aboutsummaryrefslogtreecommitdiffstats
path: root/bruiser/lua-5.3.4/src/llex.c
diff options
context:
space:
mode:
authorbloodstalker <thabogre@gmail.com>2017-05-12 12:10:08 +0000
committerbloodstalker <thabogre@gmail.com>2017-05-12 12:10:08 +0000
commit3f764e681f8d868fe725e9a573b3a1815194a559 (patch)
tree435af067340d7f027e598a79e1ec98b0c57b3e74 /bruiser/lua-5.3.4/src/llex.c
parentits a submodule (diff)
downloadmutator-3f764e681f8d868fe725e9a573b3a1815194a559.tar.gz
mutator-3f764e681f8d868fe725e9a573b3a1815194a559.zip
yup. lua. better than whatever shaky messy DSL i was gonna come up with for bruiseer.
Diffstat (limited to 'bruiser/lua-5.3.4/src/llex.c')
-rw-r--r--bruiser/lua-5.3.4/src/llex.c565
1 files changed, 565 insertions, 0 deletions
diff --git a/bruiser/lua-5.3.4/src/llex.c b/bruiser/lua-5.3.4/src/llex.c
new file mode 100644
index 0000000..7032827
--- /dev/null
+++ b/bruiser/lua-5.3.4/src/llex.c
@@ -0,0 +1,565 @@
+/*
+** $Id: llex.c,v 2.96 2016/05/02 14:02:12 roberto Exp $
+** Lexical Analyzer
+** See Copyright Notice in lua.h
+*/
+
+#define llex_c
+#define LUA_CORE
+
+#include "lprefix.h"
+
+
+#include <locale.h>
+#include <string.h>
+
+#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 currIsNewline(ls) (ls->current == '\n' || ls->current == '\r')
+
+
+/* ORDER RESERVED */
+static const char *const luaX_tokens [] = {
+ "and", "break", "do", "else", "elseif",
+ "end", "false", "for", "function", "goto", "if",
+ "in", "local", "nil", "not", "or", "repeat",
+ "return", "then", "true", "until", "while",
+ "//", "..", "...", "==", ">=", "<=", "~=",
+ "<<", ">>", "::", "<eof>",
+ "<number>", "<integer>", "<name>", "<string>"
+};
+
+
+#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;
+ if (luaZ_sizebuffer(b) >= MAX_SIZE/2)
+ lexerror(ls, "lexical element too long", 0);
+ newsize = luaZ_sizebuffer(b) * 2;
+ 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 */
+ }
+}
+
+
+const char *luaX_token2str (LexState *ls, int token) {
+ if (token < FIRST_RESERVED) { /* single-byte symbols? */
+ lua_assert(token == cast_uchar(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;
+ }
+}
+
+
+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);
+}
+
+
+/*
+** creates a new string and anchors it 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)
+*/
+TString *luaX_newstring (LexState *ls, const char *str, size_t l) {
+ lua_State *L = ls->L;
+ TValue *o; /* entry for 'str' */
+ TString *ts = luaS_newlstr(L, str, l); /* create new string */
+ setsvalue2s(L, L->top++, ts); /* temporarily anchor it in stack */
+ o = luaH_set(L, ls->h, L->top - 1);
+ if (ttisnil(o)) { /* not in use yet? */
+ /* boolean value does not need GC barrier;
+ table has no metatable, so it does not need to invalidate cache */
+ setbvalue(o, 1); /* t[string] = true */
+ luaC_checkGC(L);
+ }
+ else { /* string already present */
+ ts = tsvalue(keyfromval(o)); /* re-use value previously stored */
+ }
+ L->top--; /* remove string from stack */
+ return ts;
+}
+
+
+/*
+** 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);
+}
+
+
+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;
+ ls->envn = luaS_newliteral(L, LUA_ENV); /* get env name */
+ luaZ_resizebuffer(ls->L, ls->buff, LUA_MINBUFFER); /* initialize buffer */
+}
+
+
+
+/*
+** =======================================================
+** LEXICAL ANALYZER
+** =======================================================
+*/
+
+
+static int check_next1 (LexState *ls, int c) {
+ if (ls->current == c) {
+ next(ls);
+ return 1;
+ }
+ else return 0;
+}
+
+
+/*
+** 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;
+}
+
+
+/* LUA_NUMBER */
+/*
+** 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;
+ }
+}
+
+
+/*
+** skip a sequence '[=*[' or ']=*]'; if sequence is well formed, return
+** its number of '='s; otherwise, return a negative number (-1 iff there
+** are no '='s after initial bracket)
+*/
+static int skip_sep (LexState *ls) {
+ int 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 : (-count) - 1;
+}
+
+
+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 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;
+}
+
+
+static unsigned long readutf8esc (LexState *ls) {
+ unsigned long r;
+ int i = 4; /* chars to be removed: '\', 'u', '{', and first digit */
+ save_and_next(ls); /* skip 'u' */
+ esccheck(ls, ls->current == '{', "missing '{'");
+ r = gethexa(ls); /* must have at least one digit */
+ while ((save_and_next(ls), lisxdigit(ls->current))) {
+ i++;
+ r = (r << 4) + luaO_hexavalue(ls->current);
+ esccheck(ls, r <= 0x10FFFF, "UTF-8 value too large");
+ }
+ 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) {
+ 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, 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 '.': { /* '.', '..', '...', 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;
+ }
+ }
+ }
+ }
+}
+
+
+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;
+}
+