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 | lucenequeryfilterclass::lucenequeryfilterclass ()
|
---|
31 | : fieldedqueryfilterclass() {
|
---|
32 |
|
---|
33 |
|
---|
34 | FilterOption_t filtopt;
|
---|
35 |
|
---|
36 | // -- onePerQuery SortField, enumerated, used to list available sorting fields
|
---|
37 | filtopt.clear();
|
---|
38 | filtopt.name = "SortField";
|
---|
39 | filtopt.type = FilterOption_t::enumeratedt;
|
---|
40 | filtopt.repeatable = FilterOption_t::onePerQuery;
|
---|
41 | filtopt.defaultValue = "";
|
---|
42 | filterOptions["SortField"] = filtopt;
|
---|
43 |
|
---|
44 | // -- onePerQuery SortOder enumerated (0=ascending, 1=descending)
|
---|
45 | filtopt.clear();
|
---|
46 | filtopt.name = "SortOrder";
|
---|
47 | filtopt.type = FilterOption_t::enumeratedt;
|
---|
48 | filtopt.repeatable = FilterOption_t::onePerQuery;
|
---|
49 | filtopt.defaultValue = "ascending";
|
---|
50 | filtopt.validValues.push_back("ascending");
|
---|
51 | filtopt.validValues.push_back("descending");
|
---|
52 | filterOptions["SortOrder"] = filtopt;
|
---|
53 |
|
---|
54 | // -- onePerQuery Fuzziness string 0.0-1.0
|
---|
55 | filtopt.clear();
|
---|
56 | filtopt.name = "Fuzziness";
|
---|
57 | filtopt.type = FilterOption_t::stringt;
|
---|
58 | filtopt.repeatable = FilterOption_t::onePerQuery;
|
---|
59 | filtopt.defaultValue = "";
|
---|
60 | filterOptions["Fuzziness"] = filtopt;
|
---|
61 |
|
---|
62 | // -- onePerQuery FilterString string
|
---|
63 | filtopt.clear();
|
---|
64 | filtopt.name = "FilterString";
|
---|
65 | filtopt.type = FilterOption_t::stringt;
|
---|
66 | filtopt.repeatable = FilterOption_t::onePerQuery;
|
---|
67 | filtopt.defaultValue = "";
|
---|
68 | filterOptions["FilterString"] = filtopt;
|
---|
69 | }
|
---|
70 |
|
---|
71 | lucenequeryfilterclass::~lucenequeryfilterclass () {
|
---|
72 | }
|
---|
73 |
|
---|
74 |
|
---|
75 |
|
---|
76 | void lucenequeryfilterclass::configure (const text_t &key, const text_tarray &cfgline) {
|
---|
77 | fieldedqueryfilterclass::configure(key, cfgline);
|
---|
78 |
|
---|
79 | if (key == "textlevel") {
|
---|
80 | ((lucenesearchclass *)textsearchptr)->set_text_level(cfgline[0]);
|
---|
81 | }
|
---|
82 | }
|
---|
83 |
|
---|
84 | bool lucenequeryfilterclass::init (ostream &logout) {
|
---|
85 |
|
---|
86 | if (!fieldedqueryfilterclass::init(logout)) {
|
---|
87 | return false;
|
---|
88 | }
|
---|
89 |
|
---|
90 | text_tarray field_array;
|
---|
91 | indexfieldmap.gettoarray(field_array);
|
---|
92 | for (int i=0; i<field_array.size(); i++) {
|
---|
93 | text_t field = field_array[i];
|
---|
94 | if (field!="ZZ" && field !="ZZ") {
|
---|
95 | filterOptions["SortField"].validValues.push_back("by"+field);
|
---|
96 | }
|
---|
97 | }
|
---|
98 | return true;
|
---|
99 | }
|
---|
100 |
|
---|
101 | void lucenequeryfilterclass::set_queryparam_defaults(queryparamclass &query ) {
|
---|
102 |
|
---|
103 | fieldedqueryfilterclass::set_queryparam_defaults(query);
|
---|
104 | query.filterstring = filterOptions["FilterString"].defaultValue;
|
---|
105 | query.sortfield = filterOptions["SortField"].defaultValue;
|
---|
106 | query.sortorder = (filterOptions["SortOrder"].defaultValue == "descending");
|
---|
107 | query.fuzziness = filterOptions["Fuzziness"].defaultValue;
|
---|
108 |
|
---|
109 | }
|
---|
110 |
|
---|
111 |
|
---|
112 | void lucenequeryfilterclass::filter(const FilterRequest_t &request,
|
---|
113 | FilterResponse_t &response,
|
---|
114 | comerror_t &err, ostream &logout) {
