fix [#28658] python can assign non utf8 and crash because of string lenth limits.

add BLI_strncpy_utf8() which which ensures there are no partially copied UTF8 characters, limited by the buffer size.

2 files changed, 41 insertions, 0 deletions

diff --git a/source/blender/blenlib/BLI_string.h b/source/blender/blenlib/BLI_string.h
index be77e18c24b..c53ce9dced5 100644
--- a/source/blender/blenlib/BLI_string.h
+++ b/source/blender/blenlib/BLI_string.h
@@ -144,6 +144,7 @@ void BLI_ascii_strtoupper(char *str, int len);
 
 
 /* string_utf8.c - may move these into their own header some day - campbell */
+char *BLI_strncpy_utf8(char *dst, const char *src, size_t maxncpy);
 int BLI_utf8_invalid_byte(const char *str, int length);
 int BLI_utf8_invalid_strip(char *str, int length);
 
diff --git a/source/blender/blenlib/intern/string_utf8.c b/source/blender/blenlib/intern/string_utf8.c
index 8f7e4518e03..5c37d3003e4 100644
--- a/source/blender/blenlib/intern/string_utf8.c
+++ b/source/blender/blenlib/intern/string_utf8.c
@@ -141,3 +141,43 @@ int BLI_utf8_invalid_strip(char *str, int length)
 
 	return tot;
 }
+
+
+/* compatible with BLI_strncpy, but esnure no partial utf8 chars */
+
+/* array copied from glib's glib's gutf8.c,
+ * note: this looks to be at odd's with 'trailingBytesForUTF8',
+ * need to find out what gives here! - campbell */
+static const size_t utf8_skip_data[256] = {
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+  2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
+  3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,4,4,4,4,4,4,4,4,5,5,5,5,6,6,1,1
+};
+
+char *BLI_strncpy_utf8(char *dst, const char *src, size_t maxncpy)
+{
+	char *dst_r= dst;
+	size_t utf8_size;
+
+	/* note: currently we dont attempt to deal with invalid utf8 chars */
+
+	while(*src != '\0' && (utf8_size= utf8_skip_data[*src]) < maxncpy) {
+		maxncpy -= utf8_size;
+		switch(utf8_size) {
+			case 6: *dst ++ = *src ++;
+			case 5: *dst ++ = *src ++;
+			case 4: *dst ++ = *src ++;
+			case 3: *dst ++ = *src ++;
+			case 2: *dst ++ = *src ++;
+			case 1: *dst ++ = *src ++;
+		}
+	}
+	*dst= '\0';
+	return dst_r;
+}
+