Explain modperl that ZNC uses UTF-8 internally

Thanks to jiggawattz for showing the issue

modules/modperl/CMakeLists.txt

@@ -48,6 +48,7 @@ if(SWIG_FOUND)
 		"-I${PROJECT_BINARY_DIR}/include"
 		"-I${PROJECT_SOURCE_DIR}/include"
 		"-I${CMAKE_CURRENT_SOURCE_DIR}/.."
+		"-I${CMAKE_CURRENT_SOURCE_DIR}/include"
 		-DZNC_EXPORT_LIB_EXPORT
 		-outdir "${CMAKE_CURRENT_BINARY_DIR}"
 		-o "${CMAKE_CURRENT_BINARY_DIR}/modperl_biglib.cpp"

modules/modperl/include/perlfragments.swg

@@ -0,0 +1,15 @@
+// Make perl strings to be UTF-8, they are already UTF-8 in ZNC core
+
+%fragment("SWIG_FromCharPtrAndSize","header") {
+    SWIGINTERNINLINE SV *
+    SWIG_FromCharPtrAndSize(const char* carray, size_t size) {
+        SV *obj = sv_newmortal();
+        if (carray) {
+            sv_setpvn(obj, carray, size);
+        } else {
+            sv_setsv(obj, &PL_sv_undef);
+        }
+        SvUTF8_on(obj);
+        return obj;
+    }
+}

modules/modperl/pstring.h

@@ -61,7 +61,8 @@ class PString : public CString {
                 break;
             case STRING:
             default:
-                pSV = newSVpv(data(), length());
+                pSV = newSVpvn(data(), length());
+                SvUTF8_on(pSV);
                 break;
         }