|
a/src/rcldb/searchdata.cpp |
|
b/src/rcldb/searchdata.cpp |
1 |
#ifndef lint
|
1 |
#ifndef lint
|
2 |
static char rcsid[] = "@(#$Id: searchdata.cpp,v 1.3 2006-11-14 17:41:12 dockes Exp $ (C) 2006 J.F.Dockes";
|
2 |
static char rcsid[] = "@(#$Id: searchdata.cpp,v 1.4 2006-11-17 10:06:34 dockes Exp $ (C) 2006 J.F.Dockes";
|
3 |
#endif
|
3 |
#endif
|
4 |
/*
|
4 |
/*
|
5 |
* This program is free software; you can redistribute it and/or modify
|
5 |
* This program is free software; you can redistribute it and/or modify
|
6 |
* it under the terms of the GNU General Public License as published by
|
6 |
* it under the terms of the GNU General Public License as published by
|
7 |
* the Free Software Foundation; either version 2 of the License, or
|
7 |
* the Free Software Foundation; either version 2 of the License, or
|
|
... |
|
... |
19 |
*/
|
19 |
*/
|
20 |
|
20 |
|
21 |
// Handle translation from rcl's SearchData structures to Xapian Queries
|
21 |
// Handle translation from rcl's SearchData structures to Xapian Queries
|
22 |
|
22 |
|
23 |
#include <string>
|
23 |
#include <string>
|
24 |
#include <list>
|
24 |
#include <vector>
|
25 |
#ifndef NO_NAMESPACES
|
|
|
26 |
using namespace std;
|
|
|
27 |
#endif
|
|
|
28 |
|
25 |
|
29 |
#include "xapian.h"
|
26 |
#include "xapian.h"
|
30 |
|
27 |
|
31 |
#include "rcldb.h"
|
28 |
#include "rcldb.h"
|
32 |
#include "searchdata.h"
|
29 |
#include "searchdata.h"
|
|
... |
|
... |
34 |
#include "smallut.h"
|
31 |
#include "smallut.h"
|
35 |
#include "textsplit.h"
|
32 |
#include "textsplit.h"
|
36 |
#include "unacpp.h"
|
33 |
#include "unacpp.h"
|
37 |
#include "utf8iter.h"
|
34 |
#include "utf8iter.h"
|
38 |
|
35 |
|
|
|
36 |
#ifndef NO_NAMESPACES
|
|
|
37 |
using namespace std;
|
39 |
namespace Rcl {
|
38 |
namespace Rcl {
|
|
|
39 |
#endif
|
40 |
|
40 |
|
41 |
typedef list<SearchDataClause *>::iterator qlist_it_t;
|
41 |
typedef vector<SearchDataClause *>::iterator qlist_it_t;
|
|
|
42 |
typedef vector<SearchDataClause *>::const_iterator qlist_cit_t;
|
42 |
|
43 |
|
43 |
bool SearchData::toNativeQuery(Rcl::Db &db, void *d, const string& stemlang)
|
44 |
bool SearchData::toNativeQuery(Rcl::Db &db, void *d, const string& stemlang)
|
44 |
{
|
45 |
{
|
45 |
Xapian::Query xq;
|
46 |
Xapian::Query xq;
|
46 |
m_reason.erase();
|
47 |
m_reason.erase();
|
|
... |
|
... |
69 |
|
70 |
|
70 |
// Add the file type filtering clause if any
|
71 |
// Add the file type filtering clause if any
|
71 |
if (!m_filetypes.empty()) {
|
72 |
if (!m_filetypes.empty()) {
|
72 |
list<Xapian::Query> pqueries;
|
73 |
list<Xapian::Query> pqueries;
|
73 |
Xapian::Query tq;
|
74 |
Xapian::Query tq;
|
74 |
for (list<string>::iterator it = m_filetypes.begin();
|
75 |
for (vector<string>::iterator it = m_filetypes.begin();
|
75 |
it != m_filetypes.end(); it++) {
|
76 |
it != m_filetypes.end(); it++) {
|
76 |
string term = "T" + *it;
|
77 |
string term = "T" + *it;
|
77 |
LOGDEB(("Adding file type term: [%s]\n", term.c_str()));
|
78 |
LOGDEB(("Adding file type term: [%s]\n", term.c_str()));
|
78 |
tq = tq.empty() ? Xapian::Query(term) :
|
79 |
tq = tq.empty() ? Xapian::Query(term) :
|
79 |
Xapian::Query(Xapian::Query::OP_OR, tq, Xapian::Query(term));
|
80 |
Xapian::Query(Xapian::Query::OP_OR, tq, Xapian::Query(term));
|
|
... |
|
... |
88 |
// Add clause to current list. OR lists cant have EXCL clauses.
