Don't assume from="UTF-8" in encoding test (#7852)

* Extend the UTF-8 check for tests 2298* on old Windows
* Omit from=UTF-8 to simplify
* don't use ' within '-delimited character string

Author: MichaelChirico

inst/tests/tests.Rraw

@@ -20309,7 +20309,9 @@ local(if (utf8_check(c("\u00e4", "\u00f6", "\u00fc"))) {
   # a-umlaut, o-umlaut, u-umlaut
   eval(parse(text = '  # eval(parse()) defers parsing to runtime; see utf8_check description
     setnames(x , c("\u00e4", "\u00f6", "\u00fc"))
-    setnames(y , iconv(c("\u00f6", "\u00fc", "\u00e4"), from = "UTF-8", to = "latin1"))
+    # NB: since these can be represented e.g. in latin1, after parse(), they will be in
+    #   native encoding, _not_ necessarily UTF-8, so omit from=
+    setnames(y , iconv(c("\u00f6", "\u00fc", "\u00e4"), to = "latin1"))
     test(2298.1, rbindlist(list(x,y), use.names=TRUE), data.table("\u00e4"=c(1,6), "\u00f6"=c(2,4), "\u00fc"=c(3,5)))
     test(2298.2, rbindlist(list(y,x), use.names=TRUE), data.table("\u00f6"=c(4,2), "\u00fc"=c(5,3), "\u00e4"=c(6,1)))
     set(y, j="\u00e4", value=NULL)