Re: Dumb Windows linking question.

From: TLOlczyk (olczyk2002_at_yahoo.com)
Date: 10/06/03


Date: Mon, 06 Oct 2003 06:19:45 -0500

On Sun, 05 Oct 2003 13:42:19 GMT, alfps@start.no (Alf P. Steinbach)
wrote:

>On Sun, 05 Oct 2003 08:03:13 -0500, TLOlczyk <olczyk2002@yahoo.com> wrote:
>
>>On Sun, 05 Oct 2003 12:13:36 GMT, alfps@start.no (Alf P. Steinbach)
>>wrote:
>>
>>>On Sun, 05 Oct 2003 06:36:15 -0500, TLOlczyk <olczyk2002@yahoo.com> wrote:
>>>
>>>>This actually started out as a project to write down how to do FFI
>>>>in various languages,
>>>
>>>I don't know what FFI is.
>>>
>>>
>>Ok. You're saying "I don't understand your problem, but I am going to
>>try to solve it anyway." Do you see how *stupid* that is?
>
>Here you forgot to quote your question.
>
No. I *cited* the *very first* part of the question. The rest I did
not feel was important enough, since the raison d'etre for the rest
was the first part. ( OK. Let me simplify it for you small IQ mind,
any answer to the rest that does not lead to insight to the first
part is worthless, and given the context wrong. )

As for the rest, I presume people can read the original post.

>
>
>>>Include the header file that declares GetLogicalDrives.
>>>
>>I haven't tried it in other compielrs but PLT tells me
>>that nclude is a bad number.
>
>Here you forgot to include the "i" at the start of "include" (and
>perhaps more, such as the "#" sign before that).
As I all ready responded, I tried this because I believed
that it would be better to actually try it first ( rather than just
say it wouldn't work ). I see that I was mistaken and that
no mater how you respond to an idiot, he will still act like an idiot.

Yes I did include the #i. In fact, it is very stupid to ask, as
skipping them will not make nclude look like a bad number.

>
>
>
>>>>in both cases command line
>>>>too ), but it won't link due to an unresolved symbol error ( the
>>>>symbol being _GetLogicalDrives of course ).
>>>>I've done the above. I've replaced extern with extern "C". I've added
>>>>kernel32.dll to the compile.
>>>
>>>Most compilers and linkers won't allow you to simply add a DLL.
>>>
>>>
>>I've done it often. ( Allthough now that I think about it you may have
>>to go through a .lib file. Not a problem as only one kernel32.lib
>>comes with the compilers, and as I already said I included it. )
>
>Here you forgot what you've done or not.
>
No. You're just too dumb to understand what I've done. I've included
any combination libs and dlls. Especially since they are provided in
different mannes ( the dlls included witht the OS, the libs with the
compiler ).



Relevant Pages

  • Re: Canada - hated by terrorists, er... freedom loving fighters
    ... Roper - You idiot, if Rissia hadn't been our convenient partner in WW@ ... You're still an idiot. ... the ability to function. ... Libs in America and Libs in Britain and Canada are two entirely ...
    (alt.guitar.amps)
  • Re: Canada - hated by terrorists, er... freedom loving fighters
    ... Roper - You idiot, if Rissia hadn't been our convenient partner in WW@ ... You're still an idiot. ... the ability to function. ... Libs in America and Libs in Britain and Canada are two entirely ...
    (alt.guitar.amps)
  • Re: How to deploy private Assemblies with VC2005 SP1
    ... Holger was right....recompiling all the third party libs and dlls solved the ... After that i can ship the CRT and the manifest with my app in the ... Of course i use the new installed SP1 ...
    (microsoft.public.dotnet.languages.vc)
  • Re: passing using "string" type in DLL Interface
    ... You need two DLLs anyway if you use the typical debug/release scenario. ... those use different heaps and you must not allocate ... The problem with the compilers isn't too big. ... VC6-VC8, C#, VB and whatnot. ...
    (microsoft.public.vc.stl)
  • Re: "Process terminated by SIGINT"?
    ... how do I use the LIB to access the DLL's functions, ... For DLLs that are supposed to be used by multiple compilers, it is good practice to use OS/2 system calling convention because every OS/2 compiler needs to support that. ...
    (comp.os.os2.programmer.misc)