|Subject:||improper "require utf8" in tests|
Tests "33_non_latin_path.t" and "rt_25924_user_defined_func_unicode.t" invoke the "utf8" pragma. However, according to the doc, this pragma is supposed to be used only if the source code is in utf8, which is not the case here. So it is quite likely that accented characters in those tests are misinterpreted, and that the tests nevertheless pass by mere accident. Note : the tests invoke "utf8::upgrade(...)", but this function is always available, doesn't need any "require utf8" or "use utf8".