/*
** 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 >= MAX_INT)
lexerror(ls, "chunk has too many lines", 0);
}
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 >= LJ_MAX_LINE)
lj_lex_error(ls, ls->token, LJ_ERR_XLINES);
}
static void read_long_string(LexState *ls, TValue *tv, int sep)
{
save_and_next(ls); /* skip 2nd `[' */
if (currIsNewline(ls)) /* string starts with a newline? */
inclinenumber(ls); /* skip it */
for (;;) {
switch (ls->current) {
case END_OF_STREAM:
lj_lex_error(ls, TK_eof, tv ? LJ_ERR_XLSTR : LJ_ERR_XLCOM);
break;
case ']':
if (skip_sep(ls) == sep) {
save_and_next(ls); /* skip 2nd `]' */
goto endloop;
}
break;
case '\n':
case '\r':
save(ls, '\n');
inclinenumber(ls);
if (!tv) lj_str_resetbuf(&ls->sb); /* avoid wasting space */
break;
default:
if (tv) save_and_next(ls);
else next(ls);
break;
}
} endloop:
if (tv) {
GCstr *str = lj_parse_keepstr(ls, ls->sb.buf + (2 + (MSize)sep),
ls->sb.n - 2*(2 + (MSize)sep));
setstrV(ls->L, tv, str);
}
}
/* Setup lexer state. */
void lj_lex_setup(lua_State *L, LexState *ls)
{
ls->L = L;
ls->fs = NULL;
ls->n = 0;
ls->p = NULL;
ls->vstack = NULL;
ls->sizevstack = 0;
ls->vtop = 0;
ls->bcstack = NULL;
ls->sizebcstack = 0;
ls->lookahead = TK_eof; /* No look-ahead token. */
ls->linenumber = 1;
ls->lastline = 1;
lj_str_resizebuf(ls->L, &ls->sb, LJ_MIN_SBUF);
next(ls); /* Read-ahead first char. */
if (ls->current == 0xef && ls->n >= 2 && char2int(ls->p[0]) == 0xbb &&
char2int(ls->p[1]) == 0xbf) { /* Skip UTF-8 BOM (if buffered). */
ls->n -= 2;
ls->p += 2;
next(ls);
}
if (ls->current == '#') { /* Skip POSIX #! header line. */
do {
next(ls);
if (ls->current == END_OF_STREAM) return;
} while (!currIsNewline(ls));
inclinenumber(ls);
}
if (ls->current == LUA_SIGNATURE[0]) {
setstrV(L, L->top++, lj_err_str(L, LJ_ERR_XBCLOAD));
lj_err_throw(L, LUA_ERRSYNTAX);
}
}
/* Setup lexer state. */
int lj_lex_setup(lua_State *L, LexState *ls)
{
int header = 0;
ls->L = L;
ls->fs = NULL;
ls->n = 0;
ls->p = NULL;
ls->vstack = NULL;
ls->sizevstack = 0;
ls->vtop = 0;
ls->bcstack = NULL;
ls->sizebcstack = 0;
ls->lookahead = TK_eof; /* No look-ahead token. */
ls->linenumber = 1;
ls->lastline = 1;
lj_str_resizebuf(ls->L, &ls->sb, LJ_MIN_SBUF);
next(ls); /* Read-ahead first char. */
if (ls->current == 0xef && ls->n >= 2 && char2int(ls->p[0]) == 0xbb &&
char2int(ls->p[1]) == 0xbf) { /* Skip UTF-8 BOM (if buffered). */
ls->n -= 2;
ls->p += 2;
next(ls);
header = 1;
}
if (ls->current == '#') { /* Skip POSIX #! header line. */
do {
next(ls);
if (ls->current == END_OF_STREAM) return 0;
} while (!currIsNewline(ls));
inclinenumber(ls);
header = 1;