|
89 |
// Add clause to current list. OR lists cant have EXCL clauses.
|
89 |
bool SearchData::addClause(SearchDataClause* cl)
|
90 |
bool SearchData::addClause(SearchDataClause* cl)
|
90 |
{
|
91 |
{
|
91 |
if (m_tp == SCLT_OR && (cl->m_tp == SCLT_EXCL)) {
|
92 |
if (m_tp == SCLT_OR && (cl->m_tp == SCLT_EXCL)) {
|
92 |
LOGERR(("SearchData::addClause: cant add EXCL to OR list\n"));
|
93 |
LOGERR(("SearchData::addClause: cant add EXCL to OR list\n"));
|
|
|
94 |
m_reason = "No Negative (AND_NOT) clauses allowed in OR queries";
|
93 |
return false;
|
95 |
return false;
|
94 |
}
|
96 |
}
|
95 |
m_query.push_back(cl);
|
97 |
m_query.push_back(cl);
|
96 |
return true;
|
98 |
return true;
|
97 |
}
|
99 |
}
|
98 |
|
100 |
|
99 |
// Make me all new
|
101 |
// Make me all new
|
100 |
void SearchData::erase() {
|
102 |
void SearchData::erase() {
|
|
|
103 |
LOGDEB(("SearchData::erase\n"));
|
|
|
104 |
m_tp = SCLT_AND;
|
101 |
for (qlist_it_t it = m_query.begin(); it != m_query.end(); it++)
|
105 |
for (qlist_it_t it = m_query.begin(); it != m_query.end(); it++)
|
102 |
delete *it;
|
106 |
delete *it;
|
103 |
m_query.clear();
|
107 |
m_query.clear();
|
104 |
m_filetypes.clear();
|
108 |
m_filetypes.clear();
|
105 |
m_topdir.erase();
|
109 |
m_topdir.erase();
|
106 |
m_description.erase();
|
110 |
m_description.erase();
|
|
|
111 |
m_reason.erase();
|
107 |
}
|
112 |
}
|
108 |
|
113 |
|
109 |
// Am I a file name only search ? This is to turn off term highlighting
|
114 |
// Am I a file name only search ? This is to turn off term highlighting
|
110 |
bool SearchData::fileNameOnly() {
|
115 |
bool SearchData::fileNameOnly()
|
|
|
116 |
{
|
111 |
for (qlist_it_t it = m_query.begin(); it != m_query.end(); it++)
|
117 |
for (qlist_it_t it = m_query.begin(); it != m_query.end(); it++)
|
112 |
if (!(*it)->isFileName())
|
118 |
if (!(*it)->isFileName())
|
113 |
return false;
|
119 |
return false;
|
114 |
return true;
|
120 |
return true;
|
115 |
}
|
121 |
}
|
116 |
|
122 |
|
|
|
123 |
// Extract all terms and term groups
|
|
|
124 |
bool SearchData::getTerms(vector<string>& terms,
|
|
|
125 |
vector<vector<string> >& groups,
|
|
|
126 |
vector<int>& gslks) const
|
|
|
127 |
{
|
|
|
128 |
for (qlist_cit_t it = m_query.begin(); it != m_query.end(); it++)
|
|
|
129 |
(*it)->getTerms(terms, groups, gslks);
|
|
|
130 |
return true;
|
|
|
131 |
}
|
|
|
132 |
|
117 |
// Splitter callback for breaking a user query string into simple
|
133 |
// Splitter callback for breaking a user query string into simple
|
118 |
// terms and phrases
|
134 |
// terms and phrases.
