#19 Ability to execute a test in a sandbox

[Flid]

I try to make API Sanity Autotest to generate test in CUnit format. I faced a problem and want to asked your help.

Some test can hang (infinite loop), cause some exceptions (SEGV, FPE, BUS...) or to spoil the memory used by testing program. I think it's good idea to add a special option to xxx_run_tests functions. With this option testing program should fork and execute test in another process, wait for it to stop or stop it manually after some seconds.
Another way is to create a macros or function that can be used by test writer to execute any function in a sandbox. Not a whole testing function will be in a sandbox (in your sample: only printf, not testFPRINTF).
If this feature is really useful and can be implemented, report me and I'll do that. It's a way for me not to loose API Sanity Autotest abilities and generate high-quality tests. So I want to improve CUnit and make my work easier.

[Flid]

It's done here https://sourceforge.net/tracker/?func=detail&aid=3377087&group_id=32992&atid=407090

I need a feedback