It looks good to me. It works! - libreport compiled - all affected binaries tested (except reporter-rhsupport and reporter- kerneloops)
Regards, Jakub On Thursday 12 of July 2012 18:24:55 Denys Vlasenko wrote: > On 07/12/2012 11:15 AM, Denys Vlasenko wrote: > > On 07/11/2012 05:54 PM, Jiri Moskovcak wrote: > >> On 07/11/2012 05:31 PM, Denys Vlasenko wrote: > >>> On 07/11/2012 05:26 PM, Jiri Moskovcak wrote: > >>>>> Let's add add_text_to_problem_data() to > >>>>> libreport.1/src/include/problem_data.h > >>>>> instead of creating yet another abstraction layer. > >>>>> Abstraction layers are not zero price. > >>>> > >>>> - I don't mind the few microseconds or few bytes it will cost I really > >>>> prefer the easy to use API over that... > >>> > >>> I'm not so much concerned about overhead but by having > >>> another set of functions which do the same thing. > >>> I am an idiot. I get confused by this stuff. > >> > >> - ok, so what's the consensus here? > >> > >> - problem_data_save() definitely belongs to ABRT because it reads the > >> abrt.conf file> > > Yes. > > > >> - the rest is more or less already in libreport, > >> but I would like to see it exported thru some abrt header (which I think > >> already is)> > > #include <libreport/problem_data.h> > > > >> and also would like to rename it to problem_data_* > > > > Ok. I will prepare a patch. > > > >> - so should we rename the functions in libreport and keep just that one > >> in ABRT?> > > I think yes, this would be the best solution. > > Please review this and the following patch. > > It only renames functions in libreport, nothing else. > (Second patch does the same for arbt code).
