On Mon, 6 Aug 2001 18:45, Stefan Bodewig wrote: > > Instead we should accelerate release cycle of ant and test things > > that way. > > I don't see that happen, sorry. To do that, it would take more people > committed to the 1.x codebase.
I have a script setup for the avalon project(s) so that "releasing" is as simple as * run a script * come back in ten minutes and edit jakarta-site and do mail out * bump version in build.xml The script builds distribution, signs the distribution, constructs HEADER.html + README.html and uploads it to appropriate directory on jakarta, tag CVS module, and creates a basic announcement message. It took me a full 30 minutes to do a release and 20 of those I spent making toast ;) So as long as the CVS is kept in nice clean state and experimentation is done in proposals dir (or as a branch) then it would be simple to do. If it would make it more likely to make releases more often then I will see what I can do to convert it to ant tasks or make it portable (currently it is specific to bash + gpg + ssh with keys). Cheers, Pete *-----------------------------------------------------* | "Faced with the choice between changing one's mind, | | and proving that there is no need to do so - almost | | everyone gets busy on the proof." | | - John Kenneth Galbraith | *-----------------------------------------------------*
