home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!think.com!rpi!uwm.edu!ogicse!reed!bowman
- From: bowman@reed.edu (BoBolicious)
- Newsgroups: comp.sys.mac.oop.macapp3
- Subject: Re: C++ (was re: bedrock defection
- Message-ID: <1993Jan24.221920.13361@reed.edu>
- Date: 24 Jan 93 22:19:20 GMT
- Article-I.D.: reed.1993Jan24.221920.13361
- References: <727841301.8123714@AppleLink.Apple.COM>
- Organization: Reed College, Portland, OR
- Lines: 38
-
- In article <727841301.8123714@AppleLink.Apple.COM> ALANDAIL@AppleLink.Apple.COM (Personal Software, Alan Dail,PRT) writes:
- >I would tend to disagree. C and C++ both allow so many things to compile that
- >could easily be typos but cause the program to do the wrong thing (I do admit
- >that C++ is better than C). For example, anywhere you can use a ==, you can
- >also use = (and vice versa) and have the program compile fine (due to the
- >definition of the language), but get different results. Your program will not
- >only compile, but may even work fine in some instances. It doesn't even stand
- >out when you are looking over your code. This error would be cought in Pascal
- >(or even in Pascal '9x), but not in C or C++.
-
- MPW C++ will warn you if you use a "=" inside an if. The biggest problem
- I had switching from C to C++ was that I didn't pay enough attention to
- the warnings that CFront emits. My C code was so "warning-laden" that I
- tended to just ignore them, but CFront can be quite helpful. It could be
- better, though, I admit.
-
- >There are other many other problems like this that are part of the language and
- >not due to the compiler. This doesn't even address the other language problems
- >like having various features clash when used together or having the wrong
- >defaults when using features (why isn't virtual the default?). It's a shame
- >that Apple doesn't appear to be doing anything with Pascal '9x which appears to
- >give programmers the power of C++ without nearly as many pitfalls.
-
- Do you have any more examples? I've found C++ to be an ideal mix of flexible
- elegance & stiff type-checking. The biggest gotcha for me is usually a bad
- typecast, which if I understand correctly, Pascal allows as well. Properly
- written C++ code can pretty stable; it took many shots-to-the-foot, for
- instance, before I was fully convinced that initializing everything to nil
- and then checking everything before using it is truly The Way (tm).
-
- cheers,
- bobo In seeking the unattainable,
- bowman@reed.edu simplicity only gets in the way.
- Best personal ad ever:
- "Functionally dysfunctional bi-polar former pop-star geek with Valium habit
- seeks anorexic queen of darkness for co-dependency. Hobbies include: passing
- out, slurred speech, blurred vision and reliving past glory."
- _Willamette Week_, 1/7-13/93
-