summaryrefslogtreecommitdiffstats
path: root/tqtinterface/qt4/src/3rdparty/sqlite/tokenize.c
diff options
context:
space:
mode:
Diffstat (limited to 'tqtinterface/qt4/src/3rdparty/sqlite/tokenize.c')
-rw-r--r--tqtinterface/qt4/src/3rdparty/sqlite/tokenize.c679
1 files changed, 679 insertions, 0 deletions
diff --git a/tqtinterface/qt4/src/3rdparty/sqlite/tokenize.c b/tqtinterface/qt4/src/3rdparty/sqlite/tokenize.c
new file mode 100644
index 0000000..3f3e3cf
--- /dev/null
+++ b/tqtinterface/qt4/src/3rdparty/sqlite/tokenize.c
@@ -0,0 +1,679 @@
+/*
+** 2001 September 15
+**
+** The author disclaims copyright to this source code. In place of
+** a legal notice, here is a blessing:
+**
+** May you do good and not evil.
+** May you tqfind forgiveness for yourself and forgive others.
+** May you share freely, never taking more than you give.
+**
+*************************************************************************
+** An tokenizer for SQL
+**
+** This file tqcontains C code that splits an SQL input string up into
+** individual tokens and sends those tokens one-by-one over to the
+** parser for analysis.
+**
+** $Id: tokenize.c,v 1.68 2004/02/14 23:59:58 drh Exp $
+*/
+#include "sqliteInt.h"
+#include "os.h"
+#include <ctype.h>
+#include <stdlib.h>
+
+/*
+** All the keywords of the SQL language are stored as in a hash
+** table composed of instances of the following structure.
+*/
+typedef struct Keyword Keyword;
+struct Keyword {
+ char *zName; /* The keyword name */
+ u8 tokenType; /* Token value for this keyword */
+ u8 len; /* Length of this keyword */
+ u8 iNext; /* Index in aKeywordTable[] of next with same hash */
+};
+
+/*
+** These are the keywords
+*/
+static Keyword aKeywordTable[] = {
+ { "ABORT", TK_ABORT, },
+ { "AFTER", TK_AFTER, },
+ { "ALL", TK_ALL, },
+ { "AND", TK_AND, },
+ { "AS", TK_AS, },
+ { "ASC", TK_ASC, },
+ { "ATTACH", TK_ATTACH, },
+ { "BEFORE", TK_BEFORE, },
+ { "BEGIN", TK_BEGIN, },
+ { "BETWEEN", TK_BETWEEN, },
+ { "BY", TK_BY, },
+ { "CASCADE", TK_CASCADE, },
+ { "CASE", TK_CASE, },
+ { "CHECK", TK_CHECK, },
+ { "CLUSTER", TK_CLUSTER, },
+ { "COLLATE", TK_COLLATE, },
+ { "COMMIT", TK_COMMIT, },
+ { "CONFLICT", TK_CONFLICT, },
+ { "CONSTRAINT", TK_CONSTRAINT, },
+ { "COPY", TK_COPY, },
+ { "CREATE", TK_CREATE, },
+ { "CROSS", TK_JOIN_KW, },
+ { "DATABASE", TK_DATABASE, },
+ { "DEFAULT", TK_DEFAULT, },
+ { "DEFERRED", TK_DEFERRED, },
+ { "DEFERRABLE", TK_DEFERRABLE, },
+ { "DELETE", TK_DELETE, },
+ { "DELIMITERS", TK_DELIMITERS, },
+ { "DESC", TK_DESC, },
+ { "DETACH", TK_DETACH, },
+ { "DISTINCT", TK_DISTINCT, },
+ { "DROP", TK_DROP, },
+ { "END", TK_END, },
+ { "EACH", TK_EACH, },
+ { "ELSE", TK_ELSE, },
+ { "EXCEPT", TK_EXCEPT, },
+ { "EXPLAIN", TK_EXPLAIN, },
+ { "FAIL", TK_FAIL, },
+ { "FOR", TK_FOR, },
+ { "FOREIGN", TK_FOREIGN, },
+ { "FROM", TK_FROM, },
+ { "FULL", TK_JOIN_KW, },
+ { "GLOB", TK_GLOB, },
+ { "GROUP", TK_GROUP, },
+ { "HAVING", TK_HAVING, },
