1997-09-17 03:25:59 +08:00
|
|
|
/*
|
2011-07-15 20:48:03 +08:00
|
|
|
** $Id: llex.c,v 2.54 2011/07/15 12:30:41 roberto Exp roberto $
|
1999-02-26 05:07:26 +08:00
|
|
|
** Lexical Analyzer
|
1997-09-17 03:25:59 +08:00
|
|
|
** See Copyright Notice in lua.h
|
|
|
|
*/
|
1996-05-30 22:04:07 +08:00
|
|
|
|
1993-12-23 05:15:16 +08:00
|
|
|
|
2005-12-07 23:33:27 +08:00
|
|
|
#include <locale.h>
|
1993-12-29 00:42:29 +08:00
|
|
|
#include <string.h>
|
1993-12-23 05:15:16 +08:00
|
|
|
|
2002-12-05 01:38:31 +08:00
|
|
|
#define llex_c
|
2004-05-01 04:13:38 +08:00
|
|
|
#define LUA_CORE
|
2002-12-05 01:38:31 +08:00
|
|
|
|
2000-06-12 21:52:05 +08:00
|
|
|
#include "lua.h"
|
|
|
|
|
2009-02-20 01:33:51 +08:00
|
|
|
#include "lctype.h"
|
2002-05-08 01:36:56 +08:00
|
|
|
#include "ldo.h"
|
1997-09-17 03:25:59 +08:00
|
|
|
#include "llex.h"
|
|
|
|
#include "lobject.h"
|
|
|
|
#include "lparser.h"
|
1997-11-20 01:29:23 +08:00
|
|
|
#include "lstate.h"
|
1997-09-17 03:25:59 +08:00
|
|
|
#include "lstring.h"
|
2006-03-10 02:15:48 +08:00
|
|
|
#include "ltable.h"
|
1997-09-17 03:25:59 +08:00
|
|
|
#include "lzio.h"
|
1993-12-23 05:15:16 +08:00
|
|
|
|
1995-09-16 04:48:26 +08:00
|
|
|
|
1996-05-30 22:04:07 +08:00
|
|
|
|
2003-08-28 22:38:46 +08:00
|
|
|
#define next(ls) (ls->current = zgetc(ls->z))
|
1997-09-17 03:25:59 +08:00
|
|
|
|
2003-10-04 00:05:34 +08:00
|
|
|
|
1997-09-17 03:25:59 +08:00
|
|
|
|
2003-08-28 22:38:46 +08:00
|
|
|
#define currIsNewline(ls) (ls->current == '\n' || ls->current == '\r')
|
2003-08-21 22:16:43 +08:00
|
|
|
|
1998-05-27 21:08:34 +08:00
|
|
|
|
1999-07-23 03:29:42 +08:00
|
|
|
/* ORDER RESERVED */
|
2007-02-08 01:49:18 +08:00
|
|
|
static const char *const luaX_tokens [] = {
|
2001-11-29 04:13:13 +08:00
|
|
|
"and", "break", "do", "else", "elseif",
|
2011-02-02 22:55:17 +08:00
|
|
|
"end", "false", "for", "function", "goto", "if",
|
2001-11-29 04:13:13 +08:00
|
|
|
"in", "local", "nil", "not", "or", "repeat",
|
2004-12-04 04:54:12 +08:00
|
|
|
"return", "then", "true", "until", "while",
|
2011-06-23 22:54:48 +08:00
|
|
|
"..", "...", "==", ">=", "<=", "~=", "::", "<eof>",
|
2007-02-08 01:49:18 +08:00
|
|
|
"<number>", "<name>", "<string>"
|
2001-02-24 01:17:25 +08:00
|
|
|
};
|
1998-05-27 21:08:34 +08:00
|
|
|
|
1997-09-17 03:25:59 +08:00
|
|
|
|
2004-12-01 23:46:18 +08:00
|
|
|
#define save_and_next(ls) (save(ls, ls->current), next(ls))
|
|
|
|
|
|
|
|
|
2007-10-26 00:45:47 +08:00
|
|
|
static void lexerror (LexState *ls, const char *msg, int token);
|
|
|
|
|
|
|
|
|
2004-12-01 23:46:18 +08:00
|
|
|
static void save (LexState *ls, int c) {
|
|
|
|
Mbuffer *b = ls->buff;
|
2009-05-19 01:28:04 +08:00
|
|
|
if (luaZ_bufflen(b) + 1 > luaZ_sizebuffer(b)) {
|
2004-12-01 23:46:18 +08:00
|
|
|
size_t newsize;
|
2009-05-19 01:28:04 +08:00
|
|
|
if (luaZ_sizebuffer(b) >= MAX_SIZET/2)
|
2007-10-26 00:45:47 +08:00
|
|
|
lexerror(ls, "lexical element too long", 0);
|
2009-05-19 01:28:04 +08:00
