Commit 9b9b659f authored by unknown's avatar unknown

Merge work:/home/bk/mysql-4.0

into serg.mysql.com:/usr/home/serg/Abk/mysql-4.0
parents 5e6d9795 4866fc3b
......@@ -48703,6 +48703,11 @@ Our TODO section contains what we plan to have in 4.0. @xref{TODO MySQL 4.0}.
@itemize @bullet
@item
Fixed bug with indexless boolean fulltext search.
@item
Fixed bug that sometimes appeared when fulltext search was used
with ``const'' tables.
@item
Use @code{ORDER BY column DESC} now sorts @code{NULL} values first.
@item
Fixed bug in @code{SELECT DISTINCT ... ORDER BY DESC} optimization.
......@@ -52,24 +52,28 @@ static double _nwghts[11]={
-3.796875000000000};
static double *nwghts=_nwghts+5; /* nwghts[i] = -0.5*1.5**i */
#define FTB_FLAG_TRUNC 1 /* MUST be 1 */
#define FTB_FLAG_YES 2 /* These two - YES and NO */
#define FTB_FLAG_NO 4 /* should NEVER be set both */
typedef struct st_ftb_expr FTB_EXPR;
struct st_ftb_expr {
FTB_EXPR *up;
float weight;
int yesno;
my_off_t docid;
uint flags;
my_off_t docid[2]; /* for index search and for scan */
float cur_weight;
int yesses; /* number of "yes" words matched */
int nos; /* number of "no" words matched */
int ythresh; /* number of "yes" words in expr */
int yweaks; /* number of "yes" words for scan only */
};
typedef struct {
typedef struct st_ftb_word {
FTB_EXPR *up;
float weight;
int yesno;
int trunc;
my_off_t docid;
uint flags;
my_off_t docid[2]; /* for index search and for scan */
uint ndepth;
int len;
/* ... there can be docid cache added here. SerG */
......@@ -78,20 +82,32 @@ typedef struct {
typedef struct st_ft_info {
struct _ft_vft *please;
MI_INFO *info;
MI_INFO *info;
uint keynr;
enum { UNINITIALIZED, READY, INDEX_SEARCH, INDEX_DONE, SCAN } state;
enum { UNINITIALIZED, READY, INDEX_SEARCH, INDEX_DONE /*, SCAN*/ } state;
uint with_scan;
FTB_EXPR *root;
QUEUE queue;
FTB_WORD **list;
MEM_ROOT mem_root;
} FTB;
int FTB_WORD_cmp(void *v __attribute__((unused)), byte *a, byte *b)
int FTB_WORD_cmp(void *v __attribute__((unused)), FTB_WORD *a, FTB_WORD *b)
{
/* ORDER BY docid, ndepth DESC */
int i=CMP_NUM(((FTB_WORD *)a)->docid, ((FTB_WORD *)b)->docid);
int i=CMP_NUM(a->docid[0], b->docid[0]);
if (!i)
i=CMP_NUM(((FTB_WORD *)b)->ndepth,((FTB_WORD *)a)->ndepth);
i=CMP_NUM(b->ndepth,a->ndepth);
return i;
}
int FTB_WORD_cmp_list(void *v __attribute__((unused)), FTB_WORD **a, FTB_WORD **b)
{
/* ORDER BY word DESC, ndepth DESC */
int i=_mi_compare_text(default_charset_info, (*b)->word+1,(*b)->len-1,
(*a)->word+1,(*a)->len-1,0);
if (!i)
i=CMP_NUM((*b)->ndepth,(*a)->ndepth);
return i;
}
......@@ -120,25 +136,30 @@ void _ftb_parse_query(FTB *ftb, byte **start, byte *end,
