1997-09-16 23:25:59 +04:00
|
|
|
/*
|
2010-04-16 16:31:07 +04:00
|
|
|
** $Id: llex.c,v 2.36 2010/04/05 16:35:37 roberto Exp roberto $
|
1999-02-26 00:07:26 +03:00
|
|
|
** Lexical Analyzer
|
1997-09-16 23:25:59 +04:00
|
|
|
** See Copyright Notice in lua.h
|
|
|
|
*/
|
1996-05-30 18:04:07 +04:00
|
|
|
|
1993-12-23 00:15:16 +03:00
|
|
|
|
2005-12-07 18:33:27 +03:00
|
|
|
#include <locale.h>
|
1993-12-28 19:42:29 +03:00
|
|
|
#include <string.h>
|
1993-12-23 00:15:16 +03:00
|
|
|
|
2002-12-04 20:38:31 +03:00
|
|
|
#define llex_c
|
2004-05-01 00:13:38 +04:00
|
|
|
#define LUA_CORE
|
2002-12-04 20:38:31 +03:00
|
|
|
|
2000-06-12 17:52:05 +04:00
|
|
|
#include "lua.h"
|
|
|
|
|
2009-02-19 20:33:51 +03:00
|
|
|
#include "lctype.h"
|
2002-05-07 21:36:56 +04:00
|
|
|
#include "ldo.h"
|
1997-09-16 23:25:59 +04:00
|
|
|
#include "llex.h"
|
|
|
|
#include "lobject.h"
|
|
|
|
#include "lparser.h"
|
1997-11-19 20:29:23 +03:00
|
|
|
#include "lstate.h"
|
1997-09-16 23:25:59 +04:00
|
|
|
#include "lstring.h"
|
2006-03-09 21:15:48 +03:00
|
|
|
#include "ltable.h"
|
1997-09-16 23:25:59 +04:00
|
|
|
#include "lzio.h"
|
1993-12-23 00:15:16 +03:00
|
|
|
|
1995-09-16 00:48:26 +04:00
|
|
|
|
1996-05-30 18:04:07 +04:00
|
|
|
|
2003-08-28 18:38:46 +04:00
|
|
|
#define next(ls) (ls->current = zgetc(ls->z))
|
1997-09-16 23:25:59 +04:00
|
|
|
|
2003-10-03 20:05:34 +04:00
|
|
|
|
1997-09-16 23:25:59 +04:00
|
|
|
|
2003-08-28 18:38:46 +04:00
|
|
|
#define currIsNewline(ls) (ls->current == '\n' || ls->current == '\r')
|
2003-08-21 18:16:43 +04:00
|
|
|
|
1998-05-27 17:08:34 +04:00
|
|
|
|
1999-07-22 23:29:42 +04:00
|
|
|
/* ORDER RESERVED */
|
2007-02-07 20:49:18 +03:00
|
|
|
static const char *const luaX_tokens [] = {
|
2001-11-28 23:13:13 +03:00
|
|
|
"and", "break", "do", "else", "elseif",
|
2002-09-03 15:57:38 +04:00
|
|
|
"end", "false", "for", "function", "if",
|
2001-11-28 23:13:13 +03:00
|
|
|
"in", "local", "nil", "not", "or", "repeat",
|
2004-12-03 23:54:12 +03:00
|
|
|
"return", "then", "true", "until", "while",
|
2007-02-07 20:49:18 +03:00
|
|
|
"..", "...", "==", ">=", "<=", "~=", "<eof>",
|
|
|
|
"<number>", "<name>", "<string>"
|
2001-02-23 20:17:25 +03:00
|
|
|
};
|
1998-05-27 17:08:34 +04:00
|
|
|
|
1997-09-16 23:25:59 +04:00
|
|
|
|
2004-12-01 18:46:18 +03:00
|
|
|
#define save_and_next(ls) (save(ls, ls->current), next(ls))
|
|
|
|
|
|
|
|
|
2007-10-25 20:45:47 +04:00
|
|
|
static void lexerror (LexState *ls, const char *msg, int token);
|
|
|
|
|
|
|
|
|
2004-12-01 18:46:18 +03:00
|
|
|
static void save (LexState *ls, int c) {
|
|
|
|
Mbuffer *b = ls->buff;
|
2009-05-18 21:28:04 +04:00
|
|
|
if (luaZ_bufflen(b) + 1 > luaZ_sizebuffer(b)) {
|
2004-12-01 18:46:18 +03:00
|
|
|
size_t newsize;
|
2009-05-18 21:28:04 +04:00
|
|
|
if (luaZ_sizebuffer(b) >= MAX_SIZET/2)
|
2007-10-25 20:45:47 +04:00
|
|
|
lexerror(ls, "lexical element too long", 0);
|
2009-05-18 21:28:04 +04:00
|
|
|
newsize = luaZ_sizebuffer(b) * 2;
|
2004-12-01 18:46:18 +03:00
|
|
|
luaZ_resizebuffer(ls->L, b, newsize);
|
|
|
|
}
|
2009-05-18 21:28:04 +04:00
|
|
|
