/*
** this function is quite liberal in what it accepts, as 'luaO_str2num'
** will reject ill-formed numerals.
*/
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 part? */
check_next2(ls, "-+"); /* optional exponent sign */
if (lisxdigit(ls->current))
save_and_next(ls);
else if (ls->current == '.')
save_and_next(ls);
else break;
}
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;
}
else {
lua_assert(ttisfloat(&obj));
seminfo->r = fltvalue(&obj);
return TK_FLT;
}
}
/* LUA_NUMBER */
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 (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);
}
}
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);
}
/*
** change all characters 'from' in buffer to 'to'
*/
static void buffreplace (LexState *ls, char from, char to) {
if (from != to) {
size_t n = luaZ_bufflen(ls->buff);
char *p = luaZ_buffer(ls->buff);
while (n--)
if (p[n] == from) p[n] = to;
}
}
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 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:
save(LS, '\0', l);
luaX_lexerror(LS, "unfinished string", TK_EOS);
break; /* to avoid warnings */
case '\n':
save(LS, '\0', l);
luaX_lexerror(LS, "unfinished string", TK_STRING);
break; /* to avoid warnings */
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: {
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);
}
}
}
break;
default:
save_and_next(LS, l);
}
}
save_and_next(LS, l); /* skip delimiter */
save(LS, '\0', l);
seminfo->ts = luaS_newlstr(LS->L, luaZ_buffer(LS->buff) + 1, l - 3);
}
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);
}
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 (;;) {
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 '[':
save_and_next(LS, l);
if (LS->current == '[') {
cont++;
save_and_next(LS, l);
}
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 = luaS_newlstr(LS->L, luaZ_buffer(LS->buff) + 2, l - 5);
}
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);
}
int luaX_lex (LexState *LS, SemInfo *seminfo) {
for (;;) {
switch (LS->current) {
case '\n': {
inclinenumber(LS);
continue;
}
case '-': {
next(LS);
if (LS->current != '-') return '-';
/* else is a comment */
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 '[': {
next(LS);
if (LS->current != '[') return '[';
else {
read_long_string(LS, seminfo);
return TK_STRING;
}
}
case '=': {
next(LS);
if (LS->current != '=') return '=';
else { next(LS); return TK_EQ; }
}
case '<': {
next(LS);
if (LS->current == '<') { next(LS); return TK_SHL; }
else if (LS->current != '=') return '<';
else { next(LS); return TK_LE; }
}
case '>': {
next(LS);
if (LS->current == '>') { next(LS); return TK_SHR; }
else if (LS->current != '=') return '>';
else { next(LS); return TK_GE; }
}
case '~': {
next(LS);
if (LS->current != '=') return '~';
else { next(LS); return TK_NE; }
}
case '"':
case '\'': {
read_string(LS, LS->current, seminfo);
return TK_STRING;
}
case '.': {
next(LS);
if (LS->current == '.') {
next(LS);
if (LS->current == '.') {
next(LS);
return TK_DOTS; /* ... */
}
else return TK_CONCAT; /* .. */
}
else if (!lex_isdigit(LS->current)) return '.';
else {
return read_numeral(LS, 1, seminfo);
}
}
case EOZ: {
return TK_EOS;
}
default: {
if (isspace(LS->current)) {
next(LS);
continue;
}
else if (lex_isdigit(LS->current)) {
return (read_numeral(LS, 0, seminfo));
}
else if (lex_isalpha(LS->current) || LS->current == '_') {
char saveCh = 0;
size_t l;
TString *ts;
if (LS->current == 'L')
{
next(LS);
if (LS->current == '"')
{
read_wstring(LS, LS->current, seminfo);
return TK_WSTRING;
}
saveCh = 'L';
}
/* identifier or reserved word */
