1berry/MariaDB-server
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge abarkov@build.mysql.com:/home/bk/mysql-4.1

Browse Source 
into bar.mysql.r18.ru:/usr/home/bar/mysql-4.1


sql/sql_lex.h:
  Auto merged
This commit is contained in:
unknown 2003-03-14 18:11:20 +04:00
commit 1aeec73192
19 changed files with 250 additions and 192 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
include/m_ctype.h
View File

@ -74,6 +74,22 @@ typedef struct my_uni_idx_st
} MY_UNI_IDX;
enum my_lex_states
{
MY_LEX_START, MY_LEX_CHAR, MY_LEX_IDENT,
MY_LEX_IDENT_SEP, MY_LEX_IDENT_START,
MY_LEX_FOUND_IDENT, MY_LEX_SIGNED_NUMBER, MY_LEX_REAL, MY_LEX_HEX_NUMBER,
MY_LEX_CMP_OP, MY_LEX_LONG_CMP_OP, MY_LEX_STRING, MY_LEX_COMMENT, MY_LEX_END,
MY_LEX_OPERATOR_OR_IDENT, MY_LEX_NUMBER_IDENT, MY_LEX_INT_OR_REAL,
MY_LEX_REAL_OR_POINT, MY_LEX_BOOL, MY_LEX_EOL, MY_LEX_ESCAPE,
MY_LEX_LONG_COMMENT, MY_LEX_END_LONG_COMMENT, MY_LEX_COLON,
MY_LEX_SET_VAR, MY_LEX_USER_END, MY_LEX_HOSTNAME, MY_LEX_SKIP,
MY_LEX_USER_VARIABLE_DELIMITER, MY_LEX_SYSTEM_VAR,
MY_LEX_IDENT_OR_KEYWORD, MY_LEX_IDENT_OR_HEX, MY_LEX_IDENT_OR_BIN,
MY_LEX_STRING_OR_DELIMITER
};
typedef struct charset_info_st
{
uint number;
@ -89,6 +105,8 @@ typedef struct charset_info_st
uchar *sort_order;
uint16 *tab_to_uni;
MY_UNI_IDX *tab_from_uni;
uchar state_map[256];
uchar ident_map[256];
/* Collation routines */
uint strxfrm_multiply;

64
mysys/charset.c
View File

@ -29,7 +29,7 @@
- Initializing charset related structures
- Loading dynamic charsets
- Searching for a proper CHARSET_INFO
using charset name, collation name or collatio ID
using charset name, collation name or collation ID
- Setting server default character set
*/
@ -54,6 +54,62 @@ static void set_max_sort_char(CHARSET_INFO *cs)
}
static void init_state_maps(CHARSET_INFO *cs)
{
uint i;
uchar *state_map= cs->state_map;
uchar *ident_map= cs->ident_map;
/* Fill state_map with states to get a faster parser */
for (i=0; i < 256 ; i++)
{
if (my_isalpha(cs,i))
state_map[i]=(uchar) MY_LEX_IDENT;
else if (my_isdigit(cs,i))
state_map[i]=(uchar) MY_LEX_NUMBER_IDENT;
#if defined(USE_MB) && defined(USE_MB_IDENT)
else if (use_mb(cs) && my_ismbhead(cs, i))
state_map[i]=(uchar) MY_LEX_IDENT;
#endif
else if (!my_isgraph(cs,i))
state_map[i]=(uchar) MY_LEX_SKIP;
else
state_map[i]=(uchar) MY_LEX_CHAR;
}
state_map[(uchar)'_']=state_map[(uchar)'$']=(uchar) MY_LEX_IDENT;
state_map[(uchar)'\'']=(uchar) MY_LEX_STRING;
state_map[(uchar)'-']=state_map[(uchar)'+']=(uchar) MY_LEX_SIGNED_NUMBER;
state_map[(uchar)'.']=(uchar) MY_LEX_REAL_OR_POINT;
state_map[(uchar)'>']=state_map[(uchar)'=']=state_map[(uchar)'!']= (uchar) MY_LEX_CMP_OP;
state_map[(uchar)'<']= (uchar) MY_LEX_LONG_CMP_OP;
state_map[(uchar)'&']=state_map[(uchar)'|']=(uchar) MY_LEX_BOOL;
state_map[(uchar)'#']=(uchar) MY_LEX_COMMENT;
state_map[(uchar)';']=(uchar) MY_LEX_COLON;
state_map[(uchar)':']=(uchar) MY_LEX_SET_VAR;
state_map[0]=(uchar) MY_LEX_EOL;
state_map[(uchar)'\\']= (uchar) MY_LEX_ESCAPE;
state_map[(uchar)'/']= (uchar) MY_LEX_LONG_COMMENT;
state_map[(uchar)'*']= (uchar) MY_LEX_END_LONG_COMMENT;
state_map[(uchar)'@']= (uchar) MY_LEX_USER_END;
state_map[(uchar) '`']= (uchar) MY_LEX_USER_VARIABLE_DELIMITER;
state_map[(uchar)'"']= (uchar) MY_LEX_STRING_OR_DELIMITER;
/*
Create a second map to make it faster to find identifiers
*/
for (i=0; i < 256 ; i++)
{
ident_map[i]= (uchar) (state_map[i] == MY_LEX_IDENT ||
state_map[i] == MY_LEX_NUMBER_IDENT);
}
/* Special handling of hex and binary strings */
state_map[(uchar)'x']= state_map[(uchar)'X']= (uchar) MY_LEX_IDENT_OR_HEX;
state_map[(uchar)'b']= state_map[(uchar)'b']= (uchar) MY_LEX_IDENT_OR_BIN;
}
static void simple_cs_init_functions(CHARSET_INFO *cs)
{
@ -211,8 +267,11 @@ static void simple_cs_copy_data(CHARSET_INFO *to, CHARSET_INFO *from)
to->name= my_once_strdup(from->name,MYF(MY_WME));
if (from->ctype)
{
to->ctype= (uchar*) my_once_memdup((char*) from->ctype,
MY_CS_CTYPE_TABLE_SIZE, MYF(MY_WME));
init_state_maps(to);
}
if (from->to_lower)
to->to_lower= (uchar*) my_once_memdup((char*) from->to_lower,
MY_CS_TO_LOWER_TABLE_SIZE, MYF(MY_WME));
@ -447,7 +506,10 @@ static my_bool init_available_charsets(myf myflags)
for (cs=all_charsets; cs < all_charsets+255 ; cs++)
{
if (*cs)
{
set_max_sort_char(*cs);
init_state_maps(*cs);
}
}
strmov(get_charsets_dir(fname), MY_CHARSET_INDEX);