b->buffer[luaZ_bufflen(b)++] = cast(char, c);
|
2004-12-01 18:46:18 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
1999-11-22 16:12:07 +03:00
|
|
|
void luaX_init (lua_State *L) {
|
2000-03-03 17:58:26 +03:00
|
|
|
int i;
|
2000-02-08 19:39:42 +03:00
|
|
|
for (i=0; i<NUM_RESERVED; i++) {
|
2004-12-02 15:59:10 +03:00
|
|
|
TString *ts = luaS_new(L, luaX_tokens[i]);
|
2002-08-16 18:45:55 +04:00
|
|
|
luaS_fix(ts); /* reserved words are never collected */
|
2004-12-02 15:59:10 +03:00
|
|
|
lua_assert(strlen(luaX_tokens[i])+1 <= TOKEN_LEN);
|
2005-12-22 19:19:56 +03:00
|
|
|
ts->tsv.reserved = cast_byte(i+1); /* reserved word */
|
1998-05-27 17:08:34 +04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-09-05 00:00:28 +04:00
|
|
|
const char *luaX_token2str (LexState *ls, int token) {
|
|
|
|
if (token < FIRST_RESERVED) {
|
2004-11-24 22:16:03 +03:00
|
|
|
lua_assert(token == cast(unsigned char, token));
|
2009-02-19 20:33:51 +03:00
|
|
|
return (lisprint(token)) ? luaO_pushfstring(ls->L, LUA_QL("%c"), token) :
|
2007-08-10 00:29:15 +04:00
|
|
|
luaO_pushfstring(ls->L, "char(%d)", token);
|
2007-02-07 20:49:18 +03:00
|
|
|
}
|
|
|
|
else {
|
|
|
|
const char *s = luaX_tokens[token - FIRST_RESERVED];
|
|
|
|
if (token < TK_EOS)
|
|
|
|
return luaO_pushfstring(ls->L, LUA_QS, s);
|
|
|
|
else
|
|
|
|
return s;
|
2003-09-05 00:00:28 +04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static const char *txtToken (LexState *ls, int token) {
|
|
|
|
switch (token) {
|
2002-05-07 21:36:56 +04:00
|
|
|
case TK_NAME:
|
|
|
|
case TK_STRING:
|
|
|
|
case TK_NUMBER:
|
2003-08-28 18:38:46 +04:00
|
|
|
save(ls, '\0');
|
2007-02-07 20:49:18 +03:00
|
|
|
return luaO_pushfstring(ls->L, LUA_QS, luaZ_buffer(ls->buff));
|
2002-05-07 21:36:56 +04:00
|
|
|
default:
|
2003-09-05 00:00:28 +04:00
|
|
|
return luaX_token2str(ls, token);
|
1997-09-16 23:25:59 +04:00
|
|
|
}
|
1998-05-27 17:08:34 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2007-10-25 20:45:47 +04:00
|
|
|
static void lexerror (LexState *ls, const char *msg, int token) {
|
2008-12-26 14:55:57 +03:00
|
|
|
char buff[LUA_IDSIZE];
|
|
|
|
luaO_chunkid(buff, getstr(ls->source), LUA_IDSIZE);
|
2004-03-12 22:53:56 +03:00
|
|
|
msg = luaO_pushfstring(ls->L, "%s:%d: %s", buff, ls->linenumber, msg);
|
|
|
|
if (token)
|
2007-02-07 20:49:18 +03:00
|
|
|
luaO_pushfstring(ls->L, "%s near %s", msg, txtToken(ls, token));
|
2003-10-20 16:24:34 +04:00
|
|
|
luaD_throw(ls->L, LUA_ERRSYNTAX);
|
2001-11-16 19:29:10 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-09-05 00:00:28 +04:00
|
|
|
void luaX_syntaxerror (LexState *ls, const char *msg) {
|
2007-10-25 20:45:47 +04:00
|
|
|
lexerror(ls, msg, ls->t.token);
|
1997-09-16 23:25:59 +04:00
|
|
|
}
|
|
|
|
|
1997-04-12 19:01:49 +04:00
|
|
|
|
2010-04-16 16:31:07 +04:00
|
|
|
/*
|
|
|
|
** creates a new string and anchors it in function's table so that
|
|
|
|
** it will not be collected until the end of the function's compilation
|
|
|
|
** (by that time it should be anchored in function's prototype)
|
|
|
|
*/
|
2003-08-28 18:38:46 +04:00
