mirror of
https://gitcode.com/JianFeeeee/LuaCangjia_api.git
synced 2026-09-21 01:18:10 +00:00
vendor: 引入 Lua 5.0.3 官方源码 (lua_5.0 分支)
桥接层适配尚未完成。Lua 5.0 与 5.1+ 差异显著: - 无 linit.c/loslib.c(luaopen_os 混于 liolib.c) - 无 luaL_setfuncs/lua_getextraspace/lua_isinteger - 无 LUA_REGISTRYINDEX 现代用法差异需逐一核对 后续移植需按 ADAPTATION.md 指引重写 I/O 重定向与 typed interop 层
This commit is contained in:
JianFeeeee
793
lib/lua/llex.c
793
lib/lua/llex.c
@ -1,202 +1,138 @@
|
||||
/*
|
||||
** $Id: llex.c $
|
||||
** $Id: llex.c,v 1.119 2003/03/24 12:39:34 roberto Exp $
|
||||
** Lexical Analyzer
|
||||
** See Copyright Notice in lua.h
|
||||
*/
|
||||
|
||||
#define llex_c
|
||||
#define LUA_CORE
|
||||
|
||||
#include "lprefix.h"
|
||||
|
||||
|
||||
#include <locale.h>
|
||||
#include <ctype.h>
|
||||
#include <string.h>
|
||||
|
||||
#define llex_c
|
||||
|
||||
#include "lua.h"
|
||||
|
||||
#include "lctype.h"
|
||||
#include "ldebug.h"
|
||||
#include "ldo.h"
|
||||
#include "lgc.h"
|
||||
#include "llex.h"
|
||||
#include "lobject.h"
|
||||
#include "lparser.h"
|
||||
#include "lstate.h"
|
||||
#include "lstring.h"
|
||||
#include "ltable.h"
|
||||
#include "lzio.h"
|
||||
|
||||
|
||||
|
||||
#define next(ls) (ls->current = zgetc(ls->z))
|
||||
#define next(LS) (LS->current = zgetc(LS->z))
|
||||
|
||||
|
||||
/* minimum size for string buffer */
|
||||
#if !defined(LUA_MINBUFFER)
|
||||
#define LUA_MINBUFFER 32
|
||||
#endif
|
||||
|
||||
|
||||
#define currIsNewline(ls) (ls->current == '\n' || ls->current == '\r')
|
||||
|
||||
|
||||
/* ORDER RESERVED */
|
||||
static const char *const luaX_tokens [] = {
|
||||
static const char *const token2string [] = {
|
||||
"and", "break", "do", "else", "elseif",
|
||||
"end", "false", "for", "function", "global", "goto", "if",
|
||||
"end", "false", "for", "function", "if",
|
||||
"in", "local", "nil", "not", "or", "repeat",
|
||||
"return", "then", "true", "until", "while",
|
||||
"//", "..", "...", "==", ">=", "<=", "~=",
|
||||
"<<", ">>", "::", "<eof>",
|
||||
"<number>", "<integer>", "<name>", "<string>"
|
||||
"return", "then", "true", "until", "while", "*name",
|
||||
"..", "...", "==", ">=", "<=", "~=",
|
||||
"*number", "*string", "<eof>"
|
||||
};
|
||||
|
||||
|
||||
#define save_and_next(ls) (save(ls, ls->current), next(ls))
|
||||
|
||||
|
||||
static l_noret lexerror (LexState *ls, const char *msg, int token);
|
||||
|
||||
|
||||
static void save (LexState *ls, int c) {
|
||||
Mbuffer *b = ls->buff;
|
||||
if (luaZ_bufflen(b) + 1 > luaZ_sizebuffer(b)) {
|
||||
size_t newsize = luaZ_sizebuffer(b); /* get old size */;
|
||||
if (newsize >= (MAX_SIZE/3 * 2)) /* larger than MAX_SIZE/1.5 ? */
|
||||
lexerror(ls, "lexical element too long", 0);
|
||||
newsize += (newsize >> 1); /* new size is 1.5 times the old one */
|
||||
luaZ_resizebuffer(ls->L, b, newsize);
|
||||
}
|
||||
b->buffer[luaZ_bufflen(b)++] = cast_char(c);
|
||||
}
|
||||
|
||||
|
||||
void luaX_init (lua_State *L) {
|
||||
int i;
|
||||
TString *e = luaS_newliteral(L, LUA_ENV); /* create env name */
|
||||
luaC_fix(L, obj2gco(e)); /* never collect this name */
|
||||
for (i=0; i<NUM_RESERVED; i++) {
|
||||
TString *ts = luaS_new(L, luaX_tokens[i]);
|
||||
luaC_fix(L, obj2gco(ts)); /* reserved words are never collected */
|
||||
ts->extra = cast_byte(i+1); /* reserved word */
|
||||
TString *ts = luaS_new(L, token2string[i]);
|
||||
luaS_fix(ts); /* reserved words are never collected */
|
||||
lua_assert(strlen(token2string[i])+1 <= TOKEN_LEN);
|
||||
ts->tsv.reserved = cast(lu_byte, i+1); /* reserved word */
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
#define MAXSRC 80
|
||||
|
||||
|
||||