301
sql/sql_lex.cc
View File

@ -75,8 +75,6 @@ inline int lex_casecmp(const char *s, const char *t, uint len)
#include "lex_hash.h"
static uchar state_map[256], ident_map[256];
void lex_init(void)
{
@ -89,53 +87,6 @@ void lex_init(void)
VOID(pthread_key_create(&THR_LEX,NULL));
/* Fill state_map with states to get a faster parser */
for (i=0; i < sizeof(state_map) ; i++)
{
if (my_isalpha(system_charset_info,i))
state_map[i]=(uchar) STATE_IDENT;
else if (my_isdigit(system_charset_info,i))
state_map[i]=(uchar) STATE_NUMBER_IDENT;
#if defined(USE_MB) && defined(USE_MB_IDENT)
else if (use_mb(system_charset_info) && my_ismbhead(system_charset_info, i))
state_map[i]=(uchar) STATE_IDENT;
#endif
else if (!my_isgraph(system_charset_info,i))
state_map[i]=(uchar) STATE_SKIP;
else
state_map[i]=(uchar) STATE_CHAR;
}
state_map[(uchar)'_']=state_map[(uchar)'$']=(uchar) STATE_IDENT;
state_map[(uchar)'\'']=(uchar) STATE_STRING;
state_map[(uchar)'-']=state_map[(uchar)'+']=(uchar) STATE_SIGNED_NUMBER;
state_map[(uchar)'.']=(uchar) STATE_REAL_OR_POINT;
state_map[(uchar)'>']=state_map[(uchar)'=']=state_map[(uchar)'!']= (uchar) STATE_CMP_OP;
state_map[(uchar)'<']= (uchar) STATE_LONG_CMP_OP;
state_map[(uchar)'&']=state_map[(uchar)'|']=(uchar) STATE_BOOL;
state_map[(uchar)'#']=(uchar) STATE_COMMENT;
state_map[(uchar)';']=(uchar) STATE_COLON;
state_map[(uchar)':']=(uchar) STATE_SET_VAR;
state_map[0]=(uchar) STATE_EOL;
state_map[(uchar)'\\']= (uchar) STATE_ESCAPE;
state_map[(uchar)'/']= (uchar) STATE_LONG_COMMENT;
state_map[(uchar)'*']= (uchar) STATE_END_LONG_COMMENT;
state_map[(uchar)'@']= (uchar) STATE_USER_END;
state_map[(uchar) '`']= (uchar) STATE_USER_VARIABLE_DELIMITER;
state_map[(uchar)'"']= (uchar) STAT_STRING_OR_DELIMITER;
/*
Create a second map to make it faster to find identifiers
*/
for (i=0; i < sizeof(ident_map) ; i++)
{
ident_map[i]= (uchar) (state_map[i] == STATE_IDENT ||
state_map[i] == STATE_NUMBER_IDENT);
}
/* Special handling of hex and binary strings */
state_map[(uchar)'x']= state_map[(uchar)'X']= (uchar) STATE_IDENT_OR_HEX;
state_map[(uchar)'b']= state_map[(uchar)'b']= (uchar) STATE_IDENT_OR_BIN;
DBUG_VOID_RETURN;
}
@ -156,7 +107,7 @@ void lex_free(void)
LEX *lex_start(THD *thd, uchar *buf,uint length)
{
LEX *lex= &thd->lex;
lex->next_state=STATE_START;
lex->next_state=MY_LEX_START;
lex->end_of_query=(lex->ptr=buf)+length;
lex->yylineno = 1;
lex->select_lex.create_refs=lex->in_comment=0;
@ -458,8 +409,8 @@ inline static uint int_token(const char *str,uint length)
// yylex remember the following states from the following yylex()
// STATE_EOQ ; found end of query
// STATE_OPERATOR_OR_IDENT ; last state was an ident, text or number
// MY_LEX_EOQ ; found end of query
// MY_LEX_OPERATOR_OR_IDENT ; last state was an ident, text or number
// (which can't be followed by a signed number)
int yylex(void *arg, void *yythd)
@ -467,76 +418,79 @@ int yylex(void *arg, void *yythd)
reg1 uchar c;
int tokval;
uint length;
enum lex_states state,prev_state;
enum my_lex_states state,prev_state;
LEX *lex= &(((THD *)yythd)->lex);
YYSTYPE *yylval=(YYSTYPE*) arg;
CHARSET_INFO *cs= ((THD *) yythd)->variables.thd_charset;
uchar *state_map= cs->state_map;
uchar *ident_map= cs->ident_map;
lex->yylval=yylval; // The global state
lex->tok_start=lex->tok_end=lex->ptr;
prev_state=state=lex->next_state;
lex->next_state=STATE_OPERATOR_OR_IDENT;
lex->next_state=MY_LEX_OPERATOR_OR_IDENT;
LINT_INIT(c);
for (;;)
{
switch (state) {
case STATE_OPERATOR_OR_IDENT: // Next is operator or keyword
case STATE_START: // Start of token
case MY_LEX_OPERATOR_OR_IDENT: // Next is operator or keyword
case MY_LEX_START: // Start of token
// Skip startspace
for (c=yyGet() ; (state_map[c] == STATE_SKIP) ; c= yyGet())
for (c=yyGet() ; (state_map[c] == MY_LEX_SKIP) ; c= yyGet())
