• davidl (48/48) Jul 23 2008 consider:
• downs (4/62) Jul 23 2008 version(v1) auto cond = a == b;
• Jarrett Billingsley (5/8) Jul 23 2008 That's one solution for one case. It does not hold up in the general ca...

davidl <davidl 126.com> writes:

consider:

version(v1)
{
   enum e
   {
      a,
      b
   }
}
else
{
   enum e
   {
      a,
      b,
      c
   }
}

If my enum contains a lot of stuffs.... D version becomes a disaster.


consider:
version(v1)
{
   if ( cond1() && cond2() && a==b)
   {
      //blah
   }
}
else
{
   if ( cond1() && cond2() && a!=b)
   {
      //blah
   }
}

in C you can do it

if ( cond1() && cond2() &&
#ifdef v1
      a==b
#else
      a!=b
#endif
    )
{
    // blah
}


Any good and practical way to solve these?




-- 
使用 Opera 革命性的电子邮件客户程序: http://www.opera.com/mail/

downs <default_357-line yahoo.de> writes:

davidl wrote:
 consider:
 
 version(v1)
 {
   enum e
   {
      a,
      b
   }
 }
 else
 {
   enum e
   {
      a,
      b,
      c
   }
 }
 
 If my enum contains a lot of stuffs.... D version becomes a disaster.
 
 
 consider:
 version(v1)
 {
   if ( cond1() && cond2() && a==b)
   {
      //blah
   }
 }
 else
 {
   if ( cond1() && cond2() && a!=b)
   {
      //blah
   }
 }
 
 in C you can do it
 
 if ( cond1() && cond2() &&
 #ifdef v1
      a==b
 #else
      a!=b
 #endif
    )
 {
    // blah
 }
 
 
 Any good and practical way to solve these?
 
 
 
 

version(v1) auto cond = a == b;
else auto cond = a != b;

What's the problem again?

"Jarrett Billingsley" <kb3ctd2 yahoo.com> writes:

"downs" <default_357-line yahoo.de> wrote in message 
news:g67bqp$20ma$2 digitalmars.com...

 version(v1) auto cond = a == b;
 else auto cond = a != b;

 What's the problem again?

That's one solution for one case.  It does not hold up in the general case. 
How about the enum example?  How about differing declarations in general? 
How about trying to make class invariants work in both D1 and D2?