On Fri, Nov 30, 2018 at 09:30:36PM +0100, Alexander Zubkov wrote: > Hello, > > Yes, sounds reasonable. But I worry configure command could be > overabused with all that features, like: > configue soft keepstate keepdebug keepsomethingelse :)
I would prefer one common keep flag for all cases where a running state deviates from config value due to a user changed the running state (e.g. enable/disable, debug, ...). > May be some means to provide default configure variant could be > useful... But lets leave it for futher generations for now. That makes sense. > How this configure flag could be implemented? I see there are > reconfigure type variable in code - sould it be extended to bitmask > somehow? That is reasonable. > Or separate variable to keep this flag? I could try to > provide some patch, but I'm not sure it would be easy for me to change > parser's part in that case. It looks more complex than for adding new > protocol option. You could try it and if you hit a problem then you could just post a partial patch. -- Elen sila lumenn' omentielvo Ondrej 'Santiago' Zajicek (email: [email protected]) OpenPGP encrypted e-mails preferred (KeyID 0x11DEADC3, wwwkeys.pgp.net) "To err is human -- to blame it on a computer is even more so."
