postgresql/src/backend/utils/adt/tsrank.c

838 lines
17 KiB
C
Raw Normal View History

/*-------------------------------------------------------------------------
*
* tsrank.c
* rank tsvector by tsquery
*
* Portions Copyright (c) 1996-2007, PostgreSQL Global Development Group
*
*
* IDENTIFICATION
* $PostgreSQL: pgsql/src/backend/utils/adt/tsrank.c,v 1.5 2007/09/11 08:46:29 teodor Exp $
*
*-------------------------------------------------------------------------
*/
#include "postgres.h"
#include <math.h>
#include "tsearch/ts_type.h"
#include "tsearch/ts_utils.h"
#include "utils/array.h"
#include "miscadmin.h"
static float weights[] = {0.1, 0.2, 0.4, 1.0};
#define wpos(wep) ( w[ WEP_GETWEIGHT(wep) ] )
#define RANK_NO_NORM 0x00
#define RANK_NORM_LOGLENGTH 0x01
#define RANK_NORM_LENGTH 0x02
#define RANK_NORM_EXTDIST 0x04
#define RANK_NORM_UNIQ 0x08
#define RANK_NORM_LOGUNIQ 0x10
#define DEF_NORM_METHOD RANK_NO_NORM
static float calc_rank_or(float *w, TSVector t, TSQuery q);
static float calc_rank_and(float *w, TSVector t, TSQuery q);
/*
* Returns a weight of a word collocation
*/
static float4
word_distance(int4 w)
{
if (w > 100)
return 1e-30;
return 1.0 / (1.005 + 0.05 * exp(((float4) w) / 1.5 - 2));
}
static int
cnt_length(TSVector t)
{
WordEntry *ptr = ARRPTR(t),
*end = (WordEntry *) STRPTR(t);
int len = 0;
while (ptr < end)
{
int clen = POSDATALEN(t, ptr);
if (clen == 0)
len += 1;
else
len += clen;
ptr++;
}
return len;
}
static int
WordECompareQueryItem(char *eval, char *qval, WordEntry *ptr, QueryOperand *item)
{
if (ptr->len == item->length)
return strncmp(
eval + ptr->pos,
qval + item->distance,
item->length);
return (ptr->len > item->length) ? 1 : -1;
}
/*
* Returns a pointer to a WordEntry corresponding 'item' from tsvector 't'. 'q'
* is the TSQuery containing 'item'. Returns NULL if not found.
*/
static WordEntry *
find_wordentry(TSVector t, TSQuery q, QueryOperand *item)
{
WordEntry *StopLow = ARRPTR(t);
WordEntry *StopHigh = (WordEntry *) STRPTR(t);
WordEntry *StopMiddle;
int difference;
/* Loop invariant: StopLow <= item < StopHigh */
while (StopLow < StopHigh)
{
StopMiddle = StopLow + (StopHigh - StopLow) / 2;
difference = WordECompareQueryItem(STRPTR(t), GETOPERAND(q), StopMiddle, item);
if (difference == 0)
return StopMiddle;
else if (difference < 0)
StopLow = StopMiddle + 1;
else
StopHigh = StopMiddle;
}
return NULL;
}
/*
* sort QueryOperands by (length, word)
*/
static int
compareQueryOperand(const void *a, const void *b, void *arg)
{
char *operand = (char *) arg;
QueryOperand *qa = (*(QueryOperand **) a);
QueryOperand *qb = (*(QueryOperand **) b);
if (qa->length == qb->length)
return strncmp(operand + qa->distance,
operand + qb->distance,
qb->length);
return (qa->length > qb->length) ? 1 : -1;
}
/*
* Returns a sorted, de-duplicated array of QueryOperands in a query.
* The returned QueryOperands are pointers to the original QueryOperands
* in the query.