|
|
|
TString *luaX_newstring (LexState *ls, const char *str, size_t l) {
|
|
|
|
lua_State *L = ls->L;
|
2006-07-11 19:53:29 +04:00
|
|
|
TValue *o; /* entry for `str' */
|
2010-04-16 16:31:07 +04:00
|
|
|
TString *ts = luaS_newlstr(L, str, l); /* create new string */
|
|
|
|
setsvalue2s(L, L->top++, ts); /* temporarily anchor it in stack */
|
|
|
|
o = luaH_setstr(L, ls->fs->h, ts);
|
2009-11-17 19:33:38 +03:00
|
|
|
if (ttisnil(o)) {
|
2010-04-16 16:31:07 +04:00
|
|
|
setbvalue(o, 1); /* t[string] = true */
|
2009-11-17 19:33:38 +03:00
|
|
|
luaC_checkGC(L);
|
|
|
|
}
|
2010-04-16 16:31:07 +04:00
|
|
|
L->top--; /* remove string from stack */
|
2003-08-28 01:01:44 +04:00
|
|
|
return ts;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2010-04-16 16:31:07 +04:00
|
|
|
/*
|
|
|
|
** increment line number and skips newline sequence (any of
|
|
|
|
** \n, \r, \n\r, or \r\n)
|
|
|
|
*/
|
2003-08-28 18:38:46 +04:00
|
|
|
static void inclinenumber (LexState *ls) {
|
|
|
|
int old = ls->current;
|
|
|
|
lua_assert(currIsNewline(ls));
|
|
|
|
next(ls); /* skip `\n' or `\r' */
|
|
|
|
if (currIsNewline(ls) && ls->current != old)
|
|
|
|
next(ls); /* skip `\n\r' or `\r\n' */
|
2004-03-12 22:53:56 +03:00
|
|
|
if (++ls->linenumber >= MAX_INT)
|
|
|
|
luaX_syntaxerror(ls, "chunk has too many lines");
|
2000-05-26 18:04:04 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-08-28 18:38:46 +04:00
|
|
|
void luaX_setinput (lua_State *L, LexState *ls, ZIO *z, TString *source) {
|
2005-12-08 18:50:54 +03:00
|
|
|
ls->decpoint = '.';
|
2003-08-28 18:38:46 +04:00
|
|
|
ls->L = L;
|
|
|
|
ls->lookahead.token = TK_EOS; /* no look-ahead token */
|
|
|
|
ls->z = z;
|
|
|
|
ls->fs = NULL;
|
|
|
|
ls->linenumber = 1;
|
|
|
|
ls->lastline = 1;
|
|
|
|
ls->source = source;
|
2010-04-16 16:31:07 +04:00
|
|
|
ls->envn = luaS_new(L, "_ENV"); /* create env name */
|
2010-04-05 20:35:37 +04:00
|
|
|
luaS_fix(ls->envn); /* never collect this name */
|
2004-12-01 18:46:18 +03:00
|
|
|
luaZ_resizebuffer(ls->L, ls->buff, LUA_MINBUFFER); /* initialize buffer */
|
2003-08-28 18:38:46 +04:00
|
|
|
next(ls); /* read first char */
|
2000-05-26 18:04:04 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
|
1998-12-27 23:25:20 +03:00
|
|
|
|
1997-09-16 23:25:59 +04:00
|
|
|
/*
|
|
|
|
** =======================================================
|
1999-12-14 21:33:29 +03:00
|
|
|
** LEXICAL ANALYZER
|
1997-09-16 23:25:59 +04:00
|
|
|
** =======================================================
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
2000-09-11 21:38:42 +04:00
|
|
|
|
2005-11-08 22:45:14 +03:00
|
|
|
static int check_next (LexState *ls, const char *set) {
|
2006-09-18 20:06:41 +04:00
|
|
|
if (ls->current == '\0' || !strchr(set, ls->current))
|
2005-11-08 22:45:14 +03:00
|
|
|
return 0;
|
|
|
|
save_and_next(ls);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2010-04-16 16:31:07 +04:00
|
|
|
/*
|
|
|
|
** change all characters 'from' in buffer to 'to'
|
|
|
|
*/
|
2005-12-08 18:50:54 +03:00
|
|
|
static void buffreplace (LexState *ls, char from, char to) {
