diff options
Diffstat (limited to 'generic/regc_lex.c')
| -rw-r--r-- | generic/regc_lex.c | 41 |
1 files changed, 28 insertions, 13 deletions
diff --git a/generic/regc_lex.c b/generic/regc_lex.c index 4be02c6..132e757 100644 --- a/generic/regc_lex.c +++ b/generic/regc_lex.c @@ -742,10 +742,11 @@ lexescape( struct vars *v) { chr c; - static chr alert[] = { + int i; + static const chr alert[] = { CHR('a'), CHR('l'), CHR('e'), CHR('r'), CHR('t') }; - static chr esc[] = { + static const chr esc[] = { CHR('E'), CHR('S'), CHR('C') }; const chr *save; @@ -818,18 +819,23 @@ lexescape( RETV(PLAIN, CHR('\t')); break; case CHR('u'): - c = lexdigits(v, 16, 4, 4); + c = (uchr) lexdigits(v, 16, 1, 4); if (ISERR()) { FAILW(REG_EESCAPE); } RETV(PLAIN, c); break; case CHR('U'): - c = lexdigits(v, 16, 8, 8); + i = lexdigits(v, 16, 1, 8); if (ISERR()) { FAILW(REG_EESCAPE); } - RETV(PLAIN, c); + if (i > 0xFFFF) { + /* TODO: output a Surrogate pair + */ + i = 0xFFFD; + } + RETV(PLAIN, (uchr) i); break; case CHR('v'): RETV(PLAIN, CHR('\v')); @@ -844,7 +850,7 @@ lexescape( break; case CHR('x'): NOTE(REG_UUNPORT); - c = lexdigits(v, 16, 1, 255); /* REs >255 long outside spec */ + c = (uchr) lexdigits(v, 16, 1, 2); if (ISERR()) { FAILW(REG_EESCAPE); } @@ -866,7 +872,7 @@ lexescape( case CHR('9'): save = v->now; v->now--; /* put first digit back */ - c = lexdigits(v, 10, 1, 255); /* REs >255 long outside spec */ + c = (uchr) lexdigits(v, 10, 1, 255); /* REs >255 long outside spec */ if (ISERR()) { FAILW(REG_EESCAPE); } @@ -893,10 +899,15 @@ lexescape( case CHR('0'): NOTE(REG_UUNPORT); v->now--; /* put first digit back */ - c = lexdigits(v, 8, 1, 3); + c = (uchr) lexdigits(v, 8, 1, 3); if (ISERR()) { FAILW(REG_EESCAPE); } + if (c > 0xff) { + /* out of range, so we handled one digit too much */ + v->now--; + c >>= 3; + } RETV(PLAIN, c); break; default: @@ -909,16 +920,16 @@ lexescape( /* - lexdigits - slurp up digits and return chr value - ^ static chr lexdigits(struct vars *, int, int, int); + ^ static int lexdigits(struct vars *, int, int, int); */ -static chr /* chr value; errors signalled via ERR */ +static int /* chr value; errors signalled via ERR */ lexdigits( struct vars *v, int base, int minlen, int maxlen) { - uchr n; /* unsigned to avoid overflow misbehavior */ + int n; int len; chr c; int d; @@ -926,6 +937,10 @@ lexdigits( n = 0; for (len = 0; len < maxlen && !ATEOS(); len++) { + if (n > 0x10fff) { + /* Stop when continuing would otherwise overflow */ + break; + } c = *v->now++; switch (c) { case CHR('0'): case CHR('1'): case CHR('2'): case CHR('3'): @@ -958,7 +973,7 @@ lexdigits( ERR(REG_EESCAPE); } - return (chr)n; + return n; } /* @@ -1135,7 +1150,7 @@ newline(void) static const chr * ch(void) { - static chr chstr[] = { CHR('c'), CHR('h'), CHR('\0') }; + static const chr chstr[] = { CHR('c'), CHR('h'), CHR('\0') }; return chstr; } |