(param.trunc ? MI_MAX_KEY_BUFF :
w.len+extra));
ftbw->len=w.len+1;
ftbw->yesno=param.yesno;
ftbw->trunc=param.trunc; /* 0 or 1 */
ftbw->flags=0;
if (param.yesno>0) ftbw->flags|=FTB_FLAG_YES;
if (param.yesno<0) ftbw->flags|=FTB_FLAG_NO;
if (param.trunc) ftbw->flags|=FTB_FLAG_TRUNC;
ftbw->weight=weight;
ftbw->up=up;
ftbw->docid=HA_POS_ERROR;
ftbw->docid[0]=ftbw->docid[1]=HA_POS_ERROR;
ftbw->ndepth= (param.yesno<0) + depth;
memcpy(ftbw->word+1, w.pos, w.len);
ftbw->word[0]=w.len;
if (ftbw->yesno > 0) up->ythresh++;
if (param.yesno > 0) up->ythresh++;
queue_insert(& ftb->queue, (byte *)ftbw);
ftb->with_scan|=param.trunc;
break;
case 2: /* left bracket */
ftbe=(FTB_EXPR *)alloc_root(&ftb->mem_root, sizeof(FTB_EXPR));
ftbe->yesno=param.yesno;
ftbe->flags=0;
if (param.yesno>0) ftbe->flags|=FTB_FLAG_YES;
if (param.yesno<0) ftbe->flags|=FTB_FLAG_NO;
ftbe->weight=weight;
ftbe->up=up;
ftbe->ythresh=0;
ftbe->docid=HA_POS_ERROR;
if (ftbe->yesno > 0) up->ythresh++;
ftbe->ythresh=ftbe->yweaks=0;
ftbe->docid[0]=ftbe->docid[1]=HA_POS_ERROR;
if (param.yesno > 0) up->ythresh++;
_ftb_parse_query(ftb, start, end, ftbe, depth+1);
break;
case 3: /* right bracket */
......@@ -151,7 +172,7 @@ void _ftb_parse_query(FTB *ftb, byte **start, byte *end,
void _ftb_init_index_search(FT_INFO *ftb)
{
int i, r;
FTB_WORD *ftbw;
FTB_WORD *ftbw;
MI_INFO *info=ftb->info;
MI_KEYDEF *keyinfo;
my_off_t keyroot;
......@@ -167,18 +188,31 @@ void _ftb_init_index_search(FT_INFO *ftb)
{
ftbw=(FTB_WORD *)(ftb->queue.root[i]);
if (ftbw->flags&FTB_FLAG_TRUNC &&
(ftbw->up->ythresh > test(ftbw->flags&FTB_FLAG_YES)))
{
/* no need to search for this prefix in the index -
* it cannot ADD new matches, and to REMOVE half-matched
* rows we do scan anyway */
ftbw->docid[0]=HA_POS_ERROR;
ftbw->up->yweaks++;
continue;
}
r=_mi_search(info, keyinfo, (uchar*) ftbw->word, ftbw->len,
SEARCH_FIND | SEARCH_BIGGER, keyroot);
SEARCH_FIND | SEARCH_BIGGER, keyroot);
if (!r)
{
r=_mi_compare_text(default_charset_info,
info->lastkey+ftbw->trunc,ftbw->len-ftbw->trunc,
(uchar*) ftbw->word+ftbw->trunc,ftbw->len-ftbw->trunc,
info->lastkey + (ftbw->flags&FTB_FLAG_TRUNC),
ftbw->len - (ftbw->flags&FTB_FLAG_TRUNC),
ftbw->word + (ftbw->flags&FTB_FLAG_TRUNC),
ftbw->len - (ftbw->flags&FTB_FLAG_TRUNC),
0);
}
if (r) /* not found */
{
if (ftbw->yesno>0 && ftbw->up->up==0)
if (ftbw->flags&FTB_FLAG_YES && ftbw->up->up==0)
{ /* this word MUST BE present in every document returned,
so we can abort the search right now */
ftb->state=INDEX_DONE;
......@@ -188,7 +222,7 @@ void _ftb_init_index_search(FT_INFO *ftb)
else
{
