mariadb/storage/myisam/ft_boolean_search.c

966 lines
28 KiB
C
Raw Normal View History

/* Copyright (C) 2001-2005 MySQL AB
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; version 2 of the License.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */
/* Written by Sergei A. Golubchik, who has a shared copyright to this code */
/* TODO: add caching - pre-read several index entries at once */
/*
Added optimization for full-text queries with plus-words. It was
implemented by sharing maximal document id (max_docid) variable
inside plus subtree. max_docid could be used by any word in plus
subtree, but it could be updated by plus-word only.
The idea is: there is no need to search for docid smaller than
biggest docid inside current plus subtree.
Examples:
+word1 word2
share same max_docid
max_docid updated by word1
+word1 +(word2 word3)
share same max_docid
max_docid updated by word1
+(word1 -word2) +(+word3 word4)
share same max_docid
max_docid updated by word3
*/
2001-10-09 14:53:54 +02:00
#define FT_CORE
#include "ftdefs.h"
/* search with boolean queries */
static double _wghts[11]=
{
0.131687242798354,
0.197530864197531,
0.296296296296296,
0.444444444444444,
0.666666666666667,
1.000000000000000,
1.500000000000000,
2.250000000000000,
3.375000000000000,
5.062500000000000,
7.593750000000000};
2001-11-22 19:35:35 +01:00
static double *wghts=_wghts+5; /* wghts[i] = 1.5**i */
static double _nwghts[11]=
{
-0.065843621399177,
-0.098765432098766,
-0.148148148148148,
-0.222222222222222,
-0.333333333333334,
-0.500000000000000,
-0.750000000000000,
-1.125000000000000,
-1.687500000000000,
-2.531250000000000,
-3.796875000000000};
2001-11-22 19:35:35 +01:00
static double *nwghts=_nwghts+5; /* nwghts[i] = -0.5*1.5**i */
#define FTB_FLAG_TRUNC 1
/* At most one of the following flags can be set */
#define FTB_FLAG_YES 2
#define FTB_FLAG_NO 4
#define FTB_FLAG_WONLY 8
typedef struct st_ftb_expr FTB_EXPR;
struct st_ftb_expr
{
FTB_EXPR *up;
2004-02-15 20:36:12 +01:00
uint flags;
2003-11-20 23:15:07 +01:00
/* ^^^^^^^^^^^^^^^^^^ FTB_{EXPR,WORD} common section */
2004-02-15 20:36:12 +01:00
my_off_t docid[2];
my_off_t max_docid;
float weight;
float cur_weight;
LIST *phrase; /* phrase words */
LIST *document; /* for phrase search */
uint yesses; /* number of "yes" words matched */
uint nos; /* number of "no" words matched */
uint ythresh; /* number of "yes" words in expr */
uint yweaks; /* number of "yes" words for scan only */
};
typedef struct st_ftb_word
{
FTB_EXPR *up;
2004-02-15 20:36:12 +01:00
uint flags;
2003-11-20 23:15:07 +01:00
/* ^^^^^^^^^^^^^^^^^^ FTB_{EXPR,WORD} common section */
2004-02-15 20:36:12 +01:00
my_off_t docid[2]; /* for index search and for scan */
my_off_t key_root;
my_off_t *max_docid;
2004-02-15 20:36:12 +01:00
MI_KEYDEF *keyinfo;
struct st_ftb_word *prev;
float weight;
uint ndepth;
uint len;
uchar off;
byte word[1];
} FTB_WORD;
typedef struct st_ft_info
{
2001-10-09 14:53:54 +02:00
struct _ft_vft *please;
MI_INFO *info;
2002-04-15 22:32:40 +02:00
CHARSET_INFO *charset;
FTB_EXPR *root;
FTB_WORD **list;
FTB_WORD *last_word;
MEM_ROOT mem_root;
QUEUE queue;
TREE no_dupes;
my_off_t lastpos;
uint keynr;
uchar with_scan;
enum { UNINITIALIZED, READY, INDEX_SEARCH, INDEX_DONE } state;
} FTB;
static int FTB_WORD_cmp(my_off_t *v, FTB_WORD *a, FTB_WORD *b)
{
int i;
/* if a==curdoc, take it as a < b */
if (v && a->docid[0] == *v)
return -1;
/* ORDER BY docid, ndepth DESC */
i=CMP_NUM(a->docid[0], b->docid[0]);
if (!i)
2002-03-04 14:17:49 +01:00
i=CMP_NUM(b->ndepth,a->ndepth);
return i;
}
static int FTB_WORD_cmp_list(CHARSET_INFO *cs, FTB_WORD **a, FTB_WORD **b)
{
/* ORDER BY word DESC, ndepth DESC */
2002-11-21 14:56:48 +01:00
int i= mi_compare_text(cs, (uchar*) (*b)->word+1,(*b)->len-1,
(uchar*) (*a)->word+1,(*a)->len-1,0,0);
2002-03-04 14:17:49 +01:00
if (!i)
i=CMP_NUM((*b)->ndepth,(*a)->ndepth);
2001-10-09 14:53:54 +02:00
return i;
}
typedef struct st_my_ftb_param
{
FTB *ftb;
FTB_EXPR *ftbe;
byte *up_quot;
uint depth;
} MY_FTB_PARAM;
static int ftb_query_add_word(MYSQL_FTPARSER_PARAM *param,
char *word, int word_len,
MYSQL_FTPARSER_BOOLEAN_INFO *info)
{
MY_FTB_PARAM *ftb_param= param->mysql_ftparam;
FTB_WORD *ftbw;
FTB_EXPR *ftbe, *tmp_expr;
FT_WORD *phrase_word;
LIST *tmp_element;
int r= info->weight_adjust;
float weight= (float)
(info->wasign ? nwghts : wghts)[(r>5)?5:((r<-5)?-5:r)];
switch (info->type) {
case FT_TOKEN_WORD:
ftbw= (FTB_WORD *)alloc_root(&ftb_param->ftb->mem_root,
sizeof(FTB_WORD) +
(info->trunc ? MI_MAX_KEY_BUFF :
word_len * ftb_param->ftb->charset->mbmaxlen +
HA_FT_WLEN +
ftb_param->ftb->info->s->rec_reflength));
ftbw->len= word_len + 1;
ftbw->flags= 0;
ftbw->off= 0;
if (info->yesno > 0) ftbw->flags|= FTB_FLAG_YES;
if (info->yesno < 0) ftbw->flags|= FTB_FLAG_NO;
if (info->trunc) ftbw->flags|= FTB_FLAG_TRUNC;
ftbw->weight= weight;
ftbw->up= ftb_param->ftbe;
ftbw->docid[0]= ftbw->docid[1]= HA_OFFSET_ERROR;
ftbw->ndepth= (info->yesno < 0) + ftb_param->depth;
ftbw->key_root= HA_OFFSET_ERROR;
memcpy(ftbw->word + 1, word, word_len);
ftbw->word[0]= word_len;
if (info->yesno > 0) ftbw->up->ythresh++;
ftb_param->ftb->queue.max_elements++;
ftbw->prev= ftb_param->ftb->last_word;
ftb_param->ftb->last_word= ftbw;
ftb_param->ftb->with_scan|= (info->trunc & FTB_FLAG_TRUNC);
for (tmp_expr= ftb_param->ftbe; tmp_expr->up; tmp_expr= tmp_expr->up)
if (! (tmp_expr->flags & FTB_FLAG_YES))
break;
ftbw->max_docid= &tmp_expr->max_docid;
/* fall through */
case FT_TOKEN_STOPWORD:
if (! ftb_param->up_quot) break;
phrase_word= (FT_WORD *)alloc_root(&ftb_param->ftb->mem_root, sizeof(FT_WORD));
tmp_element= (LIST *)alloc_root(&ftb_param->ftb->mem_root, sizeof(LIST));
phrase_word->pos= word;
phrase_word->len= word_len;
tmp_element->data= (void *)phrase_word;
ftb_param->ftbe->phrase= list_add(ftb_param->ftbe->phrase, tmp_element);
/* Allocate document list at this point.
It allows to avoid huge amount of allocs/frees for each row.*/
tmp_element= (LIST *)alloc_root(&ftb_param->ftb->mem_root, sizeof(LIST));
tmp_element->data= alloc_root(&ftb_param->ftb->mem_root, sizeof(FT_WORD));
ftb_param->ftbe->document=
list_add(ftb_param->ftbe->document, tmp_element);
break;
case FT_TOKEN_LEFT_PAREN:
ftbe=(FTB_EXPR *)alloc_root(&ftb_param->ftb->mem_root, sizeof(FTB_EXPR));
ftbe->flags= 0;
if (info->yesno > 0) ftbe->flags|= FTB_FLAG_YES;
if (info->yesno < 0) ftbe->flags|= FTB_FLAG_NO;
ftbe->weight= weight;
ftbe->up= ftb_param->ftbe;
ftbe->max_docid= ftbe->ythresh= ftbe->yweaks= 0;
ftbe->docid[0]= ftbe->docid[1]= HA_OFFSET_ERROR;
ftbe->phrase= NULL;
ftbe->document= 0;
if (info->quot) ftb_param->ftb->with_scan|= 2;
if (info->yesno > 0) ftbe->up->ythresh++;
ftb_param->ftbe= ftbe;
ftb_param->depth++;
ftb_param->up_quot= info->quot;
break;
case FT_TOKEN_RIGHT_PAREN:
if (ftb_param->ftbe->document)
{
/* Circuit document list */
for (tmp_element= ftb_param->ftbe->document;
tmp_element->next; tmp_element= tmp_element->next) /* no-op */;
tmp_element->next= ftb_param->ftbe->document;
ftb_param->ftbe->document->prev= tmp_element;
}
info->quot= 0;
if (ftb_param->ftbe->up)
{
DBUG_ASSERT(ftb_param->depth);
ftb_param->ftbe= ftb_param->ftbe->up;
ftb_param->depth--;
ftb_param->up_quot= 0;
}
break;
case FT_TOKEN_EOF:
default:
break;
}
return(0);
}
static int ftb_parse_query_internal(MYSQL_FTPARSER_PARAM *param,
char *query, int len)
{
MY_FTB_PARAM *ftb_param= param->mysql_ftparam;
MYSQL_FTPARSER_BOOLEAN_INFO info;
CHARSET_INFO *cs= ftb_param->ftb->charset;
2006-01-16 15:17:45 +01:00
char **start= &query;
char *end= query + len;
FT_WORD w;
info.prev= ' ';
info.quot= 0;
while (ft_get_word(cs, start, end, &w, &info))
param->mysql_add_word(param, w.pos, w.len, &info);
return(0);
}
static void _ftb_parse_query(FTB *ftb, byte *query, uint len,
struct st_mysql_ftparser *parser)
{
MYSQL_FTPARSER_PARAM *param;
MY_FTB_PARAM ftb_param;
DBUG_ENTER("_ftb_parse_query");
DBUG_ASSERT(parser);
if (ftb->state != UNINITIALIZED)
DBUG_VOID_RETURN;
if (! (param= ftparser_call_initializer(ftb->info, ftb->keynr, 0)))
DBUG_VOID_RETURN;
ftb_param.ftb= ftb;
ftb_param.depth= 0;
ftb_param.ftbe= ftb->root;
ftb_param.up_quot= 0;
param->mysql_parse= ftb_parse_query_internal;
param->mysql_add_word= ftb_query_add_word;
param->mysql_ftparam= (void *)&ftb_param;
param->cs= ftb->charset;
param->doc= query;
param->length= len;
param->flags= 0;
param->mode= MYSQL_FTPARSER_FULL_BOOLEAN_INFO;
parser->parse(param);
DBUG_VOID_RETURN;
}
static int _ftb_no_dupes_cmp(void* not_used __attribute__((unused)),
2004-02-15 20:36:12 +01:00
const void *a,const void *b)
2002-05-07 20:27:39 +02:00
{
return CMP_NUM((*((my_off_t*)a)), (*((my_off_t*)b)));
}
/* returns 1 if the search was finished (must-word wasn't found) */
static int _ft2_search(FTB *ftb, FTB_WORD *ftbw, my_bool init_search)
{
int r;
2004-04-29 12:25:56 +02:00
int subkeys=1;
my_bool can_go_down;
MI_INFO *info=ftb->info;
uint off, extra=HA_FT_WLEN+info->s->base.rec_reflength;
byte *lastkey_buf=ftbw->word+ftbw->off;
2004-12-30 23:44:00 +01:00
LINT_INIT(off);
if (ftbw->flags & FTB_FLAG_TRUNC)
lastkey_buf+=ftbw->len;
if (init_search)
{
ftbw->key_root=info->s->state.key_root[ftb->keynr];
ftbw->keyinfo=info->s->keyinfo+ftb->keynr;
r=_mi_search(info, ftbw->keyinfo, (uchar*) ftbw->word, ftbw->len,
SEARCH_FIND | SEARCH_BIGGER, ftbw->key_root);
}
else
{
uint sflag= SEARCH_BIGGER;
if (ftbw->docid[0] < *ftbw->max_docid)
{
sflag|= SEARCH_SAME;
_mi_dpointer(info, (uchar *)(ftbw->word + ftbw->len + HA_FT_WLEN),
*ftbw->max_docid);
}
r=_mi_search(info, ftbw->keyinfo, (uchar*) lastkey_buf,
USE_WHOLE_KEY, sflag, ftbw->key_root);
}
2004-04-26 14:53:31 +02:00
can_go_down=(!ftbw->off && (init_search || (ftbw->flags & FTB_FLAG_TRUNC)));
2004-04-26 14:53:31 +02:00
/* Skip rows inserted by concurrent insert */
while (!r)
{
if (can_go_down)
{
/* going down ? */
off=info->lastkey_length-extra;
subkeys=ft_sintXkorr(info->lastkey+off);
}
if (subkeys<0 || info->lastpos < info->state->data_file_length)
break;
r= _mi_search_next(info, ftbw->keyinfo, info->lastkey,
info->lastkey_length,
SEARCH_BIGGER, ftbw->key_root);
}
2004-04-26 14:53:31 +02:00
if (!r && !ftbw->off)
{
r= mi_compare_text(ftb->charset,
info->lastkey+1,
info->lastkey_length-extra-1,
(uchar*) ftbw->word+1,
ftbw->len-1,
(my_bool) (ftbw->flags & FTB_FLAG_TRUNC),0);
}
if (r) /* not found */
{
if (!ftbw->off || !(ftbw->flags & FTB_FLAG_TRUNC))
{
2004-11-17 15:28:37 +01:00
ftbw->docid[0]=HA_OFFSET_ERROR;
if ((ftbw->flags & FTB_FLAG_YES) && ftbw->up->up==0)
{
/*
2004-02-15 20:36:12 +01:00
This word MUST BE present in every document returned,
so we can stop the search right now
*/
ftb->state=INDEX_DONE;
return 1; /* search is done */
}
else
return 0;
}
/* going up to the first-level tree to continue search there */
_mi_dpointer(info, (uchar*) (lastkey_buf+HA_FT_WLEN), ftbw->key_root);
ftbw->key_root=info->s->state.key_root[ftb->keynr];
ftbw->keyinfo=info->s->keyinfo+ftb->keynr;
ftbw->off=0;
return _ft2_search(ftb, ftbw, 0);
}
/* matching key found */
memcpy(lastkey_buf, info->lastkey, info->lastkey_length);
if (lastkey_buf == ftbw->word)
ftbw->len=info->lastkey_length-extra;
2004-04-29 12:25:56 +02:00
/* going down ? */
if (subkeys<0)
{
2004-04-29 12:25:56 +02:00
/*
yep, going down, to the second-level tree
TODO here: subkey-based optimization
*/
ftbw->off=off;
ftbw->key_root=info->lastpos;
ftbw->keyinfo=& info->s->ft2_keyinfo;
r=_mi_search_first(info, ftbw->keyinfo, ftbw->key_root);
DBUG_ASSERT(r==0); /* found something */
memcpy(lastkey_buf+off, info->lastkey, info->lastkey_length);
}
ftbw->docid[0]=info->lastpos;
if (ftbw->flags & FTB_FLAG_YES)
*ftbw->max_docid= info->lastpos;
return 0;
}
static void _ftb_init_index_search(FT_INFO *ftb)
{
int i;
FTB_WORD *ftbw;
if ((ftb->state != READY && ftb->state !=INDEX_DONE) ||
ftb->keynr == NO_SUCH_KEY)
return;
ftb->state=INDEX_SEARCH;
for (i=ftb->queue.elements; i; i--)
{
ftbw=(FTB_WORD *)(ftb->queue.root[i]);
if (ftbw->flags & FTB_FLAG_TRUNC)
{
/*
2004-02-15 20:36:12 +01:00
special treatment for truncation operator
2003-11-20 23:15:07 +01:00
1. there are some (besides this) +words
2002-05-15 20:31:24 +02:00
| no need to search in the index, it can never ADD new rows
| to the result, and to remove half-matched rows we do scan anyway
2. -trunc*
2002-05-15 20:31:24 +02:00
| same as 1.
2003-11-20 23:15:07 +01:00
3. in 1 and 2, +/- need not be on the same expr. level,
but can be on any upper level, as in +word +(trunc1* trunc2*)
4. otherwise
2002-05-15 20:31:24 +02:00
| We have to index-search for this prefix.
| It may cause duplicates, as in the index (sorted by <word,docid>)
| <aaaa,row1>
| <aabb,row2>
| <aacc,row1>
| Searching for "aa*" will find row1 twice...
*/
2003-11-20 23:15:07 +01:00
FTB_EXPR *ftbe;
for (ftbe=(FTB_EXPR*)ftbw;
ftbe->up && !(ftbe->up->flags & FTB_FLAG_TRUNC);
ftbe->up->flags|= FTB_FLAG_TRUNC, ftbe=ftbe->up)
2002-05-07 20:27:39 +02:00
{
2003-11-20 23:15:07 +01:00
if (ftbe->flags & FTB_FLAG_NO || /* 2 */
ftbe->up->ythresh - ftbe->up->yweaks >1) /* 1 */
{
FTB_EXPR *top_ftbe=ftbe->up;
ftbw->docid[0]=HA_OFFSET_ERROR;
for (ftbe=(FTB_EXPR *)ftbw;
ftbe != top_ftbe && !(ftbe->flags & FTB_FLAG_NO);
ftbe=ftbe->up)
ftbe->up->yweaks++;
2003-11-20 23:15:07 +01:00
ftbe=0;
break;
}
2002-05-07 20:27:39 +02:00
}
2003-11-20 23:15:07 +01:00
if (!ftbe)
continue;
/* 4 */
2003-11-20 23:15:07 +01:00
if (!is_tree_inited(& ftb->no_dupes))
init_tree(& ftb->no_dupes,0,0,sizeof(my_off_t),
_ftb_no_dupes_cmp,0,0,0);
else
reset_tree(& ftb->no_dupes);
}
ftbw->off=0; /* in case of reinit */
if (_ft2_search(ftb, ftbw, 1))
return;
}
queue_fix(& ftb->queue);
}
2001-10-09 14:53:54 +02:00
FT_INFO * ft_init_boolean_search(MI_INFO *info, uint keynr, byte *query,
uint query_len, CHARSET_INFO *cs)
{
FTB *ftb;
FTB_EXPR *ftbe;
FTB_WORD *ftbw;
if (!(ftb=(FTB *)my_malloc(sizeof(FTB), MYF(MY_WME))))
return 0;
ftb->please= (struct _ft_vft *) & _ft_vft_boolean;
ftb->state=UNINITIALIZED;
ftb->info=info;
ftb->keynr=keynr;
ftb->charset=cs;
DBUG_ASSERT(keynr==NO_SUCH_KEY || cs == info->s->keyinfo[keynr].seg->charset);
2002-03-04 14:17:49 +01:00
ftb->with_scan=0;
ftb->lastpos=HA_OFFSET_ERROR;
2002-05-07 20:27:39 +02:00
bzero(& ftb->no_dupes, sizeof(TREE));
ftb->last_word= 0;
2001-10-09 14:53:54 +02:00
init_alloc_root(&ftb->mem_root, 1024, 1024);
ftb->queue.max_elements= 0;
2003-12-09 20:49:48 +01:00
if (!(ftbe=(FTB_EXPR *)alloc_root(&ftb->mem_root, sizeof(FTB_EXPR))))
goto err;
ftbe->weight=1;
ftbe->flags=FTB_FLAG_YES;
ftbe->nos=1;
2002-04-26 07:56:28 +02:00
ftbe->up=0;
ftbe->max_docid= ftbe->ythresh= ftbe->yweaks= 0;
ftbe->docid[0]=ftbe->docid[1]=HA_OFFSET_ERROR;
2005-03-06 21:17:57 +01:00
ftbe->phrase= NULL;
ftbe->document= 0;
ftb->root=ftbe;
_ftb_parse_query(ftb, query, query_len, keynr == NO_SUCH_KEY ?
&ft_default_parser :
info->s->keyinfo[keynr].parser);
/*
Hack: instead of init_queue, we'll use reinit queue to be able
to alloc queue with alloc_root()
*/
if (! (ftb->queue.root= (byte **)alloc_root(&ftb->mem_root,
(ftb->queue.max_elements + 1) *
sizeof(void *))))
goto err;
reinit_queue(&ftb->queue, ftb->queue.max_elements, 0, 0,
(int (*)(void*, byte*, byte*))FTB_WORD_cmp, 0);
for (ftbw= ftb->last_word; ftbw; ftbw= ftbw->prev)
queue_insert(&ftb->queue, (byte *)ftbw);
ftb->list=(FTB_WORD **)alloc_root(&ftb->mem_root,
sizeof(FTB_WORD *)*ftb->queue.elements);
2002-03-22 18:18:06 +01:00
memcpy(ftb->list, ftb->queue.root+1, sizeof(FTB_WORD *)*ftb->queue.elements);
qsort2(ftb->list, ftb->queue.elements, sizeof(FTB_WORD *),
2002-04-15 22:32:40 +02:00
(qsort2_cmp)FTB_WORD_cmp_list, ftb->charset);
2002-04-18 16:12:29 +02:00
if (ftb->queue.elements<2) ftb->with_scan &= ~FTB_FLAG_TRUNC;
ftb->state=READY;
return ftb;
2003-12-09 20:49:48 +01:00
err:
free_root(& ftb->mem_root, MYF(0));
my_free((gptr)ftb,MYF(0));
return 0;
}
typedef struct st_my_ftb_phrase_param
{
LIST *phrase;
LIST *document;
CHARSET_INFO *cs;
uint phrase_length;
uint document_length;
uint match;
} MY_FTB_PHRASE_PARAM;
static int ftb_phrase_add_word(MYSQL_FTPARSER_PARAM *param,
char *word, int word_len,
MYSQL_FTPARSER_BOOLEAN_INFO *boolean_info __attribute__((unused)))
{
MY_FTB_PHRASE_PARAM *phrase_param= param->mysql_ftparam;
FT_WORD *w= (FT_WORD *)phrase_param->document->data;
LIST *phrase, *document;
w->pos= word;
w->len= word_len;
phrase_param->document= phrase_param->document->prev;
if (phrase_param->phrase_length > phrase_param->document_length)
{
phrase_param->document_length++;
return 0;
}
/* TODO: rewrite phrase search to avoid
comparing the same word twice. */
for (phrase= phrase_param->phrase, document= phrase_param->document->next;
phrase; phrase= phrase->next, document= document->next)
{
FT_WORD *phrase_word= (FT_WORD *)phrase->data;
FT_WORD *document_word= (FT_WORD *)document->data;
This changeset is largely a handler cleanup changeset (WL#3281), but includes fixes and cleanups that was found necessary while testing the handler changes Changes that requires code changes in other code of other storage engines. (Note that all changes are very straightforward and one should find all issues by compiling a --debug build and fixing all compiler errors and all asserts in field.cc while running the test suite), - New optional handler function introduced: reset() This is called after every DML statement to make it easy for a handler to statement specific cleanups. (The only case it's not called is if force the file to be closed) - handler::extra(HA_EXTRA_RESET) is removed. Code that was there before should be moved to handler::reset() - table->read_set contains a bitmap over all columns that are needed in the query. read_row() and similar functions only needs to read these columns - table->write_set contains a bitmap over all columns that will be updated in the query. write_row() and update_row() only needs to update these columns. The above bitmaps should now be up to date in all context (including ALTER TABLE, filesort()). The handler is informed of any changes to the bitmap after fix_fields() by calling the virtual function handler::column_bitmaps_signal(). If the handler does caching of these bitmaps (instead of using table->read_set, table->write_set), it should redo the caching in this code. as the signal() may be sent several times, it's probably best to set a variable in the signal and redo the caching on read_row() / write_row() if the variable was set. - Removed the read_set and write_set bitmap objects from the handler class - Removed all column bit handling functions from the handler class. (Now one instead uses the normal bitmap functions in my_bitmap.c instead of handler dedicated bitmap functions) - field->query_id is removed. One should instead instead check table->read_set and table->write_set if a field is used in the query. - handler::extra(HA_EXTRA_RETRIVE_ALL_COLS) and handler::extra(HA_EXTRA_RETRIEVE_PRIMARY_KEY) are removed. One should now instead use table->read_set to check for which columns to retrieve. - If a handler needs to call Field->val() or Field->store() on columns that are not used in the query, one should install a temporary all-columns-used map while doing so. For this, we provide the following functions: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->read_set); field->val(); dbug_tmp_restore_column_map(table->read_set, old_map); and similar for the write map: my_bitmap_map *old_map= dbug_tmp_use_all_columns(table, table->write_set); field->val(); dbug_tmp_restore_column_map(table->write_set, old_map); If this is not done, you will sooner or later hit a DBUG_ASSERT in the field store() / val() functions. (For not DBUG binaries, the dbug_tmp_restore_column_map() and dbug_tmp_restore_column_map() are inline dummy functions and should be optimized away be the compiler). - If one needs to temporary set the column map for all binaries (and not just to avoid the DBUG_ASSERT() in the Field::store() / Field::val() methods) one should use the functions tmp_use_all_columns() and tmp_restore_column_map() instead of the above dbug_ variants. - All 'status' fields in the handler base class (like records, data_file_length etc) are now stored in a 'stats' struct. This makes it easier to know what status variables are provided by the base handler. This requires some trivial variable names in the extra() function. - New virtual function handler::records(). This is called to optimize COUNT(*) if (handler::table_flags() & HA_HAS_RECORDS()) is true. (stats.records is not supposed to be an exact value. It's only has to be 'reasonable enough' for the optimizer to be able to choose a good optimization path). - Non virtual handler::init() function added for caching of virtual constants from engine. - Removed has_transactions() virtual method. Now one should instead return HA_NO_TRANSACTIONS in table_flags() if the table handler DOES NOT support transactions. - The 'xxxx_create_handler()' function now has a MEM_ROOT_root argument that is to be used with 'new handler_name()' to allocate the handler in the right area. The xxxx_create_handler() function is also responsible for any initialization of the object before returning. For example, one should change: static handler *myisam_create_handler(TABLE_SHARE *table) { return new ha_myisam(table); } -> static handler *myisam_create_handler(TABLE_SHARE *table, MEM_ROOT *mem_root) { return new (mem_root) ha_myisam(table); } - New optional virtual function: use_hidden_primary_key(). This is called in case of an update/delete when (table_flags() and HA_PRIMARY_KEY_REQUIRED_FOR_DELETE) is defined but we don't have a primary key. This allows the handler to take precisions in remembering any hidden primary key to able to update/delete any found row. The default handler marks all columns to be read. - handler::table_flags() now returns a ulonglong (to allow for more flags). - New/changed table_flags() - HA_HAS_RECORDS Set if ::records() is supported - HA_NO_TRANSACTIONS Set if engine doesn't support transactions - HA_PRIMARY_KEY_REQUIRED_FOR_DELETE Set if we should mark all primary key columns for read when reading rows as part of a DELETE statement. If there is no primary key, all columns are marked for read. - HA_PARTIAL_COLUMN_READ Set if engine will not read all columns in some cases (based on table->read_set) - HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS Renamed to HA_PRIMARY_KEY_REQUIRED_FOR_POSITION. - HA_DUPP_POS Renamed to HA_DUPLICATE_POS - HA_REQUIRES_KEY_COLUMNS_FOR_DELETE Set this if we should mark ALL key columns for read when when reading rows as part of a DELETE statement. In case of an update we will mark all keys for read for which key part changed value. - HA_STATS_RECORDS_IS_EXACT Set this if stats.records is exact. (This saves us some extra records() calls when optimizing COUNT(*)) - Removed table_flags() - HA_NOT_EXACT_COUNT Now one should instead use HA_HAS_RECORDS if handler::records() gives an exact count() and HA_STATS_RECORDS_IS_EXACT if stats.records is exact. - HA_READ_RND_SAME Removed (no one supported this one) - Removed not needed functions ha_retrieve_all_cols() and ha_retrieve_all_pk() - Renamed handler::dupp_pos to handler::dup_pos - Removed not used variable handler::sortkey Upper level handler changes: - ha_reset() now does some overall checks and calls ::reset() - ha_table_flags() added. This is a cached version of table_flags(). The cache is updated on engine creation time and updated on open. MySQL level changes (not obvious from the above): - DBUG_ASSERT() added to check that column usage matches what is set in the column usage bit maps. (This found a LOT of bugs in current column marking code). - In 5.1 before, all used columns was marked in read_set and only updated columns was marked in write_set. Now we only mark columns for which we need a value in read_set. - Column bitmaps are created in open_binary_frm() and open_table_from_share(). (Before this was in table.cc) - handler::table_flags() calls are replaced with handler::ha_table_flags() - For calling field->val() you must have the corresponding bit set in table->read_set. For calling field->store() you must have the corresponding bit set in table->write_set. (There are asserts in all store()/val() functions to catch wrong usage) - thd->set_query_id is renamed to thd->mark_used_columns and instead of setting this to an integer value, this has now the values: MARK_COLUMNS_NONE, MARK_COLUMNS_READ, MARK_COLUMNS_WRITE Changed also all variables named 'set_query_id' to mark_used_columns. - In filesort() we now inform the handler of exactly which columns are needed doing the sort and choosing the rows. - The TABLE_SHARE object has a 'all_set' column bitmap one can use when one needs a column bitmap with all columns set. (This is used for table->use_all_columns() and other places) - The TABLE object has 3 column bitmaps: - def_read_set Default bitmap for columns to be read - def_write_set Default bitmap for columns to be written - tmp_set Can be used as a temporary bitmap when needed. The table object has also two pointer to bitmaps read_set and write_set that the handler should use to find out which columns are used in which way. - count() optimization now calls handler::records() instead of using handler->stats.records (if (table_flags() & HA_HAS_RECORDS) is true). - Added extra argument to Item::walk() to indicate if we should also traverse sub queries. - Added TABLE parameter to cp_buffer_from_ref() - Don't close tables created with CREATE ... SELECT but keep them in the table cache. (Faster usage of newly created tables). New interfaces: - table->clear_column_bitmaps() to initialize the bitmaps for tables at start of new statements. - table->column_bitmaps_set() to set up new column bitmaps and signal the handler about this. - table->column_bitmaps_set_no_signal() for some few cases where we need to setup new column bitmaps but don't signal the handler (as the handler has already been signaled about these before). Used for the momement only in opt_range.cc when doing ROR scans. - table->use_all_columns() to install a bitmap where all columns are marked as use in the read and the write set. - table->default_column_bitmaps() to install the normal read and write column bitmaps, but not signaling the handler about this. This is mainly used when creating TABLE instances. - table->mark_columns_needed_for_delete(), table->mark_columns_needed_for_delete() and table->mark_columns_needed_for_insert() to allow us to put additional columns in column usage maps if handler so requires. (The handler indicates what it neads in handler->table_flags()) - table->prepare_for_position() to allow us to tell handler that it needs to read primary key parts to be able to store them in future table->position() calls. (This replaces the table->file->ha_retrieve_all_pk function) - table->mark_auto_increment_column() to tell handler are going to update columns part of any auto_increment key. - table->mark_columns_used_by_index() to mark all columns that is part of an index. It will also send extra(HA_EXTRA_KEYREAD) to handler to allow it to quickly know that it only needs to read colums that are part of the key. (The handler can also use the column map for detecting this, but simpler/faster handler can just monitor the extra() call). - table->mark_columns_used_by_index_no_reset() to in addition to other columns, also mark all columns that is used by the given key. - table->restore_column_maps_after_mark_index() to restore to default column maps after a call to table->mark_columns_used_by_index(). - New item function register_field_in_read_map(), for marking used columns in table->read_map. Used by filesort() to mark all used columns - Maintain in TABLE->merge_keys set of all keys that are used in query. (Simplices some optimization loops) - Maintain Field->part_of_key_not_clustered which is like Field->part_of_key but the field in the clustered key is not assumed to be part of all index. (used in opt_range.cc for faster loops) - dbug_tmp_use_all_columns(), dbug_tmp_restore_column_map() tmp_use_all_columns() and tmp_restore_column_map() functions to temporally mark all columns as usable. The 'dbug_' version is primarily intended inside a handler when it wants to just call Field:store() & Field::val() functions, but don't need the column maps set for any other usage. (ie:: bitmap_is_set() is never called) - We can't use compare_records() to skip updates for handlers that returns a partial column set and the read_set doesn't cover all columns in the write set. The reason for this is that if we have a column marked only for write we can't in the MySQL level know if the value changed or not. The reason this worked before was that MySQL marked all to be written columns as also to be read. The new 'optimal' bitmaps exposed this 'hidden bug'. - open_table_from_share() does not anymore setup temporary MEM_ROOT object as a thread specific variable for the handler. Instead we send the to-be-used MEMROOT to get_new_handler(). (Simpler, faster code) Bugs fixed: - Column marking was not done correctly in a lot of cases. (ALTER TABLE, when using triggers, auto_increment fields etc) (Could potentially result in wrong values inserted in table handlers relying on that the old column maps or field->set_query_id was correct) Especially when it comes to triggers, there may be cases where the old code would cause lost/wrong values for NDB and/or InnoDB tables. - Split thd->options flag OPTION_STATUS_NO_TRANS_UPDATE to two flags: OPTION_STATUS_NO_TRANS_UPDATE and OPTION_KEEP_LOG. This allowed me to remove some wrong warnings about: "Some non-transactional changed tables couldn't be rolled back" - Fixed handling of INSERT .. SELECT and CREATE ... SELECT that wrongly reset (thd->options & OPTION_STATUS_NO_TRANS_UPDATE) which caused us to loose some warnings about "Some non-transactional changed tables couldn't be rolled back") - Fixed use of uninitialized memory in ha_ndbcluster.cc::delete_table() which could cause delete_table to report random failures. - Fixed core dumps for some tests when running with --debug - Added missing FN_LIBCHAR in mysql_rm_tmp_tables() (This has probably caused us to not properly remove temporary files after crash) - slow_logs was not properly initialized, which could maybe cause extra/lost entries in slow log. - If we get an duplicate row on insert, change column map to read and write all columns while retrying the operation. This is required by the definition of REPLACE and also ensures that fields that are only part of UPDATE are properly handled. This fixed a bug in NDB and REPLACE where REPLACE wrongly copied some column values from the replaced row. - For table handler that doesn't support NULL in keys, we would give an error when creating a primary key with NULL fields, even after the fields has been automaticly converted to NOT NULL. - Creating a primary key on a SPATIAL key, would fail if field was not declared as NOT NULL. Cleanups: - Removed not used condition argument to setup_tables - Removed not needed item function reset_query_id_processor(). - Field->add_index is removed. Now this is instead maintained in (field->flags & FIELD_IN_ADD_INDEX) - Field->fieldnr is removed (use field->field_index instead) - New argument to filesort() to indicate that it should return a set of row pointers (not used columns). This allowed me to remove some references to sql_command in filesort and should also enable us to return column results in some cases where we couldn't before. - Changed column bitmap handling in opt_range.cc to be aligned with TABLE bitmap, which allowed me to use bitmap functions instead of looping over all fields to create some needed bitmaps. (Faster and smaller code) - Broke up found too long lines - Moved some variable declaration at start of function for better code readability. - Removed some not used arguments from functions. (setup_fields(), mysql_prepare_insert_check_table()) - setup_fields() now takes an enum instead of an int for marking columns usage. - For internal temporary tables, use handler::write_row(), handler::delete_row() and handler::update_row() instead of handler::ha_xxxx() for faster execution. - Changed some constants to enum's and define's. - Using separate column read and write sets allows for easier checking of timestamp field was set by statement. - Remove calls to free_io_cache() as this is now done automaticly in ha_reset() - Don't build table->normalized_path as this is now identical to table->path (after bar's fixes to convert filenames) - Fixed some missed DBUG_PRINT(.."%lx") to use "0x%lx" to make it easier to do comparision with the 'convert-dbug-for-diff' tool. Things left to do in 5.1: - We wrongly log failed CREATE TABLE ... SELECT in some cases when using row based logging (as shown by testcase binlog_row_mix_innodb_myisam.result) Mats has promised to look into this. - Test that my fix for CREATE TABLE ... SELECT is indeed correct. (I added several test cases for this, but in this case it's better that someone else also tests this throughly). Lars has promosed to do this.
2006-06-04 17:52:22 +02:00
if (my_strnncoll(phrase_param->cs, (uchar*) phrase_word->pos,
phrase_word->len,
(uchar*) document_word->pos, document_word->len))
return 0;
}
phrase_param->match++;
return 0;
}
static int ftb_check_phrase_internal(MYSQL_FTPARSER_PARAM *param,
char *document, int len)
{
FT_WORD word;
MY_FTB_PHRASE_PARAM *phrase_param= param->mysql_ftparam;
2006-01-16 15:17:45 +01:00
const char *docend= document + len;
while (ft_simple_get_word(phrase_param->cs, &document, docend, &word, FALSE))
{
param->mysql_add_word(param, word.pos, word.len, 0);
if (phrase_param->match)
return 1;
}
return 0;
}
2005-03-06 21:17:57 +01:00
/*
Checks if given buffer matches phrase list.
SYNOPSIS
_ftb_check_phrase()
s0 start of buffer
e0 end of buffer
phrase broken into list phrase
cs charset info
RETURN VALUE
1 is returned if phrase found, 0 else.
*/
static int _ftb_check_phrase(FTB *ftb, const byte *document, uint len,
FTB_EXPR *ftbe, struct st_mysql_ftparser *parser)
{
MY_FTB_PHRASE_PARAM ftb_param;
MYSQL_FTPARSER_PARAM *param;
DBUG_ENTER("_ftb_check_phrase");
DBUG_ASSERT(parser);
if (! (param= ftparser_call_initializer(ftb->info, ftb->keynr, 1)))
DBUG_RETURN(0);
ftb_param.phrase= ftbe->phrase;
ftb_param.document= ftbe->document;
ftb_param.cs= ftb->charset;
ftb_param.phrase_length= list_length(ftbe->phrase);
ftb_param.document_length= 1;
ftb_param.match= 0;
param->mysql_parse= ftb_check_phrase_internal;
param->mysql_add_word= ftb_phrase_add_word;
param->mysql_ftparam= (void *)&ftb_param;
param->cs= ftb->charset;
param->doc= (byte *)document;
param->length= len;
param->flags= 0;
param->mode= MYSQL_FTPARSER_WITH_STOPWORDS;
parser->parse(param);
DBUG_RETURN(ftb_param.match ? 1 : 0);
2002-04-18 16:12:29 +02:00
}
static void _ftb_climb_the_tree(FTB *ftb, FTB_WORD *ftbw, FT_SEG_ITERATOR *ftsi_orig)
2002-04-18 16:12:29 +02:00
{
FT_SEG_ITERATOR ftsi;
FTB_EXPR *ftbe;
float weight=ftbw->weight;
2002-04-18 16:12:29 +02:00
int yn=ftbw->flags, ythresh, mode=(ftsi_orig != 0);
my_off_t curdoc=ftbw->docid[mode];
struct st_mysql_ftparser *parser= ftb->keynr == NO_SUCH_KEY ?
&ft_default_parser :
ftb->info->s->keyinfo[ftb->keynr].parser;
for (ftbe=ftbw->up; ftbe; ftbe=ftbe->up)
{
ythresh = ftbe->ythresh - (mode ? 0 : ftbe->yweaks);
if (ftbe->docid[mode] != curdoc)
{
ftbe->cur_weight=0;
ftbe->yesses=ftbe->nos=0;
ftbe->docid[mode]=curdoc;
}
if (ftbe->nos)
break;
if (yn & FTB_FLAG_YES)
{
weight /= ftbe->ythresh;
ftbe->cur_weight += weight;
if ((int) ++ftbe->yesses == ythresh)
{
yn=ftbe->flags;
weight=ftbe->cur_weight*ftbe->weight;
2005-03-06 21:17:57 +01:00
if (mode && ftbe->phrase)
2002-04-18 16:12:29 +02:00
{
int not_found=1;
memcpy(&ftsi, ftsi_orig, sizeof(ftsi));
while (_mi_ft_segiterator(&ftsi) && not_found)
{
if (!ftsi.pos)
continue;
not_found = ! _ftb_check_phrase(ftb, ftsi.pos, ftsi.len,
ftbe, parser);
2002-04-18 16:12:29 +02:00
}
if (not_found) break;
} /* ftbe->quot */
}
else
break;
}
else
if (yn & FTB_FLAG_NO)
{
/*
2004-02-15 20:36:12 +01:00
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 & FTB_FLAG_YES) != 0 events. So no
already matched expression can become not-matched again.
*/
++ftbe->nos;
break;
}
else
{
if (ftbe->ythresh)
2004-02-15 20:36:12 +01:00
weight/=3;
ftbe->cur_weight += weight;
if ((int) ftbe->yesses < ythresh)
break;
2004-02-22 00:19:36 +01:00
if (!(yn & FTB_FLAG_WONLY))
2004-03-16 21:41:30 +01:00
yn= ((int) ftbe->yesses++ == ythresh) ? ftbe->flags : FTB_FLAG_WONLY ;
weight*= ftbe->weight;
}
}
}
2001-10-09 14:53:54 +02:00
int ft_boolean_read_next(FT_INFO *ftb, char *record)
{
2001-12-02 13:34:01 +01:00
FTB_EXPR *ftbe;
FTB_WORD *ftbw;
MI_INFO *info=ftb->info;
my_off_t curdoc;
if (ftb->state != INDEX_SEARCH && ftb->state != INDEX_DONE)
return -1;
/* black magic ON */
if ((int) _mi_check_index(info, ftb->keynr) < 0)
return my_errno;
if (_mi_readinfo(info, F_RDLCK, 1))
return my_errno;
/* black magic OFF */
2001-10-09 14:53:54 +02:00
if (!ftb->queue.elements)
return my_errno=HA_ERR_END_OF_FILE;
/* Attention!!! Address of a local variable is used here! See err: label */
ftb->queue.first_cmp_arg=(void *)&curdoc;
while (ftb->state == INDEX_SEARCH &&
2004-02-15 20:36:12 +01:00
(curdoc=((FTB_WORD *)queue_top(& ftb->queue))->docid[0]) !=
2004-11-17 15:28:37 +01:00
HA_OFFSET_ERROR)
2001-10-09 14:53:54 +02:00
{
while (curdoc == (ftbw=(FTB_WORD *)queue_top(& ftb->queue))->docid[0])
{
2002-04-18 16:12:29 +02:00
_ftb_climb_the_tree(ftb, ftbw, 0);
/* update queue */
_ft2_search(ftb, ftbw, 0);
2001-10-09 14:53:54 +02:00
queue_replaced(& ftb->queue);
}
ftbe=ftb->root;
if (ftbe->docid[0]==curdoc && ftbe->cur_weight>0 &&
ftbe->yesses>=(ftbe->ythresh-ftbe->yweaks) && !ftbe->nos)
{
/* curdoc matched ! */
if (is_tree_inited(&ftb->no_dupes) &&
tree_insert(&ftb->no_dupes, &curdoc, 0,
2004-02-15 20:36:12 +01:00
ftb->no_dupes.custom_arg)->count >1)
/* but it managed already to get past this line once */
2002-05-07 20:27:39 +02:00
continue;
info->lastpos=curdoc;
/* Clear all states, except that the table was updated */
info->update&= (HA_STATE_CHANGED | HA_STATE_ROW_CHANGED);
if (!(*info->read_record)(info,curdoc,record))
{
2001-10-09 14:53:54 +02:00
info->update|= HA_STATE_AKTIV; /* Record is read */
if (ftb->with_scan && ft_boolean_find_relevance(ftb,record,0)==0)
continue; /* no match */
my_errno=0;
goto err;
}
goto err;
}
}
ftb->state=INDEX_DONE;
my_errno=HA_ERR_END_OF_FILE;
err:
ftb->queue.first_cmp_arg=(void *)0;
return my_errno;
}
typedef struct st_my_ftb_find_param
2001-10-09 14:53:54 +02:00
{
FT_INFO *ftb;
FT_SEG_ITERATOR *ftsi;
} MY_FTB_FIND_PARAM;
static int ftb_find_relevance_add_word(MYSQL_FTPARSER_PARAM *param,
char *word, int len,
MYSQL_FTPARSER_BOOLEAN_INFO *boolean_info __attribute__((unused)))
{
MY_FTB_FIND_PARAM *ftb_param= param->mysql_ftparam;
FT_INFO *ftb= ftb_param->ftb;
FTB_WORD *ftbw;
int a, b, c;
for (a= 0, b= ftb->queue.elements, c= (a+b)/2; b-a>1; c= (a+b)/2)
{
ftbw= ftb->list[c];
if (mi_compare_text(ftb->charset, (uchar*)word, len,
(uchar*)ftbw->word+1, ftbw->len-1,
(my_bool)(ftbw->flags&FTB_FLAG_TRUNC), 0) > 0)
b= c;
else
a= c;
}
for (; c >= 0; c--)
{
ftbw= ftb->list[c];
if (mi_compare_text(ftb->charset, (uchar*)word, len,
(uchar*)ftbw->word + 1,ftbw->len - 1,
(my_bool)(ftbw->flags & FTB_FLAG_TRUNC), 0))
break;
if (ftbw->docid[1] == ftb->info->lastpos)
continue;
ftbw->docid[1]= ftb->info->lastpos;
_ftb_climb_the_tree(ftb, ftbw, ftb_param->ftsi);
}
return(0);
}
static int ftb_find_relevance_parse(MYSQL_FTPARSER_PARAM *param,
char *doc, int len)
{
MY_FTB_FIND_PARAM *ftb_param= param->mysql_ftparam;
FT_INFO *ftb= ftb_param->ftb;
2006-01-16 15:17:45 +01:00
char *end= doc + len;
FT_WORD w;
while (ft_simple_get_word(ftb->charset, &doc, end, &w, TRUE))
param->mysql_add_word(param, w.pos, w.len, 0);
return(0);
}
float ft_boolean_find_relevance(FT_INFO *ftb, byte *record, uint length)
{
FTB_EXPR *ftbe;
2002-04-18 16:12:29 +02:00
FT_SEG_ITERATOR ftsi, ftsi2;
my_off_t docid=ftb->info->lastpos;
MY_FTB_FIND_PARAM ftb_param;
MYSQL_FTPARSER_PARAM *param;
struct st_mysql_ftparser *parser= ftb->keynr == NO_SUCH_KEY ?
&ft_default_parser :
ftb->info->s->keyinfo[ftb->keynr].parser;
if (docid == HA_OFFSET_ERROR)
return -2.0;
2002-01-14 16:40:54 +01:00
if (!ftb->queue.elements)
return 0;
if (! (param= ftparser_call_initializer(ftb->info, ftb->keynr, 0)))
return 0;
if (ftb->state != INDEX_SEARCH && docid <= ftb->lastpos)
{
FTB_EXPR *x;
uint i;
for (i=0; i < ftb->queue.elements; i++)
{
ftb->list[i]->docid[1]=HA_OFFSET_ERROR;
for (x=ftb->list[i]->up; x; x=x->up)
x->docid[1]=HA_OFFSET_ERROR;
}
}
ftb->lastpos=docid;
if (ftb->keynr==NO_SUCH_KEY)
_mi_ft_segiterator_dummy_init(record, length, &ftsi);
else
_mi_ft_segiterator_init(ftb->info, ftb->keynr, record, &ftsi);
2002-04-18 16:12:29 +02:00
memcpy(&ftsi2, &ftsi, sizeof(ftsi));
ftb_param.ftb= ftb;
ftb_param.ftsi= &ftsi2;
param->mysql_parse= ftb_find_relevance_parse;
param->mysql_add_word= ftb_find_relevance_add_word;
param->mysql_ftparam= (void *)&ftb_param;
param->flags= 0;
param->cs= ftb->charset;
param->mode= MYSQL_FTPARSER_SIMPLE_MODE;
while (_mi_ft_segiterator(&ftsi))
{
if (!ftsi.pos)
continue;
param->doc= (byte *)ftsi.pos;
param->length= ftsi.len;
parser->parse(param);
}
ftbe=ftb->root;
if (ftbe->docid[1]==docid && ftbe->cur_weight>0 &&
ftbe->yesses>=ftbe->ythresh && !ftbe->nos)
{ /* row matched ! */
return ftbe->cur_weight;
}
else
{ /* match failed ! */
return 0.0;
}
2001-10-09 14:53:54 +02:00
}
2001-10-09 14:53:54 +02:00
void ft_boolean_close_search(FT_INFO *ftb)
{
2002-05-07 20:27:39 +02:00
if (is_tree_inited(& ftb->no_dupes))
{
delete_tree(& ftb->no_dupes);
}
2001-10-09 14:53:54 +02:00
free_root(& ftb->mem_root, MYF(0));
my_free((gptr)ftb,MYF(0));
}
2001-10-09 14:53:54 +02:00
float ft_boolean_get_relevance(FT_INFO *ftb)
{
return ftb->root->cur_weight;
}
2001-10-09 14:53:54 +02:00
void ft_boolean_reinit_search(FT_INFO *ftb)
{
_ftb_init_index_search(ftb);
2001-10-09 14:53:54 +02:00
}