June 2008 - Haskell-Cafe

Checking packages before upload to Hackage
by Henning Thielemann 20 Jun '08

20 Jun '08

Is there some command which tests a tar.gz created by 'Setup.lhs sdist' by unpacking the archive to say /tmp and compiling and documenting the sources? Hackage should recommend this tool before package upload.

3 2

Quantitative Trading Developer Position at Hutchin Hill Capital
by Neil Mehra 20 Jun '08

20 Jun '08

Job Description: Quantitative Trading Developer Description: * Newly formed multi-strategy hedge fund, Hutchin Hill Capital, located in midtown Manhattan, seeks a Quantitative Trading Developer; * Focus on building infrastructure, applications, and technical support for quantitative equity trading business; * Candidate should have demonstrated interest or passion for functional programming-with experience using Haskell, F#, Kdb+/q, OCaml, ML, and related technologies; * Position reports directly to portfolio manager, and will work closely with other technology and research professionals. Responsibilities Include: * Developing data management, quant research, and analytical tools for model-driven equity trading strategies; * Building production environment for real-time content capture, trade generation, and risk management; * Deploying equity trading technology, methods, and protocols, including electronic execution algorithms and market impact analysis; * Conducting quantitative portfolio research using cutting-edge tools and techniques. Required Experience/Education: * 2 - 5 years in IT development or quantitative analytics; * Advanced degree or equivalent experience in a technical field such as computer science, mathematics, physics, or engineering; Other Requirements: * Exceptional technical and analytical problem-solving; * Desire to master the latest technology, hardware, and quantitative methods; * Willingness to work in a fast-paced, high pressure environment; * Team-oriented and highly resourceful; * Entrepreneurial drive. Please contact quant2(a)hutchinhill.com<mailto:quant2@hutchinhill.com> with resume if interested. __________ This communication, including any attachment(s), may contain confidential and/or privileged information. If you are not the intended recipient or have received this communication in error, please destroy/delete this email and any attachments. We do not waive confidentiality or privilege by mistransmission. This communication does not constitute an offer of employment, or an offer, or the solicitation of an offer, to buy or sell any security or investment advisory services, and is not an official confirmation of any transaction.

1 0

This is a bug?
by Samuel Silva 20 Jun '08

20 Jun '08

Hello I'm using HaXml library to handle XML. How I solve this issue when call "readXml": >in <score-partwise>, In a sequence: > in <part>, In a sequence: > in <measure>, Too many elements inside <measure> at > file ../../../parsers/elite2.xml at line 75 col 15 > Found excess: > > >Done. I don't understood this error. I'm using HaXml-1.19.4 with GHC-6.8.2 -- Don't hug that, Hugs ME! Samuel Silva

2 1

Re: (Lazy) SmallCheck and peano numbers
by Levi Stephen 20 Jun '08

20 Jun '08

On Fri, Jun 20, 2008 at 3:30 AM, Benedikt Huber <benjovi(a)gmx.net> wrote: > Levi Stephen schrieb: >> >> Hi, >> >> I have the following definitions >> >> type Zero >> type Succ a >> >> so that I can muck around with a Vector type that includes its length >> encoded in its type. >> >> I was wondering whether it was possible to use SmallCheck (or >> QuickCheck) to generate random Peano numbers? Is there an issue here >> in that what I actually want to generate is a type rather than a >> value? >> >> I do have >> >> reifyInt :: Int -> (forall a. ReflectNum a => a -> b) -> b >> >> but, I'm not sure if this can help me when I need to generate other >> values based upon that type (e.g., two vectors with the same size >> type) > > Hi Levi, > > For QuickCheck, I know it is possible as long as you do not need to use type > level functions in your tests. For example, using Alfonso's type-level and > parametrized-data packages, one can write: > >> instance (Nat n, Arbitrary a) => Arbitrary (FSVec n a) where >> arbitrary = >> liftM (unsafeVector (undefined :: n)) $ >> mapM (const arbitrary) [1..toInt (undefined :: n)] > >> propLength :: forall n a. (Nat n) => FSVec n Integer -> Bool >> propLength (FSVec xs) = P.length xs == toInt (undefined :: n) > >> propLengthEqual :: forall n a. (Nat n) => >> FSVec n Integer -> FSVec n Integer -> Bool >> propLengthEqual v1 v2 = length v1 == length v2 > >> tests1 = forM_ [0..100] $ \n -> reifyIntegral n $ \(t :: ty) -> >> quickCheck (propLength :: FSVec ty Integer -> Bool) >> tests2 = forM_ [0..100] $ \n -> reifyIntegral n $ \(t :: ty) -> >> quickCheck (uncurry propLengthEqual :: >> (FSVec ty Integer,FSVec ty Integer) -> Bool) Thanks for the example code. Ideally it would be great to have n generated also. Any thoughts on whether something like propLengthEqual :: forall n a. (Nat n) => n -> FSVec n Integer -> FSVec n Integer -> Bool propLengthEqual _ v1 v 2 = length v1 == length v2 with an arbitrary instance for generate all Nat n's is possible? Is something like instance (forall n. Nat n) => Arbitrary n possible/legal haskell? and would it for the above test? > > best regards, > > benedikt > Thanks, Levi

