From: William Djaja Tjokroaminata <billtj@...>
Date: 2002-12-03T06:30:22+09:00
Subject: Re: Ruby ++, the one element and generators

Daniel Carrera <dcarrera@math.umd.edu> wrote:
> True.  The whole reason why '++' was invented in C was that it would
> reduce the number of CPU instructions:

> In C, 'a = a + 1' does this:

>    1 -> Store "1" in a memmory location.
>    + -> Add 1 to 'a' and store the result in another location.
>    = -> take the contents from this location and put them at the
>         location of 'a'.

> But 'a += 1' does this:
>    1  -> Store "1" in a memmory location.
>    += -> Add it to 'a' and put it directly in the location of 'a',
>          without the intermediate step.

> Whereas 'a++' does this:

>    ++ -> Increement 'a' by 1 and put the result directly in the
>          location of 'a'.  Without the intermediate location.


> Surely, this reason doesn't apply to Ruby. :-)

> Daniel.

Hmmm..., I don't think the explanation above is strict.  I think the
current C compilers are free to translate 'a = a + 1' to 'a++'.  Because
of this, we don't have to worry whether to write 'a = a + 1' or 'a++'.  A
good C compiler will translate either of these into an assembly/machine
equivalent of 'a++'.  You can ask/search comp.lang.c if you like.  C is
one of the languages where a lot of things are not strictly defined.

Regards,

Bill