}
if (ls->current == LUA_SIGNATURE[0]) { /* Bytecode dump. */
if (header) {
/*
** Loading bytecode with an extra header is disabled for security
** reasons. This may circumvent the usual check for bytecode vs.
** Lua code by looking at the first char. Since this is a potential
** security violation no attempt is made to echo the chunkname either.
*/
setstrV(L, L->top++, lj_err_str(L, LJ_ERR_BCBAD));
lj_err_throw(L, LUA_ERRSYNTAX);
}
return 1;
}
return 0;
}
static void read_long_string (LexState *ls, SemInfo *seminfo, int 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 */
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);
break; /* to avoid warnings */
}
case ']': {
if (skip_sep(ls) == sep) {
save_and_next(ls); /* skip 2nd ']' */
goto endloop;
}
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);
}
}
}
endloop:
if (seminfo)
seminfo->ts = luaX_newstring(ls, luaZ_buffer(ls->buff) + (2 + sep),
luaZ_bufflen(ls->buff) - 2*(2 + sep));
}
static int llex (LexState *ls, SemInfo *seminfo) {
luaZ_resetbuffer(ls->buff);
for (;;) {
switch (ls->current) {
case '\n': case '\r': { /* line breaks */
inclinenumber(ls);
break;
}
case ' ': case '\f': case '\t': case '\v': { /* spaces */
next(ls);
break;
}
case '-': { /* '-' or '--' (comment) */
next(ls);
if (ls->current != '-') return '-';
/* else is a comment */
next(ls);
if (ls->current == '[') { /* long comment? */
int sep = skip_sep(ls);
luaZ_resetbuffer(ls->buff); /* 'skip_sep' may dirty the buffer */
if (sep >= 0) {
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;
}
case '[': { /* long string or simply '[' */
int sep = skip_sep(ls);
if (sep >= 0) {
read_long_string(ls, seminfo, sep);
return TK_STRING;
}
else if (sep != -1) /* '[=...' missing second bracket */
lexerror(ls, "invalid long string delimiter", TK_STRING);
return '[';
}
case '=': {
next(ls);
if (check_next1(ls, '=')) return TK_EQ;
else return '=';
}
case '<': {
next(ls);
if (check_next1(ls, '=')) return TK_LE;
else if (check_next1(ls, '<')) return TK_SHL;
else return '<';
}
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 '/';
}
case '~': {
next(ls);
if (check_next1(ls, '=')) return TK_NE;
else return '~';
}
case ':': {
next(ls);
if (check_next1(ls, ':')) return TK_DBCOLON;
else return ':';
}
case '"': case '\'': { /* short literal strings */
read_string(ls, ls->current, seminfo);
return TK_STRING;
}
case '`': { /* relative paths */
read_string(ls, ls->current, seminfo);
return TK_PATH;
}
case '.': { /* '.', '..', '...', or number */
save_and_next(ls);
if (check_next1(ls, '.')) {
if (check_next1(ls, '.'))
return TK_DOTS; /* '...' */
else return TK_CONCAT; /* '..' */
}
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));
ts = luaX_newstring(ls, luaZ_buffer(ls->buff),
luaZ_bufflen(ls->buff));
seminfo->ts = ts;
if (isreserved(ts)) /* reserved word? */
return ts->extra - 1 + FIRST_RESERVED;
else {
return TK_NAME;
}
}
else { /* single-char tokens (+ - / ...) */
int c = ls->current;
next(ls);
return c;
}
}
}
}
}
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) {
case EOZ:
lexerror(ls, "unfinished string", TK_EOS);
break; /* to avoid warnings */
case '\n':
case '\r':
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;
}
default: {
esccheck(ls, lisdigit(ls->current), "invalid escape sequence");
c = readdecesc(ls); /* digital escape '\ddd' */
goto only_save;
}
}
read_save:
next(ls);
/* go through */
only_save:
luaZ_buffremove(ls->buff, 1); /* remove '\\' */
save(ls, c);
/* go through */
no_save: break;
}
default:
save_and_next(ls);
}
}
save_and_next(ls); /* skip delimiter */
seminfo->ts = luaX_newstring(ls, luaZ_buffer(ls->buff) + 1,
luaZ_bufflen(ls->buff) - 2);
}
