[Nix-dev] Better support for channels in nixos
Tomasz Czyż
tomasz.czyz at gmail.com
Thu Sep 1 22:36:07 CEST 2016
2016-09-01 20:23 GMT+01:00 Matthew O'Gorman <mog at rldn.net>:
> Domen Kožar <domen at dev.si> writes:
>
> > We can (and do) backport new services to stable releases.
>
> Not trying to argue, but the example I gave still stands? The service
> was not back ported.
> >
> > For "Another example would be I am developing a service that I want to
> test
> > on my system" I don't really understand what you're trying to say.
> >
> The other example, say I ported an application to nixos, like huginn,
> and I want to test it on my system easily.
>
you can do this with nix-build, you don't need all channel to install one
package.
>
> A third example would be a package that moves ridiculously quickly,
> like gitlab, stable is on 8.5, unstable is on 8.10, the current
> release is 8.11.
You can create your package set on top of nixpkgs and you can upgrade
versions by yourself and still have all the packages from the original
nixpkgs, no problem with that.
my-packages = (import <nixpkgs> {}) // {upgraded-gitlab = mkDerivation
......}
or probably better to use packageOverrides
>
> I think it would be easier to maintain a system if I could have point
> to multiple places for channels, and then in the install specifically
> describe which provider I want to use.
>
if I understand this correctly, whole point of nix is to reduce source and
have as few source of truth as possible (IMHO).
>
> --
> Matthew O'Gorman
> BM-NBUmia4p88Jny5bZbRGcMt64SWWp5WVv
> TorChat: 5w3dtk7nhkkijcpc
> Site: https://b.rldn.net
> Xim/Email: mog at rldn.net
> mog at rldn:~$ fortune wisdom -s|cowsay -f /dev/null
> ___________________________________
> < Life exists for no known purpose. >
> -----------------------------------
>
> _______________________________________________
> nix-dev mailing list
> nix-dev at lists.science.uu.nl
> http://lists.science.uu.nl/mailman/listinfo/nix-dev
>
>
--
Tomasz Czyż
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.science.uu.nl/pipermail/nix-dev/attachments/20160901/f37596db/attachment.html>
More information about the nix-dev
mailing list