[fortran,patch] Last widechar intrinsics and a few related fixes

[FX]

>    if (sslen == 0)
> -    return 1;
> +    return back ? (slen + 1) : 1;
>
> Good catch!

I wrote the testcases in such a way that they compare the result of  
simplification routines and the runtime result, which caught it.

> +++ libgfortran/gfortran.map	(working copy)
> @@ -1065,6 +1065,27 @@ GFORTRAN_1.1 {
>      _gfortran_transpose_char4;
>      _gfortran_unpack0_char4;
>      _gfortran_unpack1_char4;
> +    _gfortran_cshift0_1_char4;
>
> Recently, we decided to sort this list (see Thomas' patch/Changlog
> item).

OK, I will change that. But the list currently isn't strictly in  
alphabetic order; if I use sort on it, it changes it that way:

@@ -1051,17 +1072,17 @@ GFORTRAN_1.1 {
      _gfortran_pack_char4;
      _gfortran_pack_s_char4;
      _gfortran_reshape_char4;
-    _gfortran_select_string_char4;
      _gfortran_selected_char_kind;
+    _gfortran_select_string_char4;
      _gfortran_spread_char4;
      _gfortran_spread_char4_scalar;
-    _gfortran_st_wait;
      _gfortran_string_index_char4;
      _gfortran_string_len_trim_char4;
      _gfortran_string_minmax_char4;
      _gfortran_string_scan_char4;
      _gfortran_string_trim_char4;
      _gfortran_string_verify_char4;
+    _gfortran_st_wait;
      _gfortran_transpose_char4;
      _gfortran_unpack0_char4;
      _gfortran_unpack1_char4;

(I've committed that sorted version.)

> PS: Any plans for supporting UTF-8 source files (cf. -finput- 
> charset=)?


Yes, I intend to work on that.

FX

-- 
François-Xavier Coudert
http://www.homepages.ucl.ac.uk/~uccafco/