|
|
|
newsize = luaZ_sizebuffer(b) * 2;
|
2004-12-01 23:46:18 +08:00
|
|
|
luaZ_resizebuffer(ls->L, b, newsize);
|
|
|
|
}
|
2009-05-19 01:28:04 +08:00
|
|
|
b->buffer[luaZ_bufflen(b)++] = cast(char, c);
|
2004-12-01 23:46:18 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
1999-11-22 21:12:07 +08:00
|
|
|
void luaX_init (lua_State *L) {
|
2000-03-03 22:58:26 +08:00
|
|
|
int i;
|
2000-02-09 00:39:42 +08:00
|
|
|
for (i=0; i<NUM_RESERVED; i++) {
|
2004-12-02 20:59:10 +08:00
|
|
|
TString *ts = luaS_new(L, luaX_tokens[i]);
|
2002-08-16 22:45:55 +08:00
|
|
|
luaS_fix(ts); /* reserved words are never collected */
|
2005-12-23 00:19:56 +08:00
|
|
|
ts->tsv.reserved = cast_byte(i+1); /* reserved word */
|
1998-05-27 21:08:34 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-09-05 04:00:28 +08:00
|
|
|
const char *luaX_token2str (LexState *ls, int token) {
|
|
|
|
if (token < FIRST_RESERVED) {
|
2004-11-25 03:16:03 +08:00
|
|
|
lua_assert(token == cast(unsigned char, token));
|
2009-02-20 01:33:51 +08:00
|
|
|
return (lisprint(token)) ? luaO_pushfstring(ls->L, LUA_QL("%c"), token) :
|
2007-08-10 04:29:15 +08:00
|
|
|
luaO_pushfstring(ls->L, "char(%d)", token);
|
2007-02-08 01:49:18 +08:00
|
|
|
}
|
|
|
|
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 04:00:28 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static const char *txtToken (LexState *ls, int token) {
|
|
|
|
switch (token) {
|
2002-05-08 01:36:56 +08:00
|
|
|
case TK_NAME:
|
|
|
|
case TK_STRING:
|
|
|
|
case TK_NUMBER:
|
2003-08-28 22:38:46 +08:00
|
|
|
save(ls, '\0');
|
2007-02-08 01:49:18 +08:00
|
|
|
return luaO_pushfstring(ls->L, LUA_QS, luaZ_buffer(ls->buff));
|
2002-05-08 01:36:56 +08:00
|
|
|
default:
|
2003-09-05 04:00:28 +08:00
|
|
|
return luaX_token2str(ls, token);
|
1997-09-17 03:25:59 +08:00
|
|
|
}
|
1998-05-27 21:08:34 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2007-10-26 00:45:47 +08:00
|
|
|
static void lexerror (LexState *ls, const char *msg, int token) {
|
2008-12-26 19:55:57 +08:00
|
|
|
char buff[LUA_IDSIZE];
|
|
|
|
luaO_chunkid(buff, getstr(ls->source), LUA_IDSIZE);
|
2004-03-13 03:53:56 +08:00
|
|
|
msg = luaO_pushfstring(ls->L, "%s:%d: %s", buff, ls->linenumber, msg);
|
|
|
|
if (token)
|
2007-02-08 01:49:18 +08:00
|
|
|
luaO_pushfstring(ls->L, "%s near %s", msg, txtToken(ls, token));
|
2003-10-20 20:24:34 +08:00
|
|
|
luaD_throw(ls->L, LUA_ERRSYNTAX);
|
2001-11-17 00:29:10 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-09-05 04:00:28 +08:00
|
|
|
void luaX_syntaxerror (LexState *ls, const char *msg) {
|
2007-10-26 00:45:47 +08:00
|
|
|
lexerror(ls, msg, ls->t.token);
|
1997-09-17 03:25:59 +08:00
|
|
|
}
|
|
|
|
|
1997-04-12 23:01:49 +08:00
|
|
|
|
2010-04-16 20:31:07 +08:00
|
|
|
/*
|
|
|
|
** 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 22:38:46 +08:00
|
|
|
TString *luaX_newstring (LexState *ls, const char *str, size_t l) {
|
|
|
|
lua_State *L = ls->L;