*
* Length of the returned array is stored in *size
*/
static QueryOperand **
SortAndUniqItems(TSQuery q, int *size)
{
char *operand = GETOPERAND(q);
QueryItem * item = GETQUERY(q);
QueryOperand **res,
**ptr,
**prevptr;
ptr = res = (QueryOperand **) palloc(sizeof(QueryOperand *) * *size);
/* Collect all operands from the tree to res */
while ((*size)--)
{
if (item->type == QI_VAL)
{
*ptr = (QueryOperand *) item;
ptr++;
}
item++;
}
*size = ptr - res;
if (*size < 2)
return res;
qsort_arg(res, *size, sizeof(QueryOperand **), compareQueryOperand, (void *) operand);
ptr = res + 1;
prevptr = res;
/* remove duplicates */
while (ptr - res < *size)
{
if (compareQueryOperand((void *) ptr, (void *) prevptr, (void *) operand) != 0)
{
prevptr++;
*prevptr = *ptr;
}
ptr++;
}
*size = prevptr + 1 - res;
return res;
}
/* A dummy WordEntryPos array to use when haspos is false */
static WordEntryPosVector POSNULL = {
1, /* Number of elements that follow */
{ 0 }
};
static float
calc_rank_and(float *w, TSVector t, TSQuery q)
{
WordEntryPosVector **pos;
int i,
k,
l,
p;
WordEntry *entry;
WordEntryPos *post,
*ct;
int4 dimt,
lenct,
dist;
float res = -1.0;
QueryOperand **item;
int size = q->size;
item = SortAndUniqItems(q, &size);
if (size < 2)
{
pfree(item);
return calc_rank_or(w, t, q);
}
pos = (WordEntryPosVector **) palloc0(sizeof(WordEntryPosVector *) * q->size);
WEP_SETPOS(POSNULL.pos[0], MAXENTRYPOS - 1);
for (i = 0; i < size; i++)
{
entry = find_wordentry(t, q, item[i]);
if (!entry)
continue;
if (entry->haspos)
pos[i] = _POSVECPTR(t, entry);
else
pos[i] = &POSNULL;
dimt = pos[i]->npos;
post = pos[i]->pos;
for (k = 0; k < i; k++)
{
if (!pos[k])
continue;
lenct = pos[k]->npos;
ct = pos[k]->pos;
for (l = 0; l < dimt; l++)
{
for (p = 0; p < lenct; p++)
{
dist = Abs((int) WEP_GETPOS(post[l]) - (int) WEP_GETPOS(ct[p]));
if (dist || (dist == 0 && (pos[i] == &POSNULL || pos[k] == &POSNULL)))
{
float curw;
if (!dist)
dist = MAXENTRYPOS;
curw = sqrt(wpos(post[l]) * wpos(ct[p]) * word_distance(dist));
res = (res < 0) ? curw : 1.0 - (1.0 - res) * (1.0 - curw);
}
}
}
}
}
pfree(pos);
pfree(item);
return res;
}
static float
calc_rank_or(float *w, TSVector t, TSQuery q)
{
WordEntry *entry;
WordEntryPos *post;
int4 dimt,
j,
i;
float res = 0.0;
QueryOperand **item;
int size = q->size;
item = SortAndUniqItems(q, &size);
for (i = 0; i < size; i++)
{
float resj,
wjm;
int4 jm;
entry = find_wordentry(t, q, item[i]);
if (!entry)
continue;
if (entry->haspos)
{
dimt = POSDATALEN(t, entry);
post = POSDATAPTR(t, entry);
}
else
{
dimt = POSNULL.npos;
post = POSNULL.pos;
}
resj = 0.0;
wjm = -1.0;
jm = 0;
for (j = 0; j < dimt; j++)
{
resj = resj + wpos(post[j]) / ((j + 1) * (j + 1));
if (wpos(post[j]) > wjm)
{
wjm = wpos(post[j]);
jm = j;
}
}
/*
limit (sum(i/i^2),i->inf) = pi^2/6
resj = sum(wi/i^2),i=1,noccurence,
wi - should be sorted desc,
don't sort for now, just choose maximum weight. This should be corrected
Oleg Bartunov
*/
res = res + (wjm + resj - wjm / ((jm + 1) * (jm + 1))) / 1.64493406685;
}
if (size > 0)
res = res / size;
pfree(item);
return res;
}
static float
calc_rank(float *w, TSVector t, TSQuery q, int4 method)
{
QueryItem *item = GETQUERY(q);
float res = 0.0;
int len;
if (!t->size || !q->size)
return 0.0;
/* XXX: What about NOT? */
res = (item->type == QI_OPR && item->operator.oper == OP_AND) ?
