From: marcov@stack.nl   
      
   On 2005-04-13, Dr John Stockton  wrote:   
   > JRS:  In article , dated Wed, 13   
   > Apr 2005 10:52:18, seen in news:comp.lang.pascal.borland, Marco van de   
   > Voort  posted :   
   >   
   >>> Borland did not think about the case   
   >>>   
   >>>          const Mes : set of char = ['W', 'a', 'r', 'n', 'i', 'n','g'] ;   
   >>>          begin end.   
   >>>   
   >>> to be treated as a string.   
   >>   
   >>Well, count your blessings ;-)   
   >>   
   >>Whatever you think about the preprocessor, having that syntax _in_ language   
   >>would be horrible IMHO.   
   >   
   > The quoted code is compilable in BP7 and D3 (I'd have lined a warning   
   > about the duplicate 'n'); but,   
      
   I meant anything that makes a string out of the above construct, not   
   the construct as it stands above.   
      
   > given 'a'..'z' notation, I don't see that   
   > any improvement would be useful often enough to justify Borland making   
   > special provision.   
      
   Indeed. And bsides that it is ugly.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)