/* ** $Id: llex.c,v 2.20.1.1 2007/12/27 13:02:25 roberto Exp $ ** Lexical Analyzer ** See Copyright Notice in lua.h */ #include #include #include #define llex_c #define LUA_CORE #include "lua.h" #include "ldo.h" #include "llex.h" #include "lobject.h" #include "lparser.h" #include "lstate.h" #include "lstring.h" #include "ltable.h" #include "lzio.h" #define _HEX2INT_(v) ( (v) <= '9' ? (v)-'0' : tolower(v)-'a'+10 ) #define next(ls) (ls->current = zgetc(ls->z)) #define currIsNewline(ls) (ls->current == '\n' || ls->current == '\r') /* ORDER RESERVED */ const char *const luaX_tokens [] = { "and", "break", "continue", "do", "else", "elseif", "end", "false", "for", "function", "if", "in", "local", "nil", "not", "or", "repeat", "return", "then", "true", "until", "while", "..", "...", "==", ">=", "<=", "~=", "", "", "", "", "<<", ">>", "^^", NULL }; #define save_and_next(ls) (save(ls, ls->current), next(ls)) static void save (LexState *ls, int c) { Mbuffer *b = ls->buff; if (b->n + 1 > b->buffsize) { size_t newsize; if (b->buffsize >= MAX_SIZET/2) luaX_lexerror(ls, "lexical element too long", 0); newsize = b->buffsize * 2; luaZ_resizebuffer(ls->L, b, newsize); } b->buffer[b->n++] = cast(char, c); } void luaX_init (lua_State *L) { int i; for (i=0; itsv.reserved = cast_byte(i+1); /* reserved word */ } } #define MAXSRC 80 const char *luaX_token2str (LexState *ls, int token) { if (token < FIRST_RESERVED) { lua_assert(token == cast(unsigned char, token)); return (iscntrl(token)) ? luaO_pushfstring(ls->L, "char(%d)", token) : luaO_pushfstring(ls->L, "%c", token); } else return luaX_tokens[token-FIRST_RESERVED]; } static const char *txtToken (LexState *ls, int token) { switch (token) { case TK_NAME: case TK_STRING: case TK_NUMBER: save(ls, '\0'); return luaZ_buffer(ls->buff); default: return luaX_token2str(ls, token); } } void luaX_lexerror (LexState *ls, const char *msg, int token) { char buff[MAXSRC]; luaO_chunkid(buff, getstr(ls->source), MAXSRC); msg = luaO_pushfstring(ls->L, "%s:%d: %s", buff, ls->linenumber, msg); if (token) luaO_pushfstring(ls->L, "%s near " LUA_QS, msg, txtToken(ls, token)); luaD_throw(ls->L, LUA_ERRSYNTAX); } void luaX_syntaxerror (LexState *ls, const char *msg) { luaX_lexerror(ls, msg, ls->t.token); } TString *luaX_newstring (LexState *ls, const char *str, size_t l) { lua_State *L = ls->L; TString *ts = luaS_newlstr(L, str, l); TValue *o = luaH_setstr(L, ls->fs->h, ts); /* entry for `str' */ if (ttisnil(o)) setbvalue(o, 1); /* make sure `str' will not be collected */ return ts; } 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) luaX_syntaxerror(ls, "chunk has too many lines"); } void luaX_setinput (lua_State *L, LexState *ls, ZIO *z, TString *source) { ls->decpoint = '.'; 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; ls->refstr = 0; luaZ_resizebuffer(ls->L, ls->buff, LUA_MINBUFFER); /* initialize buffer */ next(ls); /* read first char */ } /* ** ======================================================= ** LEXICAL ANALYZER ** ======================================================= */ static int check_next (LexState *ls, const char *set) { if (!strchr(set, ls->current)) return 0; save_and_next(ls); return 1; } static void buffreplace (LexState *ls, char from, char to) { size_t n = luaZ_bufflen(ls->buff); char *p = luaZ_buffer(ls->buff); while (n--) if (p[n] == from) p[n] = to; } static void trydecpoint (LexState *ls, SemInfo *seminfo) { /* format error: try to update decimal point separator */ struct lconv *cv = localeconv(); char old = ls->decpoint; ls->decpoint = (cv ? cv->decimal_point[0] : '.'); buffreplace(ls, old, ls->decpoint); /* try updated decimal separator */ if (!luaO_str2d(luaZ_buffer(ls->buff), &seminfo->r)) { /* format error with correct decimal point: no more options */ buffreplace(ls, ls->decpoint, '.'); /* undo change (for error message) */ luaX_lexerror(ls, "malformed number", TK_NUMBER); } } /* LUA_NUMBER */ static void read_numeral (LexState *ls, SemInfo *seminfo) { lua_assert(isdigit(ls->current)); do { save_and_next(ls); } while (isdigit(ls->current) || ls->current == '.'); if (check_next(ls, "Ee")) /* `E'? */ check_next(ls, "+-"); /* optional exponent sign */ while (isalnum(ls->current) || ls->current == '_') save_and_next(ls); save(ls, '\0'); buffreplace(ls, '.', ls->decpoint); /* follow locale for decimal point */ if (!luaO_str2d(luaZ_buffer(ls->buff), &seminfo->r)) /* format error? */ trydecpoint(ls, seminfo); /* try to update decimal point separator */ } 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 cont = 0; (void)(cont); /* avoid warnings when `cont' is not used */ save_and_next(ls); /* skip 2nd `[' */ if (currIsNewline(ls)) /* string starts with a newline? */ inclinenumber(ls); /* skip it */ for (;;) { switch (ls->current) { case EOZ: luaX_lexerror(ls, (seminfo) ? "unfinished long string" : "unfinished long comment", TK_EOS); break; /* to avoid warnings */ #if defined(LUA_COMPAT_LSTR) case '[': { if (skip_sep(ls) == sep) { save_and_next(ls); /* skip 2nd `[' */ cont++; #if LUA_COMPAT_LSTR == 1 if (sep == 0) luaX_lexerror(ls, "nesting of [[...]] is deprecated", '['); #endif } break; } #endif case ']': { if (skip_sep(ls) == sep) { save_and_next(ls); /* skip 2nd `]' */ #if defined(LUA_COMPAT_LSTR) && LUA_COMPAT_LSTR == 2 cont--; if (sep == 0 && cont >= 0) break; #endif 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 read_string (LexState *ls, int del, SemInfo *seminfo) { save_and_next(ls); while (ls->current != del) { switch (ls->current) { case EOZ: luaX_lexerror(ls, "unfinished string", TK_EOS); continue; /* to avoid warnings */ case '\n': case '\r': luaX_lexerror(ls, "unfinished string", TK_STRING); continue; /* to avoid warnings */ case '\\': { int c; next(ls); /* do not save the `\' */ switch (ls->current) { 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 '\n': /* go through */ case '\r': save(ls, '\n'); inclinenumber(ls); continue; case EOZ: continue; /* will raise an error next loop */ case '$': if (del == '"') { save_and_next(ls); /* skip $ */ seminfo->ts = luaX_newstring(ls, luaZ_buffer(ls->buff) + 1, luaZ_bufflen(ls->buff) - 2); ls->refstr++; /* expect '\' anytime soon */ lua_assert(ls->lookahead.token == TK_EOS); ls->lookahead.token = TK_CONCAT; return; } /* FALLTHRU */ default: { if (!isdigit(ls->current)) save_and_next(ls); /* handles \\, \", \', and \? */ else { /* \xxx */ int i = 0; c = 0; do { c = 10*c + (ls->current-'0'); next(ls); } while (++i<3 && isdigit(ls->current)); if (c > UCHAR_MAX) luaX_lexerror(ls, "escape sequence too large", TK_STRING); save(ls, c); } continue; } // "\x00".."\xFFFF": raw hex literals // case 'x': { int i; c = 0; next(ls); // skip 'x' for( i=0; i<5 && isxdigit(ls->current); i++ ) { c = (c<<4) | _HEX2INT_(ls->current); next(ls); }; switch (i) { case 4: save( ls, (c>>8) & 0xff ); /* FALLTHRU */ case 2: save( ls, c&0xff ); break; case 5: default: luaX_lexerror(ls, "escape sequence wrong size (2 or 4 digits)", TK_STRING); } } continue; // "\u0000".."\x10FFFF": UTF-8 encoded Unicode // // Note that although codes are entered like this (must have min. four digit, // just to tease you!) the actual outcome in the string will vary between // 1..4 bytes, depending on the value range. // case 'u': { int i; c = 0; next(ls); // skip 'x' for ( i=0; i<7 && isxdigit(ls->current); i++ ) { c = (c<<4) | _HEX2INT_(ls->current); next(ls); }; if ((i!=4) && (i!=6)) { luaX_lexerror(ls, "escape sequence wrong size (4 or 6 digits)", TK_STRING); } /* http://www.cl.cam.ac.uk/~mgk25/unicode.html#utf-8 U-00000000 - U-0000007F: 0xxxxxxx U-00000080 - U-000007FF: 110xxxxx 10xxxxxx U-00000800 - U-0000FFFF: 1110xxxx 10xxxxxx 10xxxxxx U-00010000 - U-001FFFFF: 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx (U-00200000 - U-03FFFFFF: 111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx) (U-04000000 - U-7FFFFFFF: 1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx) */ if ( c <= 0x007f ) { // 1-byte save( ls, c ); } else if ( c <= 0x000007ff ) { // 2-byte save( ls, 0xc0 | ((c>>6)&0x1f) ); save( ls, 0x80 | (c&0x3f) ); } else if ( c <= 0x0000ffff ) { // 3-byte save( ls, 0xe0 | ((c>>12)&0x0f) ); save( ls, 0x80 | ((c>>6)&0x3f) ); save( ls, 0x80 | (c&0x3f) ); } else if ( c <= 0x001fffff ) { // 4-byte save( ls, 0xf0 | ((c>>18)&0x07) ); save( ls, 0x80 | ((c>>12)&0x3f) ); save( ls, 0x80 | ((c>>6)&0x3f) ); save( ls, 0x80 | (c&0x3f) ); } else { luaX_lexerror(ls, "escape sequence too large", TK_STRING); } } continue; } save(ls, c); next(ls); continue; } 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': { inclinenumber(ls); continue; } case '-': { next(ls); if (ls->current != '-') return '-'; /* else is a comment */ next(ls); if (ls->current == '[') { int sep = skip_sep(ls); luaZ_resetbuffer(ls->buff); /* `skip_sep' may dirty the buffer */ if (sep >= 0) { read_long_string(ls, NULL, sep); /* long comment */ luaZ_resetbuffer(ls->buff); continue; } } /* else short comment */ while (!currIsNewline(ls) && ls->current != EOZ) next(ls); continue; } case '/': { next(ls); if (ls->current != '/' && ls->current != '*') return '/'; /* else is a comment */ if (ls->current == '*') { char last = '\n'; /* don't remember the first asterisk, we need a different one. */ next(ls); if (currIsNewline(ls)) /* comment starts with a newline? */ inclinenumber(ls); /* skip it */ while(last != '*' || ls->current != '/') { switch (ls->current) { case EOZ: luaX_lexerror(ls, "unterminated long comment", TK_EOS); break; case '\n': case '\r': last = '\n'; /* remember the newline */ inclinenumber(ls); break; default: last = ls->current; next(ls); break; } } next(ls); continue; } /* else short comment */ while (!currIsNewline(ls) && ls->current != EOZ) next(ls); continue; } case '[': { int sep = skip_sep(ls); if (sep >= 0) { read_long_string(ls, seminfo, sep); return TK_STRING; } else if (sep == -1) return '['; else luaX_lexerror(ls, "invalid long string delimiter", TK_STRING); } /* FALLTHRU */ case '=': { next(ls); if (ls->current != '=') return '='; else { next(ls); return TK_EQ; } } /* FALLTHRU */ case '<': { next(ls); if (ls->current == '<') { next(ls); return TK_SHL; } if (ls->current != '=') return '<'; else { next(ls); return TK_LE; } } case '>': { next(ls); if (ls->current == '>') { next(ls); return TK_SHR; } if (ls->current != '=') return '>'; else { next(ls); return TK_GE; } } case '^': { next(ls); if (ls->current == '^') { next(ls); return TK_XOR; } else { return '^'; } } 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_NE; } } case '"': case '\'': { read_string(ls, ls->current, seminfo); return TK_STRING; } case '.': { save_and_next(ls); if (check_next(ls, ".")) { if (check_next(ls, ".")) return TK_DOTS; /* ... */ else return TK_CONCAT; /* .. */ } else if (!isdigit(ls->current)) return '.'; else { read_numeral(ls, seminfo); return TK_NUMBER; } } case '$': { int i = 0; next(ls); while (isdigit(ls->current)) { i = 10*i + (ls->current-'0'); next(ls); } seminfo->r = i; return '$'; } case EOZ: { return TK_EOS; } case '\\': if (ls->refstr) { ls->refstr--; ls->current = '"'; /* whacky! */ return TK_CONCAT; } /* FALLTHRU */ default: { if (isspace(ls->current)) { lua_assert(!currIsNewline(ls)); next(ls); continue; } else if (isdigit(ls->current)) { read_numeral(ls, seminfo); return TK_NUMBER; } else if (isalpha(ls->current) || ls->current == '_') { /* identifier or reserved word */ TString *ts; do { save_and_next(ls); } while (isalnum(ls->current) || ls->current == '_'); ts = luaX_newstring(ls, luaZ_buffer(ls->buff), luaZ_bufflen(ls->buff)); if (ts->tsv.reserved > 0) /* reserved word? */ return ts->tsv.reserved - 1 + FIRST_RESERVED; else { seminfo->ts = ts; return TK_NAME; } } else { int c = ls->current; 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 */ } void luaX_lookahead (LexState *ls) { lua_assert(ls->lookahead.token == TK_EOS); ls->lookahead.token = llex(ls, &ls->lookahead.seminfo); }