void luaX_checklimit (LexState *ls, int val, int limit, const char *msg) {
|
||||
if (val > limit) {
|
||||
msg = luaO_pushfstring(ls->L, "too many %s (limit=%d)", msg, limit);
|
||||
luaX_syntaxerror(ls, msg);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
void luaX_errorline (LexState *ls, const char *s, const char *token, int line) {
|
||||
lua_State *L = ls->L;
|
||||
char buff[MAXSRC];
|
||||
luaO_chunkid(buff, getstr(ls->source), MAXSRC);
|
||||
luaO_pushfstring(L, "%s:%d: %s near `%s'", buff, line, s, token);
|
||||
luaD_throw(L, LUA_ERRSYNTAX);
|
||||
}
|
||||
|
||||
|
||||
static void luaX_error (LexState *ls, const char *s, const char *token) {
|
||||
luaX_errorline(ls, s, token, ls->linenumber);
|
||||
}
|
||||
|
||||
|
||||
void luaX_syntaxerror (LexState *ls, const char *msg) {
|
||||
const char *lasttoken;
|
||||
switch (ls->t.token) {
|
||||
case TK_NAME:
|
||||
lasttoken = getstr(ls->t.seminfo.ts);
|
||||
break;
|
||||
case TK_STRING:
|
||||
case TK_NUMBER:
|
||||
lasttoken = luaZ_buffer(ls->buff);
|
||||
break;
|
||||
default:
|
||||
lasttoken = luaX_token2str(ls, ls->t.token);
|
||||
break;
|
||||
}
|
||||
luaX_error(ls, msg, lasttoken);
|
||||
}
|
||||
|
||||
|
||||
const char *luaX_token2str (LexState *ls, int token) {
|
||||
if (token < FIRST_RESERVED) { /* single-byte symbols? */
|
||||
if (lisprint(token))
|
||||
return luaO_pushfstring(ls->L, "'%c'", token);
|
||||
else /* control character */
|
||||
return luaO_pushfstring(ls->L, "'<\\%d>'", token);
|
||||
if (token < FIRST_RESERVED) {
|
||||
lua_assert(token == (unsigned char)token);
|
||||
return luaO_pushfstring(ls->L, "%c", token);
|
||||
}
|
||||
else {
|
||||
const char *s = luaX_tokens[token - FIRST_RESERVED];
|
||||
if (token < TK_EOS) /* fixed format (symbols and reserved words)? */
|
||||
return luaO_pushfstring(ls->L, "'%s'", s);
|
||||
else /* names, strings, and numerals */
|
||||
return s;
|
||||
else
|
||||
return token2string[token-FIRST_RESERVED];
|
||||
}
|
||||
|
||||
|
||||
static void luaX_lexerror (LexState *ls, const char *s, int token) {
|
||||
if (token == TK_EOS)
|
||||
luaX_error(ls, s, luaX_token2str(ls, token));
|
||||
else
|
||||
luaX_error(ls, s, luaZ_buffer(ls->buff));
|
||||
}
|
||||
|
||||
|
||||
static void inclinenumber (LexState *LS) {
|
||||
next(LS); /* skip `\n' */
|
||||
++LS->linenumber;
|
||||
luaX_checklimit(LS, LS->linenumber, MAX_INT, "lines in a chunk");
|
||||
}
|
||||
|
||||
|
||||
void luaX_setinput (lua_State *L, LexState *LS, ZIO *z, TString *source) {
|
||||
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;
|
||||
next(LS); /* read first char */
|
||||
if (LS->current == '#') {
|
||||
do { /* skip first line */
|
||||
next(LS);
|
||||
} while (LS->current != '\n' && LS->current != EOZ);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
static const char *txtToken (LexState *ls, int token) {
|
||||
switch (token) {
|
||||
case TK_NAME: case TK_STRING:
|
||||
case TK_FLT: case TK_INT:
|
||||
save(ls, '\0');
|
||||
return luaO_pushfstring(ls->L, "'%s'", luaZ_buffer(ls->buff));
|
||||
default:
|
||||
return luaX_token2str(ls, token);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
static l_noret lexerror (LexState *ls, const char *msg, int token) {
|
||||
msg = luaG_addinfo(ls->L, msg, ls->source, ls->linenumber);
|
||||
if (token)
|
||||
luaO_pushfstring(ls->L, "%s near %s", msg, txtToken(ls, token));
|
||||
luaD_throw(ls->L, LUA_ERRSYNTAX);
|
||||
}
|
||||
|
||||
|
||||
l_noret luaX_syntaxerror (LexState *ls, const char *msg) {
|
||||
lexerror(ls, msg, ls->t.token);
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** Anchors a string in scanner's table so that it will not be collected
