[mythtv] C++ style

[Dan Conti]

On Wed, 11 Feb 2004, Daniel Thor Kristjansson wrote:
>
> I just saw Ken Thompson yesterday, he invented the ++/-- on the PDP-7 in
> B. The PDP-7 did not have the PDP-11 auto-increment and auto-decrement
> address modes, the PDP-11 also did not exist yet. pcc may have
> implemented x++ as a post-increment, but there were many compilers that
> didn't do this. Only ++x was defined well until the 80's when ANSI
> standardization had to deal with the common assumption that x++ was a
> post-increment. I have used some of these annoying non-ansi compilers
> working with embedded systems, it ain't fun. If I hadn't lost my copy
> of the K&R 1st edition I'd look it up.

Funny, i checked my copy of K&R 1st edition last night and it specifically
describes i++ as being always post increment. It also does pointer
operations like

*p++ = 0;

Which i doubt they would recommend if postincrement wasn't clearly
defined. It is possible that there were broken C compilers that did not
properly implement postincrement, but from K&R on the definition is pretty
clear.

-Dan

>
> Not that this matters, with ANSI C++ or C89 post-increment is only ill
> advised for classes where you might accidentally get an expensive class
> copy with a compiler such as gcc.
>
> -- Daniel
>   << When truth is outlawed; only outlaws will tell the truth. >> - RLiegh
>
> On Thu, 12 Feb 2004, Simon Kenyon wrote:
>
> ]On Wednesday 11 February 2004 20:03, Daniel Thor Kristjansson wrote:
> ]> Classes are the real reason to prefer pre-increment over post-increment,
> ]> with integers the compiler will figure it out. There is also a backward
> ]> compatibility reason for it. In old school C ++i is defined as a
> ]> pre-increment and i++ is defined as "whatever is fastest" so there is
> ]> no actual post-increment in C. Many compilers including gcc support it
> ]> as a post-increment because that's less surprising, but you might not
> ]> want to depend on that.
> ]in any-school C i++ is post-increment
> ]has been for 20 years or more
> ]these operators derive from the pre and post increment and decrement
> ]instructions in the pdp-11
> ]--
> ]simon
> ]_______________________________________________
> ]mythtv-dev mailing list
> ]mythtv-dev at mythtv.org
> ]http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
> ]
>