On Fri, May 7, 2010 at 3:19 PM, Sebastian Dransfeld <[email protected]> wrote: > Michael Jennings wrote: >> On Friday, 07 May 2010, at 20:02:15 (+0200), >> Sebastian Dransfeld wrote: >> >>> Enlightenment SVN wrote: >>>> Log: >>>> revert broken dbus return handling, readd stringshares >>> Please! Don't mess with code you don't understand! >> >> I'm pretty sure he made those changes with Gustavo's input and >> insights, so if you all want to hash it out, fine, but don't get >> personal about it. >> >> Explain why you did what you did and why his code is wrong, just like >> Gustavo tried to explain why you were wrong. > > I just reverted the code to the state it was, which was working just fine. > > It didn't mess with the internal memory of the dbus message, as the > message would be free'd after the callback was completed. So the user of > the callback was responsible to make their own copy of the string. > > The code which was added was incomplete, as it didn't clear up all the > stringshare's it added. > > If there was a need for stringshare it could have been added to the > return call of one function, a simple switch from strdup to stringshare. > > I'm not trying to be an ass here, but I don't like it when code is added > without understanding of how the original code worked and in addition > adds bugs.
Situation already clarified at IRC, point is that both were correct in PART of the solution, that means that part of both solutions are incorrect. The old way, that Sebastian said, was wrong but just happened to work due our usage. Summary: - e_dbus part (e_hal/e_ukit) MUST get references, for the reasons I said before (as discomfitor); - e17 and its modules MUST get references if they use e_hal/e_ukit as before (and as englebass); Sebastian said he will do it. Sorry for the inconvenience due half explanations based on half reviewed code. BR, -- Gustavo Sverzut Barbieri http://profusion.mobi embedded systems -------------------------------------- MSN: [email protected] Skype: gsbarbieri Mobile: +55 (19) 9225-2202 ------------------------------------------------------------------------------ _______________________________________________ enlightenment-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