{
if (c == '\n')
lex->yylineno++;
}
lex->tok_start=lex->ptr-1; // Start of real token
state= (enum lex_states) state_map[c];
state= (enum my_lex_states) state_map[c];
break;
case STATE_ESCAPE:
case MY_LEX_ESCAPE:
if (yyGet() == 'N')
{ // Allow \N as shortcut for NULL
yylval->lex_str.str=(char*) "\\N";
yylval->lex_str.length=2;
return NULL_SYM;
}
case STATE_CHAR: // Unknown or single char token
case STATE_SKIP: // This should not happen
case MY_LEX_CHAR: // Unknown or single char token
case MY_LEX_SKIP: // This should not happen
yylval->lex_str.str=(char*) (lex->ptr=lex->tok_start);// Set to first chr
yylval->lex_str.length=1;
c=yyGet();
if (c != ')')
lex->next_state= STATE_START; // Allow signed numbers
lex->next_state= MY_LEX_START; // Allow signed numbers
if (c == ',')
lex->tok_start=lex->ptr; // Let tok_start point at next item
return((int) c);
case STATE_IDENT_OR_HEX:
case MY_LEX_IDENT_OR_HEX:
if (yyPeek() == '\'')
{ // Found x'hex-number'
state= STATE_HEX_NUMBER;
state= MY_LEX_HEX_NUMBER;
break;
}
/* Fall through */
case STATE_IDENT_OR_BIN: // TODO: Add binary string handling
case STATE_IDENT:
case MY_LEX_IDENT_OR_BIN: // TODO: Add binary string handling
case MY_LEX_IDENT:
#if defined(USE_MB) && defined(USE_MB_IDENT)
if (use_mb(system_charset_info))
if (use_mb(cs))
{
if (my_ismbhead(system_charset_info, yyGetLast()))
if (my_ismbhead(cs, yyGetLast()))
{
int l = my_ismbchar(system_charset_info,
int l = my_ismbchar(cs,
(const char *)lex->ptr-1,
(const char *)lex->end_of_query);
if (l == 0) {
state = STATE_CHAR;
state = MY_LEX_CHAR;
continue;
}
lex->ptr += l - 1;
}
while (ident_map[c=yyGet()])
{
if (my_ismbhead(system_charset_info, c))
if (my_ismbhead(cs, c))
{
int l;
if ((l = my_ismbchar(system_charset_info,
if ((l = my_ismbchar(cs,
(const char *)lex->ptr-1,
(const char *)lex->end_of_query)) == 0)
break;
@ -550,16 +504,16 @@ int yylex(void *arg, void *yythd)
length= (uint) (lex->ptr - lex->tok_start)-1;
if (lex->ignore_space)
{
for (; state_map[c] == STATE_SKIP ; c= yyGet());
for (; state_map[c] == MY_LEX_SKIP ; c= yyGet());
}
if (c == '.' && ident_map[yyPeek()])
lex->next_state=STATE_IDENT_SEP;
lex->next_state=MY_LEX_IDENT_SEP;
else
{ // '(' must follow directly if function
yyUnget();
if ((tokval = find_keyword(lex,length,c == '(')))
{
lex->next_state= STATE_START; // Allow signed numbers
lex->next_state= MY_LEX_START; // Allow signed numbers
return(tokval); // Was keyword
}
yySkip(); // next state does a unget
@ -582,30 +536,30 @@ int yylex(void *arg, void *yythd)
else
return(IDENT);
case STATE_IDENT_SEP: // Found ident and now '.'
lex->next_state=STATE_IDENT_START;// Next is an ident (not a keyword)
case MY_LEX_IDENT_SEP: // Found ident and now '.'
lex->next_state=MY_LEX_IDENT_START;// Next is an ident (not a keyword)
yylval->lex_str.str=(char*) lex->ptr;
yylval->lex_str.length=1;
c=yyGet(); // should be '.'
return((int) c);
case STATE_NUMBER_IDENT: // number or ident which num-start
while (my_isdigit(system_charset_info,(c = yyGet()))) ;
case MY_LEX_NUMBER_IDENT: // number or ident which num-start
while (my_isdigit(cs,(c = yyGet()))) ;
if (!ident_map[c])
{ // Can't be identifier
state=STATE_INT_OR_REAL;
state=MY_LEX_INT_OR_REAL;
break;
}
if (c == 'e' || c == 'E')
{
// The following test is written this way to allow numbers of type 1e1
if (my_isdigit(system_charset_info,yyPeek()) ||
if (my_isdigit(cs,yyPeek()) ||
(c=(yyGet())) == '+' || c == '-')
{ // Allow 1E+10
if (my_isdigit(system_charset_info,yyPeek())) // Number must have digit after sign
if (my_isdigit(cs,yyPeek())) // Number must have digit after sign
{
yySkip();
while (my_isdigit(system_charset_info,yyGet())) ;
while (my_isdigit(cs,yyGet())) ;
yylval->lex_str=get_token(lex,yyLength());
return(FLOAT_NUM);
}
@ -615,7 +569,7 @@ int yylex(void *arg, void *yythd)
else if (c == 'x' && (lex->ptr - lex->tok_start) == 2 &&
lex->tok_start[0] == '0' )
{ // Varbinary
while (my_isxdigit(system_charset_info,(c = yyGet()))) ;
while (my_isxdigit(cs,(c = yyGet()))) ;
if ((lex->ptr - lex->tok_start) >= 4 && !ident_map[c])
{
yylval->lex_str=get_token(lex,yyLength());
@ -627,28 +581,28 @@ int yylex(void *arg, void *yythd)
yyUnget();
}
// fall through
case STATE_IDENT_START: // Incomplete ident
case MY_LEX_IDENT_START: // Incomplete ident
#if defined(USE_MB) && defined(USE_MB_IDENT)
if (use_mb(system_charset_info))
if (use_mb(cs))
{
if (my_ismbhead(system_charset_info, yyGetLast()))
if (my_ismbhead(cs, yyGetLast()))
{
int l = my_ismbchar(system_charset_info,
int l = my_ismbchar(cs,
(const char *)lex->ptr-1,
(const char *)lex->end_of_query);
if (l == 0)
{
state = STATE_CHAR;
state = MY_LEX_CHAR;
continue;
}
lex->ptr += l - 1;
}
while (ident_map[c=yyGet()])
{
if (my_ismbhead(system_charset_info, c))
if (my_ismbhead(cs, c))
{
int l;
if ((l = my_ismbchar(system_charset_info,
if ((l = my_ismbchar(cs,
(const char *)lex->ptr-1,
(const char *)lex->end_of_query)) == 0)
break;
@ -661,28 +615,28 @@ int yylex(void *arg, void *yythd)
while (ident_map[c = yyGet()]) ;
if (c == '.' && ident_map[yyPeek()])
lex->next_state=STATE_IDENT_SEP;// Next is '.'
lex->next_state=MY_LEX_IDENT_SEP;// Next is '.'
// fall through
case STATE_FOUND_IDENT: // Complete ident
case MY_LEX_FOUND_IDENT: // Complete ident
yylval->lex_str=get_token(lex,yyLength());
if (lex->convert_set)
lex->convert_set->convert((char*) yylval->lex_str.str,lex->yytoklen);
return(IDENT);
case STATE_USER_VARIABLE_DELIMITER:
case MY_LEX_USER_VARIABLE_DELIMITER:
{
char delim= c; // Used char
lex->tok_start=lex->ptr; // Skip first `
#ifdef USE_MB
if (use_mb(system_charset_info))
if (use_mb(cs))
{
while ((c=yyGet()) && c != delim && c != (uchar) NAMES_SEP_CHAR)
{
if (my_ismbhead(system_charset_info, c))
if (my_ismbhead(cs, c))
{
int l;
if ((l = my_ismbchar(system_charset_info,
if ((l = my_ismbchar(cs,
(const char *)lex->ptr-1,
(const char *)lex->end_of_query)) == 0)
break;
@ -721,67 +675,67 @@ int yylex(void *arg, void *yythd)
yySkip(); // Skip end `
return(IDENT);
}
case STATE_SIGNED_NUMBER: // Incomplete signed number
if (prev_state == STATE_OPERATOR_OR_IDENT)
case MY_LEX_SIGNED_NUMBER: // Incomplete signed number
if (prev_state == MY_LEX_OPERATOR_OR_IDENT)
{
if (c == '-' && yyPeek() == '-' &&
(my_isspace(system_charset_info,yyPeek2()) ||
my_iscntrl(system_charset_info,yyPeek2())))
state=STATE_COMMENT;
(my_isspace(cs,yyPeek2()) ||
my_iscntrl(cs,yyPeek2())))
state=MY_LEX_COMMENT;
else
state= STATE_CHAR; // Must be operator
state= MY_LEX_CHAR; // Must be operator
break;
}
if (!my_isdigit(system_charset_info,c=yyGet()) || yyPeek() == 'x')
if (!my_isdigit(cs,c=yyGet()) || yyPeek() == 'x')
{
if (c != '.')
{
if (c == '-' && my_isspace(system_charset_info,yyPeek()))
state=STATE_COMMENT;
if (c == '-' && my_isspace(cs,yyPeek()))
state=MY_LEX_COMMENT;
else
state = STATE_CHAR; // Return sign as single char
state = MY_LEX_CHAR; // Return sign as single char
break;
}
yyUnget(); // Fix for next loop
}
while (my_isdigit(system_charset_info,c=yyGet())) ; // Incomplete real or int number
while (my_isdigit(cs,c=yyGet())) ; // Incomplete real or int number
if ((c == 'e' || c == 'E') &&
(yyPeek() == '+' || yyPeek() == '-' || my_isdigit(system_charset_info,yyPeek())))
(yyPeek() == '+' || yyPeek() == '-' || my_isdigit(cs,yyPeek())))
{ // Real number
yyUnget();
c= '.'; // Fool next test
}
// fall through
case STATE_INT_OR_REAL: // Compleat int or incompleat real
case MY_LEX_INT_OR_REAL: // Compleat int or incompleat real
if (c != '.')
{ // Found complete integer number.