|
||||
** until the end of the compilation; by that time it should be anchored
|
||||
** somewhere. It also internalizes long strings, ensuring there is only
|
||||
** one copy of each unique string.
|
||||
*/
|
||||
static TString *anchorstr (LexState *ls, TString *ts) {
|
||||
lua_State *L = ls->L;
|
||||
TValue oldts;
|
||||
int tag = luaH_getstr(ls->h, ts, &oldts);
|
||||
if (!tagisempty(tag)) /* string already present? */
|
||||
return tsvalue(&oldts); /* use stored value */
|
||||
else { /* create a new entry */
|
||||
TValue *stv = s2v(L->top.p++); /* reserve stack space for string */
|
||||
setsvalue(L, stv, ts); /* push (anchor) the string on the stack */
|
||||
luaH_set(L, ls->h, stv, stv); /* t[string] = string */
|
||||
/* table is not a metatable, so it does not need to invalidate cache */
|
||||
luaC_checkGC(L);
|
||||
L->top.p--; /* remove string from stack */
|
||||
return ts;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** Creates a new string and anchors it in scanner's table.
|
||||
*/
|
||||
TString *luaX_newstring (LexState *ls, const char *str, size_t l) {
|
||||
return anchorstr(ls, luaS_newlstr(ls->L, str, l));
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** increment line number and skips newline sequence (any of
|
||||
** \n, \r, \n\r, or \r\n)
|
||||
*/
|
||||
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' */
|
||||
if (++ls->linenumber >= INT_MAX)
|
||||
lexerror(ls, "chunk has too many lines", 0);
|
||||
}
|
||||
|
||||
|
||||
void luaX_setinput (lua_State *L, LexState *ls, ZIO *z, TString *source,
|
||||
int firstchar) {
|
||||
ls->t.token = 0;
|
||||
ls->L = L;
|
||||
ls->current = firstchar;
|
||||
ls->lookahead.token = TK_EOS; /* no look-ahead token */
|
||||
ls->z = z;
|
||||
ls->fs = NULL;
|
||||
ls->linenumber = 1;
|
||||
ls->lastline = 1;
|
||||
ls->source = source;
|
||||
/* all three strings here ("_ENV", "break", "global") were fixed,
|
||||
so they cannot be collected */
|
||||
ls->envn = luaS_newliteral(L, LUA_ENV); /* get env string */
|
||||
ls->brkn = luaS_newliteral(L, "break"); /* get "break" string */
|
||||
#if LUA_COMPAT_GLOBAL
|
||||
/* compatibility mode: "global" is not a reserved word */
|
||||
ls->glbn = luaS_newliteral(L, "global"); /* get "global" string */
|
||||
ls->glbn->extra = 0; /* mark it as not reserved */
|
||||
#endif
|
||||
luaZ_resizebuffer(ls->L, ls->buff, LUA_MINBUFFER); /* initialize buffer */
|
||||
}
|
||||
|
||||
|
||||
|
||||
/*
|
||||
** =======================================================
|
||||
@ -205,400 +141,277 @@ void luaX_setinput (lua_State *L, LexState *ls, ZIO *z, TString *source,
|
||||
*/
|
||||
|
||||
|
||||
static int check_next1 (LexState *ls, int c) {
|
||||
if (ls->current == c) {
|
||||
next(ls);
|
||||
return 1;
|
||||
}
|
||||
else return 0;
|
||||
}
|
||||
/* use buffer to store names, literal strings and numbers */
|
||||
|
||||
/* extra space to allocate when growing buffer */
|
||||
#define EXTRABUFF 32
|
||||
|
||||
/* maximum number of chars that can be read without checking buffer size */
|
||||
#define MAXNOCHECK 5
|
||||
|
||||
#define checkbuffer(LS, len) \
|
||||
if (((len)+MAXNOCHECK)*sizeof(char) > luaZ_sizebuffer((LS)->buff)) \
|
||||
luaZ_openspace((LS)->L, (LS)->buff, (len)+EXTRABUFF)
|
||||
|
||||
#define save(LS, c, l) \
|
||||
(luaZ_buffer((LS)->buff)[l++] = cast(char, c))
|
||||
#define save_and_next(LS, l) (save(LS, LS->current, l), next(LS))
|
||||
|
||||
|
||||
/*
|
||||
** Check whether current char is in set 'set' (with two chars) and
|
||||
** saves it
|
||||
*/
|
||||
static int check_next2 (LexState *ls, const char *set) {
|
||||
lua_assert(set[2] == '\0');
|
||||
if (ls->current == set[0] || ls->current == set[1]) {
|
||||
save_and_next(ls);
|
||||
return 1;
|
||||
}
|
||||
else return 0;
|
||||
static size_t readname (LexState *LS) {
|
||||
size_t l = 0;
|
||||
checkbuffer(LS, l);
|
||||
do {
|
||||
checkbuffer(LS, l);
|
||||
save_and_next(LS, l);
|
||||
} while (isalnum(LS->current) || LS->current == '_');
|
||||
save(LS, '\0', l);
|
||||
return l-1;
