> On Feb 8, 2015, at 6:53 PM, Alexander Hansen <alexanderk.han...@gmail.com> > wrote: > > On 2/7/15 6:50 AM, Daniel Johnson wrote: >> >>> On Feb 6, 2015, at 12:15 PM, Alexander Hansen <alexanderk.han...@gmail.com> >>> wrote: >>> >>> Yeah, I hear you. That’d be why we’ll have people go through pull requests >>> initially, because those show all of the changes that will be made. >>> >>>> On Feb 6, 2015, at 9:14 AM, Hisashi T Fujinaka <ht...@twofifty.com> wrote: >>>> >>>> Ideally, it would keep me from doing stupid, stupid things but so far >>>> I've found that git only allows me to commit lots of things I never >>>> meant to commit. >>>> >>>> On Fri, 6 Feb 2015, Alexander Hansen wrote: >>>> >>>>> Right. Nothing?s set in stone yet. :-) >>>>> >>>>> Ideally, it?d be good if people could update their own packages without >>>>> going through PRs, but I don?t know of a way to enforce that other than >>>>> the honor system. >>>>> >>>>>> On Feb 6, 2015, at 8:41 AM, TheSin <the...@southofheaven.org> wrote: >>>>>> >>>>>> I have mixed emotions on this one. But I don?t have a better solution >>>>>> it does give us some good extras this way, but it also puts more on the >>>>>> plates of those that can merge, and it adds a massive slow down and more >>>>>> steps to update packages, it also will make a TON of forks that some >>>>>> uses might point fink to (again plus and minus here). >>>>>> >>>>>> I?t basically like a request-tracker for EVERYTHING, and we know how >>>>>> annoying and slow the request tracker is already with less entries. >>>>>> Just want to make sure it was fully thought out, again mixed emotions so >>>>>> I?m not against it just pointing things out. >>>>>> --- >>>>>> TS >>>>>> http://www.southofheaven.org/ >>>>>> Life begins and ends with chaos, live between the chaos! >>>>>> >>>>>> On Feb 6, 2015, at 9:32 AM, Alexander Hansen >>>>>> <alexanderk.han...@gmail.com> wrote: >>>>>> >>>>>>> Yeah, at least initially. People who already have the permissions to >>>>>>> modify the fink/* family on github can make immediate modifications and >>>>>>> approve the PRs. But just because they can make immediate changes >>>>>>> doesn?t mean that they necessarily should. :-) >>>>>>> >>>>>>>> On Feb 6, 2015, at 8:27 AM, TheSin <the...@southofheaven.org> wrote: >>>>>>>> >>>>>>>> so we all fork it, make PRs and then it goes to the official repo? >>>>>>>> Just want to clarify. >>>>>>>> --- >>>>>>>> TS >>>>>>>> http://www.southofheaven.org/ >>>>>>>> Life begins and ends with chaos, live between the chaos! >>>>>>>> >>>>>>>> On Feb 6, 2015, at 9:23 AM, Alexander Hansen >>>>>>>> <alexanderk.han...@gmail.com> wrote: >>>>>>>> >>>>>>>>> 3) After the switchover, I propose that we make all changes to the >>>>>>>>> distribution via pull requests on Github. I don?t happen to recall >>>>>>>>> whether the selfupdate-git code in the development branch of that >>>>>>>>> name treats the local checkout as a fork to make this easy to do in >>>>>>>>> place. >>>>>>>> >> >> I just committed some changes to selfupdate-git. I made it clearer that >> developers can use either ssh or https for read/write access and provided >> links to info to set it up (before it said you had to use ssh) and made the >> updating procedure safer so that changes in the user's working directory >> won't be overwritten by selfupdate. It's still safer to do your work in a >> branch and rebase it onto master just before pushing but this should help >> prevent some merging issues. >> >> I haven't been able to update my git mirror of the package files because cvs >> keeps crapping out. :( I CAN do an rsync backup, which I'm still doing just >> in case cvs fails altogether and that can be used to generate a new git repo >> using cvs-fast-import. There are a number of other projects that are >> experiencing cvs failures. There's a ticket at >> https://sourceforge.net/p/forge/site-support/9618/ which I'm following. >> >> Daniel >> >> >> > Thanks, Daniel! I'd love to get us moved to git. > > Actually, I don't care about git per se, but I'd like to be on a DVCS where > we don't have to worry quite as much about a particular host provider as a > central point of failure. > > This current SF.net outage is really annoying. ;-)
Agreed. I did just get my mirroring script to work so my mirror is back up to date. And yes, while I prefer git I'd be happy with Mercurial too. Anything is better than cvs. We also have Subversion available since Github provides a bridge and if someone really wants Mercurial they could use hg-git. Daniel ------------------------------------------------------------------------------ Dive into the World of Parallel Programming. The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net/ _______________________________________________ Fink-devel mailing list Fink-devel@lists.sourceforge.net List archive: http://news.gmane.org/gmane.os.apple.fink.devel Subscription management: https://lists.sourceforge.net/lists/listinfo/fink-devel
