|
@@ -1,5 +1,5 @@
|
|
/*
|
|
/*
|
|
-** $Id: llex.c,v 2.35 2010/02/27 21:16:24 roberto Exp roberto $
|
|
|
|
|
|
+** $Id: llex.c,v 2.36 2010/04/05 16:35:37 roberto Exp roberto $
|
|
** Lexical Analyzer
|
|
** Lexical Analyzer
|
|
** See Copyright Notice in lua.h
|
|
** See Copyright Notice in lua.h
|
|
*/
|
|
*/
|
|
@@ -117,21 +117,30 @@ void luaX_syntaxerror (LexState *ls, const char *msg) {
|
|
}
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
+/*
|
|
|
|
+** 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)
|
|
|
|
+*/
|
|
TString *luaX_newstring (LexState *ls, const char *str, size_t l) {
|
|
TString *luaX_newstring (LexState *ls, const char *str, size_t l) {
|
|
lua_State *L = ls->L;
|
|
lua_State *L = ls->L;
|
|
TValue *o; /* entry for `str' */
|
|
TValue *o; /* entry for `str' */
|
|
- TString *ts = luaS_newlstr(L, str, l);
|
|
|
|
- setsvalue2s(L, L->top++, ts); /* anchor string */
|
|
|
|
- o = luaH_setstr(L, ls->fs->h, ts);
|
|
|
|
|
|
+ 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);
|
|
if (ttisnil(o)) {
|
|
if (ttisnil(o)) {
|
|
- setbvalue(o, 1); /* make sure `str' will not be collected */
|
|
|
|
|
|
+ setbvalue(o, 1); /* t[string] = true */
|
|
luaC_checkGC(L);
|
|
luaC_checkGC(L);
|
|
}
|
|
}
|
|
- L->top--;
|
|
|
|
|
|
+ L->top--; /* remove string from stack */
|
|
return ts;
|
|
return ts;
|
|
}
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
+/*
|
|
|
|
+** increment line number and skips newline sequence (any of
|
|
|
|
+** \n, \r, \n\r, or \r\n)
|
|
|
|
+*/
|
|
static void inclinenumber (LexState *ls) {
|
|
static void inclinenumber (LexState *ls) {
|
|
int old = ls->current;
|
|
int old = ls->current;
|
|
lua_assert(currIsNewline(ls));
|
|
lua_assert(currIsNewline(ls));
|
|
@@ -152,7 +161,7 @@ void luaX_setinput (lua_State *L, LexState *ls, ZIO *z, TString *source) {
|
|
ls->linenumber = 1;
|
|
ls->linenumber = 1;
|
|
ls->lastline = 1;
|
|
ls->lastline = 1;
|
|
ls->source = source;
|
|
ls->source = source;
|
|
- ls->envn = luaS_new(L, "_ENV");
|
|
|
|
|
|
+ ls->envn = luaS_new(L, "_ENV"); /* create env name */
|
|
luaS_fix(ls->envn); /* never collect this name */
|
|
luaS_fix(ls->envn); /* never collect this name */
|
|
luaZ_resizebuffer(ls->L, ls->buff, LUA_MINBUFFER); /* initialize buffer */
|
|
luaZ_resizebuffer(ls->L, ls->buff, LUA_MINBUFFER); /* initialize buffer */
|
|
next(ls); /* read first char */
|
|
next(ls); /* read first char */
|
|
@@ -176,6 +185,9 @@ static int check_next (LexState *ls, const char *set) {
|
|
}
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
+/*
|
|
|
|
+** change all characters 'from' in buffer to 'to'
|
|
|
|
+*/
|
|
static void buffreplace (LexState *ls, char from, char to) {
|
|
static void buffreplace (LexState *ls, char from, char to) {
|
|
size_t n = luaZ_bufflen(ls->buff);
|
|
size_t n = luaZ_bufflen(ls->buff);
|
|
char *p = luaZ_buffer(ls->buff);
|
|
char *p = luaZ_buffer(ls->buff);
|
|
@@ -188,11 +200,14 @@ static void buffreplace (LexState *ls, char from, char to) {
|
|
#define getlocaledecpoint() (localeconv()->decimal_point[0])
|
|
#define getlocaledecpoint() (localeconv()->decimal_point[0])
|
|
#endif
|
|
#endif
|
|
|
|
|
|
|
|
+/*
|
|
|
|
+** in case of format error, try to change decimal point separator to
|
|
|
|
+** the one defined in the current locale and check again
|
|
|
|
+*/
|
|
static void trydecpoint (LexState *ls, SemInfo *seminfo) {
|
|
static void trydecpoint (LexState *ls, SemInfo *seminfo) {
|
|