|
119 |
class wsQData : public TextSplitCB {
|
135 |
class wsQData : public TextSplitCB {
|
120 |
public:
|
136 |
public:
|
121 |
vector<string> terms;
|
137 |
vector<string> terms;
|
122 |
// Debug
|
138 |
// Debug
|
123 |
string catterms() {
|
139 |
string catterms() {
|
124 |
string s;
|
140 |
string s;
|
125 |
for (unsigned int i = 0; i < terms.size(); i++) {
|
141 |
for (unsigned int i = 0; i < terms.size(); i++)
|
126 |
s += "[" + terms[i] + "] ";
|
142 |
s += "[" + terms[i] + "] ";
|
127 |
}
|
|
|
128 |
return s;
|
143 |
return s;
|
129 |
}
|
144 |
}
|
130 |
bool takeword(const std::string &term, int , int, int) {
|
145 |
bool takeword(const std::string &term, int , int, int) {
|
131 |
LOGDEB1(("wsQData::takeword: %s\n", term.c_str()));
|
146 |
LOGDEB1(("wsQData::takeword: %s\n", term.c_str()));
|
132 |
terms.push_back(term);
|
147 |
terms.push_back(term);
|
133 |
return true;
|
148 |
return true;
|
134 |
}
|
149 |
}
|
135 |
// Decapital + deaccent all terms
|
|
|
136 |
void dumball() {
|
|
|
137 |
for (vector<string>::iterator it=terms.begin(); it !=terms.end();it++){
|
|
|
138 |
string dumb;
|
|
|
139 |
dumb_string(*it, dumb);
|
|
|
140 |
*it = dumb;
|
|
|
141 |
}
|
|
|
142 |
}
|
|
|
143 |
};
|
150 |
};
|
144 |
|
151 |
|
145 |
/** Possibly expand term into its stem siblings, make them dumb strings */
|
152 |
// This used to be a static function, but we couldn't just keep adding
|
|
|
153 |
// parameters to the interface!
|
|
|
154 |
class StringToXapianQ {
|
|
|
155 |
public:
|
|
|
156 |
StringToXapianQ(Db& db) : m_db(db) { }
|
|
|
157 |
bool translate(const string &iq,
|
|
|
158 |
const string& stemlang,
|
|
|
159 |
string &ermsg,
|
|
|
160 |
list<Xapian::Query> &pqueries,
|
|
|
161 |
int slack = 0, bool useNear = false);
|
|
|
162 |
bool getTerms(vector<string>& terms,
|
|
|
163 |
vector<vector<string> >& groups)
|
|
|
164 |
{
|
|
|
165 |
terms.insert(terms.end(), m_terms.begin(), m_terms.end());
|
|
|
166 |
groups.insert(groups.end(), m_groups.begin(), m_groups.end());
|
|
|
167 |
return true;
|
|
|
168 |
}
|
|
|
169 |
private:
|
146 |
static void maybeStemExp(Db& db, const string& stemlang, const string& term,
|
170 |
void maybeStemExp(const string& stemlang, const string& term,
|
|
|
171 |
list<string>& exp);
|
|
|
172 |
|
|
|
173 |
Db& m_db;
|
|
|
174 |
// Single terms and phrases resulting from breaking up text;
|
|
|
175 |
vector<string> m_terms;
|
|
|
176 |
vector<vector<string> > m_groups;
|
|
|
177 |
};
|
|
|
178 |
|
|
|
179 |
/** Make term dumb and possibly expand it into its stem siblings */
|
|
|
180 |
void StringToXapianQ::maybeStemExp(const string& stemlang,
|
|
|
181 |
const string& term,
|
147 |
list<string>& exp)
|
182 |
list<string>& exp)
|
148 |
{
|
183 |
{
|
149 |
LOGDEB(("maybeStemExp: [%s]\n", term.c_str()));
|
184 |
LOGDEB2(("maybeStemExp: [%s]\n", term.c_str()));
|
|
|
185 |
