On Tuesday, 14 February 2012, Kohsuke Kawaguchi <[email protected]> wrote: > > Yes. I think the main challenge is to have them take changes we need, like making various classes serializable, improving error chaining, and so on. >
One hopes that they understand these issues. > I'm all for it, but someone needs to drive this. If they are going to push all the changes we need to the upstream, I suppose we could just piggy-back on it? If not, I had some luck getting changes picked up by svnkit, though they seem to suffer from NIH when it comes to issues... I suspect test cases which fail showing the problem would have a better reception than actual patches :eek: (or perhaps they have changed their attitude recently) -Stephen > > > On 02/12/2012 02:44 AM, Christoph Kutzinski wrote: >> >> Hi, >> >> I've just seen that the Hudson guys are working on integrating SVNKit >> 1.3.7 and at the same time replacing the patched SVNKit with a vanilla one. >> >> http://java.net/projects/hudson/lists/dev/archive/2012-02/message/38 >> >> I think it would be a good idea to get inspired by this, as maintaining >> a patched SVNKit is - IMHO - not a good idea in the long run. >> >> >> cheers >> Christoph > > > -- > Kohsuke Kawaguchi | CloudBees, Inc. | http://cloudbees.com/ > Try Nectar, our professional version of Jenkins >
