SBT branch in Ivy repo

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

SBT branch in Ivy repo

Gintautas Grigelionis
I asked sbt developers [1] whether they would like to use the same Git
repo. The consensus seems to be that sbt would like to have their own
branch (or two). Would it be acceptable to graft sbt branch(es) to Ivy repo?

Gintas

[1] https://github.com/sbt/ivy/issues/28
Reply | Threaded
Open this post in threaded view
|

Re: SBT branch in Ivy repo

Stefan Bodewig
On 2018-03-25, Gintautas Grigelionis wrote:

> I asked sbt developers [1] whether they would like to use the same Git
> repo. The consensus seems to be that sbt would like to have their own
> branch (or two). Would it be acceptable to graft sbt branch(es) to Ivy repo?

I'm not sure I understand the story. This is what seems to have
happened:

* the sbt developers created a fork of Ivy 2.3

* either their changes looked to them unacceptable for Ivy, or they
  didn't want to contribute their changes to the upstream or they just
  didn't care. In either case they decided to maintain a long term fork
  over asking the Ivy devs to change Ivy so they could use an unpatched
  version.

Why would the Ivy developers want to maintain this fork for sbt?

Stefan

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: SBT branch in Ivy repo

Gintautas Grigelionis
IMHO sbt fork happened because Ivy was not using Git then.
GitHub shows clearly that the fork was created off Ivy repo clone created
by EasyAnt (which is dormant now).
So my idea is to graft their branch onto Ivy repo whereupon EasyAnt-derived
clones can be mothballed.
What happens next is an open question, sbt maintainers could either fork
the original Ivy repo or become committers.
Are there some political issues that I did not notice?

Gintas

2018-03-29 14:49 GMT+02:00 Stefan Bodewig <[hidden email]>:

> On 2018-03-25, Gintautas Grigelionis wrote:
>
> > I asked sbt developers [1] whether they would like to use the same Git
> > repo. The consensus seems to be that sbt would like to have their own
> > branch (or two). Would it be acceptable to graft sbt branch(es) to Ivy
> repo?
>
> I'm not sure I understand the story. This is what seems to have
> happened:
>
> * the sbt developers created a fork of Ivy 2.3
>
> * either their changes looked to them unacceptable for Ivy, or they
>   didn't want to contribute their changes to the upstream or they just
>   didn't care. In either case they decided to maintain a long term fork
>   over asking the Ivy devs to change Ivy so they could use an unpatched
>   version.
>
> Why would the Ivy developers want to maintain this fork for sbt?
>
> Stefan
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
>
Reply | Threaded
Open this post in threaded view
|

Re: SBT branch in Ivy repo

Stefan Bodewig
On 2018-03-29, Gintautas Grigelionis wrote:

> IMHO sbt fork happened because Ivy was not using Git then.

I don't understand. Either they have changed Ivy or could have used
binaries. If they had to change Ivy the question is what did they have
to change.

> So my idea is to graft their branch onto Ivy repo whereupon EasyAnt-derived
> clones can be mothballed.

I don't really care where the fork came from. I'd prefer a situation
where sbt didn't need to use a fork at all - or if they've got reasons
to maintain a fork, then have them do that themselves.

> What happens next is an open question, sbt maintainers could either fork
> the original Ivy repo or become committers.

For "fork the original repo" they can do so now without you or anybody
else maintaining extra branches IMHO.

You become a committer to Ivy be being involved in the Ivy community.

> Are there some political issues that I did not notice?

None that I was aware of.

There might be legal issues as changes made by the sbt maintainers have
not been contributed to Ivy explicitly and thus can not be simply copied
into an Apache repo. But this is secondary IMHO. If there is a good
reason to maintain the sbt fork inside of Ivy, then this can be
solved. So far I don't see what the good reasons might be. I may just be
slow, though.

Stefan

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]