|
2006-07-11 23:53:29 +08:00
|
|
|
TValue *o; /* entry for `str' */
|
2010-04-16 20:31:07 +08:00
|
|
|
TString *ts = luaS_newlstr(L, str, l); /* create new string */
|
|
|
|
setsvalue2s(L, L->top++, ts); /* temporarily anchor it in stack */
|
2011-01-27 00:30:02 +08:00
|
|
|
o = luaH_setstr(L, ls->fs->h, ts);
|
2009-11-18 00:33:38 +08:00
|
|
|
if (ttisnil(o)) {
|
2010-04-16 20:31:07 +08:00
|
|
|
setbvalue(o, 1); /* t[string] = true */
|
2009-11-18 00:33:38 +08:00
|
|
|
luaC_checkGC(L);
|
|
|
|
}
|
2010-04-16 20:31:07 +08:00
|
|
|
L->top--; /* remove string from stack */
|
2003-08-28 05:01:44 +08:00
|
|
|
return ts;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2010-04-16 20:31:07 +08:00
|
|
|
/*
|
|
|
|
** increment line number and skips newline sequence (any of
|
|
|
|
** \n, \r, \n\r, or \r\n)
|
|
|
|
*/
|
2003-08-28 22:38:46 +08:00
|
|
|
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' */
|
2004-03-13 03:53:56 +08:00
|
|
|
if (++ls->linenumber >= MAX_INT)
|
|
|
|
luaX_syntaxerror(ls, "chunk has too many lines");
|
2000-05-26 22:04:04 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2011-02-23 21:13:10 +08:00
|
|
|
void luaX_setinput (lua_State *L, LexState *ls, ZIO *z, TString *source,
|
|
|
|
int firstchar) {
|
2005-12-08 23:50:54 +08:00
|
|
|
ls->decpoint = '.';
|
2003-08-28 22:38:46 +08:00
|
|
|
ls->L = L;
|
2011-02-23 21:13:10 +08:00
|
|
|
ls->current = firstchar;
|
2003-08-28 22:38:46 +08:00
|
|
|
ls->lookahead.token = TK_EOS; /* no look-ahead token */
|
|
|
|
ls->z = z;
|
|
|
|
ls->fs = NULL;
|
|
|
|
ls->linenumber = 1;
|
|
|
|
ls->lastline = 1;
|
|
|
|
ls->source = source;
|
2010-09-08 03:21:39 +08:00
|
|
|
ls->envn = luaS_new(L, LUA_ENV); /* create env name */
|
2010-04-06 00:35:37 +08:00
|
|
|
luaS_fix(ls->envn); /* never collect this name */
|
2004-12-01 23:46:18 +08:00
|
|
|
luaZ_resizebuffer(ls->L, ls->buff, LUA_MINBUFFER); /* initialize buffer */
|
2000-05-26 22:04:04 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
1998-12-28 04:25:20 +08:00
|
|
|
|
1997-09-17 03:25:59 +08:00
|
|
|
/*
|
|
|
|
** =======================================================
|
1999-12-15 02:33:29 +08:00
|
|
|
** LEXICAL ANALYZER
|
1997-09-17 03:25:59 +08:00
|
|
|
** =======================================================
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
2000-09-12 01:38:42 +08:00
|
|
|
|
2005-11-09 03:45:14 +08:00
|
|
|
static int check_next (LexState *ls, const char *set) {
|
2006-09-19 00:06:41 +08:00
|
|
|
if (ls->current == '\0' || !strchr(set, ls->current))
|
2005-11-09 03:45:14 +08:00
|
|
|
return 0;
|
|
|
|
save_and_next(ls);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2011-01-27 00:30:02 +08:00
|
|
|
/*
|
2010-04-16 20:31:07 +08:00
|
|
|
** change all characters 'from' in buffer to 'to'
|
|
|
|
*/
|
2005-12-08 23:50:54 +08:00
|
|
|
static void buffreplace (LexState *ls, char from, char to) {
|
2006-02-07 02:28:16 +08:00
|
|
|
