- Haskell-Cafe - Haskell.org

llvm-3.0.1.0 installation on Mac
by Luke Evans 08 Apr '13

08 Apr '13

I'm trying to get LLVM 3.0 installed and then have the llvm-3.0.1.0 package install and bind against it with cabal. I'm on a recent 64 bit Mac which shows up in various messages like: "Target platform inferred as: x86_64-apple-darwin" I was advised on the Haskell IRC channel to install a 3.0 LLVM from a location like MacPorts first, then point cabal at the location where LLVM was installed with the appropriate config option. So, I did a "sudo ./port install llvm-3.0", and everything seemed to proceed fine. Then I did: cabal install llvm --configure-option=--with-llvm-prefix=/opt/local/libexec/llvm-3.0 This looks promising (compared to missing the config option), but eventually this happens: Building llvm-base-3.0.1.0... Preprocessing library llvm-base-3.0.1.0... dyld: Library not loaded: /Users/ian/zz64/ghc-7.4.2/libffi/build/inst/lib/libffi.5.dylib Referenced from: /private/var/folders/tz/vlbc03q5333897z201hkmcb40000gn/T/llvm-base-3.0.1.0-3941/llvm-base-3.0.1.0/dist/build/LLVM/FFI/Analysis_hsc_make Reason: image not found running dist/build/LLVM/FFI/Analysis_hsc_make failed (exit code 5) command was: dist/build/LLVM/FFI/Analysis_hsc_make >dist/build/LLVM/FFI/Analysis.hs Failed to install llvm-base-3.0.1.0 I have no idea where "/Users/ian/..." is coming from! My cabal-fu is extremely weak, so I'll plug on and see if I can figure out how cabal's analysis/build might be going wrong here. I'd be very grateful for any clues though! -- Luke

2 6

%==%, :==: and type equality in the singletons library
by Paul Brauner 08 Apr '13

08 Apr '13

Hello, from the output of -ddump-splices I dont think it is the case but I'm asking anyway: is there any way to deduce a ~ b from a :==: b? Given data T = C1 | ... | Cn I can easily derive data EqT :: T -> T -> * where EqT :: a ~ b => EqT a b eqT :: ST a -> ST b -> Maybe (EqT a b) eqT SC1 SC1 = Just EqT ... eqT SCn SCn = Just EqT eqT _ _ = Nothing but this kind of replicates the boilerplate generated by the singletons library for %==%. However I can't see how to leverage %==% to inhabit eqT since I can't deduce a ~ b from a %==% b == STrue. Any idea? If there's no way to write eqT using what singletons generates, wouldn't it make sense for it to generate something that relates :==: and ~ ? Cheers, Paul

2 2

Re: [Haskell-cafe] Prolog-style patterns
by Stephen Tetley 08 Apr '13

08 Apr '13

This is a recurring theme, see also here: http://www.haskell.org/pipermail/haskell-cafe/2009-May/061498.html On 8 April 2013 16:53, Tom Murphy <amindfv(a)gmail.com> wrote: > > Also, for some history, this was discussed a while back: > http://www.mail-archive.com/haskell@haskell.org/msg03721.html >

1 0

My first Haskell program
by John Wood 08 Apr '13

08 Apr '13

Hello, Cafe I'm new to Haskell and the mailing list, and am wondering if I could get some feedback on my first program -- a Markov text generator. The code is posted here:http://codereview.stackexchange.com/questions/24791/haskell-markov-tex… Thanks, John;

