/*
- * Copyright (C) 2012
+ * Copyright (C) 2012, 2013
* Wolfgang Bumiller
*
* Permission is hereby granted, free of charge, to any person obtaining a copy of
if (command) {
vec_pop(command);
while (vec_size(command)) {
- lex_ungetch(lex, vec_last(command));
+ lex_ungetch(lex, (unsigned char)vec_last(command));
vec_pop(command);
}
vec_free(command);
if (command) {
vec_pop(command);
while (vec_size(command)) {
- lex_ungetch(lex, vec_last(command));
+ lex_ungetch(lex, (unsigned char)vec_last(command));
vec_pop(command);
}
vec_free(command);
if (pragma) {
vec_pop(pragma);
while (vec_size(pragma)) {
- lex_ungetch(lex, vec_last(pragma));
+ lex_ungetch(lex, (unsigned char)vec_last(pragma));
vec_pop(pragma);
}
vec_free(pragma);
vec_shrinkto(lex->tok.value, 0);
vec_push(lex->frames, m);
} while (true);
+
+ return false;
}
static int GMQCC_WARN lex_finish_string(lex_file *lex, int quote)
{
+ uchar_t chr;
int ch = 0;
int nextch;
bool hex;
case '[': ch = 16; break;
case ']': ch = 17; break;
case '{':
- ch = 0;
+ chr = 0;
nextch = lex_getch(lex);
hex = (nextch == 'x');
if (!hex)
for (nextch = lex_getch(lex); nextch != '}'; nextch = lex_getch(lex)) {
if (!hex) {
if (nextch >= '0' && nextch <= '9')
- ch = ch * 10 + nextch - '0';
+ chr = chr * 10 + nextch - '0';
else {
lexerror(lex, "bad character code");
return (lex->tok.ttype = TOKEN_ERROR);
}
} else {
- if (nextch >= '0' || nextch <= '9')
- ch = ch * 16 + nextch - '0';
+ if (nextch >= '0' && nextch <= '9')
+ chr = chr * 0x10 + nextch - '0';
else if (nextch >= 'a' && nextch <= 'f')
- ch = ch * 16 + nextch - 'a' + 10;
+ chr = chr * 0x10 + nextch - 'a' + 10;
else if (nextch >= 'A' && nextch <= 'F')
- ch = ch * 16 + nextch - 'A' + 10;
+ chr = chr * 0x10 + nextch - 'A' + 10;
else {
lexerror(lex, "bad character code");
return (lex->tok.ttype = TOKEN_ERROR);
}
}
- if ( (!OPTS_FLAG(UTF8) && ch > 255) ||
- ( OPTS_FLAG(UTF8) && ch > 0x10FFFF) )
+ if (chr > 0x10FFFF || (!OPTS_FLAG(UTF8) && chr > 255))
{
lexerror(lex, "character code out of range");
return (lex->tok.ttype = TOKEN_ERROR);
}
}
- if (OPTS_FLAG(UTF8) && ch >= 128) {
- u8len = u8_fromchar((Uchar)ch, u8buf, sizeof(u8buf));
+ if (OPTS_FLAG(UTF8) && chr >= 128) {
+ u8len = u8_fromchar(chr, u8buf, sizeof(u8buf));
if (!u8len)
ch = 0;
else {
ch = u8buf[uc];
}
}
+ else
+ ch = chr;
break;
case '\n': ch = '\n'; break;
default:
break;
}
+ }
- if (ch == '.')
- {
- lex_tokench(lex, ch);
- /* peak ahead once */
- nextch = lex_getch(lex);
- if (nextch != '.') {
- lex_ungetch(lex, nextch);
- lex_endtoken(lex);
- return (lex->tok.ttype = ch);
- }
- /* peak ahead again */
- nextch = lex_getch(lex);
- if (nextch != '.') {
- lex_ungetch(lex, nextch);
- lex_ungetch(lex, '.');
- lex_endtoken(lex);
+ if (ch == '.')
+ {
+ lex_tokench(lex, ch);
+ /* peak ahead once */
+ nextch = lex_getch(lex);
+ if (nextch != '.') {
+ lex_ungetch(lex, nextch);
+ lex_endtoken(lex);
+ if (lex->flags.noops)
return (lex->tok.ttype = ch);
- }
- /* fill the token to be "..." */
- lex_tokench(lex, ch);
- lex_tokench(lex, ch);
+ else
+ return (lex->tok.ttype = TOKEN_OPERATOR);
+ }
+ /* peak ahead again */
+ nextch = lex_getch(lex);
+ if (nextch != '.') {
+ lex_ungetch(lex, nextch);
+ lex_ungetch(lex, '.');
lex_endtoken(lex);
- return (lex->tok.ttype = TOKEN_DOTS);
+ if (lex->flags.noops)
+ return (lex->tok.ttype = ch);
+ else
+ return (lex->tok.ttype = TOKEN_OPERATOR);
}
+ /* fill the token to be "..." */
+ lex_tokench(lex, ch);
+ lex_tokench(lex, ch);
+ lex_endtoken(lex);
+ return (lex->tok.ttype = TOKEN_DOTS);
}
if (ch == ',' || ch == '.') {
lex_tokench(lex, nextch);
lex_tokench(lex, thirdch);
}
+ }
+ else if (lex->flags.preprocessing &&
+ ch == '-' && isdigit(nextch))
+ {
+ lex->tok.ttype = lex_finish_digit(lex, nextch);
+ if (lex->tok.ttype == TOKEN_INTCONST)
+ lex->tok.constval.i = -lex->tok.constval.i;
+ else
+ lex->tok.constval.f = -lex->tok.constval.f;
+ lex_endtoken(lex);
+ return lex->tok.ttype;
} else
lex_ungetch(lex, nextch);
else
{
if (!lex->flags.preprocessing && strlen(lex->tok.value) > 1) {
- Uchar u8char;
+ uchar_t u8char;
/* check for a valid utf8 character */
if (!OPTS_FLAG(UTF8) || !u8_analyze(lex->tok.value, NULL, NULL, &u8char, 8)) {
- if (lexwarn(lex, WARN_MULTIBYTE_CHARACTER, "multibyte character: `%s`", lex->tok.value))
+ if (lexwarn(lex, WARN_MULTIBYTE_CHARACTER,
+ ( OPTS_FLAG(UTF8) ? "invalid multibyte character sequence `%s`"
+ : "multibyte character: `%s`" ),
+ lex->tok.value))
return (lex->tok.ttype = TOKEN_ERROR);
}
else