- /* format error: try to update decimal point separator */
|
|
|
|
char old = ls->decpoint;
|
|
char old = ls->decpoint;
|
|
ls->decpoint = getlocaledecpoint();
|
|
ls->decpoint = getlocaledecpoint();
|
|
- buffreplace(ls, old, ls->decpoint); /* try updated decimal separator */
|
|
|
|
|
|
+ buffreplace(ls, old, ls->decpoint); /* try new decimal separator */
|
|
if (!luaO_str2d(luaZ_buffer(ls->buff), &seminfo->r)) {
|
|
if (!luaO_str2d(luaZ_buffer(ls->buff), &seminfo->r)) {
|
|
/* format error with correct decimal point: no more options */
|
|
/* format error with correct decimal point: no more options */
|
|
buffreplace(ls, ls->decpoint, '.'); /* undo change (for error message) */
|
|
buffreplace(ls, ls->decpoint, '.'); /* undo change (for error message) */
|
|
@@ -218,6 +233,10 @@ static void read_numeral (LexState *ls, SemInfo *seminfo) {
|
|
}
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
+/*
|
|
|
|
+** skip a sequence '[=*=[' or ']=*]' and return its number of '='s or
|
|
|
|
+** -1 if sequence is malformed
|
|
|
|
+*/
|
|
static int skip_sep (LexState *ls) {
|
|
static int skip_sep (LexState *ls) {
|
|
int count = 0;
|
|
int count = 0;
|
|
int s = ls->current;
|
|
int s = ls->current;
|
|
@@ -248,8 +267,7 @@ static void read_long_string (LexState *ls, SemInfo *seminfo, int sep) {
|
|
}
|
|
}
|
|
break;
|
|
break;
|
|
}
|
|
}
|
|
- case '\n':
|
|
|
|
- case '\r': {
|
|
|
|
|
|
+ case '\n': case '\r': {
|
|
save(ls, '\n');
|
|
save(ls, '\n');
|
|
inclinenumber(ls);
|
|
inclinenumber(ls);
|
|
if (!seminfo) luaZ_resetbuffer(ls->buff); /* avoid wasting space */
|
|
if (!seminfo) luaZ_resetbuffer(ls->buff); /* avoid wasting space */
|
|
@@ -310,7 +328,7 @@ static int readdecesc (LexState *ls) {
|
|
|
|
|
|
|
|
|
|
static void read_string (LexState *ls, int del, SemInfo *seminfo) {
|
|
static void read_string (LexState *ls, int del, SemInfo *seminfo) {
|
|
- save_and_next(ls);
|
|
|
|
|
|
+ save_and_next(ls); /* keep delimiter (for error messages) */
|
|
while (ls->current != del) {
|
|
while (ls->current != del) {
|
|
switch (ls->current) {
|
|
switch (ls->current) {
|
|
case EOZ:
|
|
case EOZ:
|
|
@@ -335,6 +353,14 @@ static void read_string (LexState *ls, int del, SemInfo *seminfo) {
|
|
case '\n':
|
|
case '\n':
|
|
case '\r': save(ls, '\n'); inclinenumber(ls); continue;
|
|
case '\r': save(ls, '\n'); inclinenumber(ls); continue;
|
|
case EOZ: continue; /* will raise an error next loop */
|
|
case EOZ: continue; /* will raise an error next loop */
|
|
|
|
+ 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' */
|
|
|
|
+ }
|
|
default: {
|
|
default: {
|
|
if (!lisdigit(ls->current))
|
|
if (!lisdigit(ls->current))
|
|
c = ls->current; /* handles \\, \", \', and \? */
|
|
c = ls->current; /* handles \\, \", \', and \? */
|
|
@@ -361,31 +387,34 @@ static int llex (LexState *ls, SemInfo *seminfo) {
|
|
luaZ_resetbuffer(ls->buff);
|
|
luaZ_resetbuffer(ls->buff);
|
|
for (;;) {
|
|
for (;;) {
|
|
switch (ls->current) {
|
|
switch (ls->current) {
|
|
- case '\n':
|
|
|
|
- case '\r': {
|
|
|
|
|
|
+ case '\n': case '\r': { /* line breaks */
|
|
inclinenumber(ls);
|
|
inclinenumber(ls);
|
|
break;
|
|
break;
|
|
}
|
|
}
|
|
- case '-': {
|
|
|
|
|
|
+ case ' ': case '\f': case '\t': case '\v': { /* spaces */
|
|
|
|
+ next(ls);
|
|
|
|
+ break;
|
|
|
|
+ }
|
|
|
|
+ case '-': { /* '-' or '--' (comment) */
|
|
next(ls);
|
|
next(ls);
|
|
if (ls->current != '-') return '-';
|
|
if (ls->current != '-') return '-';
|
|
