What's it's purpose? I try to read the code, and as I understand it it either makes some options sticky or has no function at all. Under all circumstances, it's undocumented beyond the fact it exists.
Personally, I think it it should be removed. The main reason is that when there is problems, we compare the command line options used when invoking f-r. If there is a configuration file, the same options might give different results... I don't think the convenience of somewhat less typing is worth that confusion.
Thoughts?
Original report: https://github.com/timlau/FedoraReview/issues/15
I think it is a good idea to have a config files, so make it possible define stuff like workdir, mock settings, bz user etc. I don't like to have type every thing everytime :)
but some documentation would be nice
Perhaps, but for the moment I still think it should be removed. The reasons: - Odd semantics. Seems that some (all?) options are automatically saved, without user feedback, These are used next time, causing all sorts of troubles for user to prepared to this. - No documentation, making other problems worse,
Taken together, I think this boils down to removing the current config file code, planning for re-adding it with defined semantics, documentation and a new implementation with user feedback as required.
Removed config file in cda299a. There is still hooks in settings.py for re-implementing this.
Personally, I think most options are not really global but more per-project i. e.j the usecase is rather to make a new template with the same parameters as last time.
Login to comment on this ticket.