yylval->lex_str=get_token(lex,yyLength());
return int_token(yylval->lex_str.str,yylval->lex_str.length);
}
// fall through
case STATE_REAL: // Incomplete real number
while (my_isdigit(system_charset_info,c = yyGet())) ;
case MY_LEX_REAL: // Incomplete real number
while (my_isdigit(cs,c = yyGet())) ;
if (c == 'e' || c == 'E')
{
c = yyGet();
if (c == '-' || c == '+')
c = yyGet(); // Skip sign
if (!my_isdigit(system_charset_info,c))
if (!my_isdigit(cs,c))
{ // No digit after sign
state= STATE_CHAR;
state= MY_LEX_CHAR;
break;
}
while (my_isdigit(system_charset_info,yyGet())) ;
while (my_isdigit(cs,yyGet())) ;
yylval->lex_str=get_token(lex,yyLength());
return(FLOAT_NUM);
}
yylval->lex_str=get_token(lex,yyLength());
return(REAL_NUM);
case STATE_HEX_NUMBER: // Found x'hexstring'
case MY_LEX_HEX_NUMBER: // Found x'hexstring'
yyGet(); // Skip '
while (my_isxdigit(system_charset_info,(c = yyGet()))) ;
while (my_isxdigit(cs,(c = yyGet()))) ;
length=(lex->ptr - lex->tok_start); // Length of hexnum+3
if (!(length & 1) || c != '\'')
{
@ -794,56 +748,56 @@ int yylex(void *arg, void *yythd)
lex->yytoklen-=3;
return (HEX_NUM);
case STATE_CMP_OP: // Incomplete comparison operator
if (state_map[yyPeek()] == STATE_CMP_OP ||
state_map[yyPeek()] == STATE_LONG_CMP_OP)
case MY_LEX_CMP_OP: // Incomplete comparison operator
if (state_map[yyPeek()] == MY_LEX_CMP_OP ||
state_map[yyPeek()] == MY_LEX_LONG_CMP_OP)
yySkip();
if ((tokval = find_keyword(lex,(uint) (lex->ptr - lex->tok_start),0)))
{
lex->next_state= STATE_START; // Allow signed numbers
lex->next_state= MY_LEX_START; // Allow signed numbers
return(tokval);
}
state = STATE_CHAR; // Something fishy found
state = MY_LEX_CHAR; // Something fishy found
break;
case STATE_LONG_CMP_OP: // Incomplete comparison operator
if (state_map[yyPeek()] == STATE_CMP_OP ||
state_map[yyPeek()] == STATE_LONG_CMP_OP)
case MY_LEX_LONG_CMP_OP: // Incomplete comparison operator
if (state_map[yyPeek()] == MY_LEX_CMP_OP ||
state_map[yyPeek()] == MY_LEX_LONG_CMP_OP)
{
yySkip();
if (state_map[yyPeek()] == STATE_CMP_OP)
if (state_map[yyPeek()] == MY_LEX_CMP_OP)
yySkip();
}
if ((tokval = find_keyword(lex,(uint) (lex->ptr - lex->tok_start),0)))
{
lex->next_state= STATE_START; // Found long op
lex->next_state= MY_LEX_START; // Found long op
return(tokval);
}
state = STATE_CHAR; // Something fishy found
state = MY_LEX_CHAR; // Something fishy found
break;
case STATE_BOOL:
case MY_LEX_BOOL:
if (c != yyPeek())
{
state=STATE_CHAR;
state=MY_LEX_CHAR;
break;
}
yySkip();
tokval = find_keyword(lex,2,0); // Is a bool operator
lex->next_state= STATE_START; // Allow signed numbers
lex->next_state= MY_LEX_START; // Allow signed numbers
return(tokval);
case STAT_STRING_OR_DELIMITER:
case MY_LEX_STRING_OR_DELIMITER:
if (((THD *) yythd)->variables.sql_mode & MODE_ANSI_QUOTES)
{
state= STATE_USER_VARIABLE_DELIMITER;
state= MY_LEX_USER_VARIABLE_DELIMITER;
break;
}
/* " used for strings */
case STATE_STRING: // Incomplete text string
case MY_LEX_STRING: // Incomplete text string
if (!(yylval->lex_str.str = get_text(lex)))
{
state= STATE_CHAR; // Read char by char
state= MY_LEX_CHAR; // Read char by char
break;
}
yylval->lex_str.length=lex->yytoklen;
@ -851,16 +805,16 @@ int yylex(void *arg, void *yythd)
lex->convert_set->convert((char*) yylval->lex_str.str,lex->yytoklen);
return(TEXT_STRING);
case STATE_COMMENT: // Comment
case MY_LEX_COMMENT: // Comment
lex->select_lex.options|= OPTION_FOUND_COMMENT;
while ((c = yyGet()) != '\n' && c) ;
yyUnget(); // Safety against eof
state = STATE_START; // Try again
state = MY_LEX_START; // Try again
break;
case STATE_LONG_COMMENT: /* Long C comment? */
case MY_LEX_LONG_COMMENT: /* Long C comment? */
if (yyPeek() != '*')
{
state=STATE_CHAR; // Probable division
state=MY_LEX_CHAR; // Probable division