memcpy(ftbw->word, info->lastkey, info->lastkey_length);
ftbw->docid=info->lastpos;
ftbw->docid[0]=info->lastpos;
}
}
queue_fix(& ftb->queue);
......@@ -207,6 +241,7 @@ FT_INFO * ft_init_boolean_search(MI_INFO *info, uint keynr, byte *query,
ftb->state=UNINITIALIZED;
ftb->info=info;
ftb->keynr=keynr;
ftb->with_scan=0;
init_alloc_root(&ftb->mem_root, 1024, 1024);
......@@ -215,65 +250,75 @@ FT_INFO * ft_init_boolean_search(MI_INFO *info, uint keynr, byte *query,
*/
res=ftb->queue.max_elements=query_len/(ft_min_word_len+1);
ftb->queue.root=(byte **)alloc_root(&ftb->mem_root, (res+1)*sizeof(void*));
reinit_queue(& ftb->queue, res, 0, 0, FTB_WORD_cmp, ftb);
reinit_queue(& ftb->queue, res, 0, 0,
(int (*)(void*,byte*,byte*))FTB_WORD_cmp, ftb);
ftbe=(FTB_EXPR *)alloc_root(&ftb->mem_root, sizeof(FTB_EXPR));
ftbe->weight=1;
ftbe->yesno=ftbe->nos=1;
ftbe->flags=FTB_FLAG_YES;
ftbe->nos=1;
ftbe->up=0;
ftbe->ythresh=0;
ftbe->docid=HA_POS_ERROR;
ftbe->ythresh=ftbe->yweaks=0;
ftbe->docid[0]=ftbe->docid[1]=HA_POS_ERROR;
ftb->root=ftbe;
_ftb_parse_query(ftb, &query, query+query_len, ftbe, 0);
ftb->list=(FTB_WORD **)alloc_root(&ftb->mem_root,
sizeof(FTB_WORD *)*ftb->queue.elements);
memcpy(ftb->list, ftb->queue.root, sizeof(FTB_WORD *)*ftb->queue.elements);
qsort2(ftb->list, ftb->queue.elements, sizeof(FTB_WORD *),
(qsort2_cmp)FTB_WORD_cmp_list, 0);
if (ftb->queue.elements<2) ftb->with_scan=0;
ftb->state=READY;
return ftb;
}
void _ftb_climb_the_tree(FTB_WORD *ftbw)
void _ftb_climb_the_tree(FTB_WORD *ftbw, uint mode)
{
FTB_EXPR *ftbe;
float weight=ftbw->weight;
int yn=ftbw->yesno;
my_off_t curdoc=ftbw->docid;
int yn=ftbw->flags, ythresh;
my_off_t curdoc=ftbw->docid[mode];
for (ftbe=ftbw->up; ftbe; ftbe=ftbe->up)
{
if (ftbe->docid != curdoc)
ythresh = ftbe->ythresh - (mode ? 0 : ftbe->yweaks);
if (ftbe->docid[mode] != curdoc)
{
ftbe->cur_weight=0;
ftbe->yesses=ftbe->nos=0;
ftbe->docid=curdoc;
ftbe->docid[mode]=curdoc;
}
if (ftbe->nos)
break;
if (yn>0)
if (yn & FTB_FLAG_YES)
{
ftbe->cur_weight+=weight;
if (++ftbe->yesses == ftbe->ythresh)
if (++ftbe->yesses == ythresh)
{
yn=ftbe->yesno;
yn=ftbe->flags;
weight=ftbe->cur_weight*ftbe->weight;
}
else
break;
}
else
if (yn<0)
if (yn & FTB_FLAG_NO)
{
/* NOTE: special sort function of queue assures that all yn<0
/* NOTE: special sort function of queue assures that all
* (yn & FTB_FLAG_NO) != 0
* events for every particular subexpression will
* "auto-magically" happen BEFORE all the yn>=0 events. So no
* "auto-magically" happen BEFORE all the
* (yn & FTB_FLAG_YES) != 0 events. So no
* already matched expression can become not-matched again.