size_t n = luaZ_bufflen(ls->buff);
|
2005-12-07 23:33:27 +08:00
|
|
|
char *p = luaZ_buffer(ls->buff);
|
|
|
|
while (n--)
|
|
|
|
if (p[n] == from) p[n] = to;
|
|
|
|
}
|
|
|
|
|
2003-08-28 22:38:46 +08:00
|
|
|
|
2007-09-14 21:27:04 +08:00
|
|
|
#if !defined(getlocaledecpoint)
|
|
|
|
#define getlocaledecpoint() (localeconv()->decimal_point[0])
|
|
|
|
#endif
|
|
|
|
|
2010-12-07 05:08:36 +08:00
|
|
|
|
|
|
|
#define buff2d(b,e) luaO_str2d(luaZ_buffer(b), luaZ_bufflen(b) - 1, e)
|
|
|
|
|
2010-04-16 20:31:07 +08:00
|
|
|
/*
|
|
|
|
** 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 23:50:54 +08:00
|
|
|
static void trydecpoint (LexState *ls, SemInfo *seminfo) {
|
|
|
|
char old = ls->decpoint;
|
2007-09-14 21:27:04 +08:00
|
|
|
ls->decpoint = getlocaledecpoint();
|
2010-04-16 20:31:07 +08:00
|
|
|
buffreplace(ls, old, ls->decpoint); /* try new decimal separator */
|
2010-12-07 05:08:36 +08:00
|
|
|
if (!buff2d(ls->buff, &seminfo->r)) {
|
2005-12-08 23:50:54 +08:00
|
|
|
/* format error with correct decimal point: no more options */
|
|
|
|
buffreplace(ls, ls->decpoint, '.'); /* undo change (for error message) */
|
2007-10-26 00:45:47 +08:00
|
|
|
lexerror(ls, "malformed number", TK_NUMBER);
|
2005-12-08 23:50:54 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2000-09-12 01:38:42 +08:00
|
|
|
/* LUA_NUMBER */
|
2003-08-28 22:38:46 +08:00
|
|
|
static void read_numeral (LexState *ls, SemInfo *seminfo) {
|
2009-02-20 01:33:51 +08:00
|
|
|
lua_assert(lisdigit(ls->current));
|
2005-11-09 03:45:14 +08:00
|
|
|
do {
|
2003-08-28 22:38:46 +08:00
|
|
|
save_and_next(ls);
|
2010-09-14 05:49:58 +08:00
|
|
|
if (check_next(ls, "EePp")) /* exponent part? */
|
|
|
|
check_next(ls, "+-"); /* optional exponent sign */
|
|
|
|
} while (lislalnum(ls->current) || ls->current == '.');
|
2003-08-28 22:38:46 +08:00
|
|
|
save(ls, '\0');
|
2005-12-08 23:50:54 +08:00
|
|
|
buffreplace(ls, '.', ls->decpoint); /* follow locale for decimal point */
|
2010-12-07 05:08:36 +08:00
|
|
|
if (!buff2d(ls->buff, &seminfo->r)) /* format error? */
|
2005-12-08 23:50:54 +08:00
|
|
|
trydecpoint(ls, seminfo); /* try to update decimal point separator */
|
2000-09-12 01:38:42 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2010-04-16 20:31:07 +08:00
|
|
|
/*
|
2010-10-25 20:24:36 +08:00
|
|
|
** skip a sequence '[=*[' or ']=*]' and return its number of '='s or
|
2010-04-16 20:31:07 +08:00
|
|
|
** -1 if sequence is malformed
|
|
|
|
*/
|
2004-09-22 22:02:00 +08:00
|
|
|
static int skip_sep (LexState *ls) {
|
2003-08-30 00:48:14 +08:00
|
|
|
int count = 0;
|
|
|
|
int s = ls->current;
|
|
|
|
lua_assert(s == '[' || s == ']');
|
|
|
|
save_and_next(ls);
|
2004-09-22 22:02:00 +08:00
|
|
|
while (ls->current == '=') {
|
2003-08-30 00:48:14 +08:00
|
|
|
save_and_next(ls);
|
|
|
|
count++;
|
|
|
|
}
|
|
|
|
return (ls->current == s) ? count : (-count) - 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2004-09-22 22:02:00 +08:00