+ { "IGNORE", TK_IGNORE, },
+ { "IMMEDIATE", TK_IMMEDIATE, },
+ { "IN", TK_IN, },
+ { "INDEX", TK_INDEX, },
+ { "INITIALLY", TK_INITIALLY, },
+ { "INNER", TK_JOIN_KW, },
+ { "INSERT", TK_INSERT, },
+ { "INSTEAD", TK_INSTEAD, },
+ { "INTERSECT", TK_INTERSECT, },
+ { "INTO", TK_INTO, },
+ { "IS", TK_IS, },
+ { "ISNULL", TK_ISNULL, },
+ { "JOIN", TK_JOIN, },
+ { "KEY", TK_KEY, },
+ { "LEFT", TK_JOIN_KW, },
+ { "LIKE", TK_LIKE, },
+ { "LIMIT", TK_LIMIT, },
+ { "MATCH", TK_MATCH, },
+ { "NATURAL", TK_JOIN_KW, },
+ { "NOT", TK_NOT, },
+ { "NOTNULL", TK_NOTNULL, },
+ { "NULL", TK_NULL, },
+ { "OF", TK_OF, },
+ { "OFFSET", TK_OFFSET, },
+ { "ON", TK_ON, },
+ { "OR", TK_OR, },
+ { "ORDER", TK_ORDER, },
+ { "OUTER", TK_JOIN_KW, },
+ { "PRAGMA", TK_PRAGMA, },
+ { "PRIMARY", TK_PRIMARY, },
+ { "RAISE", TK_RAISE, },
+ { "REFERENCES", TK_REFERENCES, },
+ { "REPLACE", TK_REPLACE, },
+ { "RESTRICT", TK_RESTRICT, },
+ { "RIGHT", TK_JOIN_KW, },
+ { "ROLLBACK", TK_ROLLBACK, },
+ { "ROW", TK_ROW, },
+ { "SELECT", TK_SELECT, },
+ { "SET", TK_SET, },
+ { "STATEMENT", TK_STATEMENT, },
+ { "TABLE", TK_TABLE, },
+ { "TEMP", TK_TEMP, },
+ { "TEMPORARY", TK_TEMP, },
+ { "THEN", TK_THEN, },
+ { "TRANSACTION", TK_TRANSACTION, },
+ { "TRIGGER", TK_TRIGGER, },
+ { "UNION", TK_UNION, },
+ { "UNITQUE", TK_UNITQUE, },
+ { "UPDATE", TK_UPDATE, },
+ { "USING", TK_USING, },
+ { "VACUUM", TK_VACUUM, },
+ { "VALUES", TK_VALUES, },
+ { "VIEW", TK_VIEW, },
+ { "WHEN", TK_WHEN, },
+ { "WHERE", TK_WHERE, },
+};
+
+/*
+** This is the hash table
+*/
+#define KEY_HASH_SIZE 101
+static u8 aiHashTable[KEY_HASH_SIZE];
+
+
+/*
+** This function looks up an identifier to determine if it is a
+** keyword. If it is a keyword, the token code of that keyword is
+** returned. If the input is not a keyword, TK_ID is returned.
+*/
+int sqliteKeywordCode(const char *z, int n){
+ int h, i;
+ Keyword *p;
+ static char needInit = 1;
+ if( needInit ){
+ /* Initialize the keyword hash table */
+ sqliteOsEnterMutex();
+ if( needInit ){
+ int nk;
+ nk = sizeof(aKeywordTable)/sizeof(aKeywordTable[0]);
+ for(i=0; i<nk; i++){
+ aKeywordTable[i].len = strlen(aKeywordTable[i].zName);
+ h = sqliteHashNoCase(aKeywordTable[i].zName, aKeywordTable[i].len);
+ h %= KEY_HASH_SIZE;
+ aKeywordTable[i].iNext = aiHashTable[h];
+ aiHashTable[h] = i+1;
+ }
+ needInit = 0;
+ }
+ sqliteOsLeaveMutex();
+ }
+ h = sqliteHashNoCase(z, n) % KEY_HASH_SIZE;
+ for(i=aiHashTable[h]; i; i=p->iNext){
+ p = &aKeywordTable[i-1];
+ if( p->len==n && sqliteStrNICmp(p->zName, z, n)==0 ){
+ return p->tokenType;
+ }
+ }
+ return TK_ID;
+}
+
+
+/*
+** If X is a character that can be used in an identifier and
+** X&0x80==0 then isIdChar[X] will be 1. If X&0x80==0x80 then
+** X is always an identifier character. (Hence all UTF-8
+** characters can be part of an identifier). isIdChar[X] will
+** be 0 for every character in the lower 128 ASCII characters
+** that cannot be used as part of an identifier.
