Index: Makefile.in ================================================================== --- Makefile.in +++ Makefile.in @@ -111,11 +111,11 @@ TCC += -DSQLITE_OMIT_CURSOR # Object files for the SQLite library. # LIBOBJ = alter.lo analyze.lo attach.lo auth.lo btree.lo build.lo \ - callback.lo date.lo \ + callback.lo complete.lo date.lo \ delete.lo expr.lo func.lo hash.lo insert.lo \ main.lo opcodes.lo os_unix.lo os_win.lo \ pager.lo parse.lo pragma.lo prepare.lo printf.lo random.lo \ select.lo table.lo tokenize.lo trigger.lo update.lo \ util.lo vacuum.lo \ @@ -131,10 +131,11 @@ $(TOP)/src/auth.c \ $(TOP)/src/btree.c \ $(TOP)/src/btree.h \ $(TOP)/src/build.c \ $(TOP)/src/callback.c \ + $(TOP)/src/complete.c \ $(TOP)/src/date.c \ $(TOP)/src/delete.c \ $(TOP)/src/expr.c \ $(TOP)/src/func.c \ $(TOP)/src/hash.c \ @@ -189,11 +190,12 @@ $(TOP)/src/test4.c \ $(TOP)/src/test5.c \ $(TOP)/src/utf.c \ $(TOP)/src/util.c \ $(TOP)/src/vdbe.c \ - $(TOP)/src/md5.c + $(TOP)/src/md5.c \ + $(TOP)/src/where.c # In LIBOBJ but not TESTSRC COMMONOBJ = $(foreach obj,$(LIBOBJ),\ $(if $(findstring $(patsubst %.lo,%.c,$(obj)),$(TESTSRC)),,$(obj))) @@ -288,10 +290,13 @@ build.lo: $(TOP)/src/build.c $(HDR) $(LTCOMPILE) -c $(TOP)/src/build.c callback.lo: $(TOP)/src/callback.c $(HDR) $(LTCOMPILE) -c $(TOP)/src/callback.c + +complete.lo: $(TOP)/src/complete.c $(HDR) + $(LTCOMPILE) -c $(TOP)/src/complete.c date.lo: $(TOP)/src/date.c $(HDR) $(LTCOMPILE) -c $(TOP)/src/date.c delete.lo: $(TOP)/src/delete.c $(HDR) Index: main.mk ================================================================== --- main.mk +++ main.mk @@ -53,11 +53,11 @@ TCCX = $(TCC) $(OPTS) $(THREADSAFE) $(USLEEP) -I. -I$(TOP)/src # Object files for the SQLite library. # LIBOBJ+= alter.o analyze.o attach.o auth.o btree.o build.o \ - callback.o date.o delete.o \ + callback.o complete.o date.o delete.o \ expr.o func.o hash.o insert.o \ main.o opcodes.o os_unix.o os_win.o \ pager.o parse.o pragma.o prepare.o printf.o random.o \ select.o table.o tclsqlite.o tokenize.o trigger.o \ update.o util.o vacuum.o \ @@ -73,10 +73,11 @@ $(TOP)/src/auth.c \ $(TOP)/src/btree.c \ $(TOP)/src/btree.h \ $(TOP)/src/build.c \ $(TOP)/src/callback.c \ + $(TOP)/src/complete.c \ $(TOP)/src/date.c \ $(TOP)/src/delete.c \ $(TOP)/src/expr.c \ $(TOP)/src/func.c \ $(TOP)/src/hash.c \ @@ -219,10 +220,13 @@ build.o: $(TOP)/src/build.c $(HDR) $(TCCX) -c $(TOP)/src/build.c callback.o: $(TOP)/src/callback.c $(HDR) $(TCCX) -c $(TOP)/src/callback.c + +complete.o: $(TOP)/src/complete.c $(HDR) + $(TCCX) -c $(TOP)/src/complete.c date.o: $(TOP)/src/date.c $(HDR) $(TCCX) -c $(TOP)/src/date.c delete.o: $(TOP)/src/delete.c $(HDR) ADDED src/complete.c Index: src/complete.c ================================================================== --- /dev/null +++ src/complete.c @@ -0,0 +1,263 @@ +/* +** 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 find forgiveness for yourself and forgive others. +** May you share freely, never taking more than you give. +** +************************************************************************* +** An tokenizer for SQL +** +** This file contains C code that implements the sqlite3_complete() API. +** This code used to be part of the tokenizer.c source file. But by +** separating it out, the code will be automatically omitted from +** static links that do not use it. +** +** $Id: complete.c,v 1.1 2005/08/14 17:53:21 drh Exp $ +*/ +#include "sqliteInt.h" +#ifndef SQLITE_OMIT_COMPLETE + +/* +** This is defined in tokenize.c. We just have to import the definition. +*/ +extern const char sqlite3IsIdChar[]; +#define IdChar(C) (((c=C)&0x80)!