diff --git a/common/meson.build b/common/meson.build
index dc2f62af..6348dce5 100644
--- a/common/meson.build
+++ b/common/meson.build
@@ -5,4 +5,7 @@ libui_sources += [
 	'common/controls.c',
 	'common/errors.c',
 	'common/main.c',
+	'common/utf8.c',
+
+	'common/third_party/utf.c',
 ]
diff --git a/zOLD_common/utf.c b/common/third_party/utf.c
similarity index 100%
rename from zOLD_common/utf.c
rename to common/third_party/utf.c
diff --git a/zOLD_common/utf.h b/common/third_party/utf.h
similarity index 100%
rename from zOLD_common/utf.h
rename to common/third_party/utf.h
diff --git a/common/uipriv.h b/common/uipriv.h
index cb661dc5..7cb3b5a1 100644
--- a/common/uipriv.h
+++ b/common/uipriv.h
@@ -75,6 +75,10 @@ extern void uiprivReportError(const char *prefix, const char *msg, const char *s
 extern bool uiprivOSVtableValid(const uiControlOSVtable *osVtable, const char *func);
 extern uiControlOSVtable *uiprivCloneOSVtable(const uiControlOSVtable *osVtable);
 
+// utf8.c
+extern char *uiprivSanitizeUTF8(const char *str);
+extern void uiprivFreeUTF8(char *sanitized);
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/common/utf8.c b/common/utf8.c
new file mode 100644
index 00000000..ccc984c7
--- /dev/null
+++ b/common/utf8.c
@@ -0,0 +1,41 @@
+// 17 may 2020
+#include "uipriv.h"
+#include "third_party/utf.h"
+
+// TODO write separate tests for this file?
+// TODO ideally this functionality should really be part of utf itself, in some form or another (for instance, via utf8SanitizedLen() + the requisite loop)
+
+#define nGrow 32
+
+char *uiprivSanitizeUTF8(const char *str)
+{
+	size_t len;
+	char *out;
+	const char *s;
+	size_t i;
+	uint32_t rune;
+	char encoded[4];
+	size_t n;
+
+	// TODO can we even use strlen() with UTF-8 strings? or is '\0' == 0 == actual memory zero just a source code connection (and thus the last one isn't necessarily true)?
+	len = strlen(str);
+	out = (char *) uiprivAlloc((len + 1) * sizeof (char), "sanitized UTF-8 string");
+	s = str;
+	i = 0;
+	while (*s != '\0') {
+		s = uiprivUTF8DecodeRune(s, 0, &rune);
+		n = uiprivUTF8EncodeRune(rune, encoded);
+		if ((i + n) >= len) {
+			out = (char *) uiprivRealloc(out, (len + 1) * sizeof (char), (len + nGrow + 1) * sizeof (char), "sanitized UTF-8 string");
+			len += nGrow;
+		}
+		memcpy(out + i, encoded, n);
+		i += n;
+	}
+	return out;
+}
+
+void uiprivFreeUTF8(char *sanitized)
+{
+	uiprivFree(sanitized);
+}