+**
+** In this implementation, an identifier can be a string of
+** alphabetic characters, digits, and "_" plus any character
+** with the high-order bit set. The latter rule means that
+** any sequence of UTF-8 characters or characters taken from
+** an extended ISO8859 character set can form an identifier.
+*/
+static const char isIdChar[] = {
+/* x0 x1 x2 x3 x4 x5 x6 x7 x8 x9 xA xB xC xD xE xF */
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 0x */
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 1x */
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 2x */
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, /* 3x */
+ 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 4x */
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, /* 5x */
+ 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 6x */
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, /* 7x */
+};
+
+
+/*
+** Return the length of the token that begins at z[0].
+** Store the token type in *tokenType before returning.
+*/
+static int sqliteGetToken(const unsigned char *z, int *tokenType){
+ int i;
+ switch( *z ){
+ case ' ': case '\t': case '\n': case '\f': case '\r': {
+ for(i=1; isspace(z[i]); i++){}
+ *tokenType = TK_SPACE;
+ return i;
+ }
+ case '-': {
+ if( z[1]=='-' ){
+ for(i=2; z[i] && z[i]!='\n'; i++){}
+ *tokenType = TK_COMMENT;
+ return i;
+ }
+ *tokenType = TK_MINUS;
+ return 1;
+ }
+ case '(': {
+ *tokenType = TK_LP;
+ return 1;
+ }
+ case ')': {
+ *tokenType = TK_RP;
+ return 1;
+ }
+ case ';': {
+ *tokenType = TK_SEMI;
+ return 1;
+ }
+ case '+': {
+ *tokenType = TK_PLUS;
+ return 1;
+ }
+ case '*': {
+ *tokenType = TK_STAR;
+ return 1;
+ }
+ case '/': {
+ if( z[1]!='*' || z[2]==0 ){
+ *tokenType = TK_SLASH;
+ return 1;
+ }
+ for(i=3; z[i] && (z[i]!='/' || z[i-1]!='*'); i++){}
+ if( z[i] ) i++;
+ *tokenType = TK_COMMENT;
+ return i;
+ }
+ case '%': {
+ *tokenType = TK_REM;
+ return 1;
+ }
+ case '=': {
+ *tokenType = TK_EQ;
+ return 1 + (z[1]=='=');
+ }
+ case '<': {
+ if( z[1]=='=' ){
+ *tokenType = TK_LE;
+ return 2;
+ }else if( z[1]=='>' ){
+ *tokenType = TK_NE;
+ return 2;
+ }else if( z[1]=='<' ){
+ *tokenType = TK_LSHIFT;
+ return 2;
+ }else{
+ *tokenType = TK_LT;
+ return 1;
+ }
+ }
+ case '>': {
+ if( z[1]=='=' ){
+ *tokenType = TK_GE;
+ return 2;
+ }else if( z[1]=='>' ){
+ *tokenType = TK_RSHIFT;
+ return 2;
+ }else{
+ *tokenType = TK_GT;
+ return 1;
+ }
+ }
+ case '!': {
+ if( z[1]!='=' ){
+ *tokenType = TK_ILLEGAL;
+ return 2;
+ }else{
+ *tokenType = TK_NE;
+ return 2;
+ }
+ }
+ case '|': {
+ if( z[1]!='|' ){
+ *tokenType = TK_BITOR;
+ return 1;
+ }else{
+ *tokenType = TK_CONCAT;
+ return 2;
+ }
+ }
+ case ',': {
+ *tokenType = TK_COMMA;
+ return 1;
+ }
+ case '&': {
+ *tokenType = TK_BITAND;
+ return 1;
+ }
+ case '~': {
+ *tokenType = TK_BITNOT;
+ return 1;
+ }
+ case '\'': case '"': {
+ int delim = z[0];
+ for(i=1; z[i]; i++){
+ if( z[i]==delim ){
+ if( z[i+1]==delim ){
+ i++;
+ }else{
+ break;
+ }
+ }
+ }
+ if( z[i] ) i++;
+ *tokenType = TK_STRING;
+ return i;
+ }
+ case '.': {
