Fix wrong wxStringOutputStream optimization for UTF-8 build

We can only use utf8_length() if the conversion used with this object
uses UTF-8 too, otherwise we still need to do the conversion to find out
how many bytes does the string content take in the given encoding.

See #17985.

src/common/sstream.cpp

@@ -136,10 +136,16 @@ wxStringOutputStream::wxStringOutputStream(wxString *pString, wxMBConv& conv)
 {
     m_str = pString ? pString : &m_strInternal;
 
-#if wxUSE_UNICODE_WCHAR
-    m_pos = m_conv.FromWChar(NULL, 0, m_str->wc_str(), m_str->length());
-#elif wxUSE_UNICODE_UTF8
-    m_pos = m_str->utf8_length();
+#if wxUSE_UNICODE
+    // We can avoid doing the conversion in the common case of using UTF-8
+    // conversion in UTF-8 build, as it is exactly the same as the string
+    // length anyhow in this case.
+#if wxUSE_UNICODE_UTF8
+    if ( conv.IsUTF8() )
+        m_pos = m_str->utf8_length();
+    else
+#endif // wxUSE_UNICODE_UTF8
+        m_pos = m_conv.FromWChar(NULL, 0, m_str->wc_str(), m_str->length());
 #else // !wxUSE_UNICODE
     m_pos = m_str->length();
 #endif // wxUSE_UNICODE/!wxUSE_UNICODE