|
|
|
static void read_long_string (LexState *ls, SemInfo *seminfo, int sep) {
|
2003-08-30 00:48:14 +08:00
|
|
|
save_and_next(ls); /* skip 2nd `[' */
|
2003-08-28 22:38:46 +08:00
|
|
|
if (currIsNewline(ls)) /* string starts with a newline? */
|
|
|
|
inclinenumber(ls); /* skip it */
|
1998-12-03 23:45:15 +08:00
|
|
|
for (;;) {
|
2003-08-28 22:38:46 +08:00
|
|
|
switch (ls->current) {
|
1997-06-17 00:50:22 +08:00
|
|
|
case EOZ:
|
2007-10-26 00:45:47 +08:00
|
|
|
lexerror(ls, (seminfo) ? "unfinished long string" :
|
|
|
|
"unfinished long comment", TK_EOS);
|
2000-01-26 02:44:21 +08:00
|
|
|
break; /* to avoid warnings */
|
2005-04-28 02:37:51 +08:00
|
|
|
case ']': {
|
2004-09-22 22:02:00 +08:00
|
|
|
if (skip_sep(ls) == sep) {
|
2003-08-30 00:48:14 +08:00
|
|
|
save_and_next(ls); /* skip 2nd `]' */
|
2005-04-28 02:37:51 +08:00
|
|
|
goto endloop;
|
1995-07-07 01:47:08 +08:00
|
|
|
}
|
2005-04-28 02:37:51 +08:00
|
|
|
break;
|
|
|
|
}
|
2010-04-16 20:31:07 +08:00
|
|
|
case '\n': case '\r': {
|
2003-08-28 22:38:46 +08:00
|
|
|
save(ls, '\n');
|
|
|
|
inclinenumber(ls);
|
|
|
|
if (!seminfo) luaZ_resetbuffer(ls->buff); /* avoid wasting space */
|
2005-04-28 02:37:51 +08:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
default: {
|
2003-08-28 22:38:46 +08:00
|
|
|
if (seminfo) save_and_next(ls);
|
|
|
|
else next(ls);
|
2005-04-28 02:37:51 +08:00
|
|
|
}
|
1995-07-07 01:47:08 +08:00
|
|
|
}
|
1996-05-30 22:04:07 +08:00
|
|
|
} endloop:
|
2002-03-09 03:07:01 +08:00
|
|
|
if (seminfo)
|
2004-09-22 22:02:00 +08:00
|
|
|
seminfo->ts = luaX_newstring(ls, luaZ_buffer(ls->buff) + (2 + sep),
|
|
|
|
luaZ_bufflen(ls->buff) - 2*(2 + sep));
|
2000-01-26 02:44:21 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2011-07-09 03:17:30 +08:00
|
|
|
static void escerror (LexState *ls, int *c, int n, const char *msg) {
|
|
|
|
int i;
|
|
|
|
luaZ_resetbuffer(ls->buff); /* prepare error message */
|
|
|
|
save(ls, '\\');
|
|
|
|
for (i = 0; i < n && c[i] != EOZ; i++)
|
|
|
|
save(ls, c[i]);
|
|
|
|
lexerror(ls, msg, TK_STRING);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2009-05-19 01:28:04 +08:00
|
|
|
static int readhexaesc (LexState *ls) {
|
2011-07-15 20:30:41 +08:00
|
|
|
int c[3], i; /* keep input for error message */
|
|
|
|
int r = 0; /* result accumulator */
|
|
|
|
c[0] = 'x'; /* for error message */
|
|
|
|
for (i = 1; i < 3; i++) { /* read two hexa digits */
|
|
|
|
c[i] = next(ls);
|
|
|
|
if (!lisxdigit(c[i]))
|
|
|
|
escerror(ls, c, i + 1, "hexadecimal digit expected");
|
|
|
|
r = (r << 4) + luaO_hexavalue(c[i]);
|
2009-05-19 01:28:04 +08:00
|
|
|
}
|
2011-07-15 20:30:41 +08:00
|
|
|
return r;
|
2009-05-19 01:28:04 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static int readdecesc (LexState *ls) {
|
2011-07-15 20:30:41 +08:00
|
|
|
int c[3], i;
|
|
|
|
int r = 0; /* result accumulator */
|
|
|
|
for (i = 0; i < 3 && lisdigit(ls->current); i++) { /* read up to 3 digits */
|
|
|
|
c[i] = ls->current;
|
|
|
|
r = 10*r + c[i] - '0';
|
|
|
|
next(ls);
|
2009-05-19 01:28:04 +08:00
|
|
|
}
|
2011-07-15 20:30:41 +08:00
|
|
|
if (r > UCHAR_MAX)
|
|
|
|
escerror(ls, c, i, "decimal escape too large");
|
2011-07-09 03:17:30 +08:00
|
|
|
return r;
|
2009-05-19 01:28:04 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2003-08-28 22:38:46 +08:00
|
|
|
static void read_string (LexState *ls, int del, SemInfo *seminfo) {
|
2010-04-16 20:31:07 +08:00
|
|
|
save_and_next(ls); /* keep delimiter (for error messages) */
|
2003-08-28 22:38:46 +08:00
|
|
|
while (ls->current != del) {
|
|
|
|
switch (ls->current) {
|
2002-05-08 01:36:56 +08:00
|
|
|
case EOZ:
|
2007-10-26 00:45:47 +08:00
|
|
|
lexerror(ls, "unfinished string", TK_EOS);
|
2010-02-28 05:16:24 +08:00
|
|
|
break; /* to avoid warnings */
|
2002-05-08 01:36:56 +08:00
|
|
|
case '\n':
|
2003-08-21 22:16:43 +08:00
|
|
|
case '\r':
|
2007-10-26 00:45:47 +08:00
|
|
|
lexerror(ls, "unfinished string", TK_STRING);
|
2010-02-28 05:16:24 +08:00
|
|
|
break; /* to avoid warnings */
|
2009-05-19 01:28:04 +08:00
|
|
|
case '\\': { /* escape sequences */
|
|
|
|
int c; /* final character to be saved */
|
2003-08-28 22:38:46 +08:00
|
|
|
next(ls); /* do not save the `\' */
|
|
|
|
switch (ls->current) {
|
2011-07-15 20:48:03 +08:00
|
|
|
case 'a': c = '\a'; goto read_save;
|
|
|
|
case 'b': c = '\b'; goto read_save;
|
|
|
|
case 'f': c = '\f'; goto read_save;
|
|
|
|
case 'n': c = '\n'; goto read_save;
|
|
|
|
case 'r': c = '\r'; goto read_save;
|
|
|
|
case 't': c = '\t'; goto read_save;
|
|
|
|
case 'v': c = '\v'; goto read_save;
|
|
|
|
case 'x': c = readhexaesc(ls); goto read_save;
|
|
|
|
case '\n': case '\r':
|
|
|
|
inclinenumber(ls); c = '\n'; goto only_save;
|
|
|
|
case '\\': case '\"': case '\'':
|
|
|
|
c = ls->current; goto read_save;
|
|
|
|
case EOZ: goto no_save; /* will raise an error next loop */
|
2011-07-06 03:13:45 +08:00
|
|
|
case 'z': { /* zap following span of spaces */
|
2011-07-07 00:45:14 +08:00
|
|
|
next(ls); /* skip the 'z' */
|
2010-04-16 20:31:07 +08:00
|
|
|
while (lisspace(ls->current)) {
|
|
|
|
if (currIsNewline(ls)) inclinenumber(ls);
|
|
|
|
else next(ls);
|
|
|
|
}
|
2011-07-15 20:48:03 +08:00
|
|
|
goto no_save;
|
2010-04-16 20:31:07 +08:00
|
|
|
}
|
2001-02-24 01:17:25 +08:00
|
|
|
default: {
|
2009-02-20 01:33:51 +08:00
|
|
|
if (!lisdigit(ls->current))
|
2011-07-09 03:17:30 +08:00
|
|
|
escerror(ls, &ls->current, 1, "invalid escape sequence");
|
|
|
|
/* digital escape \ddd */
|
|
|
|
c = readdecesc(ls);
|
2011-07-15 20:48:03 +08:00
|
|
|
goto only_save;
|
2000-01-26 02:44:21 +08:00
|
|
|
}
|
|
|
|
}
|
2011-07-15 20:48:03 +08:00
|
|
|
read_save: next(ls); /* read next character */
|
|
|
|
only_save: save(ls, c); /* save 'c' */
|
|
|
|
no_save: break;
|
2003-08-28 22:38:46 +08:00
|
|
|
}
|
2000-01-26 02:44:21 +08:00
|
|
|
default:
|
2003-08-28 22:38:46 +08:00
|
|
|
save_and_next(ls);
|
2000-01-26 02:44:21 +08:00
|
|
|
}
|
|
|
|
}
|
2003-08-28 22:38:46 +08:00
|
|
|
save_and_next(ls); /* skip delimiter */
|
|
|
|
seminfo->ts = luaX_newstring(ls, luaZ_buffer(ls->buff) + 1,
|
|
|
|
luaZ_bufflen(ls->buff) - 2);
|
1995-07-07 01:47:08 +08:00
|
|
|
}
|
|
|
|
|
1997-09-17 03:25:59 +08:00
|
|
|
|
2005-12-07 23:43:05 +08:00
|
|
|
static int llex (LexState *ls, SemInfo *seminfo) {
|
2003-08-28 22:38:46 +08:00
|
|
|
luaZ_resetbuffer(ls->buff);
|
1998-12-03 23:45:15 +08:00
|
|
|
for (;;) {
|
2003-08-28 22:38:46 +08:00
|
|
|
switch (ls->current) {
|
2010-04-16 20:31:07 +08:00
|
|
|
case '\n': case '\r': { /* line breaks */
|
2003-08-28 22:38:46 +08:00
|
|
|
inclinenumber(ls);
|
2010-02-28 05:16:24 +08:00
|
|
|
break;
|
2002-06-04 04:12:21 +08:00
|
|
|
}
|
2010-04-16 20:31:07 +08:00
|
|
|
case ' ': case '\f': case '\t': case '\v': { /* spaces */
|
|
|
|
next(ls);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
case '-': { /* '-' or '--' (comment) */
|
2003-08-28 22:38:46 +08:00
|
|
|
next(ls);
|
|
|
|
if (ls->current != '-') return '-';
|
2002-03-09 03:07:01 +08:00
|
|
|
/* else is a comment */
|
2003-08-28 22:38:46 +08:00
|
|
|
next(ls);
|
2010-04-16 20:31:07 +08:00
|
|
|
if (ls->current == '[') { /* long comment? */
|
2004-09-22 22:02:00 +08:00
|
|
|
int sep = skip_sep(ls);
|
|
|
|
luaZ_resetbuffer(ls->buff); /* `skip_sep' may dirty the buffer */
|
|
|
|
if (sep >= 0) {
|
2010-04-16 20:31:07 +08:00
|
|
|
read_long_string(ls, NULL, sep); /* skip long comment */
|
|
|
|
luaZ_resetbuffer(ls->buff); /* previous call may dirty the buff. */
|
2010-02-28 05:16:24 +08:00
|
|
|
break;
|
2003-08-30 00:48:14 +08:00
|
|
|
}
|
2003-08-28 22:38:46 +08:00
|
|
|
}
|
2003-08-30 00:48:14 +08:00
|
|
|
/* else short comment */
|
|
|
|
while (!currIsNewline(ls) && ls->current != EOZ)
|
2010-04-16 20:31:07 +08:00
|
|
|
next(ls); /* skip until end of line (or end of file) */
|
2010-02-28 05:16:24 +08:00
|
|
|
break;
|
2002-06-04 04:12:21 +08:00
|
|
|
}
|
2010-04-16 20:31:07 +08:00
|
|
|
case '[': { /* long string or simply '[' */
|
2004-09-22 22:02:00 +08:00
|
|
|
int sep = skip_sep(ls);
|
|
|
|
if (sep >= 0) {
|
|
|
|
read_long_string(ls, seminfo, sep);
|
2000-03-11 02:37:44 +08:00
|
|
|
return TK_STRING;
|
1995-07-07 01:47:08 +08:00
|
|
|
}
|
2004-09-22 22:02:00 +08:00
|
|
|
else if (sep == -1) return '[';
|
2007-10-26 00:45:47 +08:00
|
|
|
else lexerror(ls, "invalid long string delimiter", TK_STRING);
|
2002-06-04 04:12:21 +08:00
|
|
|
}
|
|
|
|
case '=': {
|
2003-08-28 22:38:46 +08:00
|
|
|
next(ls);
|
|
|
|
if (ls->current != '=') return '=';
|
|
|
|
else { next(ls); return TK_EQ; }
|
2002-06-04 04:12:21 +08:00
|
|
|
}
|
|
|
|
case '<': {
|
2003-08-28 22:38:46 +08:00
|
|
|
next(ls);
|
|
|
|
if (ls->current != '=') return '<';
|
|
|
|
else { next(ls); return TK_LE; }
|
2002-06-04 04:12:21 +08:00
|
|
|
}
|
|
|
|
case '>': {
|
2003-08-28 22:38:46 +08:00
|
|
|
next(ls);
|
|
|