break;
}
yySkip(); // Skip '*'
@ -869,8 +823,8 @@ int yylex(void *arg, void *yythd)
{
ulong version=MYSQL_VERSION_ID;
yySkip();
state=STATE_START;
if (my_isdigit(system_charset_info,yyPeek()))
state=MY_LEX_START;
if (my_isdigit(cs,yyPeek()))
{ // Version number
version=strtol((char*) lex->ptr,(char**) &lex->ptr,10);
}
@ -888,88 +842,87 @@ int yylex(void *arg, void *yythd)
}
if (lex->ptr != lex->end_of_query)
yySkip(); // remove last '/'
state = STATE_START; // Try again
state = MY_LEX_START; // Try again
break;
case STATE_END_LONG_COMMENT:
case MY_LEX_END_LONG_COMMENT:
if (lex->in_comment && yyPeek() == '/')
{
yySkip();
lex->in_comment=0;
state=STATE_START;
state=MY_LEX_START;
}
else
state=STATE_CHAR; // Return '*'
state=MY_LEX_CHAR; // Return '*'
break;
case STATE_SET_VAR: // Check if ':='
case MY_LEX_SET_VAR: // Check if ':='
if (yyPeek() != '=')
{
state=STATE_CHAR; // Return ':'
state=MY_LEX_CHAR; // Return ':'
break;
}
yySkip();
return (SET_VAR);
case STATE_COLON: // optional line terminator
case MY_LEX_COLON: // optional line terminator
if (yyPeek())
{
if (((THD *)yythd)->client_capabilities & CLIENT_MULTI_QUERIES)
{
lex->found_colon=(char*)lex->ptr;
((THD *)yythd)->server_status |= SERVER_MORE_RESULTS_EXISTS;
lex->next_state=STATE_END;
lex->next_state=MY_LEX_END;
return(END_OF_INPUT);
}
else
state=STATE_CHAR; // Return ';'
state=MY_LEX_CHAR; // Return ';'
break;
}
/* fall true */
case STATE_EOL:
lex->next_state=STATE_END; // Mark for next loop
case MY_LEX_EOL:
lex->next_state=MY_LEX_END; // Mark for next loop
return(END_OF_INPUT);
case STATE_END:
lex->next_state=STATE_END;
case MY_LEX_END:
lex->next_state=MY_LEX_END;
return(0); // We found end of input last time
/* Actually real shouldn't start with . but allow them anyhow */
case STATE_REAL_OR_POINT:
if (my_isdigit(system_charset_info,yyPeek()))
state = STATE_REAL; // Real
case MY_LEX_REAL_OR_POINT:
if (my_isdigit(cs,yyPeek()))
state = MY_LEX_REAL; // Real
else
{
state = STATE_CHAR; // return '.'
lex->next_state=STATE_IDENT_START;// Next is an ident (not a keyword)
state = MY_LEX_CHAR; // return '.'
lex->next_state=MY_LEX_IDENT_START;// Next is an ident (not a keyword)
}
break;
case STATE_USER_END: // end '@' of user@hostname
case MY_LEX_USER_END: // end '@' of user@hostname
switch (state_map[yyPeek()]) {
case STATE_STRING:
case STATE_USER_VARIABLE_DELIMITER:
case STAT_STRING_OR_DELIMITER:
case MY_LEX_STRING:
case MY_LEX_USER_VARIABLE_DELIMITER:
case MY_LEX_STRING_OR_DELIMITER:
break;
case STATE_USER_END:
lex->next_state=STATE_SYSTEM_VAR;
case MY_LEX_USER_END:
lex->next_state=MY_LEX_SYSTEM_VAR;
break;
default:
lex->next_state=STATE_HOSTNAME;
lex->next_state=MY_LEX_HOSTNAME;
break;
}
yylval->lex_str.str=(char*) lex->ptr;
yylval->lex_str.length=1;
return((int) '@');
case STATE_HOSTNAME: // end '@' of user@hostname
for (c=yyGet() ;
my_isalnum(system_charset_info,c) || c == '.' || c == '_' ||
c == '$';
case MY_LEX_HOSTNAME: // end '@' of user@hostname
for (c=yyGet() ;
my_isalnum(cs,c) || c == '.' || c == '_' || c == '$';
c= yyGet()) ;
yylval->lex_str=get_token(lex,yyLength());
return(LEX_HOSTNAME);
case STATE_SYSTEM_VAR:
case MY_LEX_SYSTEM_VAR:
yylval->lex_str.str=(char*) lex->ptr;
yylval->lex_str.length=1;
lex->next_state=STATE_IDENT_OR_KEYWORD;
lex->next_state=MY_LEX_IDENT_OR_KEYWORD;
yySkip(); // Skip '@'
return((int) '@');
case STATE_IDENT_OR_KEYWORD:
case MY_LEX_IDENT_OR_KEYWORD:
/*
We come here when we have found two '@' in a row.
We should now be able to handle:
@ -978,7 +931,7 @@ int yylex(void *arg, void *yythd)
while (ident_map[c=yyGet()]) ;
if (c == '.')
lex->next_state=STATE_IDENT_SEP;
lex->next_state=MY_LEX_IDENT_SEP;
length= (uint) (lex->ptr - lex->tok_start)-1;
if ((tokval= find_keyword(lex,length,0)))
{

