Re: How can this odd warning be fixed?
From: Hendrik Schober (SpamTrap_at_gmx.de)
Date: 02/13/04
- Next message: Anders Eriksson: "CString::AllocSysString() needs to be free'd?"
- Previous message: John Carson: "Re: newbie needs quick help"
- In reply to: Ted Byers: "How can this odd warning be fixed?"
- Messages sorted by: [ date ] [ thread ]
Date: Fri, 13 Feb 2004 11:28:03 +0100
Ted Byers <r.ted.byers@sympatico.ca> wrote:
Hi Ted!
> Whenever I use VC++6 with the STL, I get the warning C4786 [...]
Yeah, that's an annoying one.
> Understand, most of the development I have done over the past few years has
> been with Borland's prodcut, along with some experimentation with Gnu's gcc:
> both of which seem to be more ANSI compliant than VC++6 (which isn't
> surprising given the age of VC++6 and when the standard was established).
> But, I have Visual Studio 6, and need to get a feel for it, so when my
> budget allows, I can upgrade to the latest version of Visual Studio with a
> minimum of pain.
If you are aiming for less painregarding the
C++ std lib, buy the much improved std lib
version for VC6 from Dinkumware. It eases
using the std lib as much as possible in VC6.
If you are aiming for the minimum of pain,
upgrade to VC7.1 right now. I have wasted a
lot of time on making templates compile with
VC6, only to find that in the end it having
it getting compiled doesn't mean it does what
it is supposed to do.
> But, the code that generates this warning would not generate such a warning
> with the other tools I use. With the compiler truncating an identifier, is
> there not a risk of name collisions?
It would, if it wasn't only in the debugging
info.
(But there is indeed another warning hinting
that the symbol got cut down. And I found VC6
sometimes calling into the wrong instance of
a template function...)
> Is there anything I can do to prevent
> VC++ from truncating names? If not, what is the best practice to work
> around this issue, especially to prevent VC++ from hiding other problems in
> all the verbiage produced by this warning?
We have a couple of setup headers that are
in all of our code. One deals with the
compiler and disables this warning for VC.
> Can I assume this problem no
> longer exists in the latest version of VC++?
That would be good, wouldn't it? However,
you cannot. We still have to turn it off
in VC7.1!
> Thanks,
>
> Ted
Schobi
-- SpamTrap@gmx.de is never read I'm Schobi at suespammers dot org "Sometimes compilers are so much more reasonable than people." Scott Meyers
- Next message: Anders Eriksson: "CString::AllocSysString() needs to be free'd?"
- Previous message: John Carson: "Re: newbie needs quick help"
- In reply to: Ted Byers: "How can this odd warning be fixed?"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
