- Haskell-Cafe - Haskell.org

Hackage down
by Dougal Stanton 19 Oct '09

19 Oct '09

Has not been responding for at least the last 12 hours. Is there somewhere to look for status reports on sysadmin details like this, so we can tell if - it's a scheduled down time - it's a problem but the admins know about it - etc etc. D -- Dougal Stanton dougal(a)dougalstanton.net // http://www.dougalstanton.net

7 6

Re: [Haskell-cafe] Error handling package
by Michael Snoyman 19 Oct '09

19 Oct '09

On Mon, Oct 19, 2009 at 3:39 PM, Jose Iborra <pepeiborra(a)gmail.com> wrote: > You may want to take a look at anoother option in Hackage, the > control-monad-exception package. > > http://pepeiborra.github.com/control-monad-exception/ > > The control-monad-exception library provides the building blocks for > > * Explicitly Typed exceptions (checked or not) > * which are composable > * and even provide stack traces (experimental feature) Jose, Thank you very much, thinks looks like just the thing I was looking for. The only thing that slightly concerns me is the transformers dependency, but it's not a serious worry. I'd love to try out the package right now, but with hackage being down I can't get the dependencies yet ;). I'll try it out as soon as I can. > > > > On 19/10/2009, at 01:00, Michael Snoyman wrote: > > (Sorry, accidently took off cafe.) >> >> On Mon, Oct 19, 2009 at 12:44 AM, Henning Thielemann < >> lemming(a)henning-thielemann.de> wrote: >> >> On Mon, 19 Oct 2009, Michael Snoyman wrote: >> >> Does the explicit-exception package provide what you need? >> >> http://hackage.haskell.org/package/explicit-exception >> >> >> I don't think so, but correct me if I'm wrong. I want to make it easy to >> chain together >> computations which could fail in different ways. For example, something >> like this: >> >> attemptReadInt :: String -> Attempt Int >> attemptLookup :: String -> [(String, String)] -> Attempt String >> attemptLookupInt :: String -> [(String, String)] -> Attempt Int >> attemptLookupInt k m = attemptLookup k m >>= attemptReadInt >> >> Now, in the explicit-exception package, I could- in this simple example- >> define >> something like: >> >> data MyErrors = KeyNotFound | InvalidInt >> >> >> type Attempt = Exceptional MyErrors >> >> True; that's what I meant by I could do this in my simple example. >> >> >> But this solution would not scale. >> >> You want to add other exceptions? The idea of my package is to make >> exceptions explicit in the type. Otherwise you would use >> extensible-exceptions. Or you could define MyErrors using an existential >> type. >> >> Which is my point. I'm trying to provide a package for non-explicit >> exceptions. To compare to other programming languages, I think your package >> is providing the equivalent of Java checked exceptions, while mine is >> providing (safe) unchecked exceptions. I say safe because you still need to >> explicitly decide to turn an Attempt into a possible runtime exception which >> will bring down your program. >> >> Defining MyErrors using an existential type would essentially recreate the >> entire attempt package; I don't see that purpose in everyone wanted >> unchecked exceptions needing to reinvent the wheel in non-compatible ways. >> If multiple libraries use attempt, they can easily have their >> possible-error-returning functions chain together safely. >> >> > I believe that control-monad-exception solves this tension between > composability and explicit exceptions. > You can have explicit exceptions which are composable: > > > data DivideByZero = DivideByZero deriving (Show, Typeable) > > data SumOverflow = SumOverflow deriving (Show, Typeable) > > > instance Exception DivideByZero > > instance Exception SumOverflow > > > data Expr = Add Expr Expr | Div Expr Expr | Val Double > > > eval (Val x) = return x > > eval (Add a1 a2) = do > > v1 <- eval a1 > > v2 <- eval a2 > > let sum = v1 + v2 > > if sum < v1 || sum < v2 then throw SumOverflow else return sum > > eval (Div a1 a2) = do > > v1 <- eval a1 > > v2 <- eval a2 > > if v2 == 0 then throw DivideByZero else return (v1 / v2) > > GHCi infers the following types > > > :t eval > > eval :: (Throws DivideByZero l, Throws SumOverflow l) => Expr -> EM l > Double > > > :t eval `catch` \ (e::DivideByZero) -> return (-1) > > .... :: Throws SumOverflow l => Expr -> EM l Double > > > :t runEM(eval `catch` \ (e::SomeException) -> return (-1)) > > .... : Expr -> Double > > >> Additionally, there's two immediate features I think I would miss from my >> package: >> >> 1) fail works properly, so an Attempt would be a valid monad response from >> people who >> use that function. >> >> As far as I understand, 'fail' is used/abused for reporting failed pattern >> matches in do notation. If a failed pattern match indicates a programming >> error, it should be a really error, and not something that must be handled >> at run-time. >> >> That's a lot of very debateable statements you just made. It might be that >> it's strongly encouraged to only use fail for failed pattern matching, but >> in practice you could use it for any monadic failure. Also, there's nothing >> stopping a user from re-throwing pattern match exceptions received in an >> Attempt. >> > > I am with Henning on 'fail'. > It must not be used as a replacement for throw, only for failed pattern > matches which are programming errors and thus unchecked exceptions. > I'm not arguing how it *should* be used, I'm arguing how it *might* be used, and I think calling error is not appropriate. An unchcked exception which can be caught if necesary (I'm assuming that's what runEMParanoid does) sounds like that right thing. Michael

1 0

Documentation (was: ANN: text 0.5, a major revision of the Unicode text library)
by John Lato 19 Oct '09

19 Oct '09

For anyone writing introductions to generic programming, take this as a plea from Haskellers everywhere. If one of the RWH authors can't understand how to make use of these techniques, what hope do the rest of us have? John Lato P.S. Some might wryly note that I'm the maintainer of a package which is also known for incomprehensible documentation. To which I would reply that our effort is much newer, I consider it a problem, and it's being worked on, contrasted to the state of GP where similarly impenetrable documentation has been and continues to be the norm. > > From: "Bryan O'Sullivan" <bos(a)serpentine.com> > > I think maybe someone else will have to take a crack at a Data instance for > Text, because the documentation for Data.Data is not written in English. In > its syntax and structure, it closely hews to what we think of as English, > but it is the kind of documentation that can only be understood by someone > who already knows what it is going to say. > > This is an exemplar of my experience with the cottage industry of generic > programming in Haskell: I'd really quite like to use the stuff, but for > goodness's sake, o beloved researchers, please aim your expository papers at > non-specialists once in a while. An endless chain of papers of the form "my > technique, which you won't understand, is better than this other technique, > which you haven't read about and won't anyway understand, in subtle ways > that you won't understand" does not feel to me like progress. > > Yours in some misery and frustration, > Bryan.

5 4

Too much strictness in binary-0.5.0.2
by Khudyakov Alexey 18 Oct '09

18 Oct '09

Hello I run into problems with new binary package. Following function reads a list of elements one by one until end of stream. List is very long (won't fit into memory). In binary-0.5.0.1 and earlier it read list lazily. Now it seems that it tries to read whole list to memory. Program does not produce any output and memory usage steadily grows. > getStream :: Get a -> Get [a] > getStream getter = do > empty <- isEmpty > if empty > then return [] > else do x <- getter > xs <- getStream getter > return (x:xs) How could I add laziness to this function to revert to old behavior. -- Khudyakov Alexey

3 3

Merging modules
by pat browne 18 Oct '09

18 Oct '09

Hi, I want to establish the strengths and weakness of the Haskell module system for the ontology merging task (I know it was not designed for this!). I wish to make a new module (MERGEDONTOLOGY) from three input modules one of which is common to the other two. The desired merge should contain the following data types: Woman FinancialBank HumanBeing RiverBank Which are all identifiable without any qualifying module names. Below is the detailed requirement and my first attempt at this task in Haskell. I would be grateful for any information on how to merge these modules giving a set of unique unqualified types (i.e. without reference to their originating modules). Regards, Pat ======================== Informal Specification ======================== The diagram and text below explain the requirement MERGEDONTOLOGY { Woman, RiverBank, FinancialBank, HumanBeing} /\ /\ / \ / \ / \ / \ / \ / \ ONTOLOGY1 ONTOLOGY2 {Woman, Bank, Person} {Woman, Bank, Human} /\ /\ \ / \ / \ / \ / \ / \ / \ / {Woman , Person} COMMMON This example includes both synonyms and homonyms. 1)The Woman sort (or data type) should be the same in all modules, there is only one Woman sort and it is named as such in each module. Hence there should be only one MERGEDONTOLOGY.Woman. 2)There is only one sort MERGEDONTOLOGY.HumanBeing, but there are 3 synonyms for it called ONTOLOGY2.Human, ONTOLOGY1.Person, and COMMON.Person. The last sentence considers ONTOLOGY1.Person and COMMON.Person as synonyms; they have different qualifiers but the intention is that they represnt the same thing. Hence should be mapped to same MERGEDONTOLOGY.HumanBeing. To do this (in Maude) COMMON.Person was renamed to ONTOLOGY2.Human which in turn was renamed to MERGEDONTOLOGY.HumanBeing. 3)The homonyms are ONTOLOGY1.Bank and ONTOLOGY2.Bank should become distinct sorts MERGEDONTOLOGY.RiverBank and MERGEDONTOLOGY.FinancialBank in the final ontology at the top of the diagram. ================================================ My first attemt at merging using Haskell modules ================================================= module COMMON where data Woman = WomanC data Person = PersonC module ONTOLOGY1 where import COMMON data Bank = BankC module ONTOLOGY2 where import COMMON data Bank = BankC module MERGEDONTOLOGY where import ONTOLOGY1 import ONTOLOGY2 If I use qualified names all the constructors are interpreted correctly MERGEDONTOLOGY> :t COMMON.WomanC COMMON.WomanC :: COMMON.Woman MERGEDONTOLOGY> :t COMMON.PersonC COMMON.PersonC :: COMMON.Person MERGEDONTOLOGY> :t ONTOLOGY2.BankC ONTOLOGY2.BankC :: ONTOLOGY2.Bank MERGEDONTOLOGY> :t ONTOLOGY1.BankC ONTOLOGY1.BankC :: ONTOLOGY1.Bank MERGEDONTOLOGY> :t COMMON.Woman However, I wish that these types and constructors be fully defined in the context of the MERGEDONTOLOGY. I have tried type synonyms and newtypes. type RiverBank = ONTOLOGY1.Bank type FinancialBank = ONTOLOGY2.Bank newtype RiverBank = BankC Int I have not explored import modes or qualified imports.

