On 18.11.2012 21.40, Ytai Ben-Tsvi wrote:
Thanks for doing that! I'll check that this new release fixes issues I've been seeing as it should. What is the process of getting this out of "experimental" status?
Well, there is no very definite process, ideally I would hear from users on various platform some feedback weather it works or not. I've run the test suite on Mac OS X (Mountain Lion) and it passes the whole test suite, same on 32 bit Ubuntu. ATM I've not had the opportunity to try it out on Windows, no FreeBSD or Solaris and no 64 bit Linux. I've not really touched the Windows backend so it should be be good but of course the refactoring of how timeout and threshold are handled plus the use (or actually non use on Windows) of pipes to wake the read from come may have exposed some latent issue with Windows backend. I hope to hear from a user some feedback on running the code on Raspberry and if that turns out ok and I don't hear anything to the contrary I will try to find a Windows to test this on and if all goes well I likely change the status within a few weeks. br Kusti