2 1

Meaning of "ribbonsPerLine" at Text.PrettyPrint.HughesPJ ?
by Alfonso Acosta 20 Jun '08

20 Jun '08

Hi, Can anyone give a good explanation of what ribbonsPerLine means? Maybe it would be better to simply ask for the meaning of ribbon in this context. The documentation is totally meaningless to me: "reibbonsPerLine: Ratio of ribbon length to line length". I asked at #haskell and frankly, I was surprised by the answer: fons: can anyone tell me what the heck does ribbonsPerLine mean in Text.PrettyPrint.HughesPJ? fons: > renderStyle style{lineLength=10, ribbonsPerLine=1} $ sep [text "foo" ,text "bar"] lambdabot: "foo bar" fons: > renderStyle style{lineLength=10, ribbonsPerLine=2} $ sep [text "foo" ,text "bar"] lambdabot: "foo\nbar" fons: ok, somehow it affects the line width ... byorgey: fons: I can't explain it, all I know is that you must set it to 1 or else it does bizarre things fons: hahah, ok fons: byorgey: that's funny considering its default value is 1.5 byorgey: if you set it to 1 then lineLength means what you think it should byorgey: fons: EXACTLY Cheers, Fons

7 7

Re: (Lazy) SmallCheck and peano numbers
by Levi Stephen 20 Jun '08

20 Jun '08

> If so, you could write a SmallCheck Series instance as follows. > > instance Serial (Vec Zero a) where > series = cons0 nil > > instance (Serial a, Serial (Vec n a)) => Serial (Vec (Succ n) a) where > series = cons2 (|>) > If we have the property > > prop_vector :: Vec (Succ (Succ Zero)) Bool -> Bool > prop_vector (V xs) = xs == xs > > we can check it, and only 2 element vectors will be tested I have some code up now at http://hpaste.org/8420 It looks pretty much like what you've written above and is what I am doing for the moment, but for various size vectors from zero to four. > Now, it seems what you really want to do is define polymorphic > properties like > > prop_poly :: Vec n Bool -> Vec n Bool -> Bool > > and have SmallCheck check all equal-sized vectors. If so, good > question! :-) Yep, this would be exactly what I'm after. Thanks for a better write up and background on what I'm after than I gave :) I'm thinking the property might be written as prop_poly:: size -> Vec size Bool -> Vec size Bool -> Bool, so that the size is generated randomly and the types take care of ensuring the vectors generated are of the same size. Levi

1 0

Re: [Haskell-cafe] Lambda and closures in PHP -- could someone please comment?
by Alexander Wagner 20 Jun '08