if (term.empty()) {
|
|
|
186 |
exp.clear();
|
|
|
187 |
return;
|
|
|
188 |
}
|
|
|
189 |
|
150 |
string term1;
|
190 |
string term1;
|
151 |
dumb_string(term, term1);
|
191 |
dumb_string(term, term1);
|
152 |
if (!stemlang.empty()) {
|
192 |
|
153 |
bool nostemexp = false;
|
193 |
bool nostemexp = stemlang.empty() ? true : false;
|
|
|
194 |
if (!nostemexp) {
|
154 |
// Check if the first letter is a majuscule in which
|
195 |
// Check if the first letter is a majuscule in which
|
155 |
// case we do not want to do stem expansion. Note that
|
196 |
// case we do not want to do stem expansion. Note that
|
156 |
// the test is convoluted and possibly problematic
|
197 |
// the test is convoluted and possibly problematic
|
157 |
if (term.length() > 0) {
|
198 |
|
158 |
string noacterm,noaclowterm;
|
199 |
string noacterm,noaclowterm;
|
159 |
if (unacmaybefold(term, noacterm, "UTF-8", false) &&
|
200 |
if (unacmaybefold(term, noacterm, "UTF-8", false) &&
|
160 |
unacmaybefold(noacterm, noaclowterm, "UTF-8", true)) {
|
201 |
unacmaybefold(noacterm, noaclowterm, "UTF-8", true)) {
|
161 |
Utf8Iter it1(noacterm);
|
202 |
Utf8Iter it1(noacterm);
|
162 |
Utf8Iter it2(noaclowterm);
|
203 |
Utf8Iter it2(noaclowterm);
|
163 |
if (*it1 != *it2)
|
204 |
if (*it1 != *it2)
|
164 |
nostemexp = true;
|
205 |
nostemexp = true;
|
165 |
}
|
|
|
166 |
}
|
206 |
}
|
167 |
LOGDEB1(("Term: %s stem expansion: %s\n",
|
207 |
LOGDEB1(("Term: %s stem expansion: %s\n", term.c_str()));
|
168 |
term.c_str(), nostemexp?"no":"yes"));
|
208 |
}
|
|
|
209 |
|
169 |
if (!nostemexp) {
|
210 |
if (nostemexp) {
|
|
|
211 |
exp = list<string>(1, term1);
|
|
|
212 |
} else {
|
170 |
exp = db.stemExpand(stemlang, term1);
|
213 |
exp = m_db.stemExpand(stemlang, term1);
|
171 |
return;
|
214 |
}
|
172 |
}
|
215 |
}
|
173 |
}
|
|
|
174 |
|
216 |
|
175 |
exp.push_back(term1);
|
217 |
/**
|
176 |
}
|
|
|
177 |
|
|
|
178 |
/** Turn string into list of xapian queries. There is little
|
218 |
* Turn string into list of xapian queries. There is little
|
179 |
* interpretation done on the string (no +term -term or filename:term
|
219 |
* interpretation done on the string (no +term -term or filename:term
|
180 |
* stuff). We just separate words and phrases, and interpret
|
220 |
* stuff). We just separate words and phrases, and interpret
|
181 |
* capitalized terms as wanting no stem expansion.
|
221 |
* capitalized terms as wanting no stem expansion.
|
182 |
* The final list contains one query for each term or phrase
|
222 |
* The final list contains one query for each term or phrase
|
183 |
* - Elements corresponding to a stem-expanded part are an OP_OR
|
223 |
* - Elements corresponding to a stem-expanded part are an OP_OR
|
184 |
* composition of the stem-expanded terms (or a single term query).
|
224 |
* composition of the stem-expanded terms (or a single term query).