2 4

x -> String
by Andrew Coppin 18 Oct '09

18 Oct '09

Is there any way that you can turn an arbitrary Haskell value into a string? I rephrase: There *is* a way to turn arbitrary values into strings. I know there is, because the GHCi debugger *does* it. The question is, does anybody know of an /easy/ way to do this? Basically, I'm writing a mutable container implementation. It can hold any type of data, but it would massively aid debugging if I could actually print out what's in it. On the other hand, I don't want to alter the entire program to have Show constraints everywhere just so I can print out some debug traces (and then alter everything back again afterwards once I'm done debugging). Anybody know of a way to do this? (As it happens, the values I'm testing with are all Showable anyway, but the type checker doesn't know that...)

8 13

Happstack + network package issue on the Mac
by Sean Leather 18 Oct '09

18 Oct '09

We have run into an issue that it seems other people have already found but has apparently not been resolved, yet. It manifests as with the error "HTTP request failed with: Unsupported socket". The problem is described in two places with different workarounds. http://code.google.com/p/happstack/issues/detail?id=88 http://groups.google.com/group/HAppS/msg/0c9a0d0fd7c6aff0 One workaround is to rewrite part of the network package, and the other is to downgrade the package. It could be the case that the rewrite came from the older version, though I'm not sure. I'm curious if it is possible to fix this, either in Happstack or the network package, once and for all. The workarounds are rather painful when so many things depend on the network package. Thanks, Sean

5 9

Reverse dependencies in Hackage
by Roel van Dijk 17 Oct '09

17 Oct '09

A few weeks ago I wrote a patch that adds reverse dependencies to hackage [1]. I have know hosted a small test hackage which demonstrates this feature. It can be found here: http://bifunctor.homelinux.net/~roel/hackage Browse to your favorite packages and find out how much other packages depend on them! I also noticed that work is underway to implement a new hackage-server [2] based on happstack. If people find this feature useful I could also write a patch against the hackage-server code base [3]. Regards, Roel van Dijk [1] - http://hackage.haskell.org/trac/hackage/ticket/576 [2] - http://sparky.haskell.org:8080/ [3] - http://code.haskell.org/hackage-server/

3 2

Monadic correctness
by Andrew Coppin 17 Oct '09

17 Oct '09

Suppose we have newtype Foo x instance Monad Foo runFoo :: Foo x -> IO x What sort of things can I do to check that I actually implemented this correctly? I mean, ignoring what makes Foo special for a moment, how can I check that it works correctly as a monad.

5 7

Genuine Need For Persistent Global State?
by Alan Carter 17 Oct '09

17 Oct '09

Hi, I've been looking at the patches given by Tom at Beware the Jabberwolk, for building Linux kernel modules in Haskell. Once I'd got Tom's stuff building, the next thing was to build a little driver which actually does something. Step by step I was making progress, and I've now got a little function which can see all the characters which are catted to the device file. That's when I got stuck. Trouble is, my function is (ultimately) being called from the C kernel stuff. It isn't on the bottom of a call graph coming from a Haskell main. A driver really needs to know where it's at. So I seem to need some kind of global, persistent state, and Control.Monad.State seems to be out because I can't pass a State around my call graph. I've thought about trying to create some State when I initialize, pass some kind of pointer back to the C shim which actually does the calling into Haskell, and then passing it back into Haskell on the writes and reads, but that seems dangerous because the Haskell garbage collector would (understandably) get the wrong idea and delete it. Also it's ugly - having to use C for something Haskell can't do. I've been looking at the Halfs Haskell file system, which surely must have solved its own version of this problem, but whatever it does in its FSState seems very complicated and is beyond my comprehension :-( So my questions are: 1) Am I right in thinking that I have a genuine need for global, persistent state? 2) Halfs works. Am I right in thinking it has (somehow) solved this problem? 3) Is there a simple way to maintain global persistent state that I can stash between calls into a function, and access without needing to pass the state in? If so, is there an example anywhere? Thanks in advance, Alan -- ... the PA system was moaning unctuously, like a lady hippopotamus reading A. E. Housman ..." -- James Blish, "They Shall Have Stars"

4 6