Re: A suggestion regarding stabel vs. experimental on the extension level

On Thu, Nov 16, 2000 at 11:27:08AM +0100, Hartmut Holzgraefe wrote: > this is somehow related to the 4.1 thread and the discussions about > which extensions should be in CVS and which should not ... > > we could maybe get rid of part of the problem if we'd create another > subdir 'ext-experimental' (or something like that), move ext/ext_skel, > ext/sekeleton/ and every extension from ext that is not ready for > prime time (like ext/dav) from ext/ to ext-experimental/ > and do *not* include ext-experimental/ in RCs and Releases I'm not sure I like the separate subdirectory system (from a cvs administrative point of view, although I recognize how trivial it is to repo-copy the files to ext/ later on). However, I like the idea that "experimental" extensions be differentiated from "stable" modules, especially for release building. I think it might be better to control what gets included by a Makefile knob. For example, defining 'EXPERIMENTAL' in etc/{foo}/Makefile would mark {foo} as experimental. The 'make dist' target could them be taught about the 'EXPERIMENTAL' tag. This method also has the advantage of it being able to be used programmatically where applicable ('#ifdef EXPERIMENTAL', for example). -- Jon Parise (jon@csh.rit.edu) . Rochester Inst. of Technology http://www.csh.rit.edu/~jon/ : Computer Science House Member