From frederic-emmanuel.picca at synchrotron-soleil.fr  Tue Dec 10 09:47:41 2019
From: frederic-emmanuel.picca at synchrotron-soleil.fr (PICCA Frederic-Emmanuel)
Date: Tue, 10 Dec 2019 09:47:41 +0000
Subject: [Haskell-beginners] how to parse
Message-ID: <A2A20EC3B8560D408356CAC2FC148E53017D846E39@SUN-DAG3.synchrotron-soleil.fr>

Hello,

I have a bunch of files names like this

<prefix>_data_00006.cbf

I would like to parse it into this type

data CbfDataFile  = CbfDataFile Text Int

where

Test is <prefix> and Int = 6

I am using attoparsec, I started to write my parser like this


parseCbfDataFile :: String -> Parser CbfDataFile
parseCbfDataFile s = do
   prefix <- ????
  string "_data_"
  v <- decimal
  string ".cbf"
  return (CbfDataFile prefix v)


So my question, is how to I write this parser. I miss for now the prefix part, and it seems also to me thaht I need something else for the numeric part.

thanks for your help

Frederic

From fa-ml at ariis.it  Tue Dec 10 10:28:03 2019
From: fa-ml at ariis.it (Francesco Ariis)
Date: Tue, 10 Dec 2019 11:28:03 +0100
Subject: [Haskell-beginners] how to parse
In-Reply-To: <A2A20EC3B8560D408356CAC2FC148E53017D846E39@SUN-DAG3.synchrotron-soleil.fr>
References: <A2A20EC3B8560D408356CAC2FC148E53017D846E39@SUN-DAG3.synchrotron-soleil.fr>
Message-ID: <20191210102803.GA9433@x60s.casa>

On Tue, Dec 10, 2019 at 09:47:41AM +0000, PICCA Frederic-Emmanuel wrote:
> I have a bunch of files names like this
> <prefix>_data_00006.cbf
>
> [...]
>
> I am using attoparsec, I started to write my parser like this
> 
> 
> parseCbfDataFile :: String -> Parser CbfDataFile
> parseCbfDataFile s = do
>    prefix <- ????
>   string "_data_"
>   v <- decimal
>   string ".cbf"
>   return (CbfDataFile prefix v)
> 
> 
> So my question, is how to I write this parser.

`manyTill` [1] should do
[1] https://hackage.haskell.org/package/attoparsec-0.13.2.2/docs/Data-Attoparsec-Combinator.html#v:manyTill

From spiridempt at gmail.com  Tue Dec 10 14:00:39 2019
From: spiridempt at gmail.com (F L)
Date: Tue, 10 Dec 2019 22:00:39 +0800
Subject: [Haskell-beginners] Why many .tix files are generated in Haskell
	Program Coverage
Message-ID: <CABjLY-mVpsbqkXnuWX8yywYjHAxEUkVNCMqqGpH8jL84U58EQA@mail.gmail.com>

I am using Haskell Program Coverage (HPC) to collect coverage information
for Pandoc (https://github.com/jgm/pandoc). However, after adding the -fhpc
option to GHC and running the program, two .tix files are generated. One is
test-pandoc.tix in the main directory and the other is pandoc.tix under the
"test" directory (test/pandoc.tix). These two files have different contents
and I just cannot see the meaning of them from their names. Moreover, I
tried to execute "hpc show xxx.tix" for them and test/pandoc.tix gives
"hpc: hash in tix file for module Main does not match hash in
./.hpc/Main.mix".

So, I wonder the specific generation rules of HPC tix files (possibly
including the number and location) and how I can get the correct coverage
information.

Thanks very much!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/beginners/attachments/20191210/aa593004/attachment.html>

From frederic-emmanuel.picca at synchrotron-soleil.fr  Tue Dec 10 20:16:48 2019
From: frederic-emmanuel.picca at synchrotron-soleil.fr (PICCA Frederic-Emmanuel)
Date: Tue, 10 Dec 2019 20:16:48 +0000
Subject: [Haskell-beginners] how to parse
In-Reply-To: <20191210102803.GA9433@x60s.casa>
References: <A2A20EC3B8560D408356CAC2FC148E53017D846E39@SUN-DAG3.synchrotron-soleil.fr>,
 <20191210102803.GA9433@x60s.casa>
Message-ID: <A2A20EC3B8560D408356CAC2FC148E53017D8471C7@SUN-DAG3.synchrotron-soleil.fr>

> `manyTill` [1] should do
> [1] https://hackage.haskell.org/package/attoparsec-0.13.2.2/docs/Data-Attoparsec-Combinator.html#v:manyTill