|
---|
115 |
|
---|
116 | outconvertclass text_t2ascii;
|
---|
117 |
|
---|
118 | response.clear ();
|
---|
119 | err = noError;
|
---|
120 | if (db_ptr == NULL) {
|
---|
121 | // most likely a configuration problem
|
---|
122 | logout << text_t2ascii
|
---|
123 | << "configuration error: queryfilter contains a null dbclass\n\n";
|
---|
124 | err = configurationError;
|
---|
125 | return;
|
---|
126 | }
|
---|
127 | if (textsearchptr == NULL) {
|
---|
128 | // most likely a configuration problem
|
---|
129 | logout << text_t2ascii
|
---|
130 | << "configuration error: queryfilter contains a null textsearchclass (lucene)\n\n";
|
---|
131 | err = configurationError;
|
---|
132 | return;
|
---|
133 | }
|
---|
134 | if (full_text_browse(request.filterResultOptions)) {
|
---|
135 | browsefilter(request, response, err, logout);
|
---|
136 | return;
|
---|
137 | }
|
---|
138 | // open the database
|
---|
139 | db_ptr->setlogout(&logout);
|
---|
140 | if (!db_ptr->opendatabase (db_filename, DB_READER, 100, false)) {
|
---|
141 | // most likely a system problem (we have already checked that the database exists)
|
---|
142 | logout << text_t2ascii
|
---|
143 | << "system problem: open on database \"" << db_filename << "\" failed\n\n";
|
---|
144 | err = systemProblem;
|
---|
145 | return;
|
---|
146 | }
|
---|
147 |
|
---|
148 |
|
---|
149 | // get the query parameters
|
---|
150 | int startresults, endresults;
|
---|
151 | vector<queryparamclass> queryfilterparams;
|
---|
152 | parse_query_params (request, queryfilterparams, startresults,
|
---|
153 | endresults, logout);
|
---|
154 |
|
---|
155 |
|
---|
156 | // do query
|
---|
157 | queryresultsclass queryresults;
|
---|
158 | do_multi_query (request, queryfilterparams, queryresults, err, logout);
|
---|
159 | response.error_message = queryresults.error_message;
|
---|
160 | if (err != noError) return;
|
---|
161 |
|
---|
162 | // assemble document results
|
---|
163 | if (need_matching_docs (request.filterResultOptions))
|
---|
164 | {
|
---|
165 | // Loop through the query results (ordered by ranking)
|
---|
166 | int resultnum = 1;
|
---|
167 | vector<text_t>::iterator docorder_iterator = queryresults.docs.docorder.begin();
|
---|
168 | while (docorder_iterator != queryresults.docs.docorder.end())
|
---|
169 | {
|
---|
170 | text_t doc_OID = (*docorder_iterator);
|
---|
171 | // logout << "Matching doc OID: " << doc_OID << endl;
|
---|
172 |
|
---|
173 | // Make sure this result is in the docset, and either in the request set or the request set is empty
|
---|
174 | docresultmap::iterator doc_result = queryresults.docs.docset.find (doc_OID);
|
---|
175 | if (doc_result != queryresults.docs.docset.end() && (request.docSet.empty() || in_set(request.docSet, doc_OID)))
|
---|
176 | {
|
---|
177 | // Add the matching document
|
---|
178 | ResultDocInfo_t resultdoc;
|
---|
179 | resultdoc.OID = doc_OID;
|
---|
180 | resultdoc.result_num = resultnum;
|
---|
181 | resultdoc.ranking = (int)((*doc_result).second.docweight * 10000.0 + 0.5);
|
---|
182 | resultdoc.num_terms_matched = (*doc_result).second.num_query_terms_matched;
|
---|
183 | response.docInfo.push_back (resultdoc);
|
---|
184 |
|
---|
185 | resultnum++;
|
---|
186 | }
|
---|
187 |
|
---|
188 | docorder_iterator++;
|
---|
189 | }
|
---|
190 | }
|
---|
191 |
|
---|
192 | // assemble the term results
|
---|
193 | if (need_term_info(request.filterResultOptions)) {
|
---|
194 | // note: the terms have already been sorted and uniqued - ?? have they??