|
2006-02-06 21:28:16 +03:00
|
|
|
size_t n = luaZ_bufflen(ls->buff);
|
2005-12-07 18:33:27 +03:00
|
|
|
char *p = luaZ_buffer(ls->buff);
|
|
|
|
while (n--)
|
|
|
|
if (p[n] == from) p[n] = to;
|
|
|
|
}
|
|
|
|
|
2003-08-28 18:38:46 +04:00
|
|
|
|
2007-09-14 17:27:04 +04:00
|
|
|
#if !defined(getlocaledecpoint)
|
|
|
|
#define getlocaledecpoint() (localeconv()->decimal_point[0])
|
|
|
|
#endif
|
|
|
|
|
2010-04-16 16:31:07 +04:00
|
|
|
/*
|
|
|
|
** in case of format error, try to change decimal point separator to
|
|
|
|
** the one defined in the current locale and check again
|
|
|
|
*/
|
2005-12-08 18:50:54 +03:00
|
|
|
static void trydecpoint (LexState *ls, SemInfo *seminfo) {
|
|
|
|
char old = ls->decpoint;
|
2007-09-14 17:27:04 +04:00
|
|
|
ls->decpoint = getlocaledecpoint();
|
2010-04-16 16:31:07 +04:00
|
|
|
buffreplace(ls, old, ls->decpoint); /* try new decimal separator */
|
2005-12-08 18:50:54 +03:00
|
|
|
if (!luaO_str2d(luaZ_buffer(ls->buff), &seminfo->r)) {
|
|
|
|
/* format error with correct decimal point: no more options */
|
|
|
|
buffreplace(ls, ls->decpoint, '.'); /* undo change (for error message) */
|
2007-10-25 20:45:47 +04:00
|
|
|
lexerror(ls, "malformed number", TK_NUMBER);
|
2005-12-08 18:50:54 +03:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2000-09-11 21:38:42 +04:00
|
|
|
/* LUA_NUMBER */
|
2003-08-28 18:38:46 +04:00
|
|
|
static void read_numeral (LexState *ls, SemInfo *seminfo) {
|
2009-02-19 20:33:51 +03:00
|
|
|
lua_assert(lisdigit(ls->current));
|
2005-11-08 22:45:14 +03:00
|
|
|
do {
|
2003-08-28 18:38:46 +04:00
|
|
|
save_and_next(ls);
|
2009-02-19 20:33:51 +03:00
|
|
|
} while (lisdigit(ls->current) || ls->current == '.');
|
2006-01-23 23:06:19 +03:00
|
|
|
if (check_next(ls, "Ee")) /* `E'? */
|
2005-11-08 22:45:14 +03:00
|
|
|
check_next(ls, "+-"); /* optional exponent sign */
|
2009-03-11 16:27:32 +03:00
|
|
|
while (lislalnum(ls->current))
|
2006-01-23 23:06:19 +03:00
|
|
|
save_and_next(ls);
|
2003-08-28 18:38:46 +04:00
|
|
|
save(ls, '\0');
|
2005-12-08 18:50:54 +03:00
|
|
|
buffreplace(ls, '.', ls->decpoint); /* follow locale for decimal point */
|
|
|
|
if (!luaO_str2d(luaZ_buffer(ls->buff), &seminfo->r)) /* format error? */
|
|
|
|
trydecpoint(ls, seminfo); /* try to update decimal point separator */
|
2000-09-11 21:38:42 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2010-04-16 16:31:07 +04:00
|
|
|
/*
|
|
|
|
** skip a sequence '[=*=[' or ']=*]' and return its number of '='s or
|
|
|
|
** -1 if sequence is malformed
|
|
|
|
*/
|
2004-09-22 18:02:00 +04:00
|
|
|
static int skip_sep (LexState *ls) {
|
2003-08-29 20:48:14 +04:00
|
|
|
int count = 0;
|
|
|
|
int s = ls->current;
|
|
|
|
lua_assert(s == '[' || s == ']');
|
|
|
|
save_and_next(ls);
|
2004-09-22 18:02:00 +04:00
|
|
|
while (ls->current == '=') {
|
2003-08-29 20:48:14 +04:00
|
|
|
save_and_next(ls);
|
|
|
|
count++;
|
|
|
|
}
|
|
|
|
return (ls->current == s) ? count : (-count) - 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2004-09-22 18:02:00 +04:00
|
|
|