15
sql/sql_lex.h
View File

@ -77,19 +77,6 @@ enum enum_sql_command {
SQLCOM_END
};
enum lex_states
{
STATE_START, STATE_CHAR, STATE_IDENT, STATE_IDENT_SEP, STATE_IDENT_START,
STATE_FOUND_IDENT, STATE_SIGNED_NUMBER, STATE_REAL, STATE_HEX_NUMBER,
STATE_CMP_OP, STATE_LONG_CMP_OP, STATE_STRING, STATE_COMMENT, STATE_END,
STATE_OPERATOR_OR_IDENT, STATE_NUMBER_IDENT, STATE_INT_OR_REAL,
STATE_REAL_OR_POINT, STATE_BOOL, STATE_EOL, STATE_ESCAPE, STATE_LONG_COMMENT,
STATE_END_LONG_COMMENT, STATE_COLON, STATE_SET_VAR, STATE_USER_END,
STATE_HOSTNAME, STATE_SKIP, STATE_USER_VARIABLE_DELIMITER, STATE_SYSTEM_VAR,
STATE_IDENT_OR_KEYWORD, STATE_IDENT_OR_HEX, STATE_IDENT_OR_BIN,
STAT_STRING_OR_DELIMITER
};
typedef List<Item> List_item;
@ -474,7 +461,7 @@ typedef struct st_lex
ulong thread_id,type;
enum_sql_command sql_command;
thr_lock_type lock_option;
enum lex_states next_state;
enum my_lex_states next_state;
enum enum_duplicates duplicates;
enum enum_tx_isolation tx_isolation;
enum enum_ha_read_modes ha_read_mode;

6
sql/sql_yacc.yy
View File

@ -3867,7 +3867,7 @@ literal:
| REAL_NUM { $$ = new Item_real($1.str, $1.length); }
| FLOAT_NUM { $$ = new Item_float($1.str, $1.length); }
| NULL_SYM { $$ = new Item_null();
Lex->next_state=STATE_OPERATOR_OR_IDENT;}
Lex->next_state=MY_LEX_OPERATOR_OR_IDENT;}
| HEX_NUM { $$ = new Item_varbinary($1.str,$1.length);}
| DATE_SYM text_literal { $$ = $2; }
| TIME_SYM text_literal { $$ = $2; }
@ -3964,8 +3964,8 @@ ident:
LEX *lex= Lex;
$$.str= lex->thd->strmake($1.str,$1.length);
$$.length=$1.length;
if (lex->next_state != STATE_END)
lex->next_state=STATE_OPERATOR_OR_IDENT;
if (lex->next_state != MY_LEX_END)
lex->next_state= MY_LEX_OPERATOR_OR_IDENT;
}
;

1
strings/ctype-big5.c
View File

@ -6244,6 +6244,7 @@ CHARSET_INFO my_charset_big5 =
sort_order_big5,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
1, /* strxfrm_multiply */
my_strnncoll_big5,
my_strnncollsp_big5,

1
strings/ctype-bin.c
View File

@ -306,6 +306,7 @@ CHARSET_INFO my_charset_bin =
bin_char_array, /* sort_order */
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_binary, /* strnncoll */
my_strnncoll_binary,

1
strings/ctype-czech.c
View File

@ -618,6 +618,7 @@ CHARSET_INFO my_charset_czech =
sort_order_czech,
tab_8859_2_uni, /* tab_to_uni */
idx_uni_8859_2, /* tab_from_uni */
"","",
4, /* strxfrm_multiply */
my_strnncoll_czech,
my_strnncollsp_czech,

1
strings/ctype-euc_kr.c
View File

@ -8652,6 +8652,7 @@ CHARSET_INFO my_charset_euc_kr =
sort_order_euc_kr,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,