|
||||
}
|
||||
|
||||
|
||||
/* LUA_NUMBER */
|
||||
/*
|
||||
** This function is quite liberal in what it accepts, as 'luaO_str2num'
|
||||
** will reject ill-formed numerals. Roughly, it accepts the following
|
||||
** pattern:
|
||||
**
|
||||
** %d(%x|%.|([Ee][+-]?))* | 0[Xx](%x|%.|([Pp][+-]?))*
|
||||
**
|
||||
** The only tricky part is to accept [+-] only after a valid exponent
|
||||
** mark, to avoid reading '3-4' or '0xe+1' as a single number.
|
||||
**
|
||||
** The caller might have already read an initial dot.
|
||||
*/
|
||||
static int read_numeral (LexState *ls, SemInfo *seminfo) {
|
||||
TValue obj;
|
||||
const char *expo = "Ee";
|
||||
int first = ls->current;
|
||||
lua_assert(lisdigit(ls->current));
|
||||
save_and_next(ls);
|
||||
if (first == '0' && check_next2(ls, "xX")) /* hexadecimal? */
|
||||
expo = "Pp";
|
||||
for (;;) {
|
||||
if (check_next2(ls, expo)) /* exponent mark? */
|
||||
check_next2(ls, "-+"); /* optional exponent sign */
|
||||
else if (lisxdigit(ls->current) || ls->current == '.') /* '%x|%.' */
|
||||
save_and_next(ls);
|
||||
else break;
|
||||
static void read_numeral (LexState *LS, int comma, SemInfo *seminfo) {
|
||||
size_t l = 0;
|
||||
checkbuffer(LS, l);
|
||||
if (comma) save(LS, '.', l);
|
||||
while (isdigit(LS->current)) {
|
||||
checkbuffer(LS, l);
|
||||
save_and_next(LS, l);
|
||||
}
|
||||
if (lislalpha(ls->current)) /* is numeral touching a letter? */
|
||||
save_and_next(ls); /* force an error */
|
||||
save(ls, '\0');
|
||||
if (luaO_str2num(luaZ_buffer(ls->buff), &obj) == 0) /* format error? */
|
||||
lexerror(ls, "malformed number", TK_FLT);
|
||||
if (ttisinteger(&obj)) {
|
||||
seminfo->i = ivalue(&obj);
|
||||
return TK_INT;
|
||||
if (LS->current == '.') {
|
||||
save_and_next(LS, l);
|
||||
if (LS->current == '.') {
|
||||
save_and_next(LS, l);
|
||||
save(LS, '\0', l);
|
||||
luaX_lexerror(LS,
|
||||
"ambiguous syntax (decimal point x string concatenation)",
|
||||
TK_NUMBER);
|
||||
}
|
||||
}
|
||||
else {
|
||||
lua_assert(ttisfloat(&obj));
|
||||
seminfo->r = fltvalue(&obj);
|
||||
return TK_FLT;
|
||||
while (isdigit(LS->current)) {
|
||||
checkbuffer(LS, l);
|
||||
save_and_next(LS, l);
|
||||
}
|
||||
if (LS->current == 'e' || LS->current == 'E') {
|
||||
save_and_next(LS, l); /* read `E' */
|
||||
if (LS->current == '+' || LS->current == '-')
|
||||
save_and_next(LS, l); /* optional exponent sign */
|
||||
while (isdigit(LS->current)) {
|
||||
checkbuffer(LS, l);
|
||||
save_and_next(LS, l);
|
||||
}
|
||||
}
|
||||
save(LS, '\0', l);
|
||||
if (!luaO_str2d(luaZ_buffer(LS->buff), &seminfo->r))
|
||||
luaX_lexerror(LS, "malformed number", TK_NUMBER);
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** read a sequence '[=*[' or ']=*]', leaving the last bracket. If