static void read_long_string (LexState *ls, SemInfo *seminfo, int sep) {
|
2003-08-29 20:48:14 +04:00
|
|
|
save_and_next(ls); /* skip 2nd `[' */
|
2003-08-28 18:38:46 +04:00
|
|
|
if (currIsNewline(ls)) /* string starts with a newline? */
|
|
|
|
inclinenumber(ls); /* skip it */
|
1998-12-03 18:45:15 +03:00
|
|
|
for (;;) {
|
2003-08-28 18:38:46 +04:00
|
|
|
switch (ls->current) {
|
1997-06-16 20:50:22 +04:00
|
|
|
case EOZ:
|
2007-10-25 20:45:47 +04:00
|
|
|
lexerror(ls, (seminfo) ? "unfinished long string" :
|
|
|
|
"unfinished long comment", TK_EOS);
|
2000-01-25 21:44:21 +03:00
|
|
|
break; /* to avoid warnings */
|
2005-04-27 22:37:51 +04:00
|
|
|
case ']': {
|
2004-09-22 18:02:00 +04:00
|
|
|
if (skip_sep(ls) == sep) {
|
2003-08-29 20:48:14 +04:00
|
|
|
save_and_next(ls); /* skip 2nd `]' */
|
2005-04-27 22:37:51 +04:00
|
|
|
goto endloop;
|
1995-07-06 21:47:08 +04:00
|
|
|
}
|
2005-04-27 22:37:51 +04:00
|
|
|
break;
|
|
|
|
}
|
2010-04-16 16:31:07 +04:00
|
|
|
case '\n': case '\r': {
|
2003-08-28 18:38:46 +04:00
|
|
|
save(ls, '\n');
|
|
|
|
inclinenumber(ls);
|
|
|
|
if (!seminfo) luaZ_resetbuffer(ls->buff); /* avoid wasting space */
|
2005-04-27 22:37:51 +04:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
default: {
|
2003-08-28 18:38:46 +04:00
|
|
|
if (seminfo) save_and_next(ls);
|
|
|
|
else next(ls);
|
2005-04-27 22:37:51 +04:00
|
|
|
}
|
1995-07-06 21:47:08 +04:00
|
|
|
}
|
1996-05-30 18:04:07 +04:00
|
|
|
} endloop:
|
2002-03-08 22:07:01 +03:00
|
|
|
if (seminfo)
|
2004-09-22 18:02:00 +04:00
|
|
|
seminfo->ts = luaX_newstring(ls, luaZ_buffer(ls->buff) + (2 + sep),
|
|
|
|
luaZ_bufflen(ls->buff) - 2*(2 + sep));
|
2000-01-25 21:44:21 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2009-05-18 21:28:04 +04:00
|
|
|
static int hexavalue (int c) {
|
|
|
|
if (lisdigit(c)) return c - '0';
|
|
|
|
else if (lisupper(c)) return c - 'A' + 10;
|
|
|
|
else return c - 'a' + 10;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static int readhexaesc (LexState *ls) {
|
|
|
|
int c1, c2 = EOZ;
|
|
|
|
if (!lisxdigit(c1 = next(ls)) || !lisxdigit(c2 = next(ls))) {
|
|
|
|
luaZ_resetbuffer(ls->buff); /* prepare error message */
|
|
|
|
save(ls, '\\'); save(ls, 'x');
|
|
|
|
if (c1 != EOZ) save(ls, c1);
|
|
|
|
if (c2 != EOZ) save(ls, c2);
|
|
|
|
lexerror(ls, "hexadecimal digit expected", TK_STRING);
|
|
|
|
}
|
|
|
|
return (hexavalue(c1) << 4) + hexavalue(c2);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static int readdecesc (LexState *ls) {
|
|
|
|
int c1 = ls->current, c2, c3;
|
|
|
|
int c = c1 - '0';
|
|
|
|
if (lisdigit(c2 = next(ls))) {
|
|
|
|
c = 10*c + c2 - '0';
|
|
|
|
if (lisdigit(c3 = next(ls))) {
|
|
|
|
c = 10*c + c3 - '0';
|
|
|
|
if (c > UCHAR_MAX) {
|
|
|
|
luaZ_resetbuffer(ls->buff); /* prepare error message */
|
|
|
|
save(ls, '\\');
|
|
|
|
save(ls, c1); save(ls, c2); save(ls, c3);
|
|
|
|
lexerror(ls, "decimal escape too large", TK_STRING);
|
|
|
|
}
|
|
|
|
return c;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
/* else, has read one character that was not a digit */
|
|
|
|
zungetc(ls->z); /* return it to input stream */
|
|
|
|
return c;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-08-28 18:38:46 +04:00
|
|
|
static void read_string (LexState *ls, int del, SemInfo *seminfo) {
|
2010-04-16 16:31:07 +04:00
|
|
|
save_and_next(ls); /* keep delimiter (for error messages) */
|
2003-08-28 18:38:46 +04:00
|
|
|
while (ls->current != del) {
|
|
|
|
switch (ls->current) {
|
2002-05-07 21:36:56 +04:00
|
|
|
case EOZ:
|
2007-10-25 20:45:47 +04:00
|
|
|
lexerror(ls, "unfinished string", TK_EOS);
|
2010-02-28 00:16:24 +03:00
|
|
|
break; /* to avoid warnings */
|
2002-05-07 21:36:56 +04:00
|
|
|
case '\n':
|
2003-08-21 18:16:43 +04:00
|
|
|
case '\r':
|
2007-10-25 20:45:47 +04:00
|
|
|
lexerror(ls, "unfinished string", TK_STRING);
|
2010-02-28 00:16:24 +03:00
|
|
|
break; /* to avoid warnings */
|
2009-05-18 21:28:04 +04:00
|
|
|
case '\\': { /* escape sequences */
|
|
|
|
int c; /* final character to be saved */
|
2003-08-28 18:38:46 +04:00
|
|
|
next(ls); /* do not save the `\' */
|
|
|
|
switch (ls->current) {
|
|
|
|
case 'a': c = '\a'; break;
|
|
|
|
case 'b': c = '\b'; break;
|
|
|
|
case 'f': c = '\f'; break;
|
|
|
|
case 'n': c = '\n'; break;
|
|
|
|
case 'r': c = '\r'; break;
|
|
|
|
case 't': c = '\t'; break;
|
|
|
|
case 'v': c = '\v'; break;
|
2009-05-18 21:28:04 +04:00
|
|
|
case 'x': c = readhexaesc(ls); break;
|
2009-02-11 21:25:20 +03:00
|
|
|
case '\n':
|
2003-08-28 18:38:46 +04:00
|
|
|
case '\r': save(ls, '\n'); inclinenumber(ls); continue;
|
|
|
|
case EOZ: continue; /* will raise an error next loop */
|
2010-04-16 16:31:07 +04:00
|
|
|
case '*': { /* skip following span of spaces */
|
|
|
|
next(ls); /* skip the '*' */
|
|
|
|
while (lisspace(ls->current)) {
|
|
|
|
if (currIsNewline(ls)) inclinenumber(ls);
|
|
|
|
else next(ls);
|
|
|
|
}
|
|
|
|
continue; /* do not save 'c' */
|
|
|
|
}
|
2001-02-23 20:17:25 +03:00
|
|
|
default: {
|
2009-02-19 20:33:51 +03:00
|
|
|
if (!lisdigit(ls->current))
|
2009-05-18 21:28:04 +04:00
|
|
|
c = ls->current; /* handles \\, \", \', and \? */
|
|
|
|
else /* digital escape \ddd */
|
|
|
|
c = readdecesc(ls);
|
|
|
|
break;
|
2000-01-25 21:44:21 +03:00
|
|
|
}
|
|
|
|
}
|
2003-08-28 18:38:46 +04:00
|
|
|
next(ls);
|
2009-05-18 21:28:04 +04:00
|
|
|
save(ls, c);
|
2010-02-28 00:16:24 +03:00
|
|
|
break;
|
2003-08-28 18:38:46 +04:00
|
|
|
}
|
2000-01-25 21:44:21 +03:00
|
|
|
default:
|
2003-08-28 18:38:46 +04:00
|
|
|
save_and_next(ls);
|
2000-01-25 21:44:21 +03:00
|
|
|
}
|
|
|
|
}
|
2003-08-28 18:38:46 +04:00
|
|
|
save_and_next(ls); /* skip delimiter */
|
|
|
|
seminfo->ts = luaX_newstring(ls, luaZ_buffer(ls->buff) + 1,
|
|
|
|
luaZ_bufflen(ls->buff) - 2);
|
1995-07-06 21:47:08 +04:00
|
|
|
}
|
|
|
|
|
1997-09-16 23:25:59 +04:00
|
|
|
|
2005-12-07 18:43:05 +03:00
|
|
|
static int llex (LexState *ls, SemInfo *seminfo) {
|
2003-08-28 18:38:46 +04:00
|
|
|
luaZ_resetbuffer(ls->buff);
|
1998-12-03 18:45:15 +03:00
|
|
|
for (;;) {
|
2003-08-28 18:38:46 +04:00
|
|
|
switch (ls->current) {
|
2010-04-16 16:31:07 +04:00
|
|
|
case '\n': case '\r': { /* line breaks */
