cvsuser 05/04/04 01:12:11
Modified: charset unicode.c
encodings utf8.c
t/op string_cs.t
Log:
fix 2 unicode issues: #34660, #34661
Revision Changes Path
1.5 +13 -4 parrot/charset/unicode.c
Index: unicode.c
===================================================================
RCS file: /cvs/public/parrot/charset/unicode.c,v
retrieving revision 1.4
retrieving revision 1.5
diff -u -r1.4 -r1.5
--- unicode.c 2 Mar 2005 17:00:49 -0000 1.4
+++ unicode.c 4 Apr 2005 08:12:09 -0000 1.5
@@ -299,10 +299,19 @@
}
static size_t
-compute_hash(Interp *interpreter, STRING *source_string)
+compute_hash(Interp *interpreter, STRING *src)
{
- UNIMPL;
- return 0;
+ String_iter iter;
+ size_t hashval = 0;
+ UINTVAL offs, c;
+
+ ENCODING_ITER_INIT(interpreter, src, &iter);
+ for (offs = 0; offs < src->strlen; ++offs) {
+ c = iter.get_and_advance(interpreter, &iter);
+ hashval += hashval << 5;
+ hashval += c;
+ }
+ return hashval;
}
CHARSET *
1.26 +3 -2 parrot/encodings/utf8.c
Index: utf8.c
===================================================================
RCS file: /cvs/public/parrot/encodings/utf8.c,v
retrieving revision 1.25
retrieving revision 1.26
diff -u -r1.25 -r1.26
--- utf8.c 7 Mar 2005 16:44:23 -0000 1.25
+++ utf8.c 4 Apr 2005 08:12:10 -0000 1.26
@@ -274,10 +274,11 @@
const STRING *s = i->str;
unsigned char *new_pos, *pos;
- assert(i->bytepos < PObj_buflen(s) - 4);
pos = (unsigned char *)s->strstart + i->bytepos;
new_pos = utf8_encode(pos, c);
i->bytepos += (new_pos - pos);
+ /* XXX possible buffer overrun exception? */
+ assert(i->bytepos < PObj_buflen(s));
i->charpos++;
}
1.9 +27 -2 parrot/t/op/string_cs.t
Index: string_cs.t
===================================================================
RCS file: /cvs/public/parrot/t/op/string_cs.t,v
retrieving revision 1.8
retrieving revision 1.9
diff -u -r1.8 -r1.9
--- string_cs.t 2 Mar 2005 13:47:33 -0000 1.8
+++ string_cs.t 4 Apr 2005 08:12:10 -0000 1.9
@@ -16,7 +16,7 @@
=cut
-use Parrot::Test tests => 28;
+use Parrot::Test tests => 30;
use Test::More;
output_is( <<'CODE', <<OUTPUT, "basic syntax" );
@@ -445,3 +445,28 @@
6
OUTPUT
+pir_output_is( <<'CODE', <<'OUTPUT', "bug #34661 literal" );
+.sub main @MAIN
+ $S0 = unicode:"\"]\nif I3 == "
+ print "ok 1\n"
+.end
+CODE
+ok 1
+OUTPUT
+
+pir_output_is( <<'CODE', <<'OUTPUT', "todo #34660 hash" );
+.sub main @MAIN
+ $P0 = new Integer
+ $P0 = 42
+ store_global "Foo", unicode:"Bar", $P0
+ print "ok 1\n"
+ $P1 = find_global "Foo", "Bar"
+ print "ok 2\n"
+ print $P1
+ print "\n"
+.end
+CODE
+ok 1
+ok 2
+42
+OUTPUT
