Send Beginners mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners
or, via email, send a message with subject or body 'help' to
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Beginners digest..."
Today's Topics:
1. Re: Matrix and types (Frederic Cogny)
2. Re: Matrix and types (mike h)
3. Re: Matrix and types (Frederic Cogny)
----------------------------------------------------------------------
Message: 1
Date: Thu, 14 Mar 2019 14:12:46 +0100
From: Frederic Cogny <[email protected]>
To: The Haskell-Beginners Mailing List - Discussion of primarily
beginner-level topics related to Haskell <[email protected]>
Subject: Re: [Haskell-beginners] Matrix and types
Message-ID:
<CAGSugssd0pPEERe7+3gy_MeHhvpRt=vdp_v1oxqwyxekq87...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
The (experimental) Static module of hmatrix seems (I've used the packaged
but not that module) to do exactly that:
http://hackage.haskell.org/package/hmatrix-0.19.0.0/docs/Numeric-LinearAlgebra-Static.html
On Thu, Mar 14, 2019, 12:37 PM Francesco Ariis <[email protected]> wrote:
> Hello Mike,
>
> On Thu, Mar 14, 2019 at 11:10:06AM +0000, mike h wrote:
> > Multiplication of two matrices is only defined when the the number of
> columns in the first matrix
> > equals the number of rows in the second matrix. i.e. c1 == r2
> >
> > So when writing the multiplication function I can check that c1 == r2
> and do something.
> > However what I really want to do, if possible, is to have the compiler
> catch the error.
>
> Type-level literals [1] or any kind of similar trickery should help you
> with having matrices checked at compile-time.
>
> [1]
> https://downloads.haskell.org/~ghc/7.10.1/docs/html/users_guide/type-level-literals.html
> _______________________________________________
> Beginners mailing list
> [email protected]
> http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners
>
--
Frederic Cogny
+33 7 83 12 61 69
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.haskell.org/pipermail/beginners/attachments/20190314/26a09d2d/attachment-0001.html>
------------------------------
Message: 2
Date: Thu, 14 Mar 2019 15:19:42 +0000
From: mike h <[email protected]>
To: [email protected], The Haskell-Beginners Mailing List - Discussion of
primarily beginner-level topics related to Haskell
<[email protected]>
Subject: Re: [Haskell-beginners] Matrix and types
Message-ID: <[email protected]>
Content-Type: text/plain; charset="utf-8"
Hi,
Thanks for the pointers. So I’ve got
data M (n :: Nat) a = M [a] deriving Show
t2 :: M 2 Int
t2 = M [1,2]
t3 :: M 3 Int
t3 = M [1,2,3]
fx :: Num a => M n a -> M n a -> M n a
fx (M xs) (M ys) = M (zipWith (+) xs ys)
and having
g = fx t2 t3
won’t compile. Which is what I want.
However…
t2 :: M 2 Int
t2 = M [1,2]
is ‘hardwired’ to 2 and clearly I could make t2 return a list of any length.
So what I then tried to look at was a general function that would take a list
of Int and create the M type using the length of the supplied list.
In other words if I supply a list, xs, of length n then I wan’t M n xs
Like this
createIntM xs = (M xs) :: M (length xs) Int
which compile and has type
λ-> :t createIntM
createIntM :: [Int] -> M (length xs) Int
and all Ms created using createIntM have the same type irrespective of the
length of the supplied list.
What’s the type jiggery I need or is this not the right way to go?
Thanks
Mike
> On 14 Mar 2019, at 13:12, Frederic Cogny <[email protected]> wrote:
>
> The (experimental) Static module of hmatrix seems (I've used the packaged but
> not that module) to do exactly that:
> http://hackage.haskell.org/package/hmatrix-0.19.0.0/docs/Numeric-LinearAlgebra-Static.html
>
> <http://hackage.haskell.org/package/hmatrix-0.19.0.0/docs/Numeric-LinearAlgebra-Static.html>
>
>
>
> On Thu, Mar 14, 2019, 12:37 PM Francesco Ariis <[email protected]
> <mailto:[email protected]>> wrote:
> Hello Mike,
>
> On Thu, Mar 14, 2019 at 11:10:06AM +0000, mike h wrote:
> > Multiplication of two matrices is only defined when the the number of
> > columns in the first matrix
> > equals the number of rows in the second matrix. i.e. c1 == r2
> >
> > So when writing the multiplication function I can check that c1 == r2 and
> > do something.
> > However what I really want to do, if possible, is to have the compiler
> > catch the error.
>
> Type-level literals [1] or any kind of similar trickery should help you
> with having matrices checked at compile-time.
