Yep, see my reply to 5302. I should have been clearer in that the only reason I'd put this in 4x is to save someone else the hassle of applying all the merges. And then only if there aren't objections or perhaps if someone volunteers to ride herd on it.
So I'll hold off for now. If someone wants to step up I could forward a patch that contains all the merges for both the original and 5488 if that would help. Erick On Tue, Dec 3, 2013 at 10:07 AM, Robert Muir <[email protected]> wrote: > Can we hold off on committing things to the stable branch until their > tests consistently pass? > > Thanks, > Robert > > On Tue, Dec 3, 2013 at 8:40 AM, Erick Erickson <[email protected]> > wrote: > > All: > > > > I have until Thursday to finish things up before I'll be unavailable for > a > > while, I'd like to leave as little as possible hanging. So. > > > > 1> As far as I know, the whole modify-config-file thing is wrapped up for > > now, anyone disagree? The current state is > > a> it's gone from 4x > > b> in trunk it's disabled by default and in its own handler > > c> I created a blocker bug for trunk so we don't lose track of this. I > > didn't assign it to myself because I won't be able to work on it for the > > foreseeable future. See SOLR-5523. > > > > If we choose to pull it out if trunk, it's quite easy, just nuke about > three > > files and do a grep on the EditFileRequestHandler to pick up the > > solrconfig.xml files it's mentioned in and remove those references. > > > > 2> Analytics component (SOLR-5302, SOLR-5488). It's on trunk but not 4x, > and > > it pretty regularly has a test failure. Steven and Houston know about it, > > but haven't got a fix for it yet. The failure isn't reproducible. I'm > sure > > they'd appreciate any coaching people would like to have on tracking this > > kind of test failure down. > > > > There are quite a number of patches, most of them having to do with tests > > that I've collected into a 4x code-base on my machine. > > > > I'd like to commit this to 4x later today. I know it'll add a bit more > noise > > to the tests. Is this objectionable? This is a new component, so I'm > pretty > > sure the failure is not a regression. > > > > I'd like to get this done today so I have a couple of days to deal with > > fallout if there is any before I'm unavailable. > > > > Erick > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
