bug#48700: guix import fails with unexpected token

[zimoun]

Hi,

On Sun, 30 May 2021 at 18:38, "Philip Munksgaard" <philip@munksgaard.me> wrote:
> As far as I can tell, this is actually an instance of
> https://issues.guix.gnu.org/48701. You can close this issue in favor of that
> one.

>From my understanding, no.  I propose to unmerge.  WDYT?

The bug#48700 is a duplicate of bug#44115 (recursive importers).  For
instance, a first attempt to fix it is proposed by patch#45984; a new
version should come soon. :-)

But bug#48701 is not related, instead it is related to the Cabal
implemented by Guix.


For instance, this works:

--8<---------------cut here---------------start------------->8---
for p in bytestring-to-vector bmp directory-tree language-c-quote 
mainland-pretty neat-interpolation pcg-random srcloc; \
do
  guix import hackage $p -r ;
done

guix import hackage futhark
--8<---------------cut here---------------end--------------->8---

i.e., the option recursive with ’futhark’ goes to ’recursive-import’
from (guix import utils) and the ’node’ is not always well-formed when
’lookup-node’ as explained in #44115 and #45984.


<http://issues.guix.gnu.org/issue/44115>
<http://issues.guix.gnu.org/issue/45984>

All the best,
simon