Re: Visual Studios Compile question

[Kris Katterjohn <katterjohn () gmail com>]

kx wrote:
> I agree. #define _CRT_SECURE_NO_DEPRECATE 1 is definitely the way to
> go, and when I removed those other two lines, I had no other warnings.
> But  #define _CRT_SECURE_NO_WARNINGS 1 is creating a new warning.
>
> I just installed VS 2005 Express fresh today and SP1 for VS 2005 and I
> have the Microsoft Platform SDK for Windows Server 2003 R2 (new XP
> laptop).
>
> I just hate using a pragma to turn off all deprecation warnings, and
> while _CRT_SECURE_NO_WARNINGS is the new way to turn off the warnings:
> http://connect.microsoft.com/VisualStudio/feedback/ViewFeedback.aspx?FeedbackID=101187
>
>
> ... it seems broke.
>
> I am wondering if you had the warnings because it wasn't SP1?
>
> If anyone has VS Express with SP1, it would be helpful to know if you
> get the CRT warnings by removing the pragma and
> _CRT_SECURE_NO_WARNINGS 1
>
> Cheers,
>  kx

Yeah, if anybody can test the different #defines and the #pragma, that'd
be awesome.  With different VS versions, too.

I really don't care which one(s) work, as long as all the unwanted
warnings go away :) I'd be happy to revert the change if we get
something that works all around.

Like I said, I unfortunately can't test anything out right now. I don't
remember if I had SP1 or not (probably no).

Below is an IM between me and Fyodor talking about it before I committed
it.  It might give you a little info about what we saw and tried; maybe
it'll give you an idea:



(19:48:03) [me]: Hey, it's Kris. I made a patch to add a couple lines to
nbase_winconfig.h. It adds an additional #define and a #pragma to
disable warnings about function deprecation (stupid MS). There's a
#define allready in there that's supposed to work, the #define I'm
putting in is supposed to work, and the #pragma is. Unfortunately, the
#define that's in there and the one I'm adding doesn't seem to work for
me. The #pragma does though. I'm sure the #define's work for other
versions or something (maybe not the Express VC++?). Is it cool to apply
to nbase?
(19:48:53) [fyodor]: are these deprecated functions like strcpy(), or
other things?
(19:49:10) [me]: Yeah, the stuff like that where they're trying to get
you to use strcpy_s() instead.
(19:49:39) [fyodor]: do they complain about strncpy() ?
(19:49:55) [fyodor]: and what are the pragma and define names?
(19:50:03) [me]: I'm not sure about that one, there are a ton of
warning. I'm pretty sure snprintf() is mentioned too.
(19:50:08) [me]: h/o, I'll get them
(19:50:37) [me]: #define _CRT_SECURE_NO_WARNINGS 1
#pragma warning(disable: 4996)
(19:51:37) [fyodor]: thanks, googling
(19:52:29) [fyodor]: what I'm looking for is a
(19:52:33) [fyodor]: list of functions affected
(19:52:47) [fyodor]: it complained about snprintf , with the 'n' ?
(19:52:59) [me]: I'm pretty sure. I can run and go check if you want
(19:53:07) [fyodor]:
http://msdn2.microsoft.com/en-us/library/wd3wzwts(VS.80).aspx
(19:53:30) [fyodor]: yeah, it looks like you are right.  If it was just
(19:53:41) [fyodor]: strcpy(), maybe we could just change
(19:53:52) [fyodor]: those to Strncpy, but since it is a huge list of
functions
(19:54:05) [fyodor]: , which in fact include strncpy, then
(19:54:17) [fyodor]: I agree with you that we should just add the
pragma.  Stupid MS!
(19:55:14) [me]: yeah, there are really tons of warnings when I build
Nmap. I'd say a third of the output is probably warnings.
(19:55:31) [fyodor]: Cool, well if you check in the pragma/define
change, that sounds great.



Thanks,
Kris Katterjohn

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://SecLists.Org