20 Jun '08

On Wednesday 18 June 2008, Luke Palmer wrote: > I don't know that much PHP, but I find the "lexical" keyword to be a > nuisance. What are the semantics if the lexical keyword is omitted? In PHP Variables are local by default, the keyword global defines them as references to the global namespace. There are some exceptions, like the availability of $this in methods or a couple of predefined superglobal variables that are implicitly global in all functions. <?php $globalVar = 1; $param = "notshared"; $localVar = "notshared"; function foo ($param) { global $globalVar; $localVar = (int) $param; return $localVar + $localVar; } ?> Given how much it works like "global", the keyword "lexical" seems quite consistent with the style of PHP. Gesundheit Wag -- There’s a phrase we live by in America: “In God We Trust”. It’s right there where Jesus would want it: on our money. - Steven Colbert

1 0

cabal-install failure
by Darrin Thompson 20 Jun '08

20 Jun '08

Cabal-install is looking good. It now, for the record, has only two deps outside of Cabal-1.4. I installed cabal-install-0.5 on ubuntu with the haskell.org linux binary for ghc 6.8.2. I then tried cabal-install yi Got this output: ... many successful installs ... Registering vty-3.0.1... Reading package info from "dist/installed-pkg-config" ... done. Saving old package config file... done. Writing new package config file... done. Downloading yi-0.3... [1 of 1] Compiling Main ( Setup.hs, dist/setup/Main.o ) Setup.hs:9:0: Warning: Deprecated use of `showPackageId' (imported from Distribution.Simple, but defined in Distribution.Package): use the Text class instead Linking dist/setup/setup ... Warning: defaultUserHooks in Setup script is deprecated. Configuring yi-0.3... Warning: Instead of 'ghc-options: -DDYNAMIC -DFRONTEND_VTY' use 'cpp-options: -DDYNAMIC -DFRONTEND_VTY' setup: alex version >=2.0.1 && <3 is required but it could not be found. cabal: Error: some packages failed to install: yi-0.3 failed during the configure step. The exception was: exit: ExitFailure 1 Is this a cabal problem of package problem? I would have expected it to fail immediately instead of discover this problem so far into process. Thanks. -- Darrin

4 9

type constructor confusion
by Stephen Howard 20 Jun '08

20 Jun '08

hello list, loading the code below into ghci gives me this error: HttpMessage.hs:36:20: Not in scope: type constructor or class `HttpRequest' The troublesome line is the definition of the cookie function at the end of the code. I've made HttpRequest and HttpResponse constructors of HttpMessage as I've sketched out an action that might return either one. Given that I thought what I was doing below was defining these as type constructors (maybe I have my vocabulary mixed up) I don't know why ghci is returning this error. module HttpMessage ( HttpMessage(..), HttpRequestMethod(..), cookie ) where import Data.Map as M data HttpHeaders = HttpHeaders !(M.Map String String ) data HttpMessage = HttpRequest { headers :: HttpHeaders, body :: String, request_method :: HttpRequestMethod, uri :: URI, http_version :: Float, cookies :: !(M.Map String String) } | HttpResponse { headers :: HttpHeaders, body :: String, status :: HttpStatus } data HttpRequestMethod = Get | Head | Post | Put | Delete | Unsupported String data HttpStatus = OK | NotFound | NotAuthorized | ServerError data URI = URI { protocol :: String, domain :: String, port :: String, path :: String, query :: !(M.Map String String), anchor :: String } cookie :: String -> HttpRequest -> Maybe String cookie n request = M.lookup n (cookies request)

4 6

Gtk2Hs and GCs
by Tim Newsham 19 Jun '08

19 Jun '08

I don't see a way to fetch an existing standard GC from a widget. In other Gtk bindings I usually do this by fetching the widget's style and then grabbing one of the gc's (such as fg_gc[STATE_NORMAL]). In Gtk2Hs docs: http://www.haskell.org/gtk2hs/docs/gtk2hs-docs-0.9.12/Graphics-UI-Gtk-Gener… it looks like you can only get Colors, not the whole GC. I can't seem to find any other way short of making a whole new GC (as is done in Gtk2Hs/demos/graphic/Drawing.hs). Am I missing something? Is there a reason the GC's arent retrievable? Tim Newsham http://www.thenewsh.com/~newsham/

2 1