TypeCompose Patch

[Conal Elliott]

Thanks for the follow-up.

Indeed the bug is fixed in 6.10 and following.

About the linear map comment, I was referring to the same bug (now fixed),
not a different one.  I described the linear maps implementation in two
recent blog posts (http://conal.net/blog).

  - Conal

On Mon, Oct 13, 2008 at 5:31 AM, Simon Peyton-Jones
<simonpj at microsoft.com>wrote:

>  I believe it's fixed in HEAD and 6.10; I recall Manuel saying he'd done
> it, the same day I told him.  (I thought I'd created a Trac ticket for it,
> but I don't seem to have done so.)   (Worth noting for future: if you create
> a ticket you can follow the history more easily.)
>
>
>
> As to the "my linear map implementation, but not when it tickles a ghc bug
> for me and for readers", do we have a Trac ticket for that?  I don't know
> what the bug is, nor whether it's fixed.
>
>
>
> Simon
>
>
>
> *From:* conal.elliott at gmail.com [mailto:conal.elliott at gmail.com] *On
> Behalf Of *Conal Elliott
> *Sent:* 12 October 2008 06:36
> *To:* Simon Peyton-Jones
> *Cc:* glasgow-haskell-bugs at haskell.org
> *Subject:* Re: TypeCompose Patch
>
>
>
> What's the status of this bug?  I'd like to know when it's safe to upgrade
> to a new ghc.  Also, I want to blog about my linear map implementation, but
> not when it tickles a ghc bug for me and for readers.
>
> Thanks,  - Conal
>
> 2008/9/25 Simon Peyton-Jones <simonpj at microsoft.com>
>
> Conal, Marc
>
> This is definitely a bug in the new equality-solving mechanism.  I can
> reproduce it readily.  I'll discuss it with Manuel tomorrow and get back to
> you. Thanks for finding it!
>
> Simon
>
>
> | -----Original Message-----
> | From: glasgow-haskell-bugs-bounces at haskell.org [mailto:
> glasgow-haskell-bugs-
> | bounces at haskell.org] On Behalf Of Marc Weber
> | Sent: 25 September 2008 01:50
> | To: glasgow-haskell-bugs at haskell.org
> | Subject: Re: TypeCompose Patch
> |
> | On Wed, Sep 24, 2008 at 02:55:01PM -0700, Conal Elliott wrote:
> | >    Eep!  That code (which uses type classes) fails to compile under
> 6.8.3
> | but
> | >    does compile under 6.9.  Looks like 6.10 broke something.
> | >
> | >    Can anyone else check whether vector-space build or breaks under
> 6.10?
> | >    The repo is
> | >
> | >        [1]http://code.haskell.org/vector-space/
> |
> | src/Data/LinearMap.hs|37 col 21 error|
> | ||     Couldn't match expected type `t' against inferred type `Basis u'
> | ||     In the second argument of `(.)', namely `basisValue'
> | ||     In the first argument of `trie', namely `(f . basisValue)'
> | ||     In the expression: trie (f . basisValue)
> |
> | Hi Conal, I'm getting the error above (current ghc from darcs)
> | If you want I can give you an ssh account where this ghc version is
> | installed within the next days.
> |
> | I also had to add a -fglasgow-exts to a MemoTrie file..
> | (Not sure which is the correct language extension which was missing)
> |
> | Sincerly
> | Marc Weber
> | _______________________________________________
> | Glasgow-haskell-bugs mailing list
> | Glasgow-haskell-bugs at haskell.org
> | http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
>
>
> _______________________________________________
> Glasgow-haskell-bugs mailing list
> Glasgow-haskell-bugs at haskell.org
> http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.haskell.org/pipermail/glasgow-haskell-bugs/attachments/20081026/8a12e9a2/attachment-0001.htm