Re: started applying Marcus' patch

On Thu, Jul 18, 2002 at 09:07:07AM -0500, Karl Fogel wrote:
>
> My mother raised me to always treat a bit test as faster than an
> arithmetic operation.
>
> Are you saying my mother is *wrong*???
>
> :-)

First people say "errno is just an int", now bit tests are the same
speed as arithmetic operations; next thing you know, people will be
saying that

        for (i = 0; i < n; i++) {
                /* stuff */
        }

is just as fast as

        i = n;
        while (i--) {
                /* stuff */
        }

> > C99 guarantees that char is _at least_ eight-bits. It can be 4711
> > bits without breaking the standard though. In practice, not much of a
> > problem though. Some PDPs have 9 bit chars, I think. But I don't
> > expect to find chars any other size than 8 anywhere where svn will
> > run.

Some old UNIVACs have 9-bit bytes. That means 36-bit words, btw. The
UNICOS I played with also had TWO different character sets: the 9-bit
one and the 6-bit one (6 chars per word!). The 6-bit one was used for
efficient terminal I/O.

> I think Brane's solution (to use the apr_isascii and isctrl tests) is
> best.

I concur.

--ben

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Jul 18 18:16:25 2002