Re: [Cs-dev] Coding Standards again
Date | 2012-05-07 17:10 |
From | john ffitch |
Subject | Re: [Cs-dev] Coding Standards again |
I have never found static analysis much of a gain. Way too many incorrect reports. Sometimes gives clues but then so does reading the code. Andres posted an analysis from google some time back, and it was very unhelpful. Looked good until one looked at the code ==John ffitch ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
Date | 2012-05-07 17:16 |
From | Andres Cabrera |
Subject | Re: [Cs-dev] Coding Standards again |
Hi, That one was done using clang. Cheers, Andrés On Mon, May 7, 2012 at 5:10 PM, john ffitch |
Date | 2012-05-08 02:17 |
From | Erik de Castro Lopo |
Subject | Re: [Cs-dev] Coding Standards again |
john ffitch wrote: > I have never found static analysis much of a gain. Way too many > incorrect reports. John, I respctfully disagree. The quality of the output of a static analysis report is highky dependant on the quality (or probably more correctly the kind) of code that goes in. Obviously for legacy code like CSound these tools throw up huge numbers of spurious warnings, often about C idioms and techniques that were common 20 years ago but which have fallen out of favour.. However, for modern code that was developed with modern compilers (and with all the compiler's warning flags enabled and all warnings fixed) there are far fewer static analysis warnings. I am also 100% certain that it is possible to write a large complicated piece of software in C or C++, that is compiler warning free and static analysis warning free. I'm also sure this code will still have bugs, but it will have fewer bugs than the same code developed without compiler warnings and static analysis. The Haskell people have an interesting view on this. They say that in order for the compiler to reject more incorrect programs it has to reject some programs that are correct, but for which it can't prove they are correct. Cheers, Erik -- ---------------------------------------------------------------------- Erik de Castro Lopo http://www.mega-nerd.com/ ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |