summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rw-r--r--ChangeLog4
-rw-r--r--re.c17
-rw-r--r--test/ruby/test_m17n.rb10
3 files changed, 27 insertions, 4 deletions
diff --git a/ChangeLog b/ChangeLog
index eac59a03a5..ea9a19ca47 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,7 @@
+Sun Dec 9 12:49:34 2007 Tanaka Akira <akr@fsij.org>
+
+ * re.c (append_utf8): check unicode range.
+
Sun Dec 9 12:39:01 2007 Nobuyoshi Nakada <nobu@ruby-lang.org>
* lib/cgi.rb (read_multipart): exclude blanks from header values.
diff --git a/re.c b/re.c
index 6187273661..1c7486068f 100644
--- a/re.c
+++ b/re.c
@@ -1425,9 +1425,22 @@ unescape_escaped_nonascii(const char **pp, const char *end, rb_encoding *enc,
}
static int
+check_unicode_range(unsigned long code, onig_errmsg_buffer err)
+{
+ if ((0xd800 <= code && code <= 0xdfff) || /* Surrogates */
+ 0x10ffff < code) {
+ strcpy(err, "invalid Unicode range");
+ return -1;
+ }
+ return 0;
+}
+
+static int
append_utf8(unsigned long uv,
VALUE buf, rb_encoding **encp, onig_errmsg_buffer err)
{
+ if (check_unicode_range(uv, err) != 0)
+ return -1;
if (uv < 0x80) {
char escbuf[5];
snprintf(escbuf, sizeof(escbuf), "\\x%02x", (int)uv);
@@ -1468,10 +1481,6 @@ unescape_unicode_list(const char **pp, const char *end,
strcpy(err, "invalid Unicode range");
return -1;
}
- if (0x10ffff < code) {
- strcpy(err, "invalid Unicode range");
- return -1;
- }
p += len;
if (append_utf8(code, buf, encp, err) != 0)
return -1;
diff --git a/test/ruby/test_m17n.rb b/test/ruby/test_m17n.rb
index 28f636fc65..f0c98eb4eb 100644
--- a/test/ruby/test_m17n.rb
+++ b/test/ruby/test_m17n.rb
@@ -446,6 +446,16 @@ class TestM17N < Test::Unit::TestCase
#assert_raise(SyntaxError) { s1, s2 = u('\xc2'), u('\xa1'); /#{s1}#{s2}/ }
end
+ def test_regexp_unicode
+ assert_nothing_raised { eval '/\u{0}/' }
+ assert_nothing_raised { eval '/\u{D7FF}/' }
+ assert_raise(SyntaxError) { eval '/\u{D800}/' }
+ assert_raise(SyntaxError) { eval '/\u{DFFF}/' }
+ assert_nothing_raised { eval '/\u{E000}/' }
+ assert_nothing_raised { eval '/\u{10FFFF}/' }
+ assert_raise(SyntaxError) { eval '/\u{110000}/' }
+ end
+
def test_regexp_mixed_unicode
assert_raise(SyntaxError) { eval(a(%{/\xc2\xa0\\u{6666}/})) }
assert_raise(SyntaxError) { eval(e(%{/\xc2\xa0\\u{6666}/})) }