- Haskell-Cafe - Haskell.org

SYB, ext1Q, and deriving custom Typeable and Data instances
by Vyacheslav Akhmechet 05 Dec '06

05 Dec '06

Hi, I'm trying to use SYB to implement some basic relational persistance. I define a PKey datatype to represent primary keys like this: newtype PKey a = PKey a deriving (Show, Typeable, Data) I then define a sample data type like this: data User = User { id :: PKey Int, firstName :: String, lastName :: String, } deriving (Show, Typeable, Data) Finally, I try to iterate over the structure like this: test = gmapQ (gshow `ext1Q` aux) where aux :: PKey a -> String aux _ = "test" The problem is that for id gmapQ calls gshow instead of my customized version (aux). I've seen a discussion on this on haskell cafe at the following URL: http://www.opensubscriber.com/message/haskell-cafe@haskell.org/2167871.html I couldn't understand the following details: 1. Do I need to derive both Typeable and Data instances? 2. Do I need to derive them for both PKey and User or just PKey? 3. Where can I find sample code to derive these instances? Documentation shows a sample of how GHC derives Data, is that a good template to use? Also, what's a good example to derive Typeable (if in fact it needs to be done manually)? Thanks, - Slava.

1 0

'+++' and 'interleave'
by Greg Fitzgerald 05 Dec '06

05 Dec '06

Could ReadP's '+++' (symmetric choice) operator be implemented for all Monads with the 'interleave' function mentioned in Backtracking, Interleaving, and Terminating Monad Transformers<http://portal.acm.org/ft_gateway.cfm?id=1086390&type=pdf&coll=GUIDE&dl=GUID…>? This paper was a tough read for me, but it looks like they solve the same problem, but one interleaves computations within a monad and the other does it from the outside. Or am I just confused? Thanks, Greg

1 0

Four Fours problem in Haskell
by Joachim Breitner 04 Dec '06

04 Dec '06

Hi, This might be interesting for some of you (I hope): http://www.joachim-breitner.de/blog/archives/210-FourFours-in-Haskell.html Preview: An interesting programming puzzle appeard in the Blogosphere: To describe every integer from 1 to 100 as a calculation involving only and exactly four fours, and standard operations. There are solutions in C# (code) and in python (code), so I tried to do mine in Haskell. This seems to be meant for Haskell. The Type system makes the program really great to read. Functional programming allows me to conveniently create all possible associations of four operands with three binary operations. (I found five. Why are the other two sources using four?) And, in this program particular, the use of the list monad to quickly generate a list of all possible combination is really a programming-time-saver. The program is not too slow either: $ time ./fourfour>/dev/null real 0m0.917s user 0m0.852s sys 0m0.040s And it could probably be sped up by using some kind of hashing or similar stuffs. Note that the program has only 73 mostly quite short lines, while the python code has 119 and the C# code has 138 of relatively long lines (comments and empty lines not counted). I think this is impressive (unless you have worked with haskell a bit, then you are used to these kind of results). Comments welcome! Thanks, Joachim -- Joachim "nomeata" Breitner mail: mail(a)joachim-breitner.de | ICQ# 74513189 | GPG-Key: 4743206C JID: joachimbreitner(a)amessage.de | http://www.joachim-breitner.de/ Debian Developer: nomeata(a)debian.org

1 0

Re: help with threadDelay
by Neil Davies 04 Dec '06

04 Dec '06

Ian/Simon(s) Thanks - looking forward to the fix. It will help with the real time enviroment that I've got. Follow on query: Is there a way of seeing the value of this interval from within the Haskell program? Helps in the calibration loop. Neil

2 3

RE: Safe forking question
by Simon Marlow 04 Dec '06

04 Dec '06

[ duh, hit send by mistake, message continues... ] Chris Kuklewicz wrote: > forkInheritIO :: IO () -> IO ThreadId -- inherits parent's > block or unblock status > > forkBlockedIO :: IO () -> IO ThreadId -- starts the action in > "block" mode. Must manually "unblock" > Either of these is certainly possible, and I agree that forkBlockedIO is more generally useful than my forkCatchIO. One tantalising possibility is that we could make forkIO have the forkInheritIO behaviour without breaking much code, since I bet forkIO is almost never called inside block. Cheers, Simon

2 1

HaXml problem
by Alexis Hazell 04 Dec '06

04 Dec '06

Hi all, Say i have data of type Content, and i want to put that data into a type which is an instance of XmlContent, so that i can use it with hPutXml. How can i do so? Does it involve the List1 type and its NonEmpty constructor? Thanks in advance, Alexis.

