[cvsnt] Mergepoint issues on 184.108.40.206 b2382
arthur.barrett at march-hare.com
Mon Jan 15 21:43:35 GMT 2007
> Arthur: if a patch of this nature is developed, and submitted
> in the cvsnt-dev list, are you (as CVSNT maintainer) prepared
> to accept it (subject to the usual inspections and revisions)?
Our policy on accepting patches is clear - if it doesn't break things then it's ok.
I suppose this one is a little ambiguous, but if it requires a specific option to "turn it on" then I think it's been clearly explained that "used correctly" the option is no more dangerous than all the other options.
A couple of additional notes:
* for a patch to be accepted it's gotta go on the cvsnt-dev list (some developers dont regularly read the support list so it's only fair that patches get seen by all the developers)
* if the patch adds options (eg: -i or -J) then it should include a patch to the cvsnt manual. I'd like to see someone summerise this thread into a short but useful explanation.
* I've already got a bunch of patches to apply to 2.5.04 which will come first - and are already badly delayed - but hopefully will happen in the next 10 days.
* EVS is the "logical" place for more "feature enhancements" - so I think it's correct that the patch process is a little more sedate in CVSNT than EVS, and equally correct that Tony Hoyle has taken the initiative and already applied this patch to EVS.
* the patches that have been applied to CVSNT in the past have had their proponents and their critics. Just because someone doesn't *like* a feature (including me) should never mean that the patch does not get accepted (as long as there is a way of avoiding the use of that feature if it is not wanted).
Finally any (non global) option on the client can be "defaulted" on the server using the CVSROOT/cvsrc file:
More information about the cvsnt