June 2009 - Haskell-Cafe

Record initialise question
by John Ky 05 Jun '09

05 Jun '09

Hi all, I have some sample code: > full = do > let myOrder = initOrder > { item = Just initItem > { itemId = "Something" > } > , operation = Just Buy > } > putStrLn $ show myOrder > return () This is just a test project, but in a real project, I would have a more elaborate structure. Notice initOrder and initItem. I actually need to know the type of field I am initialising or the name of the corresponding function that provides default values. Is there any kind of polymorphic magic that lets me do this instead? > full = do > let myOrder = init -- [1] > { item = Just init > { itemId = "Something" > } > , operation = Just Buy > } > putStrLn $ show myOrder > return () Where initOrder and initItem are both replaced with just 'init' and the compiler works out from the context (ie. the type of the field) what the types are supposed to be and therefore the actual init function to call? Thanks, -John [1] The second example is silly because there isn't actually any context at this point, but let's pretend there is.

2 1

Umlauts in command line arguments
by Günther Schmidt 05 Jun '09

05 Jun '09

Hi all, When a command line argument contains an umlaut that argument gets garbled. I'm using ghc 6.10.2 on Win XP. Are there any known solutions for this problem? Günther

4 4

Functors and the Visitor Pattern
by Tom.Amundsen 05 Jun '09

05 Jun '09

So, last night, I was having this problem with my Java code where I couldn't figure out for the life of me how to write a piece of code without a big if {} else if {} else if {} ... else {} structure. I was Googling "Java Reflection" to try to determine how to "cast to the most concerete subclass at runtime." Then it dawned on me that what I was trying to do has already been solved by using the Visitor design pattern. Then, after reading the Visitor design pattern page on Wiki, it said that the visitor pattern is essentially an implementation of a functor. Aha! It totally clicked. The Visitor pattern allows you to collect code for similar operations, while spreading apart code for similar objects. Now that really sounds like a functor! Although, now I'm second guessing myself, because I can't figure out how we could create some design pattern that simulates an applicative functor. I'm pretty sure the Visitor pattern doesn't take you this far (but I am willing to be corrected). So, is there a way to create applicative functors in non-functional languages? What would that pattern look like? - Tom -- View this message in context: http://www.nabble.com/Functors-and-the-Visitor-Pattern-tp23851113p23851113.… Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

6 7

GHCI & Curl under Windows
by Martin Huschenbett 04 Jun '09

04 Jun '09

Hi Haskellers, I've installed the newest Haskell Platform under Vista and downloaded a pre compiled version of curl-7.19.4 for MinGW. After changing the build type in curl.cabal to Simple and supplying the needed paths I could even build and install curl for haskell. If I write a program using curl and compile it with GHC everything works fine. But if I try to execute the main function of my program in GHCI I always get the following error message: Loading package curl-1.3.5 ... linking ... <interactive>: C:\Devel\Haskell\curl-1.3.5\ghc-6.10.3\HScurl-1.3.5.o: unknown symbol `__imp__curl_easy_getinfo': unable to load package `curl-1.3.5' Did anybody have the same problem and knows how to fix it? Thanks in advance, Martin.

4 3

RE: Nested Lists
by Paul Keir 04 Jun '09

04 Jun '09

Emil, Felipe, Thanks. I don't know Type Families, but take the point that the input can be parameterised with something something other than a list. i.e. (8 :+: 4 :+: 2 :+: ()) presumably has the same type as (4 :+: 2 :+: ()). My intention was to use common list functions on the sublists, but always then a concat for each level, to return to a flat list. With that in mind I made the following oddity, which in any case doesn't compile due to its use of infinite types. app (f:fs) es = appUp (f:fs) es where len = genericLength (f:fs) appUp [] es = appDown es len appUp (f:fs) es = appUp (map map fs) (f es) appDown es len = appDown (concat es) (len - 1) appDown es 0 = es Henning, I agree with you, a tree would be much better for this. Thanks. -Paul

2 1

Nested Lists
by VoidEx 04 Jun '09

04 Jun '09

You can also try this: bunch n f = takeWhile (not . null) . unfoldr (Just . (f *** id) . splitAt n) > (bunch 8 . bunch 4 . bunch 2) id [1..16] Any way, it's not possible to take list [8, 4, 2], because it's length need to be known at compile time.

1 0

Static link to packages on Hackage?
by John Goerzen 04 Jun '09

04 Jun '09

Hi, I'd like to be able to put a static link to the Haddock docs for the current version of various packages on my homepage. Right now, I can't find any such URL; they all look like: http://hackage.haskell.org/packages/archive/HSH/2.0.0/doc/html/HSH.html I'd like if there could be something like: http://hackage.haskell.org/packages/archive/HSH/latest/doc/html/HSH.html Incidentally, you can click on this "latest" URL but it doesn't go to the correct version. I can't just link to the package page either, because sometimes it won't have docs (such as shortly after I've uploaded a new version). Ideas?