It works,

thanks

From toad3k at gmail.com  Thu Dec 12 01:38:28 2019
From: toad3k at gmail.com (David McBride)
Date: Wed, 11 Dec 2019 20:38:28 -0500
Subject: [Haskell-beginners] how to parse
In-Reply-To: <A2A20EC3B8560D408356CAC2FC148E53017D8471C7@SUN-DAG3.synchrotron-soleil.fr>
References: <A2A20EC3B8560D408356CAC2FC148E53017D846E39@SUN-DAG3.synchrotron-soleil.fr>
 <20191210102803.GA9433@x60s.casa>
 <A2A20EC3B8560D408356CAC2FC148E53017D8471C7@SUN-DAG3.synchrotron-soleil.fr>
Message-ID: <CAN+Tr431fFZaAzmQd8FWmtv3VhsSaCVvuDPzqkeUALi=F3KaqQ@mail.gmail.com>

I would actually use Data.Attoparsec.ByteString.Char8.takeWhile1.

prefix <- takeWhile1 (/= '_')


On Tue, Dec 10, 2019, 15:17 PICCA Frederic-Emmanuel <
frederic-emmanuel.picca at synchrotron-soleil.fr> wrote:

> > `manyTill` [1] should do
> > [1]
> https://hackage.haskell.org/package/attoparsec-0.13.2.2/docs/Data-Attoparsec-Combinator.html#v:manyTill
>
> It works,
>
> thanks
> _______________________________________________
> Beginners mailing list
> Beginners at haskell.org
> http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/beginners/attachments/20191211/3bc8c451/attachment.html>

From frederic-emmanuel.picca at synchrotron-soleil.fr  Thu Dec 12 05:31:56 2019
From: frederic-emmanuel.picca at synchrotron-soleil.fr (PICCA Frederic-Emmanuel)
Date: Thu, 12 Dec 2019 05:31:56 +0000
Subject: [Haskell-beginners] how to parse
In-Reply-To: <CAN+Tr431fFZaAzmQd8FWmtv3VhsSaCVvuDPzqkeUALi=F3KaqQ@mail.gmail.com>
References: <A2A20EC3B8560D408356CAC2FC148E53017D846E39@SUN-DAG3.synchrotron-soleil.fr>
 <20191210102803.GA9433@x60s.casa>
 <A2A20EC3B8560D408356CAC2FC148E53017D8471C7@SUN-DAG3.synchrotron-soleil.fr>,
 <CAN+Tr431fFZaAzmQd8FWmtv3VhsSaCVvuDPzqkeUALi=F3KaqQ@mail.gmail.com>
Message-ID: <A2A20EC3B8560D408356CAC2FC148E53017D8478FF@SUN-DAG3.synchrotron-soleil.fr>

The problem is that prefix can contain also a '_'

From Leonhard.Applis at protonmail.com  Tue Dec 17 10:31:29 2019
From: Leonhard.Applis at protonmail.com (Leonhard Applis)
Date: Tue, 17 Dec 2019 10:31:29 +0000
Subject: [Haskell-beginners] Question on Lazy Evaluation, Maps & Funktions
Message-ID: <hNWIZY8vijKpBZ-9qNMolVFovKTufzKuHDPswYxHd8Ecb5GL_rL9RqT-Gy2yfE_PCpmPEWaOCSVS-yHc3OWf8RoGIPztWz8WwUa49XO90X0=@protonmail.com>

Hi there, 

I currently have a task at hand which requires to find a n-tuple of sentences s, where the biggest distance of sentences is searched(So the most dis-similar sentences shall be chosen). 

Especially with growing n, this problem is causing me trouble, as in a naive implementation the number I need to calculate the distances grows exponentially by n and linear by s.
The distance function itself is drop-in distance :: Text -> Text -> Double 
For the questions the distance can be expected to behave in constant time, and the list of n-tuples to be deepseq ready. 

Given my real world problems, this is giving me a hard time waiting. 

Coming from a more imperative environment, one approach I'd take would be to calculate a dictionary of predefined distances, so instead of calculating the distance new I'd only have to lookup the distance.
This would - in my understanding - mean that I create the dictionary which takes s * (s-1) steps to calculate the full dictionary, and then for s^n only dictionary lookups. 

But from the little I remember "Real World Haskell", the functions and variables are only evaluated once and hang around until the garbage collector comes around. 
Question 1.)is this right, and does this mean, that if there is no GC in between, the function distance(a,b) is only calculated once, and therefore the dictionary approach is mostly useless?

Question 2.) 
The function distance is symmetric, is there a way to help my Haskell (and the evaluation) to understand that it only needs to be run once? 
Iff not, does it make sense to iterate the over the tuples first and "order" them, to help distances only occur in a certain order? Is there a way to create unique, ordered n-tuples in the first place?

Additionally, I am grateful for any kind of help regarding performance-tricks to this case. 
I know that this is a typical case for multi threading, but I first want to try and find "the best" base-algorithm. Also I think it's a nice question :) 

Thank you very much
Leonhard
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/beginners/attachments/20191217/812914f9/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: publickey - Leonhard.Applis at protonmail.com - 0x807FDDF3.asc
Type: application/pgp-keys
Size: 1843 bytes
Desc: not available
URL: <http://mail.haskell.org/pipermail/beginners/attachments/20191217/812914f9/attachment.key>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 477 bytes
Desc: OpenPGP digital signature
URL: <http://mail.haskell.org/pipermail/beginners/attachments/20191217/812914f9/attachment.sig>