2 1

Vector Fabrics is hiring!
by Stefan Holdermans 08 Apr '13

08 Apr '13

[Apologies for multiple postings.] Vector Fabrics is hiring: we are looking for a top-notch programmer to extend our program-analysis and parallelization products. You design and implement algorithms to assist the programmer to create a parallel design from a sequential C or C++ program. You work with our international team of world-class computer scientists and experts in the Haskell / OCaml functional programming languages. Your work is at the forefront of technology, giving you the opportunity to publish your work in major conferences and directly cooperate with processor design companies and domain-specific application vendors. As we are a startup company, you will quickly have a major impact on our products and get to know all aspects of product creation. You will be part of a strongly committed development team and contribute to our agile development process and automated test suites. Interested? Send your CV, GitHub account or other proof of what you can do to jobs(a)vectorfabrics.com. RESPONSIBILITIES * Design and implement software optimization (e.g. parallelization) algorithms for CPUs and GPUs; * Thoroughly test your code, create automated test suites; * Contribute to our agile development planning and process; * Analyze complex customer applications for optimization opportunities and translate this to new analysis algorithms. PROFILE * Your friends and colleagues describe you as a "rockstar" programmer; your programming ability is way above average; * Demonstrable experience in design and implementation of complex software applications; prior experience in functional programming languages is preferred; * You continuously surprise us with your creative yet pragmatic solutions for complex software problems; * You are strongly committed to deliver working software as early as possible; * You work against very high quality standards. Refactoring is your bread and butter, pair-programming is how you prefer to review your code; * Whatever technologies, languages, or development environments you've been using, we expect you have mastered them in depth, and we expect that you will be able to master any technology, language, or development environment that we need in the future; * You are not afraid to get your hands dirty on low-level code, hijack a malloc() call in the standard C library, port the latest gdb debugger to Android on a jail-broken tablet? You get it done. * Excellent command of written and spoken English. EDUCATION MSc, MEng or PhD in Computer Science or significant relevant experience. ABOUT VECTOR FABRICS Vector Fabrics is a high-tech software company, developing tools for embedded multicore programming. Its technology and expertise is getting widespread recognition in the industry as being innovative and unique in their ability to address heterogeneous multicore application-specific silicon platforms. Due to the advanced nature of its tools, Vector Fabrics operates at the forefront of the next generation of embedded platforms for diverse markets ranging from supercomputers to automotive to cell phones. Vector Fabrics puts absolute priority on hiring top class individuals in key positions. Vector Fabrics’ team profile is exceptional and its ambition is to hire only individuals that match or surpass that profile. The company pays top salary and offers a challenging, engaging and stimulating work environment with a high degree of responsibility.

3 2

Compilers book in Haskell
by Andrés Sicard-Ramírez 08 Apr '13

08 Apr '13

Juan, te puede interesar On Sat, Apr 6, 2013 at 5:56 PM, Sergey Bushnyak <sergey.bushnyak(a)sigrlami.eu > wrote: > I will recommend you book "Modern Compiler Design" by Dick Grune and > others. > Besides discussing different topics, authors use Haskell as example for > describing ideas behind compilers for functional language. > -- Andrés

4 5

ANNOUNCE: binary-conduit-1.0
by Alexander V Vershilov 07 Apr '13

07 Apr '13

I'm happy to announce new binary-conduit [1] library, that's a serialization/deserialization library for using binary [2] package. [1] http://hackage.haskell.org/package/binary-conduit [2] http://hackage.haskell.org/package/binary -- Alexander

1 0

ANNOUNCE: antiquoter-0.1.0.0
by L Corbijn 07 Apr '13

07 Apr '13

Hello cafe, I'm happy to announce the release of my first package antiquoter [1], a combinator library for writing quasiquoters and antiquoters. The main aim is to simplify their definitions and reduce copy-and-paste programming. The main feature the current version is trying to solve is code duplication in antiquoters. The problem for writing antiquoters is that it need to be done twice, once for use as an expression and once for use as a pattern. These two implementations are very similar, which leads to copy-and-paste programming. For a good example of what this leads to see the antiquoters of language-c-quote[2], where almost every antiquoter is written twice. A very simple example with the rewritten version is provided in the documentation [3]. Lars [1]:http://hackage.haskell.org/package/antiquoter [2]: http://hackage.haskell.org/packages/archive/language-c-quote/0.7.1/doc/html… [3]: http://hackage.haskell.org/packages/archive/antiquoter/0.1.0.0/doc/html/Lan… [4]:https://github.com/Laar/antiquoter

2 2

Error when building executable with profiling enabled
by Nikolaos Bezirgiannis 07 Apr '13

07 Apr '13

Hello, i have an executable where i want to run profiling on. I do: ghc --make -prof -auto-all -fth Custom.hs but it does not built and the output is: bench\HLogo\Custom\Custom.hs:1:1: Dynamic linking required, but this is a non-standard build (eg. prof). You need to build the program twice: once the normal way, and then in the desired way using -osuf to set the object file suffix. I tried googling on the error with no success. What can I do to compile this file for profiling? Cheers, Nikos

3 3

code-as-config, run-time checks and error locations
by Steffen Schuldenzucker 07 Apr '13

07 Apr '13

Dear Café, I'm working on a EDSL that will include both type checks (at compile time) and semantic checks (at run time). - Semantic properties are known at compile time but feel too complex to me to be encoded in the type system. If one of the runtime checks fails, I'd like to print the location of the error, i.e. not Error: Unknown field `AMOUNT' in table `ENTRIES' (where? why?) but Error: Unknown field `AMOUNT' in table `ENTRIES' Referenced at analysis1.hs:43:7 by `sumByInvoice' which was called at analysis1.hs:66:3 by `main' ENTRIES defined at analysis1.hs:13:8 I'm not yet sure which level of granularity I want for error messages and one can probably get arbitrarily fancy on this. For the moment I think it would be enough to auto-insert the location of calls to a certain set of functions. Any experience on this? Thanks a lot. -- Steffen

3 5