2 1

Generate 50 random coordinates
by Huazhi (Hank) Gong 04 Dec '06

04 Dec '06

Hello,all My intention is to generate 50 random coordinates like (x,y). myrand :: Int myrand = randomRIO(1::Int, 100) rf=[(myrand, myrand) | a <- [1..50]] My short program is like this. However, GHCI say that the return type of randomRIO is IO a while the function defined by me is Int. Since I only need a integral type as my cooridinate, could you tell me how to fix this? Hank -- View this message in context: http://www.nabble.com/Generate-50-random-coordinates-tf2742242.html#a7651116 Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

8 8

complex state code
by Tim Newsham 04 Dec '06

04 Dec '06

I'm working on some code where I have some complex state. This is the first time I've dealt with anything like this in haskell. If anyone would review my code and let me know if there are ways I could do this better, cleaner, in a more idiomatic way, etc, I would greatly appreciate it. http://www.thenewsh.com/~newsham/Market.hs Tim Newsham http://www.thenewsh.com/~newsham/

1 0

compiler versus interpreter
by jepalomar23＠ono.com 03 Dec '06

03 Dec '06

I am programming an application with a lot of interactivity. When using x<-getChar the character corresponding to the pushed key is assigned to x immediately. Launching the module through runghc works properly. However, the compiled module through ghc requires to push return after the character key in order to complete the assignation procedure. It is a bug or there is some difference between compiler an interpreter? Thanks in advance. jepalomar

2 1

Functional GUI combinators for arbitrary graphs of components?
by Brian Hulley 03 Dec '06

03 Dec '06

Hi, I've been looking at the Fudgets research (http://www.cs.chalmers.se/ComputingScience/Research/Functional/Fudgets/ ) as an example of a purely functional gui. Afaiu a brief summary is that a Fudget can represent a GUI widget or some other object, which can have some internal state, and which receives and passes messages to the other Fudgets it is connected to. Fudgets are connected to each other by combinators eg: second >==< first -- sequential (first sends messages to second) a >*< b -- parallel There is a continuation passing implementation such that the messages are passed and internal states of Fudgets are updated (by replacing each fudget with an updated fudget) as these expressions are evaluated hence the message passing and maintenance of state doesn't involve any IORefs though interaction with external devices such as the OS uses IO. Anyway to get to my point, though all this sounds great, I'm wondering how to construct an arbitrary graph of Fudgets just from a fixed set of combinators, such that each Fudget (node in the graph) is only mentioned once in the expression. To simplify the question, assume we have the following data structure to describe the desired graph: data LinkDesc a = Series a a | Broadcast a [a] | Merge [a] a type GraphDesc a = [LinkDesc a] and a function to compute a combined object (eg Fudget) corresponding to an arbitrary graph would be: mkObj :: Ord label => Map label Obj -> GraphDesc label -> Obj mkObj = -- ??? The result of mkObj will be some expression, composed from a finite set of combinators applied to the named objects, but with the all important property that each named object appears no more than once in the expression (this is what allows Fudgets to correspond to notional objects which maintain internal state - during evaluation each fudget gets replaced by an updated fudget hence can only appear once in the whole expression). As a very simple example using existing Fudgets combinators, type Map label value = [(label, value)] mkObj [("A", objA), ("B", objB), ("C", objC)] [Series "A" "B", Series "B" "C"] === objC >==< objB >==< objA and mkObj [("A", objA), ("B", objB), ("C", objC)] [Broadcast "A" ["B", "C"]] === (objB >*< objC) >==< objA However: mkObj [("A", objA), ("B", objB), ("C", objC), ("D", objD)] [Broadcast "A" ["B", "C"], Series "B" "D", Series "A" "D"] === ??? The Fudgets library provides many combinators eg for looping, but even so, I don't think there are sufficient combinators to express the above graph (though possibly extra nodes and tagging/untagging would solve this example), or more heavily connected graphs such as: [Series "A" "B", Series "B" "C", Series "C" "D", Series "D" "B", Series "C" "A"] This problem is so general (ie converting a cyclic graph into an expression tree such that each node appears no more than once) that I'm sure someone somewhere must already have determined whether or not there can be a solution, though I don't know where to look or what to search for. Any ideas? Thanks, Brian. -- http://www.metamilk.com

3 5