|
2003-08-28 18:38:46 +04:00
|
|
|
inclinenumber(ls);
|
2010-02-28 00:16:24 +03:00
|
|
|
break;
|
2002-06-04 00:12:21 +04:00
|
|
|
}
|
2010-04-16 16:31:07 +04:00
|
|
|
case ' ': case '\f': case '\t': case '\v': { /* spaces */
|
|
|
|
next(ls);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
case '-': { /* '-' or '--' (comment) */
|
2003-08-28 18:38:46 +04:00
|
|
|
next(ls);
|
|
|
|
if (ls->current != '-') return '-';
|
2002-03-08 22:07:01 +03:00
|
|
|
/* else is a comment */
|
2003-08-28 18:38:46 +04:00
|
|
|
next(ls);
|
2010-04-16 16:31:07 +04:00
|
|
|
if (ls->current == '[') { /* long comment? */
|
2004-09-22 18:02:00 +04:00
|
|
|
int sep = skip_sep(ls);
|
|
|
|
luaZ_resetbuffer(ls->buff); /* `skip_sep' may dirty the buffer */
|
|
|
|
if (sep >= 0) {
|
2010-04-16 16:31:07 +04:00
|
|
|
read_long_string(ls, NULL, sep); /* skip long comment */
|
|
|
|
luaZ_resetbuffer(ls->buff); /* previous call may dirty the buff. */
|
2010-02-28 00:16:24 +03:00
|
|
|
break;
|
2003-08-29 20:48:14 +04:00
|
|
|
}
|
2003-08-28 18:38:46 +04:00
|
|
|
}
|
2003-08-29 20:48:14 +04:00
|
|
|
/* else short comment */
|
|
|
|
while (!currIsNewline(ls) && ls->current != EOZ)
|
2010-04-16 16:31:07 +04:00
|
|
|
next(ls); /* skip until end of line (or end of file) */
|
2010-02-28 00:16:24 +03:00
|
|
|
break;
|
2002-06-04 00:12:21 +04:00
|
|
|
}
|
2010-04-16 16:31:07 +04:00
|
|
|
case '[': { /* long string or simply '[' */
|
2004-09-22 18:02:00 +04:00
|
|
|
int sep = skip_sep(ls);
|
|
|
|
if (sep >= 0) {
|
|
|
|
read_long_string(ls, seminfo, sep);
|
2000-03-10 21:37:44 +03:00
|
|
|
return TK_STRING;
|
1995-07-06 21:47:08 +04:00
|
|
|
}
|
2004-09-22 18:02:00 +04:00
|
|
|
else if (sep == -1) return '[';
|
2007-10-25 20:45:47 +04:00
|
|
|
else lexerror(ls, "invalid long string delimiter", TK_STRING);
|
2002-06-04 00:12:21 +04:00
|
|
|
}
|
|
|
|
case '=': {
|
2003-08-28 18:38:46 +04:00
|
|
|
next(ls);
|
|
|
|
if (ls->current != '=') return '=';
|
|
|
|
else { next(ls); return TK_EQ; }
|
2002-06-04 00:12:21 +04:00
|
|
|
}
|
|
|
|
case '<': {
|
2003-08-28 18:38:46 +04:00
|
|
|
next(ls);
|
|
|
|
if (ls->current != '=') return '<';
|
|
|
|
else { next(ls); return TK_LE; }
|
2002-06-04 00:12:21 +04:00
|
|
|
}
|
|
|
|
case '>': {
|
2003-08-28 18:38:46 +04:00
|
|
|
next(ls);
|
|
|
|
if (ls->current != '=') return '>';
|
|
|
|
else { next(ls); return TK_GE; }
|
2002-06-04 00:12:21 +04:00
|
|
|
}
|
|
|
|
case '~': {
|
2003-08-28 18:38:46 +04:00
|
|
|
next(ls);
|
|
|
|
if (ls->current != '=') return '~';
|
|
|
|
else { next(ls); return TK_NE; }
|
2002-06-04 00:12:21 +04:00
|
|
|
}
|
2010-04-16 16:31:07 +04:00
|
|
|
case '"': case '\'': { /* short literal strings */
|
2003-08-28 18:38:46 +04:00
|
|
|
read_string(ls, ls->current, seminfo);
|
2000-03-10 21:37:44 +03:00
|
|
|
return TK_STRING;
|
2002-06-04 00:12:21 +04:00
|
|
|
}
|
2010-04-16 16:31:07 +04:00
|
|
|
case '.': { /* '.', '..', '...', or number */
|
2003-08-28 18:38:46 +04:00
|
|
|
save_and_next(ls);
|
2005-11-08 22:45:14 +03:00
|
|
|
if (check_next(ls, ".")) {
|
|
|
|
if (check_next(ls, "."))
