Re: [RFC] CVS directory structure

On 15 Nov 2003 at 16:13, Pierre-Alain Joye wrote: > On Sat, 15 Nov 2003 16:08:36 +0100 > Tobias Schlitt <tobias@schlitt.info> wrote: > > > On Sat, 2003-11-15 at 15:02, Xavier Noguer wrote: > > > > > I'm -1 on this. CVS is a tool for source control. Doing > > > development directly > > > on your CVS dir is a bad idea. > > > > That's exactly the point. CVS is initially made for developing > > inside it's checkout directories. That's just your local copy, which > > you will commit to the central repository. So copiing files away > > from your sandbox nd recopiing is double effort, without win. > > Developing here == keeping files history, not run them. Nothing more. > > pear install package.xml should be used, or a little bat to sync the > dev tree (your running script) and the cvs repo helps. You mean you also check out source from CVS into one dir, then install it, modify in that location and afterwards copy everything back to the CVS-dir to check it in? And how do you make sure you don't forget a file or something? For me, forgetting to include a file or so is one major point that all IDEs I know of solve by working in the CVS-checkout-dir directly. And currently I can't see any bad side about it. Stefan