+ *tokenType = TK_DOT;
+ return 1;
+ }
+ case '0': case '1': case '2': case '3': case '4':
+ case '5': case '6': case '7': case '8': case '9': {
+ *tokenType = TK_INTEGER;
+ for(i=1; isdigit(z[i]); i++){}
+ if( z[i]=='.' && isdigit(z[i+1]) ){
+ i += 2;
+ while( isdigit(z[i]) ){ i++; }
+ *tokenType = TK_FLOAT;
+ }
+ if( (z[i]=='e' || z[i]=='E') &&
+ ( isdigit(z[i+1])
+ || ((z[i+1]=='+' || z[i+1]=='-') && isdigit(z[i+2]))
+ )
+ ){
+ i += 2;
+ while( isdigit(z[i]) ){ i++; }
+ *tokenType = TK_FLOAT;
+ }
+ return i;
+ }
+ case '[': {
+ for(i=1; z[i] && z[i-1]!=']'; i++){}
+ *tokenType = TK_ID;
+ return i;
+ }
+ case '?': {
+ *tokenType = TK_VARIABLE;
+ return 1;
+ }
+ default: {
+ if( (*z&0x80)==0 && !isIdChar[*z] ){
+ break;
+ }
+ for(i=1; (z[i]&0x80)!=0 || isIdChar[z[i]]; i++){}
+ *tokenType = sqliteKeywordCode((char*)z, i);
+ return i;
+ }
+ }
+ *tokenType = TK_ILLEGAL;
+ return 1;
+}
+
+/*
+** Run the parser on the given SQL string. The parser structure is
+** passed in. An STQLITE_ status code is returned. If an error occurs
+** and pzErrMsg!=NULL then an error message might be written into
+** memory obtained from malloc() and *pzErrMsg made to point to that
+** error message. Or maybe not.
+*/
+int sqliteRunParser(Parse *pParse, const char *zSql, char **pzErrMsg){
+ int nErr = 0;
+ int i;
+ void *pEngine;
+ int tokenType;
+ int lastTokenParsed = -1;
+ sqlite *db = pParse->db;
+ extern void *sqliteParserAlloc(void*(*)(int));
+ extern void sqliteParserFree(void*, void(*)(void*));
+ extern int sqliteParser(void*, int, Token, Parse*);
+
+ db->flags &= ~STQLITE_Interrupt;
+ pParse->rc = STQLITE_OK;
+ i = 0;
+ pEngine = sqliteParserAlloc((void*(*)(int))malloc);
+ if( pEngine==0 ){
+ sqliteSetString(pzErrMsg, "out of memory", (char*)0);
+ return 1;
+ }
+ pParse->sLastToken.dyn = 0;
+ pParse->zTail = zSql;
+ while( sqlite_malloc_failed==0 && zSql[i]!=0 ){
+ assert( i>=0 );
+ pParse->sLastToken.z = &zSql[i];
+ assert( pParse->sLastToken.dyn==0 );
+ pParse->sLastToken.n = sqliteGetToken((unsigned char*)&zSql[i], &tokenType);
+ i += pParse->sLastToken.n;
+ switch( tokenType ){
+ case TK_SPACE:
+ case TK_COMMENT: {
+ if( (db->flags & STQLITE_Interrupt)!=0 ){
+ pParse->rc = STQLITE_INTERRUPT;
+ sqliteSetString(pzErrMsg, "interrupt", (char*)0);
+ goto abort_parse;
+ }
+ break;
+ }
+ case TK_ILLEGAL: {
+ sqliteSetNString(pzErrMsg, "unrecognized token: \"", -1,
+ pParse->sLastToken.z, pParse->sLastToken.n, "\"", 1, 0);
+ nErr++;
+ goto abort_parse;
+ }
+ case TK_SEMI: {
+ pParse->zTail = &zSql[i];
+ /* Fall thru into the default case */
+ }
+ default: {
+ sqliteParser(pEngine, tokenType, pParse->sLastToken, pParse);
+ lastTokenParsed = tokenType;
+ if( pParse->rc!=STQLITE_OK ){
+ goto abort_parse;
+ }
+ break;
+ }
+ }
+ }
+abort_parse:
+ if( zSql[i]==0 && nErr==0 && pParse->rc==STQLITE_OK ){
+ if( lastTokenParsed!=TK_SEMI ){
+ sqliteParser(pEngine, TK_SEMI, pParse->sLastToken, pParse);
+ pParse->zTail = &zSql[i];
+ }
+ sqliteParser(pEngine, 0, pParse->sLastToken, pParse);
+ }
+ sqliteParserFree(pEngine, free);
+ if( pParse->rc!=STQLITE_OK && pParse->rc!=STQLITE_DONE && pParse->zErrMsg==0 ){
+ sqliteSetString(&pParse->zErrMsg, sqlite_error_string(pParse->rc),
+ (char*)0);
+ }
+ if( pParse->zErrMsg ){
+ if( pzErrMsg && *pzErrMsg==0 ){
+ *pzErrMsg = pParse->zErrMsg;
+ }else{
+ sqliteFree(pParse->zErrMsg);
+ }
+ pParse->zErrMsg = 0;
+ if( !nErr ) nErr++;
+ }
+ if( pParse->pVdbe && pParse->nErr>0 ){
+ sqliteVdbeDelete(pParse->pVdbe);
+ pParse->pVdbe = 0;
+ }
+ if( pParse->pNewTable ){
+ sqliteDeleteTable(pParse->db, pParse->pNewTable);
+ pParse->pNewTable = 0;
+ }
+ if( pParse->pNewTrigger ){
+ sqliteDeleteTrigger(pParse->pNewTrigger);
+ pParse->pNewTrigger = 0;
+ }
+ if( nErr>0 && (pParse->rc==STQLITE_OK || pParse->rc==STQLITE_DONE) ){
+ pParse->rc = STQLITE_ERROR;
+ }
+ return nErr;
+}
+
+/*
+** Token types used by the sqlite_complete() routine. See the header
+** comments on that procedure for additional information.
+*/
+#define tkEXPLAIN 0
+#define tkCREATE 1
+#define tkTEMP 2
+#define tkTRIGGER 3
+#define tkEND 4
+#define tkSEMI 5
+#define tkWS 6
+#define tkOTHER 7
+
+/*
+** Return TRUE if the given SQL string ends in a semicolon.
+**
+** Special handling is require for CREATE TRIGGER statements.
+** Whenever the CREATE TRIGGER keywords are seen, the statement
+** must end with ";END;".
+**
+** This implementation uses a state machine with 7 states:
+**
+** (0) START At the beginning or end of an SQL statement. This routine
+** returns 1 if it ends in the START state and 0 if it ends
+** in any other state.
+**
+** (1) EXPLAIN The keyword EXPLAIN has been seen at the beginning of
+** a statement.
+**
+** (2) CREATE The keyword CREATE has been seen at the beginning of a
+** statement, possibly preceeded by EXPLAIN and/or followed by
+** TEMP or TEMPORARY
+**
+** (3) NORMAL We are in the middle of statement which ends with a single
+** semicolon.
+**
+** (4) TRIGGER We are in the middle of a trigger definition that must be
+** ended by a semicolon, the keyword END, and another semicolon.
+**
+** (5) SEMI We've seen the first semicolon in the ";END;" that occurs at
+** the end of a trigger definition.
+**
+** (6) END We've seen the ";END" of the ";END;" that occurs at the end
+** of a trigger difinition.
+**
+** Transitions between states above are determined by tokens extracted
+** from the input. The following tokens are significant:
+**
+** (0) tkEXPLAIN The "explain" keyword.
+** (1) tkCREATE The "create" keyword.
+** (2) tkTEMP The "temp" or "temporary" keyword.
+** (3) tkTRIGGER The "trigger" keyword.
+** (4) tkEND The "end" keyword.
+** (5) tkSEMI A semicolon.
+** (6) tkWS Whitespace
+** (7) tkOTHER Any other SQL token.
+**
+** Whitespace never causes a state transition and is always ignored.
+*/
+int sqlite_complete(const char *zSql){
+ u8 state = 0; /* Current state, using numbers defined in header comment */
+ u8 token; /* Value of the next token */
+
+ /* The following matrix defines the transition from one state to another
+ ** according to what token is seen. trans[state][token] returns the
+ ** next state.
