:Matthew Dillon <dil...@apollo.backplane.com> wrote:
:>:> NEW
:>:>
:>:> #define btokup(addr) (&kmemusage[((caddr_t)(addr) - kmembase) >>
PAGE_SHIFT])
:>:
:>:The added parentheses don't make any difference, semantically. This
:>:change probably wouldn't meet the criteria spelled out in style(9):
:>
:> Then style(9) needs to be updated, because we have to add parenthesis
:> to be able to not get warnings with -Wall.
:
:I'll support that. The example given in style(9):
:
: a = b->c[0] + ~d == (e || f) || g && h ? i : j >> 1;
:
:should rate as an entry in the Obfuscated C competition rather than
:an example of maintainable code.
:
:style(9) should emphasize legibility and maintainability, rather than
:minimizing the number of extraneous (from the compiler's perspective)
:parenthesis. The code you're writing has to be maintained for many years
:- and the maintainers will not always have your in-depth expertise.
:The code also forms a `reference implementation' for someone who wants
:to do something similar.
:
:Peter
As far as parenthesis go, it's irrelevant because -Wall pretty much
covers the most common mistakes. If your code compiles without generating
a warning, your parenthesization is in good shape.
Braces and indentation and other purely visual effects are a different
matter.
-Matt
Matthew Dillon
<dil...@backplane.com>
To Unsubscribe: send mail to majord...@freebsd.org
with "unsubscribe freebsd-current" in the body of the message
