Giorgio,
IMO main hindrance for the Axis2/C development is not having good test
framework and test code. I am guilty as anyone else for not having tests. I
think, developing a test framework and tests (both Unit and Integration
tests) is vital for longevity of the project. Most of current issues could
have been addressed if we have a good test coverage. Just my 2 cents.
thanks,
Dinesh.
--
Be happy
On Sun, Jul 12, 2015 at 3:41 PM, Giorgio Zoppi <giorgio.zo...@gmail.com>
wrote:
> Dear all,
> ai have done my homeworks this week about axis2c this has led to the
> conclusion that 1.7 is not in anyway a viable way to start the axis effort
> at Apache.
> Too many leaks and too many bugged features.
> The Jira has more or less 174 bugs most of them are leaks in somewhat
> conditions. The path that we might follow is branch pre 1.6.0 revision
> for leading to a 1.8,0 and start eliminating things for reducing the code
> base until we high quality.
> Support just two O.S: Linux and Windows.
> + I will branch to github a 1.6.0 release and change the system to cmake,
> eliminating:
> - we dont need the internal parser
> - until being stable we dont need apmq
> we can think to have a odd number as development version and a pair number
> as stable version. The idea is to have something stable just reducing code
> base.
> Best Regards,
> Giorgio.
>
>
>