+ */
+ static const u8 trans[7][8] = {
+ /* Token: */
+ /* State: ** EXPLAIN CREATE TEMP TRIGGER END SEMI WS OTHER */
+ /* 0 START: */ { 1, 2, 3, 3, 3, 0, 0, 3, },
+ /* 1 EXPLAIN: */ { 3, 2, 3, 3, 3, 0, 1, 3, },
+ /* 2 CREATE: */ { 3, 3, 2, 4, 3, 0, 2, 3, },
+ /* 3 NORMAL: */ { 3, 3, 3, 3, 3, 0, 3, 3, },
+ /* 4 TRIGGER: */ { 4, 4, 4, 4, 4, 5, 4, 4, },
+ /* 5 SEMI: */ { 4, 4, 4, 4, 6, 5, 5, 4, },
+ /* 6 END: */ { 4, 4, 4, 4, 4, 0, 6, 4, },
+ };
+
+ while( *zSql ){
+ switch( *zSql ){
+ case ';': { /* A semicolon */
+ token = tkSEMI;
+ break;
+ }
+ case ' ':
+ case '\r':
+ case '\t':
+ case '\n':
+ case '\f': { /* White space is ignored */
+ token = tkWS;
+ break;
+ }
+ case '/': { /* C-style comments */
+ if( zSql[1]!='*' ){
+ token = tkOTHER;
+ break;
+ }
+ zSql += 2;
+ while( zSql[0] && (zSql[0]!='*' || zSql[1]!='/') ){ zSql++; }
+ if( zSql[0]==0 ) return 0;
+ zSql++;
+ token = tkWS;
+ break;
+ }
+ case '-': { /* SQL-style comments from "--" to end of line */
+ if( zSql[1]!='-' ){
+ token = tkOTHER;
+ break;
+ }
+ while( *zSql && *zSql!='\n' ){ zSql++; }
+ if( *zSql==0 ) return state==0;
+ token = tkWS;
+ break;
+ }
+ case '[': { /* Microsoft-style identifiers in [...] */
+ zSql++;
+ while( *zSql && *zSql!=']' ){ zSql++; }
+ if( *zSql==0 ) return 0;
+ token = tkOTHER;
+ break;
+ }
+ case '"': /* single- and double-quoted strings */
+ case '\'': {
+ int c = *zSql;
+ zSql++;
+ while( *zSql && *zSql!=c ){ zSql++; }
+ if( *zSql==0 ) return 0;
+ token = tkOTHER;
+ break;
+ }
+ default: {
+ if( isIdChar[(u8)*zSql] ){
+ /* Keywords and unquoted identifiers */
+ int nId;
+ for(nId=1; isIdChar[(u8)zSql[nId]]; nId++){}
+ switch( *zSql ){
+ case 'c': case 'C': {
+ if( nId==6 && sqliteStrNICmp(zSql, "create", 6)==0 ){
+ token = tkCREATE;
+ }else{
+ token = tkOTHER;
+ }
+ break;
+ }
+ case 't': case 'T': {
+ if( nId==7 && sqliteStrNICmp(zSql, "trigger", 7)==0 ){
+ token = tkTRIGGER;
+ }else if( nId==4 && sqliteStrNICmp(zSql, "temp", 4)==0 ){
+ token = tkTEMP;
+ }else if( nId==9 && sqliteStrNICmp(zSql, "temporary", 9)==0 ){
+ token = tkTEMP;
+ }else{
+ token = tkOTHER;
+ }
+ break;
+ }
+ case 'e': case 'E': {
+ if( nId==3 && sqliteStrNICmp(zSql, "end", 3)==0 ){
+ token = tkEND;
+ }else if( nId==7 && sqliteStrNICmp(zSql, "explain", 7)==0 ){
+ token = tkEXPLAIN;
+ }else{
+ token = tkOTHER;
+ }
+ break;
+ }
+ default: {
+ token = tkOTHER;
+ break;
+ }
+ }
+ zSql += nId-1;
+ }else{
+ /* Operators and special symbols */
+ token = tkOTHER;
+ }
+ break;
+ }
+ }
+ state = trans[state][token];
+ zSql++;
+ }
+ return state==0;
+}