[Nix-dev] unable to expose libraries internal to GHC?

[Corey O'Connor]

The NixOS module I use to configure a global haskell environment is:

* https://github.com/coreyoconnor/nix_configs/blob/master/haskell-dev.nix

I then use the normal cabal-install commands to do development. This has
been sufficient to pass the vty and yi tests. Some more tests are required
before production approval, but good so far!

Cheers,
Corey

-Corey O'Connor
coreyoconnor at gmail.com
http://corebotllc.com/


On Tue, Apr 15, 2014 at 11:15 PM, Raahul Kumar <raahul.kumar at gmail.com>wrote:

> Corey, would you mind posting your working vesion? If you can actually
> build GHC 7.8, I would be very interested!
>
> Aloha,
> RK.
>
>
> On Wed, Apr 16, 2014 at 3:07 AM, Corey O'Connor <coreyoconnor at gmail.com>wrote:
>
>> Thanks for the help Peter.
>>
>> I tried using ghcWithPackagesOld and, in this case, that worked fine.
>> Cool!
>>
>> This bug is unfortunate. I was pretty happy to see terminfo bundled with
>> GHC. This meant the hardest part of installing vty, installing terminfo,
>> would already be done in 7.8. Darn, no such luck.
>>
>> Cheers,
>> Corey
>>
>> -Corey O'Connor
>> coreyoconnor at gmail.com
>> http://corebotllc.com/
>>
>>
>> On Mon, Apr 14, 2014 at 5:25 AM, Peter Simons <simons at cryp.to> wrote:
>>
>>> Hi Corey
>>>
>>> GHC 7.8.x installs a copy of xhtml, but it doesn't make that library
>>> visible to users.
>>>
>>> This choice causes us serious trouble that we cannot easily resolve: we
>>> can't install your own copy of xhtml, because it collides with GHC's
>>> copy (as you have found out). Yet, we cannot access GHC's copy either,
>>> so we *need* to install our own version. Duh!
>>>
>>> I've opened a ticket [1] for this issue a while ago, but GHC HQ hasn't
>>> addressed it yet (probably because I didn't do a good job explaining why
>>> this is issue is significant).
>>>
>>> Anyway, you can use 'ghcWithPackagesOld' instead of 'ghcWithPackages' to
>>> create your environment. ghcWithPackagesOld simply ignores all
>>> collisions and chose a "random" version of the colliding paths to
>>> include in the derivation. This sucks from an aesthetics point of view,
>>> but it's probably going to work fine.
>>>
>>> I hope this helps,
>>> Peter
>>>
>>>
>>> [1] https://ghc.haskell.org/trac/ghc/ticket/8919
>>>
>>> _______________________________________________
>>> nix-dev mailing list
>>> nix-dev at lists.science.uu.nl
>>> http://lists.science.uu.nl/mailman/listinfo/nix-dev
>>>
>>
>>
>> _______________________________________________
>> nix-dev mailing list
>> nix-dev at lists.science.uu.nl
>> http://lists.science.uu.nl/mailman/listinfo/nix-dev
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.science.uu.nl/pipermail/nix-dev/attachments/20140416/756dff18/attachment.html