|
||||
** sequence is well formed, return its number of '='s + 2; otherwise,
|
||||
** return 1 if it is a single bracket (no '='s and no 2nd bracket);
|
||||
** otherwise (an unfinished '[==...') return 0.
|
||||
*/
|
||||
static size_t skip_sep (LexState *ls) {
|
||||
size_t count = 0;
|
||||
int s = ls->current;
|
||||
lua_assert(s == '[' || s == ']');
|
||||
save_and_next(ls);
|
||||
while (ls->current == '=') {
|
||||
save_and_next(ls);
|
||||
count++;
|
||||
}
|
||||
return (ls->current == s) ? count + 2
|
||||
: (count == 0) ? 1
|
||||
: 0;
|
||||
}
|
||||
|
||||
|
||||
static void read_long_string (LexState *ls, SemInfo *seminfo, size_t sep) {
|
||||
int line = ls->linenumber; /* initial line (for error message) */
|
||||
save_and_next(ls); /* skip 2nd '[' */
|
||||
if (currIsNewline(ls)) /* string starts with a newline? */
|
||||
inclinenumber(ls); /* skip it */
|
||||
static void read_long_string (LexState *LS, SemInfo *seminfo) {
|
||||
int cont = 0;
|
||||
size_t l = 0;
|
||||
checkbuffer(LS, l);
|
||||
save(LS, '[', l); /* save first `[' */
|
||||
save_and_next(LS, l); /* pass the second `[' */
|
||||
if (LS->current == '\n') /* string starts with a newline? */
|
||||
inclinenumber(LS); /* skip it */
|
||||
for (;;) {
|
||||
switch (ls->current) {
|
||||
case EOZ: { /* error */
|
||||
const char *what = (seminfo ? "string" : "comment");
|
||||
const char *msg = luaO_pushfstring(ls->L,
|
||||
"unfinished long %s (starting at line %d)", what, line);
|
||||
lexerror(ls, msg, TK_EOS);
|
||||
checkbuffer(LS, l);
|
||||
switch (LS->current) {
|
||||
case EOZ:
|
||||
save(LS, '\0', l);
|
||||
luaX_lexerror(LS, (seminfo) ? "unfinished long string" :
|
||||
"unfinished long comment", TK_EOS);
|
||||
break; /* to avoid warnings */
|
||||
}
|
||||
case ']': {
|
||||
if (skip_sep(ls) == sep) {
|
||||
save_and_next(ls); /* skip 2nd ']' */
|
||||
goto endloop;
|
||||
case '[':
|
||||
save_and_next(LS, l);
|
||||
if (LS->current == '[') {
|
||||
cont++;
|
||||
save_and_next(LS, l);
|
||||
}
|
||||
break;
|
||||
}
|
||||
case '\n': case '\r': {
|
||||
save(ls, '\n');
|
||||
inclinenumber(ls);
|
||||
if (!seminfo) luaZ_resetbuffer(ls->buff); /* avoid wasting space */
|
||||
break;
|
||||
}
|
||||
default: {
|
||||
if (seminfo) save_and_next(ls);
|
||||
else next(ls);
|
||||
}
|
||||
continue;
|
||||
case ']':
|
||||
save_and_next(LS, l);
|
||||
if (LS->current == ']') {
|
||||
if (cont == 0) goto endloop;
|
||||
cont--;
|
||||
save_and_next(LS, l);
|
||||
}
|
||||
continue;
|
||||
case '\n':
|
||||
save(LS, '\n', l);
|
||||
inclinenumber(LS);
|
||||
if (!seminfo) l = 0; /* reset buffer to avoid wasting space */
|
||||
continue;
|
||||
default:
|
||||
save_and_next(LS, l);
|
||||
}
|
||||
} endloop:
|
||||
save_and_next(LS, l); /* skip the second `]' */
|
||||
save(LS, '\0', l);
|
||||
if (seminfo)
|
||||
seminfo->ts = luaX_newstring(ls, luaZ_buffer(ls->buff) + sep,
|
||||
luaZ_bufflen(ls->buff) - 2 * sep);
|
||||
seminfo->ts = luaS_newlstr(LS->L, luaZ_buffer(LS->buff) + 2, l - 5);
|
||||
}
|
||||
|
||||
|
||||
static void esccheck (LexState *ls, int c, const char *msg) {
|
||||
if (!c) {
|
||||
if (ls->current != EOZ)
|
||||
save_and_next(ls); /* add current to buffer for error message */
|
||||
lexerror(ls, msg, TK_STRING);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
static int gethexa (LexState *ls) {
|
||||
save_and_next(ls);
|
||||
esccheck (ls, lisxdigit(ls->current), "hexadecimal digit expected");
|
||||
return luaO_hexavalue(ls->current);
|
||||
}
|
||||
|
||||
|
||||
static int readhexaesc (LexState *ls) {
|
||||
int r = gethexa(ls);
|
||||
r = (r << 4) + gethexa(ls);
|
||||
luaZ_buffremove(ls->buff, 2); /* remove saved chars from buffer */
|
||||
return r;
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** When reading a UTF-8 escape sequence, save everything to the buffer
|
||||
** for error reporting in case of errors; 'i' counts the number of
|
||||
** saved characters, so that they can be removed if case of success.
|
||||
*/
|
||||
static l_uint32 readutf8esc (LexState *ls) {
|
||||
l_uint32 r;
|
||||
int i = 4; /* number of chars to be removed: start with #"\u{X" */
|
||||
save_and_next(ls); /* skip 'u' */
|
||||
esccheck(ls, ls->current == '{', "missing '{'");
|
||||
r = cast_uint(gethexa(ls)); /* must have at least one digit */
|
||||
while (cast_void(save_and_next(ls)), lisxdigit(ls->current)) {
|
||||
i++;
|
||||
esccheck(ls, r <= (0x7FFFFFFFu >> 4), "UTF-8 value too large");
|
||||
r = (r << 4) + luaO_hexavalue(ls->current);
|
||||
}
|
||||
esccheck(ls, ls->current == '}', "missing '}'");
|
||||
next(ls); /* skip '}' */
|
||||
luaZ_buffremove(ls->buff, i); /* remove saved chars from buffer */
|
||||
return r;
|
||||
}
|
||||
|
||||
|
||||
static void utf8esc (LexState *ls) {
|
||||
char buff[UTF8BUFFSZ];
|
||||
int n = luaO_utf8esc(buff, readutf8esc(ls));
|
||||
for (; n > 0; n--) /* add 'buff' to string */
|
||||
save(ls, buff[UTF8BUFFSZ - n]);
|
||||
}
|
||||
|
||||
|
||||
static int readdecesc (LexState *ls) {
|
||||
int i;
|
||||
int r = 0; /* result accumulator */
|
||||
for (i = 0; i < 3 && lisdigit(ls->current); i++) { /* read up to 3 digits */
|
||||
r = 10*r + ls->current - '0';
|
||||
save_and_next(ls);
|
||||
}
|
||||
esccheck(ls, r <= UCHAR_MAX, "decimal escape too large");
|
||||
luaZ_buffremove(ls->buff, i); /* remove read digits from buffer */
|
||||
return r;
|
||||
}
|
||||
|
||||
|
||||
static void read_string (LexState *ls, int del, SemInfo *seminfo) {
|
||||
save_and_next(ls); /* keep delimiter (for error messages) */
|
||||
while (ls->current != del) {
|
||||
switch (ls->current) {
|
||||
static void read_string (LexState *LS, int del, SemInfo *seminfo) {
|
||||
size_t l = 0;
|
||||
checkbuffer(LS, l);
|
||||
save_and_next(LS, l);
|
||||
while (LS->current != del) {
|
||||
checkbuffer(LS, l);
|
||||
switch (LS->current) {
|
||||
case EOZ:
|
||||
lexerror(ls, "unfinished string", TK_EOS);
|
||||
save(LS, '\0', l);
|
||||
luaX_lexerror(LS, "unfinished string", TK_EOS);
|
||||
break; /* to avoid warnings */
|
||||
case '\n':
|
||||
case '\r':
|
||||
lexerror(ls, "unfinished string", TK_STRING);
|
||||
save(LS, '\0', l);
|
||||
luaX_lexerror(LS, "unfinished string", TK_STRING);
|
||||
break; /* to avoid warnings */
|
||||
case '\\': { /* escape sequences */
|
||||
int c; /* final character to be saved */
|
||||
save_and_next(ls); /* keep '\\' for error messages */
|
||||
switch (ls->current) {
|
||||
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 'u': utf8esc(ls); goto no_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 */
|
||||
case 'z': { /* zap following span of spaces */
|
||||
luaZ_buffremove(ls->buff, 1); /* remove '\\' */
|
||||
next(ls); /* skip the 'z' */
|
||||
while (lisspace(ls->current)) {
|
||||
if (currIsNewline(ls)) inclinenumber(ls);
|
||||
else next(ls);
|
||||
}
|
||||
goto no_save;
|
||||
}
|
||||
case '\\':
|
||||
next(LS); /* do not save the `\' */
|
||||
switch (LS->current) {
|
||||
case 'a': save(LS, '\a', l); next(LS); break;
|
||||
case 'b': save(LS, '\b', l); next(LS); break;
|
||||
case 'f': save(LS, '\f', l); next(LS); break;
|
||||
case 'n': save(LS, '\n', l); next(LS); break;
|
||||
case 'r': save(LS, '\r', l); next(LS); break;
|
||||
case 't': save(LS, '\t', l); next(LS); break;
|
||||
case 'v': save(LS, '\v', l); next(LS); break;
|
||||
case '\n': save(LS, '\n', l); inclinenumber(LS); break;
|
||||
case EOZ: break; /* will raise an error next loop */
|
||||
default: {
|
||||
esccheck(ls, lisdigit(ls->current), "invalid escape sequence");
|
||||
c = readdecesc(ls); /* digital escape '\ddd' */
|
||||
goto only_save;
|
||||
if (!isdigit(LS->current))
|
||||
save_and_next(LS, l); /* handles \\, \", \', and \? */
|
||||
else { /* \xxx */
|
||||
int c = 0;
|
||||
int i = 0;
|
||||
do {
|
||||
c = 10*c + (LS->current-'0');
|
||||
next(LS);
|
||||
} while (++i<3 && isdigit(LS->current));
|
||||
if (c > UCHAR_MAX) {
|
||||
save(LS, '\0', l);
|
||||
luaX_lexerror(LS, "escape sequence too large", TK_STRING);
|
||||
}
|
||||
save(LS, c, l);
|
||||
}
|
||||
}
|
||||
}
|
||||
read_save:
|
||||