|
185 |
* - Elements corresponding to a phrase are an OP_PHRASE composition of the
|
225 |
* - Elements corresponding to a phrase are an OP_PHRASE composition of the
|
186 |
* phrase terms (no stem expansion in this case)
|
226 |
* phrase terms (no stem expansion in this case)
|
187 |
* @return the subquery count (either or'd stem-expanded terms or phrase word
|
227 |
* @return the subquery count (either or'd stem-expanded terms or phrase word
|
188 |
* count)
|
228 |
* count)
|
189 |
*/
|
229 |
*/
|
190 |
static bool stringToXapianQueries(const string &iq,
|
230 |
bool StringToXapianQ::translate(const string &iq,
|
191 |
const string& stemlang,
|
231 |
const string& stemlang,
|
192 |
Db& db,
|
|
|
193 |
string &ermsg,
|
232 |
string &ermsg,
|
194 |
list<Xapian::Query> &pqueries,
|
233 |
list<Xapian::Query> &pqueries,
|
195 |
int slack = 0, bool useNear = false)
|
234 |
int slack, bool useNear)
|
196 |
{
|
235 |
{
|
197 |
string qstring = iq;
|
236 |
string qstring = iq;
|
198 |
bool opt_stemexp = !stemlang.empty();
|
237 |
bool opt_stemexp = !stemlang.empty();
|
199 |
ermsg.erase();
|
238 |
ermsg.erase();
|
|
|
239 |
m_terms.clear();
|
|
|
240 |
m_groups.clear();
|
200 |
|
241 |
|
201 |
// Split into words and phrases (word1 word2 "this is a phrase"):
|
242 |
// Split into words and phrases (word1 word2 "this is a phrase"):
|
202 |
list<string> phrases;
|
243 |
list<string> phrases;
|
203 |
stringToStrings(qstring, phrases);
|
244 |
stringToStrings(qstring, phrases);
|
204 |
|
245 |
|
|
... |
|
... |
229 |
case 0: continue;// ??
|
270 |
case 0: continue;// ??
|
230 |
case 1: // Not a real phrase: one term
|
271 |
case 1: // Not a real phrase: one term
|
231 |
{
|
272 |
{
|
232 |
string term = splitData.terms.front();
|
273 |
string term = splitData.terms.front();
|
233 |
list<string> exp;
|
274 |
list<string> exp;
|
234 |
maybeStemExp(db, stemlang, term, exp);
|
275 |
maybeStemExp(stemlang, term, exp);
|
235 |
// Push either term or OR of stem-expanded set
|
276 |
// Push either term or OR of stem-expanded set
|
236 |
pqueries.push_back(Xapian::Query(Xapian::Query::OP_OR,
|
277 |
pqueries.push_back(Xapian::Query(Xapian::Query::OP_OR,
|
237 |
exp.begin(), exp.end()));
|
278 |
exp.begin(), exp.end()));
|
|
|
279 |
m_terms.insert(m_terms.end(), exp.begin(), exp.end());
|
238 |
}
|
280 |
}
|
239 |
break;
|
281 |
break;
|
240 |
|
282 |
|
241 |
default:
|
283 |
default:
|
242 |
// Phrase/near
|
284 |
// Phrase/near
|
243 |
Xapian::Query::op op = useNear ? Xapian::Query::OP_NEAR :
|
285 |
Xapian::Query::op op = useNear ? Xapian::Query::OP_NEAR :
|
244 |
Xapian::Query::OP_PHRASE;
|
286 |
Xapian::Query::OP_PHRASE;
|
245 |
list<Xapian::Query> orqueries;
|
287 |
list<Xapian::Query> orqueries;
|
246 |
bool hadmultiple = false;
|
288 |
bool hadmultiple = false;
|
247 |
string nolang, lang;
|
289 |
string nolang, lang;
|
|
|
290 |
vector<string> dumbterms;
|
248 |
for (vector<string>::iterator it = splitData.terms.begin();
|
291 |
for (vector<string>::iterator it = splitData.terms.begin();
|
249 |
it != splitData.terms.end(); it++) {
|
292 |
it != splitData.terms.end(); it++) {
|
250 |
list<string>exp;
|
293 |
list<string>exp;
|
251 |
lang = (op == Xapian::Query::OP_PHRASE || hadmultiple) ?
|
294 |
lang = (op == Xapian::Query::OP_PHRASE || hadmultiple) ?
|
252 |
nolang : stemlang;
|
295 |
nolang : stemlang;
|
253 |
maybeStemExp(db, lang, *it, exp);
|
296 |
maybeStemExp(lang, *it, exp);
|
|
|
297 |
dumbterms.insert(dumbterms.end(), exp.begin(), exp.end());
|
|
|
298 |
#ifdef XAPIAN_NEAR_EXPAND_SINGLE_BUF
|
254 |
if (exp.size() > 1)
|
299 |
if (exp.size() > 1)
|
255 |
hadmultiple = true;
|
300 |
hadmultiple = true;
|
|
|
301 |
#endif
|
256 |
orqueries.push_back(Xapian::Query(Xapian::Query::OP_OR,
|
302 |
orqueries.push_back(Xapian::Query(Xapian::Query::OP_OR,
|
257 |
exp.begin(), exp.end()));
|
303 |
exp.begin(), exp.end()));
|
258 |
}
|
304 |
}
|
259 |
pqueries.push_back(Xapian::Query(op,
|
305 |
pqueries.push_back(Xapian::Query(op,
|
260 |
orqueries.begin(),
|
306 |
orqueries.begin(),
|
261 |
orqueries.end(),
|
307 |
orqueries.end(),
|
262 |
splitData.terms.size() + slack));
|
308 |
splitData.terms.size() + slack));
|
|
|
309 |
m_groups.push_back(dumbterms);
|
263 |
}
|
310 |
}
|
264 |
}
|
311 |
}
|
265 |
} catch (const Xapian::Error &e) {
|
312 |
} catch (const Xapian::Error &e) {
|
266 |
ermsg = e.get_msg();
|
313 |
ermsg = e.get_msg();
|
267 |
} catch (const string &s) {
|
314 |
} catch (const string &s) {
|
|
... |
|
... |
280 |
|
327 |
|
281 |
// Translate a simple OR, AND, or EXCL search clause.
|
328 |
// Translate a simple OR, AND, or EXCL search clause.
|
282 |
bool SearchDataClauseSimple::toNativeQuery(Rcl::Db &db, void *p,
|
329 |
bool SearchDataClauseSimple::toNativeQuery(Rcl::Db &db, void *p,
|
283 |
const string& stemlang)
|
330 |
const string& stemlang)
|
284 |
{
|
331 |
{
|
|
|
332 |
m_terms.clear();
|
|
|
333 |
m_groups.clear();
|
285 |
Xapian::Query *qp = (Xapian::Query *)p;
|
334 |
Xapian::Query *qp = (Xapian::Query *)p;
|
286 |
*qp = Xapian::Query();
|
335 |
*qp = Xapian::Query();
|
287 |
|
336 |
|
288 |
Xapian::Query::op op;
|
337 |
Xapian::Query::op op;
|
289 |
switch (m_tp) {
|
338 |
switch (m_tp) {
|
290 |
case SCLT_AND: op = Xapian::Query::OP_AND; break;
|
339 |
case SCLT_AND: op = Xapian::Query::OP_AND; break;
|
|
|
340 |
// EXCL will be set with AND_NOT in the list. So it's an OR list here
|
291 |
case SCLT_OR:
|
341 |
case SCLT_OR:
|
292 |
case SCLT_EXCL: op = Xapian::Query::OP_OR; break;
|
342 |
case SCLT_EXCL: op = Xapian::Query::OP_OR; break;
|
293 |
default:
|
343 |
default:
|
294 |
LOGERR(("SearchDataClauseSimple: bad m_tp %d\n", m_tp));
|
344 |
LOGERR(("SearchDataClauseSimple: bad m_tp %d\n", m_tp));
|
295 |
return false;
|
345 |
return false;
|
296 |
}
|
346 |
}
|
297 |
list<Xapian::Query> pqueries;
|
347 |
list<Xapian::Query> pqueries;
|
|
|
348 |
StringToXapianQ tr(db);
|
298 |
if (!stringToXapianQueries(m_text, stemlang, db, m_reason, pqueries))
|
349 |
if (!tr.translate(m_text, stemlang, m_reason, pqueries))
|
299 |
return false;
|
350 |
return false;
|
300 |
if (pqueries.empty()) {
|
351 |
if (pqueries.empty()) {
|
301 |
LOGERR(("SearchDataClauseSimple: resolved to null query\n"));
|
352 |
LOGERR(("SearchDataClauseSimple: resolved to null query\n"));
|
302 |
return true;
|
353 |
return true;
|
303 |
}
|
354 |
}
|
|
|
355 |
tr.getTerms(m_terms, m_groups);
|
304 |
*qp = Xapian::Query(op, pqueries.begin(), pqueries.end());
|
356 |
*qp = Xapian::Query(op, pqueries.begin(), pqueries.end());
|
305 |
return true;
|
357 |
return true;
|
306 |
}
|
358 |
}
|
307 |
|
359 |
|
308 |
// Translate a FILENAME search clause.