calc_rank_and(w, t, q) : calc_rank_or(w, t, q);
if (res < 0)
res = 1e-20;
if ((method & RANK_NORM_LOGLENGTH) && t->size > 0)
res /= log((double) (cnt_length(t) + 1)) / log(2.0);
if (method & RANK_NORM_LENGTH)
{
len = cnt_length(t);
if (len > 0)
res /= (float) len;
}
if ((method & RANK_NORM_UNIQ) && t->size > 0)
res /= (float) (t->size);
if ((method & RANK_NORM_LOGUNIQ) && t->size > 0)
res /= log((double) (t->size + 1)) / log(2.0);
return res;
}
static float *
getWeights(ArrayType *win)
{
static float ws[lengthof(weights)];
int i;
float4 *arrdata;
if (win == 0)
return weights;
if (ARR_NDIM(win) != 1)
ereport(ERROR,
(errcode(ERRCODE_ARRAY_SUBSCRIPT_ERROR),
errmsg("array of weight must be one-dimensional")));
if (ArrayGetNItems(ARR_NDIM(win), ARR_DIMS(win)) < lengthof(weights))
ereport(ERROR,
(errcode(ERRCODE_ARRAY_SUBSCRIPT_ERROR),
errmsg("array of weight is too short")));
if (ARR_HASNULL(win))
ereport(ERROR,
(errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
errmsg("array of weight must not contain nulls")));
arrdata = (float4 *) ARR_DATA_PTR(win);
for (i = 0; i < lengthof(weights); i++)
{
ws[i] = (arrdata[i] >= 0) ? arrdata[i] : weights[i];
if (ws[i] > 1.0)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("weight out of range")));
}
return ws;
}
Datum
ts_rank_wttf(PG_FUNCTION_ARGS)
{
ArrayType *win = (ArrayType *) PG_DETOAST_DATUM(PG_GETARG_DATUM(0));
TSVector txt = PG_GETARG_TSVECTOR(1);
TSQuery query = PG_GETARG_TSQUERY(2);
int method = PG_GETARG_INT32(3);
float res;
res = calc_rank(getWeights(win), txt, query, method);
PG_FREE_IF_COPY(win, 0);
PG_FREE_IF_COPY(txt, 1);
PG_FREE_IF_COPY(query, 2);
PG_RETURN_FLOAT4(res);
}
Datum
ts_rank_wtt(PG_FUNCTION_ARGS)
{
ArrayType *win = (ArrayType *) PG_DETOAST_DATUM(PG_GETARG_DATUM(0));
TSVector txt = PG_GETARG_TSVECTOR(1);
TSQuery query = PG_GETARG_TSQUERY(2);
float res;
res = calc_rank(getWeights(win), txt, query, DEF_NORM_METHOD);
PG_FREE_IF_COPY(win, 0);
PG_FREE_IF_COPY(txt, 1);
PG_FREE_IF_COPY(query, 2);
PG_RETURN_FLOAT4(res);
}
Datum
ts_rank_ttf(PG_FUNCTION_ARGS)
{
TSVector txt = PG_GETARG_TSVECTOR(0);
TSQuery query = PG_GETARG_TSQUERY(1);
int method = PG_GETARG_INT32(2);
float res;
res = calc_rank(getWeights(NULL), txt, query, method);
PG_FREE_IF_COPY(txt, 0);
PG_FREE_IF_COPY(query, 1);
PG_RETURN_FLOAT4(res);
}
Datum
ts_rank_tt(PG_FUNCTION_ARGS)
{
TSVector txt = PG_GETARG_TSVECTOR(0);
TSQuery query = PG_GETARG_TSQUERY(1);
float res;
res = calc_rank(getWeights(NULL), txt, query, DEF_NORM_METHOD);
PG_FREE_IF_COPY(txt, 0);
PG_FREE_IF_COPY(query, 1);
PG_RETURN_FLOAT4(res);
}
typedef struct
{
QueryItem **item;
int16 nitem;
uint8 wclass;
int32 pos;
} DocRepresentation;
static int
compareDocR(const void *va, const void *vb)
{
DocRepresentation *a = (DocRepresentation *) va;
DocRepresentation *b = (DocRepresentation *) vb;
if (a->pos == b->pos)
return 0;
return (a->pos > b->pos) ? 1 : -1;
}
static bool
checkcondition_QueryOperand(void *checkval, QueryOperand *val)
{
return (bool) (val->istrue);
}
static void
reset_istrue_flag(TSQuery query)
{
QueryItem *item = GETQUERY(query);
int i;
/* reset istrue flag */
for (i = 0; i < query->size; i++)
{
if (item->type == QI_VAL)
item->operand.istrue = 0;
item++;
}
}
typedef struct
{
int pos;
int p;
int q;
DocRepresentation *begin;
DocRepresentation *end;
} Extention;
static bool
Cover(DocRepresentation *doc, int len, TSQuery query, Extention *ext)
{
DocRepresentation *ptr;
int lastpos = ext->pos;
int i;
bool found = false;
/* since this function recurses, it could be driven to stack overflow.