next(ls);
|
||||
/* go through */
|
||||
only_save:
|
||||
luaZ_buffremove(ls->buff, 1); /* remove '\\' */
|
||||
save(ls, c);
|
||||
/* go through */
|
||||
no_save: break;
|
||||
}
|
||||
break;
|
||||
default:
|
||||
save_and_next(ls);
|
||||
save_and_next(LS, l);
|
||||
}
|
||||
}
|
||||
save_and_next(ls); /* skip delimiter */
|
||||
seminfo->ts = luaX_newstring(ls, luaZ_buffer(ls->buff) + 1,
|
||||
luaZ_bufflen(ls->buff) - 2);
|
||||
save_and_next(LS, l); /* skip delimiter */
|
||||
save(LS, '\0', l);
|
||||
seminfo->ts = luaS_newlstr(LS->L, luaZ_buffer(LS->buff) + 1, l - 3);
|
||||
}
|
||||
|
||||
|
||||
static int llex (LexState *ls, SemInfo *seminfo) {
|
||||
luaZ_resetbuffer(ls->buff);
|
||||
int luaX_lex (LexState *LS, SemInfo *seminfo) {
|
||||
for (;;) {
|
||||
switch (ls->current) {
|
||||
case '\n': case '\r': { /* line breaks */
|
||||
inclinenumber(ls);
|
||||
break;
|
||||
switch (LS->current) {
|
||||
|
||||
case '\n': {
|
||||
inclinenumber(LS);
|
||||
continue;
|
||||
}
|
||||
case ' ': case '\f': case '\t': case '\v': { /* spaces */
|
||||
next(ls);
|
||||
break;
|
||||
}
|
||||
case '-': { /* '-' or '--' (comment) */
|
||||
next(ls);
|
||||
if (ls->current != '-') return '-';
|
||||
case '-': {
|
||||
next(LS);
|
||||
if (LS->current != '-') return '-';
|
||||
/* else is a comment */
|
||||
next(ls);
|
||||
if (ls->current == '[') { /* long comment? */
|
||||
size_t sep = skip_sep(ls);
|
||||
luaZ_resetbuffer(ls->buff); /* 'skip_sep' may dirty the buffer */
|
||||
if (sep >= 2) {
|
||||
read_long_string(ls, NULL, sep); /* skip long comment */
|
||||
luaZ_resetbuffer(ls->buff); /* previous call may dirty the buff. */
|
||||
break;
|
||||
}
|
||||
}
|
||||
/* else short comment */
|
||||
while (!currIsNewline(ls) && ls->current != EOZ)
|
||||
next(ls); /* skip until end of line (or end of file) */
|
||||
break;
|
||||
next(LS);
|
||||
if (LS->current == '[' && (next(LS), LS->current == '['))
|
||||
read_long_string(LS, NULL); /* long comment */
|
||||
else /* short comment */
|
||||
while (LS->current != '\n' && LS->current != EOZ)
|
||||
next(LS);
|
||||
continue;
|
||||
}
|
||||
case '[': { /* long string or simply '[' */
|
||||
size_t sep = skip_sep(ls);
|
||||
if (sep >= 2) {
|
||||
read_long_string(ls, seminfo, sep);
|
||||
case '[': {
|
||||
next(LS);
|
||||
if (LS->current != '[') return '[';
|
||||
else {
|
||||
read_long_string(LS, seminfo);
|
||||
return TK_STRING;
|
||||
}
|
||||
else if (sep == 0) /* '[=...' missing second bracket? */
|
||||
lexerror(ls, "invalid long string delimiter", TK_STRING);
|
||||
return '[';
|
||||
}
|
||||
case '=': {
|
||||
next(ls);
|
||||
if (check_next1(ls, '=')) return TK_EQ; /* '==' */
|
||||
else return '=';
|
||||
next(LS);
|
||||
if (LS->current != '=') return '=';
|
||||
else { next(LS); return TK_EQ; }
|
||||
}
|
||||
case '<': {
|
||||
next(ls);
|
||||
if (check_next1(ls, '=')) return TK_LE; /* '<=' */
|
||||
else if (check_next1(ls, '<')) return TK_SHL; /* '<<' */
|
||||
else return '<';
|
||||
next(LS);
|
||||
if (LS->current != '=') return '<';
|
||||
else { next(LS); return TK_LE; }
|
||||
}
|
||||
case '>': {
|
||||
next(ls);
|
||||
if (check_next1(ls, '=')) return TK_GE; /* '>=' */
|
||||
else if (check_next1(ls, '>')) return TK_SHR; /* '>>' */
|
||||
else return '>';
|
||||
}
|
||||
case '/': {
|
||||
next(ls);
|
||||
if (check_next1(ls, '/')) return TK_IDIV; /* '//' */
|
||||
else return '/';
|
||||
next(LS);
|
||||
if (LS->current != '=') return '>';
|
||||
else { next(LS); return TK_GE; }
|
||||
}
|
||||
case '~': {
|
||||
next(ls);
|
||||
if (check_next1(ls, '=')) return TK_NE; /* '~=' */
|
||||
else return '~';
|
||||
next(LS);
|
||||
if (LS->current != '=') return '~';
|
||||
else { next(LS); return TK_NE; }
|
||||
}
|
||||
case ':': {
|
||||
next(ls);
|
||||
if (check_next1(ls, ':')) return TK_DBCOLON; /* '::' */
|
||||
else return ':';
|
||||
}
|
||||
case '"': case '\'': { /* short literal strings */
|
||||
read_string(ls, ls->current, seminfo);
|
||||
case '"':
|
||||
case '\'': {
|
||||
read_string(LS, LS->current, seminfo);
|
||||
return TK_STRING;
|
||||
}
|
||||
case '.': { /* '.', '..', '...', or number */
|
||||
save_and_next(ls);
|
||||
if (check_next1(ls, '.')) {
|
||||
if (check_next1(ls, '.'))