|
360 |
// Translate a FILENAME search clause.
|
|
... |
|
... |
317 |
// Build a query out of the matching file name terms.
|
369 |
// Build a query out of the matching file name terms.
|
318 |
*qp = Xapian::Query(Xapian::Query::OP_OR, names.begin(), names.end());
|
370 |
*qp = Xapian::Query(Xapian::Query::OP_OR, names.begin(), names.end());
|
319 |
return true;
|
371 |
return true;
|
320 |
}
|
372 |
}
|
321 |
|
373 |
|
322 |
// Translate NEAR or PHRASE clause. We're not handling the distance parameter
|
374 |
// Translate NEAR or PHRASE clause.
|
323 |
// yet.
|
|
|
324 |
bool SearchDataClauseDist::toNativeQuery(Rcl::Db &db, void *p,
|
375 |
bool SearchDataClauseDist::toNativeQuery(Rcl::Db &db, void *p,
|
325 |
const string& stemlang)
|
376 |
const string& stemlang)
|
326 |
{
|
377 |
{
|
|
|
378 |
m_terms.clear();
|
|
|
379 |
m_groups.clear();
|
|
|
380 |
|
327 |
Xapian::Query *qp = (Xapian::Query *)p;
|
381 |
Xapian::Query *qp = (Xapian::Query *)p;
|
328 |
*qp = Xapian::Query();
|
382 |
*qp = Xapian::Query();
|
329 |
|
383 |
|
330 |
list<Xapian::Query> pqueries;
|
384 |
list<Xapian::Query> pqueries;
|
331 |
Xapian::Query nq;
|
385 |
Xapian::Query nq;
|
|
|
386 |
|
|
|
387 |
// Use stringToXapianQueries to lowercase and simplify the phrase
|
|
|
388 |
// terms etc. The result should be a single element list
|
332 |
string s = string("\"") + m_text + string("\"");
|
389 |
string s = string("\"") + m_text + string("\"");
|
333 |
bool useNear = m_tp == SCLT_NEAR;
|
390 |
bool useNear = m_tp == SCLT_NEAR;
|
334 |
|
391 |
StringToXapianQ tr(db);
|
335 |
// Use stringToXapianQueries anyway to lowercase and simplify the
|
392 |
if (!tr.translate(s, stemlang, m_reason, pqueries, m_slack, useNear))
|
336 |
// phrase terms etc. The result should be a single element list
|
|
|
337 |
if (!stringToXapianQueries(s, stemlang, db, m_reason, pqueries,
|
|
|
338 |
m_slack, useNear))
|
|
|
339 |
return false;
|
393 |
return false;
|
340 |
if (pqueries.empty()) {
|
394 |
if (pqueries.empty()) {
|
341 |
LOGERR(("SearchDataClauseDist: resolved to null query\n"));
|
395 |
LOGERR(("SearchDataClauseDist: resolved to null query\n"));
|
342 |
return true;
|
396 |
return true;
|
343 |
}
|
397 |
}
|
|
|
398 |
tr.getTerms(m_terms, m_groups);
|
344 |
*qp = *pqueries.begin();
|
399 |
*qp = *pqueries.begin();
|
345 |
return true;
|
400 |
return true;
|
346 |
}
|
401 |
}
|
347 |
|
402 |
|
348 |
} // Namespace Rcl
|
403 |
} // Namespace Rcl
|