24
strings/ctype-extra.c
View File

@ -2818,6 +2818,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_latin1,
tab_8859_1_uni, /* tab_to_uni */
idx_uni_8859_1, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -2869,6 +2870,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_cp1251,
tab_cp1251_uni, /* tab_to_uni */
idx_uni_cp1251, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -2919,6 +2921,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_cp1257,
tab_cp1257_uni, /* tab_to_uni */
idx_uni_cp1257, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -2969,6 +2972,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_croat,
tab_8859_2_uni, /* tab_to_uni */
idx_uni_8859_2, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3020,6 +3024,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_danish,
tab_8859_1_uni, /* tab_to_uni */
idx_uni_8859_1, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3070,6 +3075,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_dec8,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3120,6 +3126,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_dos,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3170,6 +3177,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_estonia,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3221,6 +3229,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_german1,
tab_8859_1_uni, /* tab_to_uni */
idx_uni_8859_1, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3271,6 +3280,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_greek,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3321,6 +3331,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_hebrew,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3371,6 +3382,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_hp8,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3421,6 +3433,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_hungarian,
tab_8859_2_uni, /* tab_to_uni */
idx_uni_8859_2, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3471,6 +3484,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_koi8_ru,
tab_koi8_r_uni, /* tab_to_uni */
idx_uni_koi8_r, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3521,6 +3535,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_koi8_ukr,
tab_koi8_u_uni, /* tab_to_uni */
idx_uni_koi8_u, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3572,6 +3587,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_latin2,
tab_8859_2_uni, /* tab_to_uni */
idx_uni_8859_2, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3622,6 +3638,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_latin5,
tab_8859_9_uni, /* tab_to_uni */
idx_uni_8859_9, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3673,6 +3690,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_swe7,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3724,6 +3742,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_usa7,
tab_us_ascii_uni, /* tab_to_uni */
idx_uni_us_ascii, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3774,6 +3793,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_win1250,
tab_cp1250_uni, /* tab_to_uni */
idx_uni_cp1250, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3824,6 +3844,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_win1251ukr,
tab_cp1251_uni, /* tab_to_uni */
idx_uni_cp1251, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3874,6 +3895,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_armscii8,
tab_armscii_8_uni, /* tab_to_uni */
idx_uni_armscii_8, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3924,6 +3946,7 @@ CHARSET_INFO compiled_charsets[] = {
sort_order_win1251,
tab_cp1251_uni, /* tab_to_uni */
idx_uni_cp1251, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,/* strnncollsp */
@ -3973,6 +3996,7 @@ CHARSET_INFO compiled_charsets[] = {
NULL,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0,
NULL, /* strnncoll */
NULL, /* strnncollsp */

1
strings/ctype-gb2312.c
View File

@ -5702,6 +5702,7 @@ CHARSET_INFO my_charset_gb2312 =
sort_order_gb2312,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,

1
strings/ctype-gbk.c
View File

@ -9899,6 +9899,7 @@ CHARSET_INFO my_charset_gbk =
sort_order_gbk,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
1, /* strxfrm_multiply */
my_strnncoll_gbk,
my_strnncollsp_gbk,

1
strings/ctype-latin1.c
View File

@ -188,6 +188,7 @@ CHARSET_INFO my_charset_latin1 =
sort_order_latin1,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
2, /* strxfrm_multiply */
my_strnncoll_simple,
my_strnncollsp_simple,

1
strings/ctype-latin1_de.c
View File

@ -359,6 +359,7 @@ CHARSET_INFO my_charset_latin1_de =
sort_order_latin1_de,
tab_8859_1_uni, /* tab_to_uni */
idx_uni_8859_1, /* tab_from_uni */
"","",
2, /* strxfrm_multiply */
my_strnncoll_latin1_de,
my_strnncollsp_latin1_de,

1
strings/ctype-sjis.c
View File

@ -4486,6 +4486,7 @@ CHARSET_INFO my_charset_sjis =
sort_order_sjis,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
1, /* strxfrm_multiply */
my_strnncoll_sjis,
my_strnncollsp_sjis,

1
strings/ctype-tis620.c
View File

@ -709,6 +709,7 @@ CHARSET_INFO my_charset_tis620 =
sort_order_tis620,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
4, /* strxfrm_multiply */
my_strnncoll_tis620,
my_strnncollsp_tis620,

1
strings/ctype-ujis.c
View File

@ -8443,6 +8443,7 @@ CHARSET_INFO my_charset_ujis =
sort_order_ujis,
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
0, /* strxfrm_multiply */
my_strnncoll_simple,/* strnncoll */
my_strnncollsp_simple,

2
strings/ctype-utf8.c
View File

@ -1988,6 +1988,7 @@ CHARSET_INFO my_charset_utf8 =
to_upper_utf8, /* sort_order */
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
1, /* strxfrm_multiply */
my_strnncoll_utf8, /* strnncoll */
my_strnncollsp_utf8,
@ -3095,6 +3096,7 @@ CHARSET_INFO my_charset_ucs2 =
to_upper_ucs2, /* sort_order */
NULL, /* tab_to_uni */
NULL, /* tab_from_uni */
"","",
1, /* strxfrm_multiply */
my_strnncoll_ucs2, /* strnncoll */
my_strnncoll_ucs2,

1
strings/ctype-win1250ch.c
View File

@ -653,6 +653,7 @@ CHARSET_INFO my_charset_win1250ch =
sort_order_win1250ch,
tab_cp1250_uni, /* tab_to_uni */
idx_uni_cp1250, /* tab_from_uni */
"","",
2, /* strxfrm_multiply */
my_strnncoll_win1250ch,
my_strnncollsp_win1250ch,