|
---|
195 |
|
---|
196 | TermInfo_t terminfo;
|
---|
197 | bool terms_first = true;
|
---|
198 |
|
---|
199 | termfreqclassarray::iterator terms_here = queryresults.terms.begin();
|
---|
200 | termfreqclassarray::iterator terms_end = queryresults.terms.end();
|
---|
201 |
|
---|
202 | while (terms_here != terms_end) {
|
---|
203 | terminfo.clear();
|
---|
204 | terminfo.term = (*terms_here).termstr;
|
---|
205 | terminfo.freq = (*terms_here).termfreq;
|
---|
206 | // lucene doesn't return any termvariants at this stage,
|
---|
207 | // so make sure the original term is set
|
---|
208 | terminfo.matchTerms.push_back(terminfo.term);
|
---|
209 |
|
---|
210 | // this bit gets the matchTerms ie the equivalent (stem/casefold) terms
|
---|
211 | if (terms_first) {
|
---|
212 | text_tset::iterator termvariants_here = queryresults.termvariants.begin();
|
---|
213 | text_tset::iterator termvariants_end = queryresults.termvariants.end();
|
---|
214 | while (termvariants_here != termvariants_end) {
|
---|
215 | terminfo.matchTerms.push_back (*termvariants_here);
|
---|
216 | ++termvariants_here;
|
---|
217 | }
|
---|
218 | }
|
---|
219 | terms_first = false;
|
---|
220 |
|
---|
221 | response.termInfo.push_back (terminfo);
|
---|
222 |
|
---|
223 | ++terms_here;
|
---|
224 | }
|
---|
225 |
|
---|
226 | // add the stop words
|
---|
227 | text_tset::iterator stopwords_here = queryresults.stopwords.begin();
|
---|
228 | text_tset::iterator stopwords_end = queryresults.stopwords.end();
|
---|
229 | while (stopwords_here != stopwords_end) {
|
---|
230 | response.stopwords.insert(*stopwords_here);
|
---|
231 | ++stopwords_here;
|
---|
232 | }
|
---|
233 | }
|
---|
234 |
|
---|
235 | db_ptr->closedatabase(); // Important that local library doesn't leave any files open
|
---|
236 | response.numDocs = queryresults.docs_matched;
|
---|
237 | response.isApprox = queryresults.is_approx;
|
---|
238 | }
|
---|
239 |
|
---|
240 | void lucenequeryfilterclass::browsefilter(const FilterRequest_t &request,
|
---|
241 | FilterResponse_t &response,
|
---|
242 | comerror_t &err, ostream &logout) {
|
---|
243 |
|
---|
244 | outconvertclass text_t2ascii;
|
---|
245 |
|
---|
246 | // get the query parameters
|
---|
247 | int startresults, endresults;
|
---|
248 |
|
---|
249 | vector<queryparamclass> queryfilterparams;
|
---|
250 | parse_query_params (request, queryfilterparams, startresults,
|
---|
251 | endresults, logout);
|
---|
252 |
|
---|
253 | vector<queryparamclass>::const_iterator query_here = queryfilterparams.begin();
|
---|
254 |
|
---|
255 | // do query
|
---|
256 | queryresultsclass queryresults;
|
---|
257 | queryresults.clear();
|
---|
258 |
|
---|
259 | int numDocs = endresults-startresults;
|
---|
260 | textsearchptr->setcollectdir (collectdir);
|
---|
261 |
|
---|
262 | if (!((lucenesearchclass*)textsearchptr)->browse_search((*query_here), startresults, numDocs, queryresults)) {
|
---|
263 | // most likely a system problem
|
---|
264 | logout << text_t2ascii
|
---|
265 | << "system problem: could not do full text browse with lucene for index \""
|
---|
266 | << (*query_here).index << (*query_here).subcollection
|
---|
267 | << (*query_here).language << "\".\n\n";
|
---|
268 | err = systemProblem;
|
---|
269 | return;
|
---|
270 | }
|
---|
271 |
|
---|
272 | // assemble the term results
|
---|
273 | TermInfo_t terminfo;
|
---|
274 |
|
---|
275 | termfreqclassarray::iterator terms_here = queryresults.terms.begin();
|
---|
276 | termfreqclassarray::iterator terms_end = queryresults.terms.end();
|
---|
277 |
|
---|
278 | while (terms_here != terms_end) {
|
---|
279 | terminfo.clear();
|
---|
280 | terminfo.term = (*terms_here).termstr;
|
---|
281 | terminfo.freq = (*terms_here).termfreq;
|
---|
282 |
|
---|
283 | response.termInfo.push_back (terminfo);
|
---|
284 |
|
---|
285 | ++terms_here;
|
---|
286 | }
|
---|
287 |
|
---|
288 |
|
---|
289 | }
|
---|
290 |