static int llex(LexState *ls, TValue *tv)
{
lj_str_resetbuf(&ls->sb);
for (;;) {
if (lj_char_isident(ls->current)) {
GCstr *s;
if (lj_char_isdigit(ls->current)) { /* Numeric literal. */
lex_number(ls, tv);
return TK_number;
}
/* Identifier or reserved word. */
do {
save_and_next(ls);
} while (lj_char_isident(ls->current));
s = lj_parse_keepstr(ls, ls->sb.buf, ls->sb.n);
setstrV(ls->L, tv, s);
if (s->reserved > 0) /* Reserved word? */
return TK_OFS + s->reserved;
return TK_name;
}
switch (ls->current) {
case '\n':
case '\r':
inclinenumber(ls);
continue;
case ' ':
case '\t':
case '\v':
case '\f':
next(ls);
continue;
case '-':
next(ls);
if (ls->current != '-') return '-';
/* else is a comment */
next(ls);
if (ls->current == '[') {
int sep = skip_sep(ls);
lj_str_resetbuf(&ls->sb); /* `skip_sep' may dirty the buffer */
if (sep >= 0) {
read_long_string(ls, NULL, sep); /* long comment */
lj_str_resetbuf(&ls->sb);
continue;
}
}
/* else short comment */
while (!currIsNewline(ls) && ls->current != END_OF_STREAM)
next(ls);
continue;
case '[': {
int sep = skip_sep(ls);
if (sep >= 0) {
read_long_string(ls, tv, sep);
return TK_string;
} else if (sep == -1) {
return '[';
} else {
lj_lex_error(ls, TK_string, LJ_ERR_XLDELIM);
continue;
}
}
case '=':
next(ls);
if (ls->current != '=') return '='; else { next(ls); return TK_eq; }
case '<':
next(ls);
if (ls->current != '=') return '<'; else { next(ls); return TK_le; }
case '>':
next(ls);
if (ls->current != '=') return '>'; else { next(ls); return TK_ge; }
case '~':
next(ls);
if (ls->current != '=') return '~'; else { next(ls); return TK_ne; }
case ':':
next(ls);
if (ls->current != ':') return ':'; else { next(ls); return TK_label; }
case '"':
case '\'':
read_string(ls, ls->current, tv);
return TK_string;
case '.':
save_and_next(ls);
if (ls->current == '.') {
next(ls);
if (ls->current == '.') {
next(ls);
return TK_dots; /* ... */
}
return TK_concat; /* .. */
} else if (!lj_char_isdigit(ls->current)) {
return '.';
} else {
lex_number(ls, tv);
return TK_number;
}
case END_OF_STREAM:
return TK_eof;
default: {
int c = ls->current;
next(ls);
return c; /* Single-char tokens (+ - / ...). */
}
}
}
}
static void read_string(LexState *ls, int delim, TValue *tv)
{
save_and_next(ls);
while (ls->current != delim) {
switch (ls->current) {
case END_OF_STREAM:
lj_lex_error(ls, TK_eof, LJ_ERR_XSTR);
continue;
case '\n':
case '\r':
lj_lex_error(ls, TK_string, LJ_ERR_XSTR);
continue;
case '\\': {
int c = next(ls); /* Skip the '\\'. */
switch (c) {
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;
case 'x': /* Hexadecimal escape '\xXX'. */
c = (next(ls) & 15u) << 4;
if (!lj_char_isdigit(ls->current)) {
if (!lj_char_isxdigit(ls->current)) goto err_xesc;
c += 9 << 4;
}
c += (next(ls) & 15u);
if (!lj_char_isdigit(ls->current)) {
if (!lj_char_isxdigit(ls->current)) goto err_xesc;
c += 9;
}
break;
case 'z': /* Skip whitespace. */
next(ls);
while (lj_char_isspace(ls->current))
if (currIsNewline(ls)) inclinenumber(ls); else next(ls);
continue;
case '\n': case '\r': save(ls, '\n'); inclinenumber(ls); continue;
case '\\': case '\"': case '\'': break;
case END_OF_STREAM: continue;
default:
if (!lj_char_isdigit(c))
goto err_xesc;
c -= '0'; /* Decimal escape '\ddd'. */
if (lj_char_isdigit(next(ls))) {
c = c*10 + (ls->current - '0');
if (lj_char_isdigit(next(ls))) {
c = c*10 + (ls->current - '0');
if (c > 255) {
err_xesc:
lj_lex_error(ls, TK_string, LJ_ERR_XESC);
}
next(ls);
}
}
save(ls, c);
continue;
}
save(ls, c);
next(ls);
continue;
}
default:
save_and_next(ls);
break;
}
}
save_and_next(ls); /* skip delimiter */
setstrV(ls->L, tv, lj_parse_keepstr(ls, ls->sb.buf + 1, ls->sb.n - 2));
}
