unofficial mirror of guix-devel@gnu.org 
 help / color / mirror / code / Atom feed
* Lisp: Package cl-cffi-gtk
@ 2018-10-09 19:41 Pierre Neidhardt
  2018-10-10  5:51 ` Andy Patterson
  0 siblings, 1 reply; 5+ messages in thread
From: Pierre Neidhardt @ 2018-10-09 19:41 UTC (permalink / raw)
  To: guix-devel, Andy Patterson

[-- Attachment #1: Type: text/plain, Size: 350 bytes --]

Hi,

I'm trying to package cl-cffi-gtk, which should not be so hard, except
that according to

	https://github.com/Ferada/cl-cffi-gtk/blob/master/INSTALL

we need to push all .asd to asdf:*central-registry*.

This should work, but I wonder how to do this properly with the current
ASDF build system.

Andy?

--
Pierre Neidhardt
https://ambrevar.xyz/

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 487 bytes --]

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Lisp: Package cl-cffi-gtk
  2018-10-09 19:41 Lisp: Package cl-cffi-gtk Pierre Neidhardt
@ 2018-10-10  5:51 ` Andy Patterson
  2018-10-10 13:16   ` Pierre Neidhardt
  0 siblings, 1 reply; 5+ messages in thread
From: Andy Patterson @ 2018-10-10  5:51 UTC (permalink / raw)
  To: Pierre Neidhardt; +Cc: guix-devel

Hi Pierre,

On Tue, 09 Oct 2018 21:41:05 +0200
Pierre Neidhardt <mail@ambrevar.xyz> wrote:

> Hi,
> 
> I'm trying to package cl-cffi-gtk, which should not be so hard, except
> that according to
> 
> 	https://github.com/Ferada/cl-cffi-gtk/blob/master/INSTALL
> 
> we need to push all .asd to asdf:*central-registry*.

That file is just trying to be helpful by providing an easy way to set
things up and make them work.  It also says
> or the corresponding symlinks are created
which is effectively what we will do with the build system.

> 
> This should work, but I wonder how to do this properly with the
> current ASDF build system.

The project doesn't seem to be doing anything special; we just need to
make sure that those dependencies are created as packages first, since
the build system maps asd systems to packages one to one.  The slynk
package may be a helpful resource for seeing how to do that.

Hoping that helps,

--
Andy

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Lisp: Package cl-cffi-gtk
  2018-10-10  5:51 ` Andy Patterson
@ 2018-10-10 13:16   ` Pierre Neidhardt
  2018-10-11  7:30     ` Andy Patterson
  0 siblings, 1 reply; 5+ messages in thread
From: Pierre Neidhardt @ 2018-10-10 13:16 UTC (permalink / raw)
  To: Andy Patterson; +Cc: guix-devel

[-- Attachment #1: Type: text/plain, Size: 499 bytes --]

Alright, I've packaged it as you suggested.  See the wip-nextbrowser branch.
If was slightly simpler as for Sly.

That said, I wonder if this is not over-complicated.
Wouldn't it be easier to mimic Quicklisp there and automatically push all .asd
to the central registry?  Or make the central registry configurable, so that
packages can decide what goes in there?
That would replace all those "define"s with a single line I think.

Thoughts?

-- 
Pierre Neidhardt
https://ambrevar.xyz/

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 487 bytes --]

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Lisp: Package cl-cffi-gtk
  2018-10-10 13:16   ` Pierre Neidhardt
@ 2018-10-11  7:30     ` Andy Patterson
  2018-10-11  8:48       ` Pierre Neidhardt
  0 siblings, 1 reply; 5+ messages in thread
From: Andy Patterson @ 2018-10-11  7:30 UTC (permalink / raw)
  To: Pierre Neidhardt; +Cc: guix-devel

Hi,

On Wed, 10 Oct 2018 15:16:33 +0200
Pierre Neidhardt <mail@ambrevar.xyz> wrote:

> Alright, I've packaged it as you suggested.  See the wip-nextbrowser
> branch. If was slightly simpler as for Sly.
> 
> That said, I wonder if this is not over-complicated.
> Wouldn't it be easier to mimic Quicklisp there and automatically push
> all .asd to the central registry?  

We don't want to use the central registry at all as it's the older way
to tell asdf where systems live, and it's less flexible.  I understand
what you're getting at though.  I do think there are some advantages to
mapping systems to packages.  It allows the guix dependency graph to
exactly match the real dependency graph between systems, for example.
The #:asd-system-name and #:asd-file arguments could become complicated
if we're going to support multiple systems per package.

If you feel that it's worth it to do this, I could have a go at
implementing it so that we can see what it would look like (I might do
this anyway).  Of course, you're also free to do so as well, and I can
help if you need.

--
Andy

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Lisp: Package cl-cffi-gtk
  2018-10-11  7:30     ` Andy Patterson
@ 2018-10-11  8:48       ` Pierre Neidhardt
  0 siblings, 0 replies; 5+ messages in thread
From: Pierre Neidhardt @ 2018-10-11  8:48 UTC (permalink / raw)
  To: Andy Patterson; +Cc: guix-devel

[-- Attachment #1: Type: text/plain, Size: 348 bytes --]

Hi,

OK, I understand your points.

Even if we don't end up using this approach, it would be interesting to give it
a shot, at least to make sure that we are heading in the right direction.

That said, this can probably wait and I'd put the recursive importer higher on
the priority list :)

-- 
Pierre Neidhardt
https://ambrevar.xyz/

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 487 bytes --]

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2018-10-11  8:48 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-10-09 19:41 Lisp: Package cl-cffi-gtk Pierre Neidhardt
2018-10-10  5:51 ` Andy Patterson
2018-10-10 13:16   ` Pierre Neidhardt
2018-10-11  7:30     ` Andy Patterson
2018-10-11  8:48       ` Pierre Neidhardt

Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/guix.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).