[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH] mailmanclient
From: |
Leo Famulari |
Subject: |
Re: [PATCH] mailmanclient |
Date: |
Wed, 28 Sep 2016 13:48:50 -0400 |
User-agent: |
Mutt/1.7.0 (2016-08-17) |
On Wed, Sep 28, 2016 at 05:25:52PM +0000, ng0 wrote:
> Leo Famulari <address@hidden> writes:
>
> > [ Unknown signature status ]
> > On Wed, Sep 28, 2016 at 05:02:56PM +0000, ng0 wrote:
> >> Subject: [PATCH] gnu: Add python-mailmanclient.
> >>
> >> * gnu/packages/mail.scm (python-mailmanclient): New variable.
> >
> >> +(define-public python2-mailmanclient
> >> + (let ((base (package-with-python2
> >> + (strip-python2-variant python-mailmanclient))))
> >> + (package (inherit base)
> >> + (native-inputs
> >> + `(("python2-setuptools" ,python2-setuptools))))))
> >
> > This variant will not be used unless there is a python2-variant
> > properties field in python-mailman, like this:
> >
> > (properties `((python2-variant . ,(delay python2-mailmanclient))))
> >
> > You can see the recent commit "gnu: Add python-q. (e8c9b01090)" for an
> > example in context.
>
> Is this necessary for all python2 package variants? I rarely see this.
For any use of the python2-variant system, yes. I replied to a lot of
your submitted packages a couple days ago to point out that they needed
this change.
If you are just doing a plain package-with-python2 transformation, then
it's not required. Python2-variant is used when package-with-python2 is
not enough, because there are extra inputs, arguments, etc.