[ciphershed] Re: Git logistics

  • From: Rocki Hack <rocki.hack@xxxxxxxxx>
  • To: ciphershed@xxxxxxxxxxxxx
  • Date: Tue, 24 Jun 2014 19:31:06 +0200

I think that the history should be included and I prefer the clean way by
resetting the repo once again.
We need trust and not conspiracy theories why we merged the history back-in.
It's easier to audit code if the commit history is linear.


2014-06-24 7:23 GMT+02:00 Stephen R Guglielmo <srguglielmo@xxxxxxxxx>:

> On Mon, Jun 23, 2014 at 9:43 AM, Jason Pyeron <jpyeron@xxxxxxxx> wrote:
> >> -----Original Message-----
> >> From: Stephen R Guglielmo
> >> Sent: Monday, June 23, 2014 9:38
> >>
> >> On Fri, Jun 20, 2014 at 12:13 PM, Jason Pyeron
> >> <jpyeron@xxxxxxxx> wrote:
> >> > I am trying to get a straw poll out there on how many
> >> people have started
> >> > hacking against any of the commits in the git repo?
> >> >
> >> > If you have any pending edits can you say which commit its against?
> >> >
> >> > It is looking like importing the history and linking it is
> >> a little hacky, and I
> >> > am strongly opposed to rebooting the repository.
> >> > By hacky, it requires a double merge, but if anyone else is
> >> branching or working
> >> > off of a commit prior to the double merge, the history gets
> >> lost. To proceed
> >> > each uncommited change set will have to be stashed and then
> >> pulled to work of
> >> > the latest.
> >> >
> >> > -Jason
> >>
> >> My fork is against the current version published in GitHub, a03e565.I
> >> plan on adding an upstream remote and keeping my fork synced with the
> >> upstream...
> >>
> >> I'm not sure what you mean about the double merge? No history
> >> should be lost?
> >
> > See: http://marc.info/?l=git&m=140312217717968&w=2
> >
> > Nothing is lost, it just is awkward getting there.
>
> Ah, I see what you were referring to. I'm honestly not sure. It seems
> pretty hacky. Should we bother with getting the history in the repo?
> The ability to `git blame` would be awesome (for licensing purposes),
> but I too am against rebooting the repo.
>
>

Other related posts: