Danek Duvall writes: > On Fri, Apr 04, 2008 at 12:10:28AM -0400, Richard Lowe wrote: > > > Do we want to ship a default system-wide hgrc with a sensible config > > for merge (mergetools.hgrc, but slightly adjusted I would imagine)? > > > > I'm not sure I have much of an opinion either way, at this point, but > > it would seem defaulting to trying meld (when meld is in SFW) would be > > sensible. > > I would tend not to, if for no other reason than nothing's actually shipped > yet.
If "nothing" refers to meld, then that's a pretty good reason not to bother right now. > I also don't want to run afoul of the "no new config files in /etc" > crowd. And I don't want to create a new package. The "no new config files in /etc" rule doesn't quite exist. (If it does, please point me to it.) The real rule is that you have to integrate with SMF to the extent necessary to make system administration possible, and if you're doing something to control access, you need to have interfaces that can be audited. Furthermore, if you have something that's configurable, you should consider how to get something better than "vi" as your interface. But common sense and engineering rule here. If Mercurial has the ability to use a system-wide hgrc in a sensible way, and if supplying one will make the system easier to use, then *do it*. Please don't be afraid of doing the right thing. -- James Carlson, Solaris Networking <james.d.carlson at sun.com> Sun Microsystems / 35 Network Drive 71.232W Vox +1 781 442 2084 MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677