* (though any decent compiler will optimize away the tail-recursion. */
check_stack_depth();
reset_istrue_flag(query);
ext->p = 0x7fffffff;
ext->q = 0;
ptr = doc + ext->pos;
/* find upper bound of cover from current position, move up */
while (ptr - doc < len)
{
for (i = 0; i < ptr->nitem; i++)
{
if(ptr->item[i]->type == QI_VAL)
ptr->item[i]->operand.istrue = 1;
}
if (TS_execute(GETQUERY(query), NULL, false, checkcondition_QueryOperand))
{
if (ptr->pos > ext->q)
{
ext->q = ptr->pos;
ext->end = ptr;
lastpos = ptr - doc;
found = true;
}
break;
}
ptr++;
}
if (!found)
return false;
reset_istrue_flag(query);
ptr = doc + lastpos;
/* find lower bound of cover from found upper bound, move down */
while (ptr >= doc + ext->pos)
{
for (i = 0; i < ptr->nitem; i++)
if(ptr->item[i]->type == QI_VAL)
ptr->item[i]->operand.istrue = 1;
if (TS_execute(GETQUERY(query), NULL, true, checkcondition_QueryOperand))
{
if (ptr->pos < ext->p)
{
ext->begin = ptr;
ext->p = ptr->pos;
}
break;
}
ptr--;
}
if (ext->p <= ext->q)
{
/*
* set position for next try to next lexeme after begining of founded
* cover
*/
ext->pos = (ptr - doc) + 1;
return true;
}
ext->pos++;
return Cover(doc, len, query, ext);
}
static DocRepresentation *
get_docrep(TSVector txt, TSQuery query, int *doclen)
{
QueryItem *item = GETQUERY(query);
WordEntry *entry;
WordEntryPos *post;
int4 dimt,
j,
i;
int len = query->size * 4,
cur = 0;
DocRepresentation *doc;
char *operand;
doc = (DocRepresentation *) palloc(sizeof(DocRepresentation) * len);
operand = GETOPERAND(query);
reset_istrue_flag(query);
for (i = 0; i < query->size; i++)
{
QueryOperand *curoperand;
if (item[i].type != QI_VAL)
continue;
curoperand = &item[i].operand;
if(item[i].operand.istrue)
continue;
entry = find_wordentry(txt, query, curoperand);
if (!entry)
continue;
if (entry->haspos)
{
dimt = POSDATALEN(txt, entry);
post = POSDATAPTR(txt, entry);
}
else
{
dimt = POSNULL.npos;
post = POSNULL.pos;
}
while (cur + dimt >= len)
{
len *= 2;
doc = (DocRepresentation *) repalloc(doc, sizeof(DocRepresentation) * len);
}
for (j = 0; j < dimt; j++)
{
if (j == 0)
{
int k;
doc[cur].nitem = 0;
doc[cur].item = (QueryItem **) palloc(sizeof(QueryItem *) * query->size);
for (k = 0; k < query->size; k++)
{
QueryOperand *kptr = &item[k].operand;
QueryOperand *iptr = &item[i].operand;
if (k == i ||
(item[k].type == QI_VAL &&
compareQueryOperand(&kptr, &iptr, operand) == 0))
{
/* if k == i, we've already checked above that it's type == Q_VAL */
doc[cur].item[doc[cur].nitem] = item + k;
doc[cur].nitem++;
item[k].operand.istrue = 1;
}
}
}
else
{
doc[cur].nitem = doc[cur - 1].nitem;
doc[cur].item = doc[cur - 1].item;
}
doc[cur].pos = WEP_GETPOS(post[j]);
doc[cur].wclass = WEP_GETWEIGHT(post[j]);
cur++;
}
}
*doclen = cur;
if (cur > 0)
{
qsort((void *) doc, cur, sizeof(DocRepresentation), compareDocR);
return doc;
}
pfree(doc);
return NULL;
}
static float4
calc_rank_cd(float4 *arrdata, TSVector txt, TSQuery query, int method)
{
DocRepresentation *doc;
int len,
i,
doclen = 0;
Extention ext;
double Wdoc = 0.0;
double invws[lengthof(weights)];
double SumDist = 0.0,
PrevExtPos = 0.0,
CurExtPos = 0.0;
int NExtent = 0;
