On 09/10/2012 12:52 PM, Mike
Milinkovich wrote:
Mickael,
I
don't think anyone is against hosting a Sonar instance at
Eclipse, but it is not the #1 priority at the moment.
I'm a bit frustrated: this sentence sounds like "code quality is not
a priority". Just like a few monthes ago someone decided
"performance tests on platform are not #1 priority"...
What are #1, #2... #10 priorities at the moment ? Which is the rank
of having tools to make quality better?
Even if it's low priority, has anyone evaluated the cost of
installing Sonar? If it's only about a few minutes to get started,
there is probably no need for this task to be considered as #1
priority before someone work on it.
That
could change if it turns out that its faster to get better
metrics using a tool like Sonar than porting the older
Eclipse platform testing frameworks to CBI.
I'll work on it and hopefully will come with some concrete stuff to
show that code analysis is worth the effort.
In the meantime, here are the generalities about long-term goodness
of static analysis:
http://www.eclipsecon.org/2012/sessions/how-profit-static-analysis
Thank
you for your kind offer to set up reports for the Platform.
If we get Sonar up and running, I suspect that we would
definitely like your help in doing that.
The question now is more "what can we do to get Foundation set up a
minimal Sonar install?" The request on Bugzilla shows that I'm not
the only one to interest in code quality and think about it as a
good way to make better software, it's a pity there has not been
progress on that question.
|