On 10/4/06, Tommy Pettersson <[EMAIL PROTECTED]> wrote:
It has been proposed (a long time ago) to change the way darcs
versions are numbered, and I've been thinking (for a long time :-)
about it.
The way I've been using is simply an extrapolation of how it was
done before me, supported by my interpretation of David's
expressed intentions at some point very long ago. That scheme is
<major>.<odd=experimental/even=stable>.<minor>
But now we have darcs-unstable, so there is no need for a
parallel "experimental stable" branch version, unless we'd start
tagging and releasing experimental versions from darcs-unstable.
At some point a new patch theory will enter the picture. It has
already been talked about as darcs 2, not 1.1.0. All this speaks
in favor of dropping the odd/even numbering scheme.
The proposed scheme is
<major>.<minor>.<micro>
Just an idea to throw in more options and create confusion and debates :)
I sort of like the way the lambdabot version number is done.
<major>p<# of patches>. Perhaps it's too ambiguous with some patches
going immediately into stable and some going into unstable then
migrating to stable? So maybe <micro> could be the number of patches
since <minor> was incremented? Just a thought. Seems like it might
be good to work in the # of patches but I'm not sure how.
Jason
_______________________________________________
darcs-devel mailing list
[email protected]
http://www.abridgegame.org/cgi-bin/mailman/listinfo/darcs-devel