2 2

TASE 2009 - Call for Participation
by CRACIUN F. 04 Jun '09

04 Jun '09

TASE 2009 - CALL FOR PARTICIPATION ********************************************************** * 3rd IEEE International Symposium on * Theoretical Aspects of Software Engineering * (TASE 2009) * 29-31 July 2009, Tianjin, China * http://www.dur.ac.uk/ieee.tase2009 * * Early Registration Deadline : 18 June 2009 * For more information email: IEEE.TASE2009(a)durham.ac.uk ***********************=********************************** TASE 2009 Invited Speakers ========================== Kim G. Larsen, Aalborg University, Denmark Zhong Shao, Yale University, USA Jin-Song Dong, National University of Singapore TASE 2009 Programme ==================== Day 1: 29 July 2009 ------------------- Invited Talk: Verification and Performance Analysis of Embedded Systems Kim G. Larsen (Aalborg University) Session 1 : Real-Time and Embedded Systems Improving Responsiveness of Hard Real-Time Embedded Systems Hugh Anderson (Wellington Institute of Technology) and Siau-Cheng KHOO (National University of Singapore) Environmental Simulation of Real-Time Systems with Nested Interrupts Guoqiang Li (Shanghai Jiao Tong University), Shoji Yuen (Nagoya University) and Masakazu Adachi (Toyota Central R&D Labs. INC.). Semantics for Communicating Actors with Interdependent Real-Time Deadlines Istv¨¢n Knoll (Aalborg University), Anders P. Ravn (Aalborg University) and Arne Skou (Aalborg University). An Efficient Algorithm for Finding Empty Space for Reconfigurable Systems Zhenhua Duan (Xidian University) and Yan Xiao (Xidian University). Session 2 : Semantics State Visibility and Communication in Unifying Theories of Programming Andrew Butterfield (Trinity College Dublin), Pawel Gancarski (Trinity College Dublin) and Jim Woodcock (University of York). Semantics of Metamodels in UML Lijun Shan (National University of Defence Technology) and Hong Zhu (Oxford Brookes University). Refinement Algebra with Explicit Probabilism Tahiry Rabehaja (UNU/IIST) and Jeffrey Sanders (UNU/IIST). Session 3 : Model Checking Environment Abstraction with State Clustering and Parameter Truncating Hong Pan (Institute of Software, Chinese Academy of Sciences), Yi Lv (Institute of Software, Chinese Academy of Sciences) and Huimin Lin (Institute of Software, Chinese Academy of Sciences). Verification of Population Ring Protocols in PAT Yang Liu (National University of Singapore), Jun Pang (University of Luxembourg), Jun Sun (National University of Singapore) and Jianhua Zhao (Nanjing University). Bounded Model Checking of ACTL Formulae Wei Chen (Institute of Software, Chinese Academy of Sciences) and Wenhui Zhang (Institute of Software, Chinese Academy of Sciences). Day 2, 30 July 2009 ------------------- Invited Talk: Modular Development of Certified System Software Zhong Shao (Yale University) Session 4: Specification and Security Coarse Grained Retrenchment and the Mondex Denial of Service Attacks Richard Banach (Manchester University). Specifying and Enforcing Constraints of Artifact Life Cycles Xiangpeng Zhao (Peking University), Jianwen Su (University of California at Santa Barbara), Hongli Yang (Beijing University of Technology) and Zongyan Qiu (Peking University). Consistency Checking for LSC Specifications Hai-Feng Guo (University of Nebraska at Omaha), Wen Zheng (University of Nebraska at Omaha) and Mahadevan Subramaniam (University of Nebraska at Omaha). Integrating Specification and Programs for System Modeling and Verification Jun Sun (National University of Singapore), Yang Liu (National University of Singapore), Jin Song Dong (National University of Singapore) and Chunqing Chen (National University of Singapore). Session 5 : Software Testing I A Framework and Language Support for Automatic Dynamic Testing of Workflow Management Systems Gwan-Hwan Hwang (National Taiwan Normal University), Che-Sheng Lin (National Taiwan Normal University), Li-Te Tsao (National Taiwan Normal University), Kuei-Huan Chen (National Taiwan Normal University) and Yan-You Li (National Taiwan Normal University). Fault-based Test Case Generation for Component Connectors Bernhard Aichernig (Graz University of Technology), Farhad Arbab (CWI), Lacramioara Astefanoaei (CWI), Frank de Boer (CWI), Meng Sun (CWI) and Jan Rutten (CWI). Test Data Generation for Derived Types in C Program Zheng Wang (East China Normal University), Xiao Yu (East China Normal University), Tao Sun (East China Normal University), Geguang Pu (East China Normal University) and Zuohua Ding (Zhejiang Sci-Tech University). Session 6 : Software Models Program Repair as Sound Optimization of Broken Programs Bernd Fischer (University of Southampton), Ando Saabas (Tallinn University of Technology) and Tarmo Uustalu (Tallinn University of Technology). Modeling Web Applications and Generating Tests: A Combination and Interactions-guided Approach Bo Song (Shanghai University) and Huaikou Miao (Shanghai University). Merging of Use Case Models: Semantic Foundations Stephen Barrett (Concordia University), Daniel Sinnig (Concordia University), Patrice Chalin (Concordia University) and Greg Butler (Concordia University). Day 3, 31 July 2009 ------------------- Invited Tutorial: Towards Expressive Specification and Efficient Model Checking Jin Song Dong (National University of Singapore) Session 7 : Verification Verifying Semistructured Data Normalization using SWRL Yuan Fang Li (University of Queensland), Jing Sun (University of Auckland), Gillian Dobbie (University of Auckland), Scott Uk-Jin Lee (University of Auckland) and Hai H. Wang (Aston University). Verifying Self-stabilizing Population Protocols with Coq Yuxin Deng (Shanghai Jiao Tong University) and Jean-fran?ois Monin (Universit¨¦ de Grenoble 1). The Logical Approach to Low-level Stack Reasoning Xinyu Jiang (University of Science and Technology of China), Yu Guo (University of Science and Technology of China) and Yiyun Chen (University of Science and Technology of China). Constructing Program Invariants via Solving QBF Shikun Chen (National University of Defence Technology), Zhoujun Li (Beihang University) and Mengjun Li (National University of Defence Technology). Session 8 : Concurrency Using Architectural Constraints for Deadlock-Freedom of Component Systems with Multiway Cooperation Moritz Martens (University of Mannheim) and Mila Majster-Cederbaum (University of Mannheim). Formal Reasoning about Concurrent Assembly Code with Reentrant Locks Ming Fu (University of Science and Technology of China), Yu Zhang (University of Science and Technology of China) and Yong Li (University of Science and Technology of China). Algorithms for Computing Weak Bisimulation Equivalence Weisong Li (Institute of Software, Chinese Academy of Sciences). Session 9 : Software Testing II Interpreting a Successful Testing Process: Risk and Actual Coverage Marielle Stoelinga (University of Twente) and Mark Timmer (University of Twente). Automated Test Case Generation based on Coverage Analysis Tim A. Majchrzak (University of Muenster) and Herbert Kuchen (University of Muenster). Exploring Topological Structure of Boolean Expressions for Test Data Selection Lian Yu (Peking University), Wei Zhao (IBM China Research Lab), Xiangdong Fan (Peking University) and Jun Zhu (IBM China Research Lab). On Testing 1-Safe Petri Nets Guy-Vincent Jourdan (University of Ottawa) and Gregor von Bochmann (University of Ottawa). The programme also include two poster sessions, comprising of 23 selected poster presentations. This list of posters can be found at http://www.dur.ac.uk/ieee.tase2009/

1 0

understanding regex libs
by Simon Michael 04 Jun '09

04 Jun '09

Hi Chris.. thanks for your extensive work on haskell regex libs. I'm looking for a good way to make my regex-using app more portable to windows. I couldn't figure out the difference between the regex-pcre and regex- pcre-builtin on hackage. Could you clarify ? Best regards, -Simon

3 4

Text formatting question
by Evan Klitzke 04 Jun '09

04 Jun '09

I'm writing code with hslogger, and I'm finding myself frequently writing things like this: infoM $ printf "%s saw %s with %s" (show x) (show y) (show z) On #haskell I asked about why you can't have use %s with any instance of (Show a), and augustss, the Text.Printf maintainer, pointed out that it's not possible in Haskell 98 because of overlapping instances. So I'm trying to figure out the best way to work around this in my code. Some of the logging statements can get kind of long, so the ability to elide the calls to `show` would be nice. So what's the best way to do this? I was thinking of making my own variant of printf that only accepted the %s formatter and took (Show a)'s as arguments, but there might be a simpler way? Another thing is the code is already fairly GHC specific at this point, so if there are GHC extensions that might be useful here that I don't know about, I'm interested in hearing about them. Thanks. -- Evan Klitzke <evan(a)eklitzke.org> :wq

2 1