Indeed, this is currently just to get an idea about whether it is technically possible at all and how 'expensive' it is.
I'd like to look at the size of the git repo. There might be binaries, jars etc which got committed over time. Those might need to get filtered out to streamline the repo etc. We also might do a full import and then git-filter it in a second step to create a baseline repo for further development. We also need to check whether the author and contributor flags are properly moved over by the import. We don't like to loose any IP provenance... Etc, etc. To be honest: I was a bit worried when the github import blew up. That was actually the main reason why I started a local import. LieGrue, strub On Monday, 19 September 2016, 15:08, Geertjan Wielenga <geertjan.wiele...@googlemail.com> wrote: > > >On 19.09.2016 11:42 PM, Mark Struberg wrote: >If I succeed then I'll push it to github. > > >From the NetBeans side, we assume this is basically an experiment at this >stage - we will look at the results and this is a very interesting experiment. > > >There could be different approaches. For example, when doing a conversion of >this scale, we think it could be appropriate to consider splitting the >repository into several smaller repositories. It is not completely clear how >to split it, we could discuss different approaches, but we could start with a >split per cluster and then doing adjustments as needed. > > >In all cases, we would strongly suggest to keep history, but we assume you are >keeping it. > > >Thanks, > > >Geertjan > > > > > > > >On Sun, Sep 18, 2016 at 11:42 PM, Mark Struberg <strub...@yahoo.de.invalid> >wrote: > >quick update. >> >> >>hg cloned it locally and now batch importing over here. >>The hg clone took me 3 hours and the import is now running for 4 hours. >> >>Maybe that is the problem why github failed. >> >>I'm now at 89000/303000 in the import step. >> >>If I succeed then I'll push it to github. >> >>LieGrue, >>strub >> >> >> >> >> >> >>> On Sunday, 18 September 2016, 18:24, Mark Struberg >>> <strub...@yahoo.de.INVALID> wrote: >>> > btw, I tried to import the netbeans-core repo from hg into github and it >>> > failed. >>> Now trying to manually import it... >>> >>> Did anybody else do that? >>> >>> >>> LieGrue, >>> strub >>> >>> >>> >>> >>>> On Sunday, 18 September 2016, 18:22, Dennis E. Hamilton >>> <dennis.hamil...@acm.org> wrote: >>>> > +1 >>>> >>>> >>>>> -----Original Message----- >>>>> From: Greg Stein [mailto:gst...@gmail.com] >>>>> Sent: Sunday, September 18, 2016 00:46 >>>>> To: general@incubator.apache.org >>>>> Subject: Re: [DISCUSS] Apache NetBeans Incubator Proposal >>>>> >>>>> On Sat, Sep 17, 2016 at 9:54 AM, Geertjan Wielenga < >>>>> geertjan.wielenga@googlemail. com> wrote: >>>>> >>>>> > Hi all, >>>>> > >>>>> > Can we be specific about what info is needed, or what further >>> details >>>>> > specifically, before going into a vote for acceptance of the >>> proposal? >>>>> My >>>>> > concern is that each question we answer is answered by further >>>>> questions to >>>>> > answer. Maybe we could do a phone conference with the NetBeans >>>>> > infrastructure side together with the Apache infrastructure side. >>>>> Maybe we >>>>> > can work through the infrastructure challenges during incubation. >>>>> > >>>>> >>>>> Generally speaking, Apache would prefer to slow things down and use a >>>>> mailing list, rather than to have a phone conference. The email is >>>>> documented for everybody to review, to participate, and to record for >>>>> future examination. A phone conference probably wouldn't resolve >>> many >>>>> questions/concerns anyway, simply because much of that comes from >>>>> considered thought. A phone call is "THINK NOW. RESPOND. OOPS. >>> MISSED >>>>> YOUR >>>>> CHANCE." ... Mailing lists give people time to think. >>>>> >>>>> There is no rush, no dates, no deadlines at the ASF. It may take >>> longer >>>>> via >>>>> mailing lists, but it means that the larger community can be involved, >>>>> can >>>>> review, and can be archived. >>>>> >>>>> If one question turns into three ... well, that is deliberation. As >>>>> David >>>>> noted else-thread, we rarely get such a large, well-established >>>>> community >>>>> arriving at the Incubator. That necessitates a bit more inquiry than >>>>> most >>>>> other entrants receive. Layers of the onion get peeled, and new >>>>> questions >>>>> arrive. More layers unpeeled ... >>>>> >>>>> And to point to the elephant in the room: I bet there are people >>>>> concerned >>>>> given the recent misadventures of AOO [and Oracle's donations of >>> these >>>>> two >>>>> projects]. Personally, I think it is hogwash, and don't believe >>> any >>>>> concern >>>>> applies here, as the communities and the userbase are very different. >>>>> BUT, >>>>> temporally, there is a conflation of the donations of these two >>>>> projects. I >>>>> suspect that will cause a few people to slow down and ask more >>>>> questions. >>>>> >>>>> Cheers, >>>>> -g >>>> >>>> >>>> ------------------------------ ------------------------------ --------- >>>> To unsubscribe, e-mail: general-unsubscribe@incubator. apache.org >>>> For additional commands, e-mail: general-help@incubator.apache. org >>> >>>> >>> >>> ------------------------------ ------------------------------ --------- >>> To unsubscribe, e-mail: general-unsubscribe@incubator. apache.org >>> For additional commands, e-mail: general-help@incubator.apache. org >>> >> >>------------------------------ ------------------------------ --------- >>To unsubscribe, e-mail: general-unsubscribe@incubator. apache.org >>For additional commands, e-mail: general-help@incubator.apache. org >> >> > > > --------------------------------------------------------------------- To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org For additional commands, e-mail: general-h...@incubator.apache.org
