On Sunday, July 19, 2015 at 2:39:48 PM UTC-4, jean-pierre.muench wrote:
>
>  I think our aim should really be to pass all builds with all sorts of 
> -Wall (MSVC, Intel C++, GCC) however I think this may get interesting on at 
> least Windows. But each warning has a reason and as a security library we 
> should be notified about anything that may cause unusual behavior. So I 
> think enabling Wall on GCC (+Clang) is a good direction for our library. If 
> we "killed" all warnings on Clang and GCC and I get back in involvement we 
> may also want to increase warning levels on Windows and Intel, although I 
> think standard libraries also cause some of the warnings...
>
Yeah, I'm a big fan of warnings. The compiler creates a rich context for 
analysis, and all projects should be leveraging them.

Code under my purview in real life has to clean compile under -Wall -Wextra 
-Wconversion. Its a security gate, so I have the authority to reject 
anything that does not meet standards.

But its painful to elevate those warnings in a mature project that did not 
start with them. In this regard, there's nothing special about Crypto++. 
OpenSSL, GnuPG, and others suffer the same.

Related, we'll be doing more with Coverity in the future, too. They provide 
their service for free of FOSS projects. https://scan.coverity.com/projects.

Jeff

-- 
-- 
You received this message because you are subscribed to the "Crypto++ Users" 
Google Group.
To unsubscribe, send an email to [email protected].
More information about Crypto++ and this group is available at 
http://www.cryptopp.com.
--- 
You received this message because you are subscribed to the Google Groups 
"Crypto++ Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to