On Sun, 3 Jun 2001, Sergey S. Khoroshavin wrote:
> As I said, it was UNchanged circle30bpl18. I haven't modified anything
> in it. Just compiled and run.
Where did you obtain it? I would suggest downloading a fresh copy from
http://www.circlemud.org/pub/CircleMUD/3.x/, setting it up in a separate
directory so you don't overwrite any work you may have done (or the bad
copy[1] you have), and doing a fresh compile from there. If the problem
persists, come back and we'll try to diagnose it.
As it stands, there's no code in that particular section that should be
sensitive to platform or compiler, so--barring exceptionally buggy
compilers[2]--the code should work exactly the same on every platform.
On top of that, I am personally aware of several Muds that are presently
running that were compiled with MSVC++ 5 and up, and I have not seen nor
heard of similar problems.
I don't know what more to tell you than that, sorry.
-dak
[1] Assuming that a "bad copy" is the cause of your problems.
[2] That's not a Microsoft knock: even they manage to be mostly ANSI C
compliant. It'd take an exceptionally buggy compiler, more so than
any I've seen recently. (As I have cited several times before, the
worst I have seen was several years ago with Solaris cc, which
generated bad code for "char ch = 0;" -- I never delved too deeply
into the problem with the generated code since so often with
professional programming it's solely about getting things to run.
The only thing I can think of is that it stupidly would cast the 0
to (char*) and then dereference that, turning the code into the
rather stupid "char ch = *((char*) 0);". Why it would do this is
probably the real stumper.)
--
+---------------------------------------------------------------+
| FAQ: http://qsilver.queensu.ca/~fletchra/Circle/list-faq.html |
| Archives: http://post.queensu.ca/listserv/wwwarch/circle.html |
+---------------------------------------------------------------+
This archive was generated by hypermail 2b30 : 12/05/01 PST