Re: feature branch bumps

+1 I like it. Thanks to both of your for working through this. And thanks to Otto for all the heavy lifting. I'm willing to do whatever is needed to help the process. On Mon, Sep 25, 2017 at 8:52 PM, Matt Foley wrote: > Hi all, > Otto and I had an off-line discussion about

Re: feature branch bumps

Hi all, Otto and I had an off-line discussion about this, and we think we have a constructive suggestion that will allow chunking the feature branch to some extent, which will of course make it easier to review. Otto is willing to make a series of PRs, each of which must be reviewed and

Re: feature branch bumps

My thought was that in answering things in the wiki, it would build out the ‘guide’ there. But I was just taking a stab at that. I am open to whatever the group thinks would work best. On September 21, 2017 at 14:47:10, Nick Allen (n...@nickallen.org) wrote: > What I would like to do, and

Re: feature branch bumps

> What I would like to do, and what it seems to me that we need to do ( again I apologize if I am wrong ), is to catch up on the confluence and where things currently stand, and then move the discussion on from there. Perfect, thanks. I think you are completely right. I will review that in

Re: feature branch bumps

Nick, thanks for taking the time to reply, I have no doubts about your motivations and intent. Hopefully we get through this point by point stuff, which is always difficult on email. https://cwiki.apache.org/confluence/display/METRON/Metron+Extension+System+and+Parser+Extensions This is my

Re: feature branch bumps

Omit On September 20, 2017 at 19:20:27, Otto Fowler (ottobackwa...@gmail.com) wrote: > With the bundles, archetype, plug-in and parser moves gone 777 will be > much smaller though. > > Also I did not mean to admit Mike’s doc reviews, sorry. > > On September 20, 2017 at 18:04:20, Otto Fowler

Re: feature branch bumps

With the bundles, archetype, plug-in and parser moves gone 777 will be much smaller though. Also I did not mean to admit Mike’s doc reviews, sorry. On September 20, 2017 at 18:04:20, Otto Fowler (ottobackwa...@gmail.com) wrote: > I am not taking any of this as starting to do a flame war :) And

Re: feature branch bumps

I am not taking any of this as starting to do a flame war :) And your concern is about review not merging right? If it is all reviewed and accepted, why would 6 merges be preferable? That doesn’t make sense. Breaking it up only makes sense with regards to review, so that is how I’ll take it.

Re: feature branch bumps

I was just responding to your statement "Well, if there is an alternative merge strategy, I’m all ears." I understand that you disagree, but this is an alternative strategy. I think this approach is feasible and I outlined how here [1]. The criticism of this approach is the amount of effort and

Re: feature branch bumps

“Bite off small chunks” is what I keep hearing. I have no idea how to do that from an already integrated and working branch. Do you mean I… - create patch files of whole directories and do a pr per directory, but the build doesn’t work? On September 20, 2017 at 15:07:56, Nick Allen

Re: feature branch bumps

But wait, I thought we had established that this was such a fundamental change that it was hard to chunk it out and keep master working. Jon On Wed, Sep 20, 2017 at 3:08 PM Nick Allen wrote: > > Otto: Well, if there is an alternative merge strategy, I’m all ears. > > Yes,

Re: feature branch bumps

> Otto: Well, if there is an alternative merge strategy, I’m all ears. Yes, the alternative strategy is what I mentioned. :) Copied in below. > Nick: Are you going to bite-off small chunks of the feature branch and introduce PRs against master? On Wed, Sep 20, 2017 at 12:02 PM Otto Fowler

Re: feature branch bumps

Well, if there is an alternative merge strategy, I’m all ears. As it stands, the only complete code / conceptual review has been the bundles-lib by mattf. So there is still a great deal of review activity to do. Also conceptually there is ( might not be complete ) * the discuss thread topic from

Re: feature branch bumps

So, were the functionality is not FB related, I have been doing PR’s against master ( such as hdfs service ability to set permissions ). I don’t think we have talked about the end game PR from feature to master, I am don’t know how we would do multiple PRs to bring it down once ‘accepted’. I

Re: feature branch bumps

So it sounds like these PRs do move us closer to bringing the two branches together. But I think I am missing your high-level approach though. How are we going to get all of the functionality from the feature branch into master? Are you going to bite-off small chunks of the feature branch and

Re: feature branch bumps

Right now, these two PR’s are stacked, the UI depends on the BundleSystem changes. I would rather bring them down to feature before I do master integration, than integrate master into feature and then bring it out to the stacked PR’s. Simple as that. On September 20, 2017 at 08:35:09, Nick

Re: feature branch bumps

Hi Otto - What is the plan for bringing the feature branch and master together? Do these PRs move us closer to bringing the two branches together? Thanks On Wed, Sep 20, 2017 at 8:19 AM Otto Fowler wrote: > Can I get a bump on