l = readname(LS, saveCh);
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 (+ - / ...) */
}
}
}
}
}
static void read_wstring (LexState *LS, int del, SemInfo *seminfo) {
size_t l = 0;
checkbuffer(LS, l * 2);
wsave_and_next(LS, l);
while (LS->current != del) {
checkbuffer(LS, l * 2);
switch (LS->current) {
case EOZ:
wsave(LS, '\0', l);
luaX_lexerror(LS, "unfinished string", TK_EOS);
break; /* to avoid warnings */
case '\n':
wsave(LS, '\0', l);
luaX_lexerror(LS, "unfinished string", TK_STRING);
break; /* to avoid warnings */
case '\\':
next(LS); /* do not save the `\' */
switch (LS->current) {
case 'a': wsave(LS, '\a', l); next(LS); break;
case 'b': wsave(LS, '\b', l); next(LS); break;
case 'f': wsave(LS, '\f', l); next(LS); break;
case 'n': wsave(LS, '\n', l); next(LS); break;
case 'r': wsave(LS, '\r', l); next(LS); break;
case 't': wsave(LS, '\t', l); next(LS); break;
case 'v': wsave(LS, '\v', l); next(LS); break;
case '\n': wsave(LS, '\n', l); inclinenumber(LS); break;
case EOZ: break; /* will raise an error next loop */
case 'x': {
int ch;
next(LS);
ch = tolower(LS->current);
if (!lex_isdigit(ch) && !(ch >= 'a' && ch <= 'f') )
save(LS, 'x', l); /* handles \\, \", \', and \? */
else { /* \xxx */
int c = 0;
int i = 0;
int numDigits = 4;
do {
ch = tolower(LS->current);
if (lex_isdigit(ch))
c = 16*c + (ch-'0');
else if (ch >= 'a' && ch <= 'f')
c = 16*c + (ch-'a') + 10;
next(LS);
ch = tolower(LS->current);
} while (++i<numDigits && (lex_isdigit(ch) || (ch >= 'a' && ch <= 'f')));
wsave(LS, c, l);
}
break;
}
default: {
if (!lex_isdigit(LS->current))
wsave_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 && lex_isdigit(LS->current));
if (c > UCHAR_MAX) {
wsave(LS, '\0', l);
luaX_lexerror(LS, "escape sequence too large", TK_STRING);
}
wsave(LS, c, l);
}
}
}
break;
default:
wsave_and_next(LS, l);
}
}
wsave_and_next(LS, l); /* skip delimiter */
wsave(LS, '\0', l);
seminfo->ts = luaS_newlwstr(LS->L, (const lua_WChar*)(luaZ_buffer(LS->buff) + 1 * 2), (l - 3 * 2) / 2);
}
/* LUA_NUMBER */
static int read_numeral (LexState *LS, int period, SemInfo *seminfo) {
int isReal = 0;
int startsWithZero = LS->current == '0';
size_t l = 0;
checkbuffer(LS, l);
if (period) {
save(LS, '.', l);
isReal = 1;
}
if (startsWithZero) {
next(LS);
if (LS->current == 'x') {
/* Process a hex number */
int ch = 0;
int c = 0;
int i = 0;
int numDigits = 8;
next(LS);
do {
ch = tolower(LS->current);
if (lex_isdigit(ch))
c = 16*c + (ch-'0');
else if (ch >= 'a' && ch <= 'f')
c = 16*c + (ch-'a') + 10;
next(LS);
ch = tolower(LS->current);
} while (++i<numDigits && (lex_isdigit(ch) || (ch >= 'a' && ch <= 'f')));
seminfo->r = c;
return TK_NUMBER;
} else {
checkbuffer(LS, 1);
save(LS, '0', l);
}
}
while (lex_isdigit(LS->current)) {
checkbuffer(LS, l);
save_and_next(LS, l);
}
if (LS->current == '.') {
isReal = 1;
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);
}
}
while (lex_isdigit(LS->current)) {
checkbuffer(LS, l);
save_and_next(LS, l);
}
if (LS->current == 'e' || LS->current == 'E') {
isReal = 1;
save_and_next(LS, l); /* read `E' */
if (LS->current == '+' || LS->current == '-')
save_and_next(LS, l); /* optional exponent sign */
while (lex_isdigit(LS->current)) {
checkbuffer(LS, l);
save_and_next(LS, l);
}
}
save(LS, '\0', l);
if (isReal) {
if (!luaO_str2d(luaZ_buffer(LS->buff), &seminfo->r))
luaX_lexerror(LS, "malformed number", TK_NUMBER);
return TK_NUMBER;
} else {
if (!luaO_str2d(luaZ_buffer(LS->buff), &seminfo->r))
luaX_lexerror(LS, "malformed integer", TK_NUMBER);
return TK_NUMBER;
}
}
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));
}