>
> [1]
> https://downloads.haskell.org/~ghc/7.10.1/docs/html/users_guide/type-level-literals.html
>
> <https://downloads.haskell.org/~ghc/7.10.1/docs/html/users_guide/type-level-literals.html>
> _______________________________________________
> Beginners mailing list
> [email protected] <mailto:[email protected]>
> http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners
> <http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners>
> --
> Frederic Cogny
> +33 7 83 12 61 69
> _______________________________________________
> Beginners mailing list
> [email protected]
> http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.haskell.org/pipermail/beginners/attachments/20190314/2163ff21/attachment-0001.html>
------------------------------
Message: 3
Date: Fri, 15 Mar 2019 12:37:34 +0100
From: Frederic Cogny <[email protected]>
To: mike h <[email protected]>
Cc: The Haskell-Beginners Mailing List - Discussion of primarily
beginner-level topics related to Haskell <[email protected]>
Subject: Re: [Haskell-beginners] Matrix and types
Message-ID:
<cagsugstdt3ert9tmtaxthhmas+br66j-gz3_cpt-ewomxce...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
I'm not sure I understand your question Mike.
Are you saying createIntM behaves as desired but the data constructor M
could let you build a data M with the wrong type. for instance M [1,2] :: M
1 Int ?
If that is your question, then one way to handle this is to have a separate
module where you define the data type and the proper constructor (here
M and createIntM) but where you do not expose the type constructor. so
something like
module MyModule
( M -- as opposed to M(..) to not expose the type constructor
, createIntM
) where
Then, outside of MyModule, you can not create an incorrect lentgh annotated
list since the only way to build it is through createIntM
Does that make sense?
On Thu, Mar 14, 2019 at 4:19 PM mike h <[email protected]> wrote:
> Hi,
> Thanks for the pointers. So I’ve got
>
> data M (n :: Nat) a = M [a] deriving Show
>
> t2 :: M 2 Int
> t2 = M [1,2]
>
> t3 :: M 3 Int
> t3 = M [1,2,3]
>
> fx :: Num a => M n a -> M n a -> M n a
> fx (M xs) (M ys) = M (zipWith (+) xs ys)
>
> and having
> g = fx t2 t3
>
> won’t compile. Which is what I want.
> However…
>
> t2 :: M 2 Int
> t2 = M [1,2]
>
> is ‘hardwired’ to 2 and clearly I could make t2 return a list of
> any length.
> So what I then tried to look at was a general function that would take a
> list of Int and create the M type using the length of the supplied list.
> In other words if I supply a list, xs, of length n then I wan’t M n xs
> Like this
>
> createIntM xs = (M xs) :: M (length xs) Int
>
> which compile and has type
> λ-> :t createIntM
> createIntM :: [Int] -> M (length xs) Int
>
> and all Ms created using createIntM have the same type irrespective of
> the length of the supplied list.
>
> What’s the type jiggery I need or is this not the right way to go?
>
> Thanks
>
> Mike
>
>
>
>
> On 14 Mar 2019, at 13:12, Frederic Cogny <[email protected]> wrote:
>
> The (experimental) Static module of hmatrix seems (I've used the packaged
> but not that module) to do exactly that:
> http://hackage.haskell.org/package/hmatrix-0.19.0.0/docs/Numeric-LinearAlgebra-Static.html
>
>
>
> On Thu, Mar 14, 2019, 12:37 PM Francesco Ariis <[email protected]> wrote:
>
>> Hello Mike,
>>
>> On Thu, Mar 14, 2019 at 11:10:06AM +0000, mike h wrote:
>> > Multiplication of two matrices is only defined when the the number of
>> columns in the first matrix
>> > equals the number of rows in the second matrix. i.e. c1 == r2
>> >
>> > So when writing the multiplication function I can check that c1 == r2
>> and do something.
>> > However what I really want to do, if possible, is to have the compiler
>> catch the error.
>>
>> Type-level literals [1] or any kind of similar trickery should help you
>> with having matrices checked at compile-time.
>>
>> [1]
>> https://downloads.haskell.org/~ghc/7.10.1/docs/html/users_guide/type-level-literals.html
>> _______________________________________________
>> Beginners mailing list
>> [email protected]
>> http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners
>>
> --
> Frederic Cogny
> +33 7 83 12 61 69
> _______________________________________________
> Beginners mailing list
> [email protected]
> http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners
>
>
> --
Frederic Cogny
+33 7 83 12 61 69
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.haskell.org/pipermail/beginners/attachments/20190315/ce93d70f/attachment.html>
------------------------------
Subject: Digest Footer
_______________________________________________
Beginners mailing list
[email protected]
http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners
------------------------------
End of Beginners Digest, Vol 129, Issue 4
*****************************************
