summaryrefslogtreecommitdiff
path: root/enc
diff options
context:
space:
mode:
authornobu <nobu@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>2017-03-09 02:04:10 +0000
committernobu <nobu@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>2017-03-09 02:04:10 +0000
commit4171ed6c21a36d88cfc5fe5152a6b58a75cf6240 (patch)
tree7e2e4ce7c0bd435f00c4f26e79f023782fd6f67a /enc
parente65c9bd17749dc50f5ebd514dd6fc1d6ee99643b (diff)
fix UTF-32 valid_encoding?
* enc/utf_32be.c (utf32be_mbc_enc_len): check arguments precisely. [ruby-core:79966] [Bug #13292] * enc/utf_32le.c (utf32le_mbc_enc_len): ditto. * regenc.h (UNICODE_VALID_CODEPOINT_P): predicate for valid Unicode codepoints. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57816 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Diffstat (limited to 'enc')
-rw-r--r--enc/utf_32be.c18
-rw-r--r--enc/utf_32le.c18
2 files changed, 30 insertions, 6 deletions
diff --git a/enc/utf_32be.c b/enc/utf_32be.c
index 995c9d8..17841e5 100644
--- a/enc/utf_32be.c
+++ b/enc/utf_32be.c
@@ -30,11 +30,23 @@
#include "regenc.h"
#include "iso_8859.h"
+static OnigCodePoint utf32be_mbc_to_code(const UChar* p, const UChar* end, OnigEncoding enc);
static int
-utf32be_mbc_enc_len(const UChar* p ARG_UNUSED, const OnigUChar* e ARG_UNUSED,
- OnigEncoding enc ARG_UNUSED)
+utf32be_mbc_enc_len(const UChar* p ARG_UNUSED, const OnigUChar* e,
+ OnigEncoding enc)
{
- return 4;
+ if (e < p) {
+ return ONIGENC_CONSTRUCT_MBCLEN_INVALID();
+ }
+ else if (e-p < 4) {
+ return ONIGENC_CONSTRUCT_MBCLEN_NEEDMORE(4-(int)(e-p));
+ }
+ else {
+ OnigCodePoint c = utf32be_mbc_to_code(p, e, enc);
+ if (!UNICODE_VALID_CODEPOINT_P(c))
+ return ONIGENC_CONSTRUCT_MBCLEN_INVALID();
+ return ONIGENC_CONSTRUCT_MBCLEN_CHARFOUND(4);
+ }
}
static int
diff --git a/enc/utf_32le.c b/enc/utf_32le.c
index e255f0e..18b798f 100644
--- a/enc/utf_32le.c
+++ b/enc/utf_32le.c
@@ -30,11 +30,23 @@
#include "regenc.h"
#include "iso_8859.h"
+static OnigCodePoint utf32le_mbc_to_code(const UChar* p, const UChar* end, OnigEncoding enc);
static int
-utf32le_mbc_enc_len(const UChar* p ARG_UNUSED, const OnigUChar* e ARG_UNUSED,
- OnigEncoding enc ARG_UNUSED)
+utf32le_mbc_enc_len(const UChar* p ARG_UNUSED, const OnigUChar* e,
+ OnigEncoding enc)
{
- return 4;
+ if (e < p) {
+ return ONIGENC_CONSTRUCT_MBCLEN_INVALID();
+ }
+ else if (e-p < 4) {
+ return ONIGENC_CONSTRUCT_MBCLEN_NEEDMORE(4-(int)(e-p));
+ }
+ else {
+ OnigCodePoint c = utf32le_mbc_to_code(p, e, enc);
+ if (!UNICODE_VALID_CODEPOINT_P(c))
+ return ONIGENC_CONSTRUCT_MBCLEN_INVALID();
+ return ONIGENC_CONSTRUCT_MBCLEN_CHARFOUND(4);
+ }
}
static int