*/
++ftbe->nos;
break;
}
else
/* if (yn==0) */
{
ftbe->cur_weight+=weight;
if (ftbe->yesses < ftbe->ythresh)
if (ftbe->yesses < ythresh)
break;
yn= (ftbe->yesses++ == ftbe->ythresh) * ftbe->yesno;
yn= (ftbe->yesses++ == ythresh) ? ftbe->flags : 0 ;
weight*=ftbe->weight;
}
}
......@@ -303,11 +348,11 @@ int ft_boolean_read_next(FT_INFO *ftb, char *record)
return my_errno=HA_ERR_END_OF_FILE;
while(ftb->state == INDEX_SEARCH &&
(curdoc=((FTB_WORD *)queue_top(& ftb->queue))->docid) != HA_POS_ERROR)
(curdoc=((FTB_WORD *)queue_top(& ftb->queue))->docid[0]) != HA_POS_ERROR)
{
while (curdoc==(ftbw=(FTB_WORD *)queue_top(& ftb->queue))->docid)
while (curdoc==(ftbw=(FTB_WORD *)queue_top(& ftb->queue))->docid[0])
{
_ftb_climb_the_tree(ftbw);
_ftb_climb_the_tree(ftbw,0);
/* update queue */
r=_mi_search(info, keyinfo, (uchar*) ftbw->word, USE_WHOLE_KEY,
......@@ -315,14 +360,16 @@ int ft_boolean_read_next(FT_INFO *ftb, char *record)
if (!r)
{
r=_mi_compare_text(default_charset_info,
info->lastkey+ftbw->trunc,ftbw->len-ftbw->trunc,
(uchar*) ftbw->word+ftbw->trunc,
ftbw->len-ftbw->trunc,0);
info->lastkey + (ftbw->flags&FTB_FLAG_TRUNC),
ftbw->len - (ftbw->flags&FTB_FLAG_TRUNC),
ftbw->word + (ftbw->flags&FTB_FLAG_TRUNC),
ftbw->len - (ftbw->flags&FTB_FLAG_TRUNC),
0);
}
if (r) /* not found */
{
ftbw->docid=HA_POS_ERROR;
if (ftbw->yesno>0 && ftbw->up->up==0)
ftbw->docid[0]=HA_POS_ERROR;
if (ftbw->flags&FTB_FLAG_YES && ftbw->up->up==0)
{ /* this word MUST BE present in every document returned,
so we can stop the search right now */
ftb->state=INDEX_DONE;
......@@ -331,22 +378,24 @@ int ft_boolean_read_next(FT_INFO *ftb, char *record)
else
{
memcpy(ftbw->word, info->lastkey, info->lastkey_length);
ftbw->docid=info->lastpos;
ftbw->docid[0]=info->lastpos;
}
queue_replaced(& ftb->queue);
}
ftbe=ftb->root;
if (ftbe->docid==curdoc && ftbe->cur_weight>0 &&
ftbe->yesses>=ftbe->ythresh && !ftbe->nos)
if (ftbe->docid[0]==curdoc && ftbe->cur_weight>0 &&
ftbe->yesses>=(ftbe->ythresh-ftbe->yweaks) && !ftbe->nos)
{
/* curdoc matched ! */
info->lastpos=curdoc;
info->update&= (HA_STATE_CHANGED | HA_STATE_ROW_CHANGED); /* why is this ? */
info->lastpos=curdoc;
if (!(*info->read_record)(info,curdoc,record))
{
info->update|= HA_STATE_AKTIV; /* Record is read */
if (ftb->with_scan && ft_boolean_find_relevance(ftb,record,0)==0)
continue; /* no match */
return 0;
}
return my_errno;
......@@ -370,30 +419,13 @@ float ft_boolean_find_relevance(FT_INFO *ftb, byte *record, uint length)
return -2.0;
if (!ftb->queue.elements)
return 0;
if (ftb->state == READY || ftb->state == INDEX_DONE)
{
for (i=1; i<=ftb->queue.elements; i++)
{
ftbw=(FTB_WORD *)(ftb->queue.root[i]);
ftbw->docid=HA_POS_ERROR;
for (ftbe=ftbw->up; ftbe; ftbe=ftbe->up)
{
if (ftbe->docid != HA_POS_ERROR)
{
ftbe->cur_weight=0;
ftbe->yesses=ftbe->nos=0;
ftbe->docid=HA_POS_ERROR;
}
else
break;
}
}
queue_fix(& ftb->queue);
#if 0
if (ftb->state == READY || ftb->state == INDEX_DONE)
