tokenize.c

sqlite数据库管理系统开放源码

/*
** 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 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,        },
  { "UNIQUE",            TK_UNIQUE,       },
  { "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;
          }
        }
      }