Patch for PRs 34429 and 34896

[Tobias Burnus]

Paul,

Paul Richard Thomas wrote:
> It helps for you to see the patch.....:)
>   
Indeed.

Comments:


   /* Except in the case of the length being a function, where symbol
      association looks after itself, deal with character functions
      after the specification statements.  */
-  if (gfc_matching_function
-       && !(len && len->expr_type != EXPR_VARIABLE
-                && len->expr_type != EXPR_OP))
+  if (gfc_matching_function)
     {
+      gfc_free_expr (len);
       gfc_undo_symbols ();
       return MATCH_YES;
     }

I believe that the comment needs to updated as well.

Ditto:

       /* Except in the case of a function valued character length,
         delay matching the function characteristics until after the
         specification block by signalling kind=-1.  */
-      if (!(current_ts.type == BT_CHARACTER


Otherwise the patch looks OK.

Tobias