresults.docs);
|
---|
| 358 | }
|
---|
| 359 | multiresults.docs_matched = multiresults.docs.docset.size();
|
---|
| 360 | multiresults.is_approx = Exact;
|
---|
| 361 | }
|
---|
| 362 | }
|
---|
| 363 |
|
---|
| 364 | // combine the term information
|
---|
| 365 | if (need_term_info (request.filterResultOptions)) {
|
---|
| 366 | // append the terms
|
---|
| 367 | multiresults.orgterms.insert(multiresults.orgterms.end(),
|
---|
| 368 | thisqueryresults.orgterms.begin(),
|
---|
| 369 | thisqueryresults.orgterms.end());
|
---|
| 370 |
|
---|
| 371 |
|
---|
| 372 | // add the term variants -
|
---|
| 373 | text_tset::iterator termvar_here = thisqueryresults.termvariants.begin();
|
---|
| 374 | text_tset::iterator termvar_end = thisqueryresults.termvariants.end();
|
---|
| 375 | while (termvar_here != termvar_end) {
|
---|
| 376 | multiresults.termvariants.insert(*termvar_here);
|
---|
[9620] | 377 | ++termvar_here;
|
---|
[1324] | 378 | }
|
---|
| 379 | }
|
---|
| 380 |
|
---|
[9620] | 381 | ++query_here;
|
---|
[1324] | 382 | }
|
---|
| 383 |
|
---|
| 384 | // sort and unique the query terms
|
---|
| 385 | multiresults.sortuniqqueryterms ();
|
---|
| 386 | }
|
---|
| 387 |
|
---|
| 388 |
|
---|
| 389 |
|
---|