Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commit21b748e

Browse files
committed
1 Fix problem with lost precision in rank with OR-ed lexemes
2 Allow tsquery_in to input void tsquery: resolve dump/restore problem with tsquery
1 parentfbff2e9 commit21b748e

File tree

3 files changed

+49
-23
lines changed

3 files changed

+49
-23
lines changed

‎contrib/tsearch2/expected/tsearch2.out

Lines changed: 9 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -746,21 +746,21 @@ select count(*) FROM test_tsvector WHERE a @@ to_tsquery('copyright');
746746
(1 row)
747747

748748
select rank(' a:1 s:2C d g'::tsvector, 'a | s');
749-
rank
750-
------
751-
0.28
749+
rank
750+
-----------
751+
0.0911891
752752
(1 row)
753753

754754
select rank(' a:1 s:2B d g'::tsvector, 'a | s');
755-
rank
756-
------
757-
0.46
755+
rank
756+
----------
757+
0.151982
758758
(1 row)
759759

760760
select rank(' a:1 s:2 d g'::tsvector, 'a | s');
761-
rank
762-
------
763-
0.19
761+
rank
762+
-----------
763+
0.0607927
764764
(1 row)
765765

766766
select rank(' a:1 s:2C d g'::tsvector, 'a & s');

‎contrib/tsearch2/query.c

Lines changed: 17 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -55,6 +55,7 @@ Datumto_tsquery_current(PG_FUNCTION_ARGS);
5555
/* parser's states */
5656
#defineWAITOPERAND 1
5757
#defineWAITOPERATOR2
58+
#defineWAITFIRSTOPERAND 3
5859

5960
/*
6061
* node of query tree, also used
@@ -137,6 +138,7 @@ gettoken_query(QPRS_STATE * state, int4 *val, int4 *lenval, char **strval, int2
137138
{
138139
switch (state->state)
139140
{
141+
caseWAITFIRSTOPERAND:
140142
caseWAITOPERAND:
141143
if (*(state->buf)=='!')
142144
{
@@ -159,14 +161,16 @@ gettoken_query(QPRS_STATE * state, int4 *val, int4 *lenval, char **strval, int2
159161
elseif (*(state->buf)!=' ')
160162
{
161163
state->valstate.prsbuf=state->buf;
162-
state->state=WAITOPERATOR;
163164
if (gettoken_tsvector(&(state->valstate)))
164165
{
165166
*strval=state->valstate.word;
166167
*lenval=state->valstate.curpos-state->valstate.word;
167168
state->buf=get_weight(state->valstate.prsbuf,weight);
169+
state->state=WAITOPERATOR;
168170
returnVAL;
169171
}
172+
elseif (state->state==WAITFIRSTOPERAND )
173+
returnEND;
170174
else
171175
ereport(ERROR,
172176
(errcode(ERRCODE_SYNTAX_ERROR),
@@ -596,7 +600,7 @@ static QUERYTYPE *
596600

597601
/* init state */
598602
state.buf=buf;
599-
state.state=WAITOPERAND;
603+
state.state=WAITFIRSTOPERAND;
600604
state.count=0;
601605
state.num=0;
602606
state.str=NULL;
@@ -616,10 +620,13 @@ static QUERYTYPE *
616620
/* parse query & make polish notation (postfix, but in reverse order) */
617621
makepol(&state,pushval);
618622
pfree(state.valstate.word);
619-
if (!state.num)
620-
ereport(ERROR,
621-
(errcode(ERRCODE_SYNTAX_ERROR),
622-
errmsg("empty query")));
623+
if (!state.num) {
624+
elog(NOTICE,"Query doesn't contain lexem(s)");
625+
query= (QUERYTYPE*)palloc(HDRSIZEQT );
626+
query->len=HDRSIZEQT;
627+
query->size=0;
628+
returnquery;
629+
}
623630

624631
/* make finish struct */
625632
commonlen=COMPUTESIZE(state.num,state.sumlen);
@@ -905,6 +912,10 @@ to_tsquery(PG_FUNCTION_ARGS)
905912
PG_FREE_IF_COPY(in,1);
906913

907914
query=queryin(str,pushval_morph,PG_GETARG_INT32(0));
915+
916+
if (query->size==0 )
917+
PG_RETURN_POINTER(query);
918+
908919
res=clean_fakeval_v2(GETQUERY(query),&len);
909920
if (!res)
910921
{

‎contrib/tsearch2/rank.c

Lines changed: 23 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -257,7 +257,7 @@ calc_rank_or(float *w, tsvector * t, QUERYTYPE * q)
257257
int4dimt,
258258
j,
259259
i;
260-
floatres=-1.0;
260+
floatres=0.0;
261261
ITEM**item;
262262
intsize=q->size;
263263

@@ -266,6 +266,8 @@ calc_rank_or(float *w, tsvector * t, QUERYTYPE * q)
266266

267267
for (i=0;i<size;i++)
268268
{
269+
floatresj,wjm;
270+
int4jm;
269271
entry=find_wordentry(t,q,item[i]);
270272
if (!entry)
271273
continue;
@@ -281,14 +283,27 @@ calc_rank_or(float *w, tsvector * t, QUERYTYPE * q)
281283
post=POSNULL+1;
282284
}
283285

284-
for (j=0;j<dimt;j++)
285-
{
286-
if (res<0)
287-
res=wpos(post[j]);
288-
else
289-
res=1.0- (1.0-res)* (1.0-wpos(post[j]));
290-
}
286+
resj=0.0;
287+
wjm=-1.0;
288+
jm=0;
289+
for (j=0;j<dimt;j++)
290+
{
291+
resj=resj+wpos(post[j])/((j+1)*(j+1));
292+
if (wpos(post[j])>wjm ) {
293+
wjm=wpos(post[j]);
294+
jm=j;
295+
}
296+
}
297+
/*
298+
limit (sum(i/i^2),i->inf) = pi^2/6
299+
resj = sum(wi/i^2),i=1,noccurence,
300+
wi - should be sorted desc,
301+
don't sort for now, just choose maximum weight. This should be corrected
302+
Oleg Bartunov
303+
*/
304+
res=res+ (wjm+resj-wjm/((jm+1)*(jm+1)))/1.64493406685;
291305
}
306+
res=res /size;
292307
pfree(item);
293308
returnres;
294309
}

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp