ltree support for multibyte encodings. Patch was made by

laser <laserlist@pgsqldb.com> with some editorization by me.
This commit is contained in:
Teodor Sigaev 2008-06-30 18:30:48 +00:00
parent 995fb74202
commit 8eee65c996
5 changed files with 136 additions and 87 deletions

View File

@ -1,13 +1,14 @@
/*
* op function for ltree and lquery
* Teodor Sigaev <teodor@stack.net>
* $PostgreSQL: pgsql/contrib/ltree/lquery_op.c,v 1.12 2008/05/12 00:00:42 alvherre Exp $
* $PostgreSQL: pgsql/contrib/ltree/lquery_op.c,v 1.13 2008/06/30 18:30:48 teodor Exp $
*/
#include "postgres.h"
#include <ctype.h>
#include "utils/array.h"
#include "utils/formatting.h"
#include "ltree.h"
PG_FUNCTION_INFO_V1(ltq_regex);
@ -32,23 +33,24 @@ static char *
getlexeme(char *start, char *end, int *len)
{
char *ptr;
while (start < end && *start == '_')
start++;
int charlen;
while (start < end && (charlen = pg_mblen(start)) == 1 && t_iseq(start,'_') )
start += charlen;
ptr = start;
if (ptr == end)
if (ptr >= end)
return NULL;
while (ptr < end && *ptr != '_')
ptr++;
while (ptr < end && !( (charlen = pg_mblen(ptr)) == 1 && t_iseq(ptr, '_') ) )
ptr += charlen;
*len = ptr - start;
return start;
}
bool
compare_subnode(ltree_level * t, char *qn, int len, int (*cmpptr) (const char *, const char *, size_t), bool anyend)
compare_subnode(ltree_level * t, char *qn, int len, int (*cmpptr) (const char *, const char *, size_t), bool anyend)
{
char *endt = t->name + t->len;
char *endq = qn + len;
@ -85,6 +87,21 @@ bool
return true;
}
int
ltree_strncasecmp(const char *a, const char *b, size_t s)
{
char *al = str_tolower(a, s);
char *bl = str_tolower(b, s);
int res;
res = strncmp(al, bl,s);
pfree(al);
pfree(bl);
return res;
}
static bool
checkLevel(lquery_level * curq, ltree_level * curt)
{
@ -94,7 +111,7 @@ checkLevel(lquery_level * curq, ltree_level * curt)
for (i = 0; i < curq->numvar; i++)
{
cmpptr = (curvar->flag & LVAR_INCASE) ? pg_strncasecmp : strncmp;
cmpptr = (curvar->flag & LVAR_INCASE) ? ltree_strncasecmp : strncmp;
if (curvar->flag & LVAR_SUBLEXEME)
{

View File

@ -1,17 +1,19 @@
/* $PostgreSQL: pgsql/contrib/ltree/ltree.h,v 1.20 2008/05/12 00:00:42 alvherre Exp $ */
/* $PostgreSQL: pgsql/contrib/ltree/ltree.h,v 1.21 2008/06/30 18:30:48 teodor Exp $ */
#ifndef __LTREE_H__
#define __LTREE_H__
#include "postgres.h"
#include "fmgr.h"
#include "tsearch/ts_locale.h"
typedef struct
{
uint8 len;
uint16 len;
char name[1];
} ltree_level;
#define LEVEL_HDRSIZE (sizeof(uint8))
#define LEVEL_HDRSIZE (offsetof(ltree_level,name))
#define LEVEL_NEXT(x) ( (ltree_level*)( ((char*)(x)) + MAXALIGN(((ltree_level*)(x))->len + LEVEL_HDRSIZE) ) )
typedef struct
@ -21,7 +23,7 @@ typedef struct
char data[1];
} ltree;
#define LTREE_HDRSIZE MAXALIGN(VARHDRSZ + sizeof(uint16))
#define LTREE_HDRSIZE MAXALIGN( offsetof(ltree, data) )
#define LTREE_FIRST(x) ( (ltree_level*)( ((char*)(x))+LTREE_HDRSIZE ) )
@ -30,12 +32,12 @@ typedef struct
typedef struct
{
int4 val;
uint8 len;
uint16 len;
uint8 flag;
char name[1];
} lquery_variant;
#define LVAR_HDRSIZE MAXALIGN(sizeof(uint8)*2 + sizeof(int4))
#define LVAR_HDRSIZE MAXALIGN(offsetof(lquery_variant, name))
#define LVAR_NEXT(x) ( (lquery_variant*)( ((char*)(x)) + MAXALIGN(((lquery_variant*)(x))->len) + LVAR_HDRSIZE ) )
#define LVAR_ANYEND 0x01
@ -52,7 +54,7 @@ typedef struct
char variants[1];
} lquery_level;
#define LQL_HDRSIZE MAXALIGN( sizeof(uint16)*5 )
#define LQL_HDRSIZE MAXALIGN( offsetof(lquery_level,variants) )
#define LQL_NEXT(x) ( (lquery_level*)( ((char*)(x)) + MAXALIGN(((lquery_level*)(x))->totallen) ) )
#define LQL_FIRST(x) ( (lquery_variant*)( ((char*)(x))+LQL_HDRSIZE ) )
@ -73,12 +75,12 @@ typedef struct
char data[1];
} lquery;
#define LQUERY_HDRSIZE MAXALIGN(VARHDRSZ + 3*sizeof(uint16))
#define LQUERY_HDRSIZE MAXALIGN( offsetof(lquery, data) )
#define LQUERY_FIRST(x) ( (lquery_level*)( ((char*)(x))+LQUERY_HDRSIZE ) )
#define LQUERY_HASNOT 0x01
#define ISALNUM(x) ( isalnum((unsigned char)(x)) || (x) == '_' )
#define ISALNUM(x) ( t_isalpha(x) || t_isdigit(x) || ( pg_mblen(x) == 1 && t_iseq((x), '_') ) )
/* full text query */
@ -156,9 +158,10 @@ bool ltree_execute(ITEM * curitem, void *checkval,
int ltree_compare(const ltree * a, const ltree * b);
bool inner_isparent(const ltree * c, const ltree * p);
bool compare_subnode(ltree_level * t, char *q, int len,
int (*cmpptr) (const char *, const char *, size_t), bool anyend);
bool compare_subnode(ltree_level * t, char *q, int len,
int (*cmpptr) (const char *, const char *, size_t), bool anyend);
ltree *lca_inner(ltree ** a, int len);
int ltree_strncasecmp(const char *a, const char *b, size_t s);
#define PG_GETARG_LTREE(x) ((ltree*)DatumGetPointer(PG_DETOAST_DATUM(PG_GETARG_DATUM(x))))
#define PG_GETARG_LTREE_COPY(x) ((ltree*)DatumGetPointer(PG_DETOAST_DATUM_COPY(PG_GETARG_DATUM(x))))

View File

@ -1,7 +1,7 @@
/*
* in/out function for ltree and lquery
* Teodor Sigaev <teodor@stack.net>
* $PostgreSQL: pgsql/contrib/ltree/ltree_io.c,v 1.16 2008/05/12 00:00:43 alvherre Exp $
* $PostgreSQL: pgsql/contrib/ltree/ltree_io.c,v 1.17 2008/06/30 18:30:48 teodor Exp $
*/
#include "postgres.h"
@ -25,15 +25,16 @@ Datum lquery_out(PG_FUNCTION_ARGS);
#define UNCHAR ereport(ERROR, \
(errcode(ERRCODE_SYNTAX_ERROR), \
errmsg("syntax error at position %d near \"%c\"", \
(int)(ptr-buf), *ptr)));
errmsg("syntax error at position %d", \
pos)));
typedef struct
{
char *start;
int len;
int len; /* length in bytes */
int flag;
int wlen; /* length in characters */
} nodeitem;
#define LTPRS_WAITNAME 0
@ -51,24 +52,30 @@ ltree_in(PG_FUNCTION_ARGS)
int state = LTPRS_WAITNAME;
ltree *result;
ltree_level *curlevel;
int charlen;
int pos=0;
ptr = buf;
while (*ptr)
{
if (*ptr == '.')
charlen = pg_mblen(ptr);
if ( charlen == 1 && t_iseq(ptr, '.') )
num++;
ptr++;
ptr+=charlen;
}
list = lptr = (nodeitem *) palloc(sizeof(nodeitem) * (num + 1));
ptr = buf;
while (*ptr)
{
charlen = pg_mblen(ptr);
if (state == LTPRS_WAITNAME)
{
if (ISALNUM(*ptr))
if (ISALNUM(ptr))
{
lptr->start = ptr;
lptr->wlen = 0;
state = LTPRS_WAITDELIM;
}
else
@ -76,40 +83,43 @@ ltree_in(PG_FUNCTION_ARGS)
}
else if (state == LTPRS_WAITDELIM)
{
if (*ptr == '.')
if ( charlen == 1 && t_iseq(ptr, '.') )
{
lptr->len = ptr - lptr->start;
if (lptr->len > 255)
if (lptr->wlen > 255)
ereport(ERROR,
(errcode(ERRCODE_NAME_TOO_LONG),
errmsg("name of level is too long"),
errdetail("Name length is %d, must "
"be < 256, in position %d.",
lptr->len, (int) (lptr->start - buf))));
lptr->wlen, pos)));
totallen += MAXALIGN(lptr->len + LEVEL_HDRSIZE);
lptr++;
state = LTPRS_WAITNAME;
}
else if (!ISALNUM(*ptr))
else if (!ISALNUM(ptr))
UNCHAR;
}
else
/* internal error */
elog(ERROR, "internal error in parser");
ptr++;
ptr+=charlen;
lptr->wlen++;
pos++;
}
if (state == LTPRS_WAITDELIM)
{
lptr->len = ptr - lptr->start;
if (lptr->len > 255)
if (lptr->wlen > 255)
ereport(ERROR,
(errcode(ERRCODE_NAME_TOO_LONG),
errmsg("name of level is too long"),
errdetail("Name length is %d, must "
"be < 256, in position %d.",
lptr->len, (int) (lptr->start - buf))));
lptr->wlen, pos)));
totallen += MAXALIGN(lptr->len + LEVEL_HDRSIZE);
lptr++;
@ -127,7 +137,7 @@ ltree_in(PG_FUNCTION_ARGS)
lptr = list;
while (lptr - list < result->numlevel)
{
curlevel->len = (uint8) lptr->len;
curlevel->len = (uint16) lptr->len;
memcpy(curlevel->name, lptr->start, lptr->len);
curlevel = LEVEL_NEXT(curlevel);
lptr++;
@ -198,15 +208,23 @@ lquery_in(PG_FUNCTION_ARGS)
lquery_variant *lrptr = NULL;
bool hasnot = false;
bool wasbad = false;
int charlen;
int pos=0;
ptr = buf;
while (*ptr)
{
if (*ptr == '.')
num++;
else if (*ptr == '|')
numOR++;
ptr++;
charlen = pg_mblen(ptr);
if ( charlen == 1 )
{
if (t_iseq(ptr, '.'))
num++;
else if (t_iseq(ptr, '|'))
numOR++;
}
ptr+=charlen;
}
num++;
@ -214,16 +232,18 @@ lquery_in(PG_FUNCTION_ARGS)
ptr = buf;
while (*ptr)
{
charlen = pg_mblen(ptr);
if (state == LQPRS_WAITLEVEL)
{
if (ISALNUM(*ptr))
if (ISALNUM(ptr))
{
GETVAR(curqlevel) = lptr = (nodeitem *) palloc0(sizeof(nodeitem) * (numOR + 1));
lptr->start = ptr;
state = LQPRS_WAITDELIM;
curqlevel->numvar = 1;
}
else if (*ptr == '!')
else if (charlen==1 && t_iseq(ptr, '!'))
{
GETVAR(curqlevel) = lptr = (nodeitem *) palloc0(sizeof(nodeitem) * (numOR + 1));
lptr->start = ptr + 1;
@ -232,14 +252,14 @@ lquery_in(PG_FUNCTION_ARGS)
curqlevel->flag |= LQL_NOT;
hasnot = true;
}
else if (*ptr == '*')
else if (charlen==1 && t_iseq(ptr, '*'))
state = LQPRS_WAITOPEN;
else
UNCHAR;
}
else if (state == LQPRS_WAITVAR)
{
if (ISALNUM(*ptr))
if (ISALNUM(ptr))
{
lptr++;
lptr->start = ptr;
@ -251,61 +271,61 @@ lquery_in(PG_FUNCTION_ARGS)
}
else if (state == LQPRS_WAITDELIM)
{
if (*ptr == '@')
if (charlen==1 && t_iseq(ptr, '@'))
{
if (lptr->start == ptr)
UNCHAR;
lptr->flag |= LVAR_INCASE;
curqlevel->flag |= LVAR_INCASE;
}
else if (*ptr == '*')
else if (charlen==1 && t_iseq(ptr, '*'))
{
if (lptr->start == ptr)
UNCHAR;
lptr->flag |= LVAR_ANYEND;
curqlevel->flag |= LVAR_ANYEND;
}
else if (*ptr == '%')
else if (charlen==1 && t_iseq(ptr, '%'))
{
if (lptr->start == ptr)
UNCHAR;
lptr->flag |= LVAR_SUBLEXEME;
curqlevel->flag |= LVAR_SUBLEXEME;
}
else if (*ptr == '|')
else if (charlen==1 && t_iseq(ptr, '|'))
{
lptr->len = ptr - lptr->start -
((lptr->flag & LVAR_SUBLEXEME) ? 1 : 0) -
((lptr->flag & LVAR_INCASE) ? 1 : 0) -
((lptr->flag & LVAR_ANYEND) ? 1 : 0);
if (lptr->len > 255)
if (lptr->wlen > 255)
ereport(ERROR,
(errcode(ERRCODE_NAME_TOO_LONG),
errmsg("name of level is too long"),
errdetail("Name length is %d, must "
"be < 256, in position %d.",
lptr->len, (int) (lptr->start - buf))));
lptr->wlen, pos)));
state = LQPRS_WAITVAR;
}
else if (*ptr == '.')
else if (charlen==1 && t_iseq(ptr, '.'))
{
lptr->len = ptr - lptr->start -
((lptr->flag & LVAR_SUBLEXEME) ? 1 : 0) -
((lptr->flag & LVAR_INCASE) ? 1 : 0) -
((lptr->flag & LVAR_ANYEND) ? 1 : 0);
if (lptr->len > 255)
if (lptr->wlen > 255)
ereport(ERROR,
(errcode(ERRCODE_NAME_TOO_LONG),
errmsg("name of level is too long"),
errdetail("Name length is %d, must "
"be < 256, in position %d.",
lptr->len, (int) (lptr->start - buf))));
lptr->wlen, pos)));
state = LQPRS_WAITLEVEL;
curqlevel = NEXTLEV(curqlevel);
}
else if (ISALNUM(*ptr))
else if (ISALNUM(ptr))
{
if (lptr->flag)
UNCHAR;
@ -315,9 +335,9 @@ lquery_in(PG_FUNCTION_ARGS)
}
else if (state == LQPRS_WAITOPEN)
{
if (*ptr == '{')
if (charlen==1 && t_iseq(ptr, '{'))
state = LQPRS_WAITFNUM;
else if (*ptr == '.')
else if (charlen==1 && t_iseq(ptr, '.'))
{
curqlevel->low = 0;
curqlevel->high = 0xffff;
@ -329,9 +349,9 @@ lquery_in(PG_FUNCTION_ARGS)
}
else if (state == LQPRS_WAITFNUM)
{
if (*ptr == ',')
if (charlen==1 && t_iseq(ptr, ','))
state = LQPRS_WAITSNUM;
else if (isdigit((unsigned char) *ptr))
else if (t_isdigit(ptr))
{
curqlevel->low = atoi(ptr);
state = LQPRS_WAITND;
@ -341,12 +361,12 @@ lquery_in(PG_FUNCTION_ARGS)
}
else if (state == LQPRS_WAITSNUM)
{
if (isdigit((unsigned char) *ptr))
if (t_isdigit(ptr))
{
curqlevel->high = atoi(ptr);
state = LQPRS_WAITCLOSE;
}
else if (*ptr == '}')
else if (charlen==1 && t_iseq(ptr, '}'))
{
curqlevel->high = 0xffff;
state = LQPRS_WAITEND;
@ -356,26 +376,26 @@ lquery_in(PG_FUNCTION_ARGS)
}
else if (state == LQPRS_WAITCLOSE)
{
if (*ptr == '}')
if (charlen==1 && t_iseq(ptr, '}'))
state = LQPRS_WAITEND;
else if (!isdigit((unsigned char) *ptr))
else if (!t_isdigit(ptr))
UNCHAR;
}
else if (state == LQPRS_WAITND)
{
if (*ptr == '}')
if (charlen==1 && t_iseq(ptr, '}'))
{
curqlevel->high = curqlevel->low;
state = LQPRS_WAITEND;
}
else if (*ptr == ',')
else if (charlen==1 && t_iseq(ptr, ','))
state = LQPRS_WAITSNUM;
else if (!isdigit((unsigned char) *ptr))
else if (!t_isdigit(ptr))
UNCHAR;
}
else if (state == LQPRS_WAITEND)
{
if (*ptr == '.')
if (charlen==1 && t_iseq(ptr, '.'))
{
state = LQPRS_WAITLEVEL;
curqlevel = NEXTLEV(curqlevel);
@ -386,7 +406,11 @@ lquery_in(PG_FUNCTION_ARGS)
else
/* internal error */
elog(ERROR, "internal error in parser");
ptr++;
ptr+=charlen;
if ( state == LQPRS_WAITDELIM )
lptr->wlen++;
pos++;
}
if (state == LQPRS_WAITDELIM)
@ -407,13 +431,13 @@ lquery_in(PG_FUNCTION_ARGS)
errmsg("syntax error"),
errdetail("Unexpected end of line.")));
if (lptr->len > 255)
if (lptr->wlen > 255)
ereport(ERROR,
(errcode(ERRCODE_NAME_TOO_LONG),
errmsg("name of level is too long"),
errdetail("Name length is %d, must "
"be < 256, in position %d.",
lptr->len, (int) (lptr->start - buf))));
lptr->wlen, pos)));
}
else if (state == LQPRS_WAITOPEN)
curqlevel->high = 0xffff;

View File

@ -1,7 +1,7 @@
/*
* txtquery io
* Teodor Sigaev <teodor@stack.net>
* $PostgreSQL: pgsql/contrib/ltree/ltxtquery_io.c,v 1.15 2008/05/12 00:00:43 alvherre Exp $
* $PostgreSQL: pgsql/contrib/ltree/ltxtquery_io.c,v 1.16 2008/06/30 18:30:48 teodor Exp $
*/
#include "postgres.h"
@ -59,49 +59,53 @@ typedef struct
static int4
gettoken_query(QPRS_STATE * state, int4 *val, int4 *lenval, char **strval, uint16 *flag)
{
while (1)
int charlen;
for(;;)
{
charlen = pg_mblen(state->buf);
switch (state->state)
{
case WAITOPERAND:
if (*(state->buf) == '!')
if (charlen==1 && t_iseq(state->buf, '!'))
{
(state->buf)++;
*val = (int4) '!';
return OPR;
}
else if (*(state->buf) == '(')
else if (charlen==1 && t_iseq(state->buf, '('))
{
state->count++;
(state->buf)++;
return OPEN;
}
else if (ISALNUM(*(state->buf)))
else if (ISALNUM(state->buf))
{
state->state = INOPERAND;
*strval = state->buf;
*lenval = 1;
*lenval = charlen;
*flag = 0;
}
else if (!isspace((unsigned char) *(state->buf)))
else if (!t_isspace(state->buf))
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("operand syntax error")));
break;
case INOPERAND:
if (ISALNUM(*(state->buf)))
if (ISALNUM(state->buf))
{
if (*flag)
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
errmsg("modificators syntax error")));
(*lenval)++;
*lenval += charlen;
}
else if (*(state->buf) == '%')
else if (charlen==1 && t_iseq(state->buf, '%'))
*flag |= LVAR_SUBLEXEME;
else if (*(state->buf) == '@')
else if (charlen==1 && t_iseq(state->buf, '@'))
*flag |= LVAR_INCASE;
else if (*(state->buf) == '*')
else if (charlen==1 && t_iseq(state->buf, '*'))
*flag |= LVAR_ANYEND;
else
{
@ -110,14 +114,14 @@ gettoken_query(QPRS_STATE * state, int4 *val, int4 *lenval, char **strval, uint1
}
break;
case WAITOPERATOR:
if (*(state->buf) == '&' || *(state->buf) == '|')
if (charlen==1 && ( t_iseq(state->buf, '&') || t_iseq(state->buf, '|') ))
{
state->state = WAITOPERAND;
*val = (int4) *(state->buf);
(state->buf)++;
return OPR;
}
else if (*(state->buf) == ')')
else if (charlen==1 && t_iseq(state->buf, ')'))
{
(state->buf)++;
state->count--;
@ -125,14 +129,15 @@ gettoken_query(QPRS_STATE * state, int4 *val, int4 *lenval, char **strval, uint1
}
else if (*(state->buf) == '\0')
return (state->count) ? ERR : END;
else if (*(state->buf) != ' ')
else if (charlen==1 && !t_iseq(state->buf, ' '))
return ERR;
break;
default:
return ERR;
break;
}
(state->buf)++;
state->buf += charlen;
}
return END;
}

View File

@ -1,7 +1,7 @@
/*
* txtquery operations with ltree
* Teodor Sigaev <teodor@stack.net>
* $PostgreSQL: pgsql/contrib/ltree/ltxtquery_op.c,v 1.8 2008/05/12 00:00:43 alvherre Exp $
* $PostgreSQL: pgsql/contrib/ltree/ltxtquery_op.c,v 1.9 2008/06/30 18:30:48 teodor Exp $
*/
#include "postgres.h"
@ -57,7 +57,7 @@ checkcondition_str(void *checkval, ITEM * val)
char *op = ((CHKVAL *) checkval)->operand + val->distance;
int (*cmpptr) (const char *, const char *, size_t);
cmpptr = (val->flag & LVAR_INCASE) ? pg_strncasecmp : strncmp;
cmpptr = (val->flag & LVAR_INCASE) ? ltree_strncasecmp : strncmp;
while (tlen > 0)
{
if (val->flag & LVAR_SUBLEXEME)