Experimental Physics and
Industrial Control System

Kay-Uwe Kasemir <[email protected]> · Thu, 29 Sep 2005 11:06:58 -0400

On Sep 29, 2005, at 10:13 , Andrew Johnson wrote:

I am not providing cast operators, and to some extent I agree about

the difficulty reading code that uses them. However the assignment

operator= is an integral part of C++ that the compiler will create

for you (usually wrongly) if you don't explicitly tell it not to.

.. right, which makes it more of a C--

class StringReader {
...
    // Individual byte access
    virtual char at(size_t pos) const = 0;

Yes, that looks good to me.

At one point in time, actually the Oak Ridge EPICS meeting,
we were gathered at Big Ed's Pizza, a place that I have not
visited ever since, but not because of any bad feelings,
there simply never was an opportunity for it.
Anyway, we discussed some EPICS 'OSI' ideas. One issue
was signed vs. unsigned for index variables and sizes.
I was asked what I used at the time in the archiver code,
and mentioned size_t, and the result was grave concern
because who knows how big size_t is, if it's signed or not, ...
So is it now OK to use size_t?
I assume you must still account for the unlikely possibility
that it's signed, so you always check
    if (xx < 0  ||  xx >= length)
and then you get the warning about x always being positive
on most compilers?
Well, it's only 11:00 and now I get this urge for oil dripping
pepperoni pizza.

-Kay

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System