|
||||
return TK_DOTS; /* '...' */
|
||||
else return TK_CONCAT; /* '..' */
|
||||
case '.': {
|
||||
next(LS);
|
||||
if (LS->current == '.') {
|
||||
next(LS);
|
||||
if (LS->current == '.') {
|
||||
next(LS);
|
||||
return TK_DOTS; /* ... */
|
||||
}
|
||||
else return TK_CONCAT; /* .. */
|
||||
}
|
||||
else if (!isdigit(LS->current)) return '.';
|
||||
else {
|
||||
read_numeral(LS, 1, seminfo);
|
||||
return TK_NUMBER;
|
||||
}
|
||||
else if (!lisdigit(ls->current)) return '.';
|
||||
else return read_numeral(ls, seminfo);
|
||||
}
|
||||
case '0': case '1': case '2': case '3': case '4':
|
||||
case '5': case '6': case '7': case '8': case '9': {
|
||||
return read_numeral(ls, seminfo);
|
||||
}
|
||||
case EOZ: {
|
||||
return TK_EOS;
|
||||
}
|
||||
default: {
|
||||
if (lislalpha(ls->current)) { /* identifier or reserved word? */
|
||||
TString *ts;
|
||||
do {
|
||||
save_and_next(ls);
|
||||
} while (lislalnum(ls->current));
|
||||
/* find or create string */
|
||||
ts = luaS_newlstr(ls->L, luaZ_buffer(ls->buff),
|
||||
luaZ_bufflen(ls->buff));
|
||||
if (isreserved(ts)) /* reserved word? */
|
||||
return ts->extra - 1 + FIRST_RESERVED;
|
||||
else {
|
||||
seminfo->ts = anchorstr(ls, ts);
|
||||
return TK_NAME;
|
||||
}
|
||||
if (isspace(LS->current)) {
|
||||
next(LS);
|
||||
continue;
|
||||
}
|
||||
else { /* single-char tokens ('+', '*', '%', '{', '}', ...) */
|
||||
int c = ls->current;
|
||||
next(ls);
|
||||
return c;
|
||||
else if (isdigit(LS->current)) {
|
||||
read_numeral(LS, 0, seminfo);
|
||||
return TK_NUMBER;
|
||||
}
|
||||
else if (isalpha(LS->current) || LS->current == '_') {
|
||||
/* identifier or reserved word */
|
||||
size_t l = readname(LS);
|
||||
TString *ts = luaS_newlstr(LS->L, luaZ_buffer(LS->buff), l);
|
||||
if (ts->tsv.reserved > 0) /* reserved word? */
|
||||
return ts->tsv.reserved - 1 + FIRST_RESERVED;
|
||||
seminfo->ts = ts;
|
||||
return TK_NAME;
|
||||
}
|
||||
else {
|
||||
int c = LS->current;
|
||||
if (iscntrl(c))
|
||||
luaX_error(LS, "invalid control char",
|
||||
luaO_pushfstring(LS->L, "char(%d)", c));
|
||||
next(LS);
|
||||
return c; /* single-char tokens (+ - / ...) */
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
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 */
|
||||
}
|
||||
|
||||
|
||||
int luaX_lookahead (LexState *ls) {
|
||||
lua_assert(ls->lookahead.token == TK_EOS);
|
||||
ls->lookahead.token = llex(ls, &ls->lookahead.seminfo);
|
||||
return ls->lookahead.token;
|
||||
}
|
||||
|
||||
#undef next
|
||||
|
||||
Reference in New Issue
Block a user