|
---|
291 | // lucenesearchptr and db_ptr are assumed to be valid
|
---|
292 | void lucenequeryfilterclass::do_multi_query (const FilterRequest_t &request,
|
---|
293 | const vector<queryparamclass> &query_params,
|
---|
294 | queryresultsclass &multiresults,
|
---|
295 | comerror_t &err, ostream &logout) {
|
---|
296 | outconvertclass text_t2ascii;
|
---|
297 |
|
---|
298 | err = noError;
|
---|
299 | textsearchptr->setcollectdir (collectdir);
|
---|
300 | multiresults.clear();
|
---|
301 |
|
---|
302 | vector<queryparamclass>::const_iterator query_here = query_params.begin();
|
---|
303 | vector<queryparamclass>::const_iterator query_end = query_params.end();
|
---|
304 | while (query_here != query_end) {
|
---|
305 | queryresultsclass thisqueryresults;
|
---|
306 | if (!textsearchptr->search((*query_here), thisqueryresults)) {
|
---|
307 | // most likely a system problem
|
---|
308 | logout << text_t2ascii
|
---|
309 | << "system problem: could not do search with lucene for index \""
|
---|
310 | << (*query_here).index << (*query_here).level
|
---|
311 | << (*query_here).subcollection
|
---|
312 | << (*query_here).language << "\".\n\n";
|
---|
313 | err = systemProblem;
|
---|
314 | return;
|
---|
315 | }
|
---|
316 |
|
---|
317 | // check for syntax error
|
---|
318 | if (thisqueryresults.syntax_error==true) {
|
---|
319 | logout << text_t2ascii
|
---|
320 | << "syntax problem: invalid query string \""
|
---|
321 | << (*query_here).querystring<<"\".\n";
|
---|
322 | err = syntaxError;
|
---|
323 | return;
|
---|
324 | }
|
---|
325 | // combine the results
|
---|
326 | if (need_matching_docs (request.filterResultOptions)) {
|
---|
327 |
|
---|
328 | if (query_params.size() == 1) {
|
---|
329 | multiresults.error_message = thisqueryresults.error_message;
|
---|
330 | multiresults.docs = thisqueryresults.docs; // just one set of results
|
---|
331 | multiresults.docs_matched = thisqueryresults.docs_matched;
|
---|
332 | multiresults.is_approx = thisqueryresults.is_approx;
|
---|
333 |
|
---|
334 | } else {
|
---|
335 | if ((*query_here).combinequery == "and") {
|
---|
336 | multiresults.docs.combine_and (thisqueryresults.docs);
|
---|
337 | } else if ((*query_here).combinequery == "or") {
|
---|
338 | multiresults.docs.combine_or (thisqueryresults.docs);
|
---|
339 | } else if ((*query_here).combinequery == "not") {
|
---|
340 | multiresults.docs.combine_not (thisqueryresults.docs);
|
---|
341 | }
|
---|
342 | multiresults.docs_matched = multiresults.docs.docset.size();
|
---|
343 | multiresults.is_approx = Exact;
|
---|
344 | }
|
---|
345 | }
|
---|
346 |
|
---|
347 | // combine the term information
|
---|
348 | if (need_term_info (request.filterResultOptions)) {
|
---|
349 | // append the terms
|
---|
350 | multiresults.orgterms.insert(multiresults.orgterms.end(),
|
---|
351 | thisqueryresults.orgterms.begin(),
|
---|
352 | thisqueryresults.orgterms.end());
|
---|
353 |
|
---|
354 |
|
---|
355 | // add the term variants -
|
---|
356 | text_tset::iterator termvar_here = thisqueryresults.termvariants.begin();
|
---|
357 | text_tset::iterator termvar_end = thisqueryresults.termvariants.end();
|
---|
358 | while (termvar_here != termvar_end) {
|
---|
359 | multiresults.termvariants.insert(*termvar_here);
|
---|
360 | ++termvar_here;
|
---|
361 | }
|
---|
362 |
|
---|
363 | // add the stop words
|
---|
364 | text_tset::iterator stopwords_here = thisqueryresults.stopwords.begin();
|
---|
365 | text_tset::iterator stopwords_end = thisqueryresults.stopwords.end();
|
---|
366 | while (stopwords_here != stopwords_end) {
|
---|
367 | multiresults.stopwords.insert(*stopwords_here);
|
---|
368 | ++stopwords_here;
|
---|
369 | }
|
---|
370 | }
|
---|
371 |
|
---|
372 | ++query_here;
|
---|
373 | }
|
---|
374 |
|
---|
375 | // sort and unique the query terms
|
---|
376 | multiresults.sortuniqqueryterms ();
|
---|
377 | }
|
---|
378 |
|
---|
379 |
|
---|