=0 || (c>0x1f && sqlite3IsIdChar[c-0x20])) + + +/* +** Token types used by the sqlite3_complete() routine. See the header +** comments on that procedure for additional information. +*/ +#define tkSEMI 0 +#define tkWS 1 +#define tkOTHER 2 +#define tkEXPLAIN 3 +#define tkCREATE 4 +#define tkTEMP 5 +#define tkTRIGGER 6 +#define tkEND 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) NORMAL We are in the middle of statement which ends with a single +** semicolon. +** +** (2) EXPLAIN The keyword EXPLAIN has been seen at the beginning of +** a statement. +** +** (3) CREATE The keyword CREATE has been seen at the beginning of a +** statement, possibly preceeded by EXPLAIN and/or followed by +** TEMP or TEMPORARY +** +** (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) tkSEMI A semicolon. +** (1) tkWS Whitespace +** (2) tkOTHER Any other SQL token. +** (3) tkEXPLAIN The "explain" keyword. +** (4) tkCREATE The "create" keyword. +** (5) tkTEMP The "temp" or "temporary" keyword. +** (6) tkTRIGGER The "trigger" keyword. +** (7) tkEND The "end" keyword. +** +** Whitespace never causes a state transition and is always ignored. +** +** If we compile with SQLITE_OMIT_TRIGGER, all of the computation needed +** to recognize the end of a trigger can be omitted. All we have to do +** is look for a semicolon that is not part of an string or comment. +*/ +int sqlite3_complete(const char *zSql){ + u8 state = 0; /* Current state, using numbers defined in header comment */ + u8 token; /* Value of the next token */ + +#ifndef SQLITE_OMIT_TRIGGER + /* A complex statement machine used to detect the end of a CREATE TRIGGER + ** statement. This is the normal case. + */ + static const u8 trans[7][8] = { + /* Token: */ + /* State: ** SEMI WS OTHER EXPLAIN CREATE TEMP TRIGGER END */ + /* 0 START: */ { 0, 0, 1, 2, 3, 1, 1, 1, }, + /* 1 NORMAL: */ { 0, 1, 1, 1, 1, 1, 1, 1, }, + /* 2 EXPLAIN: */ { 0, 2, 1, 1, 3, 1, 1, 1, }, + /* 3 CREATE: */ { 0, 3, 1, 1, 1, 3, 4, 1, }, + /* 4 TRIGGER: */ { 5, 4, 4, 4, 4, 4, 4, 4, }, + /* 5 SEMI: */ { 5, 5, 4, 4, 4, 4, 4, 6, }, + /* 6 END: */ { 0, 6, 4, 4, 4, 4, 4, 4, }, + }; +#else + /* If triggers are not suppored by this compile then the statement machine + ** used to detect the end of a statement is much simplier + */ + static const u8 trans[2][3] = { + /* Token: */ + /* State: ** SEMI WS OTHER */ + /* 0 START: */ { 0, 0, 1, }, + /* 1 NORMAL: */ { 0, 1, 1, }, + }; +#endif /* SQLITE_OMIT_TRIGGER */ + + 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 '`': /* Grave-accent quoted symbols used by MySQL */ + 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: { + int c; + if( IdChar((u8)*zSql) ){ + /* Keywords and unquoted identifiers */ + int nId; + for(nId=1; IdChar(zSql[nId]); nId++){} +#ifdef SQLITE_OMIT_TRIGGER + token = tkOTHER; +#else + switch( *zSql ){ + case 'c': case 'C': { + if( nId==6 && sqlite3StrNICmp(zSql, "create", 6)==0 ){ + token = tkCREATE; + }else{ + token = tkOTHER; + } + break; + } + case 't': case 'T': { + if( nId==7 && sqlite3StrNICmp(zSql, "trigger", 7)==0 ){ + token = tkTRIGGER; + }else if( nId==4 && sqlite3StrNICmp(zSql, "temp", 4)==0 ){ + token = tkTEMP; + }else if( nId==9 && sqlite3StrNICmp(zSql, "temporary", 9)==0 ){ + token = tkTEMP; + }else{ + token = tkOTHER; + } + break; + } + case 'e': case 'E': { + if( nId==3 && sqlite3StrNICmp(zSql, "end", 3)==0 ){ + token = tkEND; + }else +#ifndef