for (i = 0; i < lengthof(weights); i++)
{
invws[i] = ((double) ((arrdata[i] >= 0) ? arrdata[i] : weights[i]));
if (invws[i] > 1.0)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("weight out of range")));
invws[i] = 1.0 / invws[i];
}
doc = get_docrep(txt, query, &doclen);
if (!doc)
return 0.0;
MemSet(&ext, 0, sizeof(Extention));
while (Cover(doc, doclen, query, &ext))
{
double Cpos = 0.0;
double InvSum = 0.0;
int nNoise;
DocRepresentation *ptr = ext.begin;
while (ptr <= ext.end)
{
InvSum += invws[ptr->wclass];
ptr++;
}
Cpos = ((double) (ext.end - ext.begin + 1)) / InvSum;
/*
* if doc are big enough then ext.q may be equal to ext.p due to limit
* of posional information. In this case we approximate number of
* noise word as half cover's length
*/
nNoise = (ext.q - ext.p) - (ext.end - ext.begin);
if (nNoise < 0)
nNoise = (ext.end - ext.begin) / 2;
Wdoc += Cpos / ((double) (1 + nNoise));
CurExtPos = ((double) (ext.q + ext.p)) / 2.0;
if (NExtent > 0 && CurExtPos > PrevExtPos /* prevent devision by
* zero in a case of
multiple lexize */ )
SumDist += 1.0 / (CurExtPos - PrevExtPos);
PrevExtPos = CurExtPos;
NExtent++;
}
if ((method & RANK_NORM_LOGLENGTH) && txt->size > 0)
Wdoc /= log((double) (cnt_length(txt) + 1));
if (method & RANK_NORM_LENGTH)
{
len = cnt_length(txt);
if (len > 0)
Wdoc /= (double) len;
}
if ((method & RANK_NORM_EXTDIST) && SumDist > 0)
Wdoc /= ((double) NExtent) / SumDist;
if ((method & RANK_NORM_UNIQ) && txt->size > 0)
Wdoc /= (double) (txt->size);
if ((method & RANK_NORM_LOGUNIQ) && txt->size > 0)
Wdoc /= log((double) (txt->size + 1)) / log(2.0);
pfree(doc);
return (float4) Wdoc;
}
Datum
ts_rankcd_wttf(PG_FUNCTION_ARGS)
{
ArrayType *win = (ArrayType *) PG_DETOAST_DATUM(PG_GETARG_DATUM(0));
TSVector txt = PG_GETARG_TSVECTOR(1);
TSQuery query = PG_GETARG_TSQUERY_COPY(2); /* copy because we modify the istrue-flag */
int method = PG_GETARG_INT32(3);
float res;
res = calc_rank_cd(getWeights(win), txt, query, method);
PG_FREE_IF_COPY(win, 0);
PG_FREE_IF_COPY(txt, 1);
PG_FREE_IF_COPY(query, 2);
PG_RETURN_FLOAT4(res);
}
Datum
ts_rankcd_wtt(PG_FUNCTION_ARGS)
{
ArrayType *win = (ArrayType *) PG_DETOAST_DATUM(PG_GETARG_DATUM(0));
TSVector txt = PG_GETARG_TSVECTOR(1);
TSQuery query = PG_GETARG_TSQUERY_COPY(2); /* copy because we modify the istrue-flag */
float res;
res = calc_rank_cd(getWeights(win), txt, query, DEF_NORM_METHOD);
PG_FREE_IF_COPY(win, 0);
PG_FREE_IF_COPY(txt, 1);
PG_FREE_IF_COPY(query, 2);
PG_RETURN_FLOAT4(res);
}
Datum
ts_rankcd_ttf(PG_FUNCTION_ARGS)
{
TSVector txt = PG_GETARG_TSVECTOR(0);
TSQuery query = PG_GETARG_TSQUERY_COPY(1); /* copy because we modify the istrue-flag */
int method = PG_GETARG_INT32(2);
float res;
res = calc_rank_cd(getWeights(NULL), txt, query, method);
PG_FREE_IF_COPY(txt, 0);
PG_FREE_IF_COPY(query, 1);
PG_RETURN_FLOAT4(res);
}
Datum
ts_rankcd_tt(PG_FUNCTION_ARGS)
{
TSVector txt = PG_GETARG_TSVECTOR(0);
TSQuery query = PG_GETARG_TSQUERY_COPY(1); /* copy because we modify the istrue-flag */
float res;
res = calc_rank_cd(getWeights(NULL), txt, query, DEF_NORM_METHOD);
PG_FREE_IF_COPY(txt, 0);
PG_FREE_IF_COPY(query, 1);
PG_RETURN_FLOAT4(res);
}