|
if (ls->current != '=') return '>';
|
|
|
|
else { next(ls); return TK_GE; }
|
2002-06-04 04:12:21 +08:00
|
|
|
}
|
|
|
|
case '~': {
|
2003-08-28 22:38:46 +08:00
|
|
|
next(ls);
|
|
|
|
if (ls->current != '=') return '~';
|
|
|
|
else { next(ls); return TK_NE; }
|
2002-06-04 04:12:21 +08:00
|
|
|
}
|
2011-06-23 22:54:48 +08:00
|
|
|
case ':': {
|
|
|
|
next(ls);
|
|
|
|
if (ls->current != ':') return ':';
|
|
|
|
else { next(ls); return TK_DBCOLON; }
|
|
|
|
}
|
2010-04-16 20:31:07 +08:00
|
|
|
case '"': case '\'': { /* short literal strings */
|
2003-08-28 22:38:46 +08:00
|
|
|
read_string(ls, ls->current, seminfo);
|
2000-03-11 02:37:44 +08:00
|
|
|
return TK_STRING;
|
2002-06-04 04:12:21 +08:00
|
|
|
}
|
2010-04-16 20:31:07 +08:00
|
|
|
case '.': { /* '.', '..', '...', or number */
|
2003-08-28 22:38:46 +08:00
|
|
|
save_and_next(ls);
|
2005-11-09 03:45:14 +08:00
|
|
|
if (check_next(ls, ".")) {
|
|
|
|
if (check_next(ls, "."))
|
2010-04-16 20:31:07 +08:00
|
|
|
return TK_DOTS; /* '...' */
|
|
|
|
else return TK_CONCAT; /* '..' */
|
1993-12-23 05:15:16 +08:00
|
|
|
}
|
2009-02-20 01:33:51 +08:00
|
|
|
else if (!lisdigit(ls->current)) return '.';
|
2010-04-16 20:31:07 +08:00
|
|
|
/* 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 04:12:21 +08:00
|
|
|
}
|
|
|
|
case EOZ: {
|
2000-03-11 02:37:44 +08:00
|
|
|
return TK_EOS;
|
2002-06-04 04:12:21 +08:00
|
|
|
}
|
2001-02-24 01:17:25 +08:00
|
|
|
default: {
|
2010-04-16 20:31:07 +08:00
|
|
|
if (lislalpha(ls->current)) { /* identifier or reserved word? */
|
2003-08-28 22:38:46 +08:00
|
|
|
TString *ts;
|
|
|
|
do {
|
|
|
|
save_and_next(ls);
|
2009-03-11 21:27:32 +08:00
|
|
|
} while (lislalnum(ls->current));
|
2003-08-28 22:38:46 +08:00
|
|
|
ts = luaX_newstring(ls, luaZ_buffer(ls->buff),
|
|
|
|
luaZ_bufflen(ls->buff));
|
2010-11-19 02:38:44 +08:00
|
|
|
seminfo->ts = ts;
|
2002-08-16 22:45:55 +08:00
|
|
|
if (ts->tsv.reserved > 0) /* reserved word? */
|
|
|
|
return ts->tsv.reserved - 1 + FIRST_RESERVED;
|
2003-09-05 04:00:28 +08:00
|
|
|
else {
|
|
|
|
return TK_NAME;
|
|
|
|
}
|
1997-07-02 03:32:41 +08:00
|
|
|
}
|
2010-04-16 20:31:07 +08:00
|
|
|
else { /* single-char tokens (+ - / ...) */
|
2003-08-28 22:38:46 +08:00
|
|
|
int c = ls->current;
|
|
|
|
next(ls);
|
2010-04-16 20:31:07 +08:00
|
|
|
return c;
|
2001-02-24 01:17:25 +08:00
|
|
|
}
|
|
|
|
}
|
1993-12-23 05:15:16 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
1996-05-30 22:04:07 +08:00
|
|
|
|
2005-12-07 23:43:05 +08:00
|
|
|
|
|
|
|
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 */
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2007-05-12 01:28:56 +08:00
|
|
|
int luaX_lookahead (LexState *ls) {
|
2005-12-07 23:43:05 +08:00
|
|
|
lua_assert(ls->lookahead.token == TK_EOS);
|
|
|
|
ls->lookahead.token = llex(ls, &ls->lookahead.seminfo);
|
2007-05-12 01:28:56 +08:00
|
|
|
return ls->lookahead.token;
|
2005-12-07 23:43:05 +08:00
|
|
|
}
|
|
|
|
|