/* else is a comment */
|
|
/* else is a comment */
|
|
next(ls);
|
|
next(ls);
|
|
- if (ls->current == '[') {
|
|
|
|
|
|
+ if (ls->current == '[') { /* long comment? */
|
|
int sep = skip_sep(ls);
|
|
int sep = skip_sep(ls);
|
|
luaZ_resetbuffer(ls->buff); /* `skip_sep' may dirty the buffer */
|
|
luaZ_resetbuffer(ls->buff); /* `skip_sep' may dirty the buffer */
|
|
if (sep >= 0) {
|
|
if (sep >= 0) {
|
|
- read_long_string(ls, NULL, sep); /* long comment */
|
|
|
|
- luaZ_resetbuffer(ls->buff);
|
|
|
|
|
|
+ read_long_string(ls, NULL, sep); /* skip long comment */
|
|
|
|
+ luaZ_resetbuffer(ls->buff); /* previous call may dirty the buff. */
|
|
break;
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
/* else short comment */
|
|
/* else short comment */
|
|
while (!currIsNewline(ls) && ls->current != EOZ)
|
|
while (!currIsNewline(ls) && ls->current != EOZ)
|
|
- next(ls);
|
|
|
|
|
|
+ next(ls); /* skip until end of line (or end of file) */
|
|
break;
|
|
break;
|
|
}
|
|
}
|
|
- case '[': {
|
|
|
|
|
|
+ case '[': { /* long string or simply '[' */
|
|
int sep = skip_sep(ls);
|
|
int sep = skip_sep(ls);
|
|
if (sep >= 0) {
|
|
if (sep >= 0) {
|
|
read_long_string(ls, seminfo, sep);
|
|
read_long_string(ls, seminfo, sep);
|
|
@@ -414,39 +443,30 @@ static int llex (LexState *ls, SemInfo *seminfo) {
|
|
if (ls->current != '=') return '~';
|
|
if (ls->current != '=') return '~';
|
|
else { next(ls); return TK_NE; }
|
|
else { next(ls); return TK_NE; }
|
|
}
|
|
}
|
|
- case '"':
|
|
|
|
- case '\'': {
|
|
|
|
|
|
+ case '"': case '\'': { /* short literal strings */
|
|
read_string(ls, ls->current, seminfo);
|
|
read_string(ls, ls->current, seminfo);
|
|
return TK_STRING;
|
|
return TK_STRING;
|
|
}
|
|
}
|
|
- case '.': {
|
|
|
|
|
|
+ case '.': { /* '.', '..', '...', or number */
|
|
save_and_next(ls);
|
|
save_and_next(ls);
|
|
if (check_next(ls, ".")) {
|
|
if (check_next(ls, ".")) {
|
|
if (check_next(ls, "."))
|
|
if (check_next(ls, "."))
|
|
- return TK_DOTS; /* ... */
|
|
|
|
- else return TK_CONCAT; /* .. */
|
|
|
|
|
|
+ return TK_DOTS; /* '...' */
|
|
|
|
+ else return TK_CONCAT; /* '..' */
|
|
}
|
|
}
|
|
else if (!lisdigit(ls->current)) return '.';
|
|
else if (!lisdigit(ls->current)) return '.';
|
|
- else {
|
|
|
|
- read_numeral(ls, seminfo);
|
|
|
|
- return TK_NUMBER;
|
|
|
|
- }
|
|
|
|
|
|
+ /* 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;
|
|
}
|
|
}
|
|
case EOZ: {
|
|
case EOZ: {
|
|
return TK_EOS;
|
|
return TK_EOS;
|
|
}
|
|
}
|
|
default: {
|
|
default: {
|
|
- if (lisspace(ls->current)) {
|
|
|
|
- lua_assert(!currIsNewline(ls));
|
|
|
|
- next(ls);
|
|
|
|
- break;
|
|
|
|
- }
|
|
|
|
- else if (lisdigit(ls->current)) {
|
|
|
|
- read_numeral(ls, seminfo);
|
|
|
|
- return TK_NUMBER;
|
|
|
|
- }
|
|
|
|
- else if (lislalpha(ls->current)) {
|
|
|
|
- /* identifier or reserved word */
|
|
|
|
|
|
+ if (lislalpha(ls->current)) { /* identifier or reserved word? */
|
|
TString *ts;
|
|
TString *ts;
|
|
do {
|
|
do {
|
|
save_and_next(ls);
|
|
save_and_next(ls);
|
|
@@ -460,10 +480,10 @@ static int llex (LexState *ls, SemInfo *seminfo) {
|
|
return TK_NAME;
|
|
return TK_NAME;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
- else {
|
|
|
|
|
|
+ else { /* single-char tokens (+ - / ...) */
|
|
int c = ls->current;
|
|
int c = ls->current;
|
|
next(ls);
|
|
next(ls);
|
|
- return c; /* single-char tokens (+ - / ...) */
|
|
|
|
|
|
+ return c;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|