SQLITE_OMIT_EXPLAIN + if( nId==7 && sqlite3StrNICmp(zSql, "explain", 7)==0 ){ + token = tkEXPLAIN; + }else +#endif + { + token = tkOTHER; + } + break; + } + default: { + token = tkOTHER; + break; + } + } +#endif /* SQLITE_OMIT_TRIGGER */ + zSql += nId-1; + }else{ + /* Operators and special symbols */ + token = tkOTHER; + } + break; + } + } + state = trans[state][token]; + zSql++; + } + return state==0; +} + +#ifndef SQLITE_OMIT_UTF16 +/* +** This routine is the same as the sqlite3_complete() routine described +** above, except that the parameter is required to be UTF-16 encoded, not +** UTF-8. +*/ +int sqlite3_complete16(const void *zSql){ + sqlite3_value *pVal; + char const *zSql8; + int rc = 0; + + pVal = sqlite3ValueNew(); + sqlite3ValueSetStr(pVal, -1, zSql, SQLITE_UTF16NATIVE, SQLITE_STATIC); + zSql8 = sqlite3ValueText(pVal, SQLITE_UTF8); + if( zSql8 ){ + rc = sqlite3_complete(zSql8); + } + sqlite3ValueFree(pVal); + return rc; +} +#endif /* SQLITE_OMIT_UTF16 */ +#endif /* SQLITE_OMIT_COMPLETE */ Index: src/tokenize.c ================================================================== --- src/tokenize.c +++ src/tokenize.c @@ -13,11 +13,11 @@ ** ** This file contains 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 1.105 2005/08/13 18:15:43 drh Exp $ +** $Id: tokenize.c,v 1.106 2005/08/14 17:53:21 drh Exp $ */ #include "sqliteInt.h" #include "os.h" #include #include @@ -36,13 +36,13 @@ #include "keywordhash.h" /* ** 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&0x80==0 then sqlite3IsIdChar[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 +** characters can be part of an identifier). sqlite3IsIdChar[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 @@ -53,21 +53,21 @@ ** Ticket #1066. the SQL standard does not allow '$' in the ** middle of identfiers. But many SQL implementations do. ** SQLite will allow '$' in identifiers for compatibility. ** But the feature is undocumented. */ -static const char isIdChar[] = { +const char sqlite3IsIdChar[] = { /* x0 x1 x2 x3 x4 x5 x6 x7 x8 x9 xA xB xC xD xE xF */ 0, 0, 0, 0, 1, 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 */ }; -#define IdChar(C) (((c=C)&0x80)!=0 || (c>0x1f && isIdChar[c-0x20])) +#define IdChar(C) (((c=C)&0x80)!=0 || (c>0x1f && sqlite3IsIdChar[c-0x20])) /* ** Return the length of the token that begins at z[0]. ** Store the token type in *tokenType before returning. */ @@ -422,244 +422,5 @@ if( nErr>0 && (pParse->rc==SQLITE_OK || pParse->rc==SQLITE_DONE) ){ pParse->rc = SQLITE_ERROR; } return nErr; } - -/* The sqlite3_complete() API may be omitted (to save code space) by -** defining the following symbol. -*/ -#ifndef SQLITE_OMIT_COMPLETE - -/* -** Token types used by the sqlite3_complete() routine. See the header -** comments on that procedure for additional information. -*/ -#define tkSEMI 0 -#define tkWS 1 -#define tkOTHER 2 -#define tkEXPLAIN 3 -#define tkCREATE 4 -#define tkTEMP 5 -#define tkTRIGGER 6 -#define tkEND 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) NORMAL We are in the middle of statement which ends with a single -** semicolon. -** -** (2) EXPLAIN The keyword EXPLAIN has been seen at the beginning of -** a statement. -** -** (3) CREATE The keyword CREATE has been seen at the beginning of a -** statement, possibly preceeded by EXPLAIN and/or followed by -** TEMP or TEMPORARY -** -** (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) tkSEMI A semicolon. -** (1) tkWS Whitespace -** (2) tkOTHER Any other SQL token. -** (3) tkEXPLAIN The "explain" keyword. -** (4) tkCREATE The "create" keyword. -** (5) tkTEMP The "temp" or "temporary" keyword. -** (6) tkTRIGGER The "trigger" keyword. -** (7) tkEND The "end" keyword. -** -** Whitespace never causes a state transition and is always ignored. -** -** If we compile with SQLITE_OMIT_TRIGGER, all of the computation needed -** to recognize the end of a trigger can be omitted. All we have to do -** is look for a semicolon that is not part of an string or comment. -*/ -int sqlite3_complete(const char *zSql){ - u8 state = 0; /* Current state, using numbers defined in header comment */ - u8 token; /* Value of the next token */ - -#ifndef SQLITE_OMIT_TRIGGER - /* A complex statement machine used to detect the end of a CREATE TRIGGER - ** statement. This is the normal case. - */ - static const u8 trans[7][8] = { - /* Token: */ - /* State: ** SEMI WS OTHER EXPLAIN CREATE TEMP TRIGGER END */ - /* 0 START: */ { 0, 0, 1, 2, 3, 1, 1, 1, }, - /* 1 NORMAL: */ { 0, 1, 1, 1, 1, 1, 1, 1, }, - /* 2 EXPLAIN: */ { 0, 2, 1, 1, 3, 1, 1, 1, }, - /* 3 CREATE: */ { 0, 3, 1, 1, 1, 3, 4, 1, }, - /* 4 TRIGGER: */ { 5, 4, 4, 4, 4, 4, 4, 4, }, - /* 5 SEMI: */ { 5, 5, 4, 4, 4, 4, 4, 6, }, - /* 6 END: */ { 0, 6, 4, 4, 4, 4, 4, 4, }, - }; -#else - /* If triggers are not suppored by this compile then the statement machine - ** used to detect the end of a statement is much simplier - */ - static const u8 trans[2][3] = { - /* Token: */ - /* State: ** SEMI WS OTHER */ - /* 0 START: */ { 0, 0, 1, }, - /* 1 NORMAL: */ { 0, 1, 1, }, - }; -#endif /* SQLITE_OMIT_TRIGGER */ - - 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 '`': /* Grave-accent quoted symbols used by MySQL */ - 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: { - int c; - if( IdChar((u8)*zSql) ){ - /* Keywords and unquoted identifiers */ - int nId; - for(nId=1; IdChar(zSql[nId]); nId++){} -#ifdef SQLITE_OMIT_TRIGGER - token = tkOTHER; -#else - switch( *zSql ){ - case 'c': case 'C': { - if( nId==6 && sqlite3StrNICmp(zSql, "create", 6)==0 ){ - token = tkCREATE; - }else{ - token = tkOTHER; - } - break; - } - case 't': case 'T': { - if( nId==7 && sqlite3StrNICmp(zSql, "trigger", 7)==0 ){ - token = tkTRIGGER; - }else if( nId==4 && sqlite3StrNICmp(zSql, "temp", 4)==0 ){ - token = tkTEMP; - }else if( nId==9 && sqlite3StrNICmp(zSql, "temporary", 9)==0 ){ - token = tkTEMP; - }else{ - token = tkOTHER; - } - break; - } - case 'e': case 'E': { - if( nId==3 && sqlite3StrNICmp(zSql, "end", 3)==0 ){ - token = tkEND; - }else -#ifndef SQLITE_OMIT_EXPLAIN - if( nId==7 && sqlite3StrNICmp(zSql, "explain", 7)==0 ){ - token = tkEXPLAIN; - }else -#endif - { - token = tkOTHER; - } - break; - } - default: { - token = tkOTHER; - break; - } - } -#endif /* SQLITE_OMIT_TRIGGER */ - zSql += nId-1; - }else{ - /* Operators and special symbols */ - token = tkOTHER; - } - break; - } - } - state = trans[state][token]; - zSql++; - } - return state==0; -} - -#ifndef SQLITE_OMIT_UTF16 -/* -** This routine is the same as the sqlite3_complete() routine described -** above, except that the parameter is required to be UTF-16 encoded, not -** UTF-8. -*/ -int sqlite3_complete16(const void *zSql){ - sqlite3_value *pVal; - char const *zSql8; - int rc = 0; - - pVal = sqlite3ValueNew(); - sqlite3ValueSetStr(pVal, -1, zSql, SQLITE_UTF16NATIVE, SQLITE_STATIC); - zSql8 = sqlite3ValueText(pVal, SQLITE_UTF8); - if( zSql8 ){ - rc = sqlite3_complete(zSql8); - } - sqlite3ValueFree(pVal); - return rc; -} -#endif /* SQLITE_OMIT_UTF16 */ -#endif /* SQLITE_OMIT_COMPLETE */