|
2010-04-16 16:31:07 +04:00
|
|
|
return TK_DOTS; /* '...' */
|
|
|
|
else return TK_CONCAT; /* '..' */
|
1993-12-23 00:15:16 +03:00
|
|
|
}
|
2009-02-19 20:33:51 +03:00
|
|
|
else if (!lisdigit(ls->current)) return '.';
|
2010-04-16 16:31:07 +04:00
|
|
|
/* else go through */
|
|
|
|
}
|
|
|
|
case '0': case '1': case '2': case '3': case '4':
|
|
|
|
case '5': case '6': case '7': case '8': case '9': {
|
|
|
|
read_numeral(ls, seminfo);
|
|
|
|
return TK_NUMBER;
|
2002-06-04 00:12:21 +04:00
|
|
|
}
|
|
|
|
case EOZ: {
|
2000-03-10 21:37:44 +03:00
|
|
|
return TK_EOS;
|
2002-06-04 00:12:21 +04:00
|
|
|
}
|
2001-02-23 20:17:25 +03:00
|
|
|
default: {
|
2010-04-16 16:31:07 +04:00
|
|
|
if (lislalpha(ls->current)) { /* identifier or reserved word? */
|
2003-08-28 18:38:46 +04:00
|
|
|
TString *ts;
|
|
|
|
do {
|
|
|
|
save_and_next(ls);
|
2009-03-11 16:27:32 +03:00
|
|
|
} while (lislalnum(ls->current));
|
2003-08-28 18:38:46 +04:00
|
|
|
ts = luaX_newstring(ls, luaZ_buffer(ls->buff),
|
|
|
|
luaZ_bufflen(ls->buff));
|
2002-08-16 18:45:55 +04:00
|
|
|
if (ts->tsv.reserved > 0) /* reserved word? */
|
|
|
|
return ts->tsv.reserved - 1 + FIRST_RESERVED;
|
2003-09-05 00:00:28 +04:00
|
|
|
else {
|
|
|
|
seminfo->ts = ts;
|
|
|
|
return TK_NAME;
|
|
|
|
}
|
1997-07-01 23:32:41 +04:00
|
|
|
}
|
2010-04-16 16:31:07 +04:00
|
|
|
else { /* single-char tokens (+ - / ...) */
|
2003-08-28 18:38:46 +04:00
|
|
|
int c = ls->current;
|
|
|
|
next(ls);
|
2010-04-16 16:31:07 +04:00
|
|
|
return c;
|
2001-02-23 20:17:25 +03:00
|
|
|
}
|
|
|
|
}
|
1993-12-23 00:15:16 +03:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
1996-05-30 18:04:07 +04:00
|
|
|
|
2005-12-07 18:43:05 +03:00
|
|
|
|
|
|
|
void luaX_next (LexState *ls) {
|
|
|
|
ls->lastline = ls->linenumber;
|
|
|
|
if (ls->lookahead.token != TK_EOS) { /* is there a look-ahead token? */
|
|
|
|
ls->t = ls->lookahead; /* use this one */
|
|
|
|
ls->lookahead.token = TK_EOS; /* and discharge it */
|
|
|
|
}
|
|
|
|
else
|
|
|
|
ls->t.token = llex(ls, &ls->t.seminfo); /* read next token */
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2007-05-11 21:28:56 +04:00
|
|
|
int luaX_lookahead (LexState *ls) {
|
2005-12-07 18:43:05 +03:00
|
|
|
lua_assert(ls->lookahead.token == TK_EOS);
|
|
|
|
ls->lookahead.token = llex(ls, &ls->lookahead.seminfo);
|
2007-05-11 21:28:56 +04:00
|
|
|
return ls->lookahead.token;
|
2005-12-07 18:43:05 +03:00
|
|
|
}
|
|
|
|
|