ftb->state=SCAN;
}
else if (ftb->state != SCAN)
return -3.0;
#endif
if (ftb->keynr==NO_SUCH_KEY)
_mi_ft_segiterator_dummy_init(record, length, &ftsi);
......@@ -408,32 +440,34 @@ float ft_boolean_find_relevance(FT_INFO *ftb, byte *record, uint length)
end=ftsi.pos+ftsi.len;
while (ft_simple_get_word((byte **)&ftsi.pos,(byte *)end,&word))
{
uint a, b, c;
for (a=1, b=ftb->queue.elements+1, c=(a+b)/2; b-a>1; c=(a+b)/2)
int a, b, c;
for (a=0, b=ftb->queue.elements, c=(a+b)/2; b-a>1; c=(a+b)/2)
{
ftbw=(FTB_WORD *)(ftb->queue.root[c]);
ftbw=(FTB_WORD *)(ftb->list[c]);
if (_mi_compare_text(default_charset_info, word.pos,word.len,
(uchar*) ftbw->word+1,ftbw->len-1,ftbw->trunc) >0)
(uchar*) ftbw->word+1,ftbw->len-1,
(ftbw->flags&FTB_FLAG_TRUNC) ) >0)
b=c;
else
a=c;
}
for (; c; c--)
for (; c>=0; c--)
{
ftbw=(FTB_WORD *)(ftb->queue.root[c]);
ftbw=(FTB_WORD *)(ftb->list[c]);
if (_mi_compare_text(default_charset_info, word.pos,word.len,
(uchar*) ftbw->word+1,ftbw->len-1,ftbw->trunc))
(uchar*) ftbw->word+1,ftbw->len-1,
(ftbw->flags&FTB_FLAG_TRUNC) ))
break;
if (ftbw->docid == docid)
if (ftbw->docid[1] == docid)
continue;
ftbw->docid=docid;
_ftb_climb_the_tree(ftbw);
ftbw->docid[1]=docid;
_ftb_climb_the_tree(ftbw,1);
}
}
}
ftbe=ftb->root;
if (ftbe->docid==docid && ftbe->cur_weight>0 &&
if (ftbe->docid[1]==docid && ftbe->cur_weight>0 &&
ftbe->yesses>=ftbe->ythresh && !ftbe->nos)
{ /* row matched ! */
return ftbe->cur_weight;
......
......@@ -48,7 +48,14 @@ Full-text search in MySQL implements vector space model
select * from t1 where MATCH(a,b) AGAINST("+search -(support vector)" IN BOOLEAN MODE);
a b
Function MATCH ... AGAINST() is used to do a search
select *, MATCH(a,b) AGAINST("support collections" IN BOOLEAN MODE) as x from t1;
select *, MATCH(a,b) AGAINST("support collections" IN BOOLEAN MODE) as x from t1;
a b x
MySQL has now support for full-text search 1
Full-text indexes are called collections 1
Only MyISAM tables support collections 2
Function MATCH ... AGAINST() is used to do a search 0
Full-text search in MySQL implements vector space model 0
select *, MATCH(a,b) AGAINST("collections support" IN BOOLEAN MODE) as x from t1;
a b x
MySQL has now support for full-text search 1
Full-text indexes are called collections 1
......
......@@ -27,7 +27,8 @@ select * from t1 where MATCH(a,b) AGAINST("+support +collections" IN BOOLEAN MOD
select * from t1 where MATCH(a,b) AGAINST("+search" IN BOOLEAN MODE);
select * from t1 where MATCH(a,b) AGAINST("+search +(support vector)" IN BOOLEAN MODE);
select * from t1 where MATCH(a,b) AGAINST("+search -(support vector)" IN BOOLEAN MODE);
select *, MATCH(a,b) AGAINST("support collections" IN BOOLEAN MODE) as x from t1;
select *, MATCH(a,b) AGAINST("support collections" IN BOOLEAN MODE) as x from t1;
select *, MATCH(a,b) AGAINST("collections support" IN BOOLEAN MODE) as x from t1;
# boolean w/o index:
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment