January 2015 - Haskell-Cafe

ANNOUNCE: picoparsec-0.1
by Mario Blazevic 13 Jan '15

13 Jan '15

The newly released package picoparsec [1] is a fork of attoparsec that works with a much wider variety of input types than just ByteString and Text. Parsers written with Picoparsec will generally perform a bit slower than Attoparsec on strict ByteString and Text, somewhat faster than Attoparsec on lazy Text, and much faster than Parsec on any of the mutually-supported input types. That being said, the main point of the library is to enable the writing of generic parsers that are not constrained to any single input type. Depending on the parser primitives used, Picoparsec will take any input whose type is an instance of either NullMonoid, LeftGCDMonoid, FactorialMonoid, or TextualMonoid. These are some of the subclasses of Monoid defined in the monoid-subclasses package [2]. Some of the input types that are fully supported out of the box would be: - Text - String - Seq Char - Vector Char - ByteStringUTF8, a thin newtype wrapper around ByteString - Stateful (Map String a) Text, if you need to parse Text and keep track of the the symbol table state - OffsetPositioned String, in case you need to keep track of the current input position - LinePositioned Text, to keep track of the current line and column This release can be seen as the belated completion of the work described in my Haskell Symposium 2013 presentation [3]. The name Picoparsec seemed to fit in just the right place between Parsec and Attoparsec, especially since Nanoparsec [4] has already been taken. [1] http://hackage.haskell.org/package/picoparsec [2] http://hackage.haskell.org/package/monoid-subclasses [3] https://github.com/blamario/monoid-subclasses/wiki/Files/HaskellSymposium20… [4] http://hackage.haskell.org/package/nanoparsec

1 0

Vacancies: 4 PhD students Software Technology Utrecht University
by Johan Jeuring 13 Jan '15

13 Jan '15

============================================================ 4 x PhD position in Software Technology ============================================================ The research group of Software Technology is part of the Software Systems division of in the department of Information and Computer Science at the Utrecht University. We focus our research on functional programming, compiler construction, program analysis, validation, and verification. Financed by the Netherlands Organisation for Scientific Research (NWO), the EU, Technology Foundation STW and Utrecht University we currently have job openings for four PhD researchers (PhD students) in Software Technology. We are looking for PhD students to work on some of the following topics: * Version control of structured data The theory and practice underlying structure-aware version control systems capable of handling more than just text files. * Intelligent tutoring technologies Technologies for tutoring subjects such as functional programming, statistics, algebra, etc. * Serious games Domain-specific languages and technologies for specifying strategies for serious games. * iTasks iTasks is a formalism for specifying distributed tasks. Specify and test properties of iTasks, and give run-time feedback for iTasks. Besides research, the successful candidate will be expected to help supervise MSc students and assist teaching courses. We prefer candidates to start no later than September 2015. --------------------------- What we are looking for --------------------------- The candidate should have an MSc in Computer Science, be highly motivated, speak and write English well. Furthermore the successful candidate should be proficient in reporting scientific findings. Knowledge of and experience with at least some of the following areas (depending on the topic chosen to work on for your PhD) is essential: * functional programming, such as Haskell or ML; * datatype generic programming; * intelligent tutoring systems; * serious games; * strategies, rewriting, parsing; * modern version control systems such as git, mercurial, or darcs. ------------------ What we offer ------------------ The candidate is offered a full-time position for 4 years. A part-time position of at least 0.8 fte may also be possible. Salary starts at 2083 euro and increases to 2664 euro gross per month in the fourth year of the appointment. The salary is supplemented with a holiday bonus of 8% and an end-of-year bonus of 8.3% per year. In addition we offer: a pension scheme, partially paid parental leave, flexible employment conditions. Conditions are based on the Collective Labour Agreement Dutch Universities. The research group will provide the candidate with necessary support on all aspects of the project. ---------------- How to apply ---------------- To apply please attach a letter of motivation, a curriculum vitae, and (email) addresses of two referees. Make sure to also include a transcript of the courses you have followed (at bachelor and master level), with the grades you obtained, and to include a sample of your scientific writing, such as your master thesis. It is possible to apply for this position if you are close to obtaining your Master's. In that case include a letter of your supervisor with an estimate of your progress, and do not forget to include at least a sample of your technical writing skills. Application deadline is March 8, 2015. You can apply online through the University's website: http://tinyurl.com/qhlco6s --------------------------- Additional information --------------------------- If you have any questions regarding these positions, please contact Johan Jeuring +31 (0)640010053 J.T.Jeuring(a)uu.nl Wouter Swierstra +31 (0)30 253 9207 w.s.swierstra(a)uu.nl

1 0

May I have some examples on Kinds and Data.Type.Equality?
by Magicloud Magiclouds 12 Jan '15

12 Jan '15

Hi, I am trying to learn some advanced stuff of Haskell. But haskell wiki seems only giving some concepts and so. May I have some code example? Also other documents are welcomed. Thanks. -- 竹密岂妨流水过山高哪阻野云飞 And for G+, please use magiclouds#gmail.com.

2 1

Re: [Haskell-cafe] Typesafe API calls
by Fabian Bergmark 12 Jan '15

12 Jan '15

I have uploaded a new version of the library https://hackage.haskell.org/package/apis-0.0.1 I'have included 15 more APIs from Facebook and Dropbox. I have also uploaded documentation to Hackage. I have been working on creating an autotools installation script that builds a makepkg package with the FFI bindings for Python and Java. I have tried it on Arch and it seems to work. If you want to try it out, check out the github repository. Building on top of this, Im working on a library that uses Pipes to create producers (triggers) and consumers (actions), allowing one to create effects (recipes) of the kind, when I am tagged in a photo on Facebook, notify me via email. This very much resembles IFTTT but with static and typesafe compositions. Im experimenting with using my javascript interpreter to allow for more dynamic compositions. 2014-12-18 21:59 GMT+01:00 Jonathan Paugh <jpaugh(a)gmx.us>: > Hello Fabian, > > I have been dreaming of something like this for Haskell since I wrote a > (non type-safe) python library to access any REST API generically[1]. > This comes from a different (and more ambitious) angle than I had imagined. > > I don't have time to test this just now, but I will keep it bookmarked. > Thanks for sharing your hard work! > > Regards, > Jonathan > > [1]: http://github.com/jpaugh/agithub (unmaintained, but maybe interesting) > > On 12/12/2014 07:01 AM, Fabian Bergmark wrote:> I just released a > library on hackage >> (https://hackage.haskell.org/package/apis-0.0.0) The goal of the >> library is to provide typesafe API calls to a large amount of existing >> APIs. By typesafe I mean that, for each API the library provides two >> types, Input and Output, and a function :: Input -> IO Output. >> Internally the library knows how to serialize Input to make an HTTP >> request and how to deserialize the response into Output. >> >> Whats cool about this library is that it uses Template Haskell to >> generate the API interfaces. A combination of Open Data Tables, an >> internal YQL engine and JSON Schemas, all written in Haskell makes >> this possible. >> >> Right now only two APIs are included (Swedish weather data and Swedish >> public announcement), but more APIs are soon to follow. >> >> I have also implemented FFI exports of all types and functions so >> that, in combination with SWIG, this library can be used in almost any >> language. So far I have tried Java and Python with success. >> >> If anyone would like to try it and respond with feedback it would be >> greatly appreciated! If you want any specific API supported, I can try >> to import that right away >> >> Fabian Bergmark >> >

1 1

why can it never go smoothly, language-c-0.4.7 edition
by briand＠aracnet.com 12 Jan '15

12 Jan '15

Hi, setup-Simple-Cabal-1.22.0.0-x86_64-linux-ghc-7.8.4: The program 'happy' is required but it could not be found. so i installed happy. then it wanted alex. which i was also able to install. and then language-c-0.4.7 successfully installed. I have seen this sort of thing many times across all sorts of different packages. So my question is : are these dependency bugs that should be reported to the package maintainer ? or is that certain packages can't be built as a dependency target for some reason ? Thanks, Brian

2 1

ghc-mod split feature
by Tobias Pflug 11 Jan '15

11 Jan '15

Hi, i just wanted to try some of the ghcmod features that I haven't used before in order to hopefully add them to my vim haskell workflow somehow. What I apparently can't get to work is the split command: $ ghc-mod split src/Foo.hs Foo 47 7 which points to exactly the code listed in the ghc-mod help: f :: [a] -> a f x = _ It does however not yield any output at all. Am I missing something obvious here ? I'm using version 5.2.1.1 Thanks, Tobi

2 1

new haskell music track: Mother Is Waiting
by Anton Kholomiov 11 Jan '15

11 Jan '15

Dear Haskellers I'd like to present yet another piece of music made in Haskell. https://soundcloud.com/anton-kho/mother-is-waiting It's very soothing piece that can be used as a helper for dreaming or meditation. It's dedicated to my friend. She's waiting for her second child to be born. This time it contains no samples. All sounds are made out of waves and filters. So I can attach the source code right in the mail. The track relies on new versions of the packages `csound-expression`, `csound-sampler` and `csound-catalog`. So you can do > cabal update > cabal install csound-sampler csound-catalog Ad the we can load the file in the ghci and run the main function. > ghci Mother.hs > main and the Csound should be installed to render the synths. Cheers, Anton

4 3

The amount of CPP we have to use is getting out of hand
by Johan Tibell 09 Jan '15

09 Jan '15

Hi, (This was initially written as a Google+ post, but I'm reposting it here to raise awareness of the issue.) The amount of CPP we have to use in Haskell is getting a bit out of hand. Here are the number of modules, per library, that use CPP for some of the libraries I maintain: containers 18/18 hashable 4/5 unordered-containers 6/9 network 3/7 cassava 4/16 cabal/cabal-install 13/75 cabal/Cabal 7/78 ekg 1/15 If this doesn't look like a lot to you (I hope it does!) consider than some languages don't use CPP at all (e.g. Java). CPP really sucks from a maintenance perspective: * It's not Haskell, but this bizarre string concatenation language. * The code is harder to read, bitrots more easily, and is harder to test. * The code can't be compiled without using Cabal (which generates some of the CPP macros for us.) This hurts e.g. ad-hoc testing/benchmarking. There are a couple of reasons we use CPP, but the main one is breaking changes in GHC and libraries we depend on. We need to reduce these kind of breakages in the future. Dealing with breakages and maintaining the resulting CPP-ed code is costing us time we could spend on other things, such as improving our libraries or writing new ones. I for one would like to get on with writing applications instead of spending time on run-of-the-mill libraries. Often these breaking changes are done in the name of "making things cleaner". Breaking changes, no matter how well-intended, doesn't make code cleaner, it makes it less clean*. Users end up having to use *both* the old "unclean" API *and* the new "clean" API. The right way to move to evolve an new API is to add new functions and data types, not modify old ones, whenever possible. * It takes about 3 major GHC releases (~3 years) before you can remove the CPP, but since new things keep breaking all the time you always have a considerable amount of CPP. -- Johan

9 10

PSA regarding Cabal-1.22 and ghc-mod
by Anthony Cowley 09 Jan '15

09 Jan '15

I was lucky to avoid this myself, but I keep seeing it crop up: ghc-mod doesn't work with GHC < 7.10 and Cabal-1.22. If you are still on GHC 7.8.3 or 7.8.4, you may want to hold off on updating cabal or cabal-install. This isn't really my news to report, but folks need to know. See this issue for more information: https://github.com/kazu-yamamoto/ghc-mod/issues/417#issuecomment-68819938 Anthony

1 0

Re: [Haskell-cafe] Installing ihaskell on nixos
by info＠rotnetix.com 08 Jan '15

08 Jan '15

Thank you for the tip. I did that but then the ihaskell notebook failed to actually load haskell. so I added the following to the config file: c = get_config() c.KernelManager.kernel_cmd = [exe, 'kernel', '{connection_file}'] c.Session.key = b'' c.Session.keyfile = b'' # Syntax highlight properly in Haskell notebooks. c.NbConvertBase.default_language = "haskell" # Where to look for templates. template_path = "/".join(__file__.split("/")[:-1] + ["templates"]) c.TemplateExporter.template_path = [template_path] And every time I run the ihaskell notebook / console it also adds: exe = '/nix/store/hlkvkavkgm3hj08ffdlziq30r7zyr060-haskell-ihaskell-ghc7.8.3-0.4.3.0-shared/bin/IHaskell'.replace(' ', '\\ ') So I now have a whole list of these at the top. However now the kernel keeps crashing with : 2015-01-09 03:47:51.673 [NotebookApp] KernelRestarter: restarting kernel (1/5) WARNING:root:kernel 0d8057d4-b51e-4acc-a262-ceee9a84edee restarted IHaskell: Failed to load interface for ‘IHaskell.Display’ Use -v to see a list of the files searched for. Failed to load interface for ‘IHaskell.Display’ Use -v to see a list of the files searched for. Failed to load interface for ‘IHaskell.IPython.Stdin’ Use -v to see a list of the files searched for. Any ideas? On Thursday, January 8, 2015 at 8:09:03 PM UTC+11, Utku Demir wrote: > > I reproduced the issue, had ~/.ipython/profile_haskell directory, but not > ipython_config.py inside it after installation. > > I just created an empty file with `touch > ~/.ipython/profile_haskell/ipython_config.py` and then iHaskell started > successfully. > > On Thu Jan 08 2015 at 10:07:36 EET <in...(a)rotnetix.com <javascript:>> > wrote: > >> I am trying to set up ihaskell on EC2. I start with the NixOs EC2 >> instance (from http://nixos.org/nixos/download.html) and then basically >> install the ihaskell package using nix-env -i ihaskell. The installation >> runs fine but when I try to start the notebook with ihaskell notebook i get >> the following: >> >> Updating IPython profile. >> Loading profile from >> /nix/store/hlkvkavkgm3hj08ffdlziq30r7zyr060-haskell-ihaskell-ghc7.8.3-0.4.3.0-shared/share/x86_64-linux-ghc-7.8.3/ihaskell-0.4.3.0/profile/profile.tar >> IHaskell: >> Ran commands: >> mkdir -p /root/.ihaskell >> mkdir -p /root/.ihaskell/notebooks >> mkdir -p /root/.ihaskell >> /nix/store/hnmgn0jrpzlp5d91m5xawpr2iqsjxh7j-ipython-2.3.1/bin/ipython >> locate profile haskell >> which >> /nix/store/hnmgn0jrpzlp5d91m5xawpr2iqsjxh7j-ipython-2.3.1/bin/ipython >> /nix/store/hnmgn0jrpzlp5d91m5xawpr2iqsjxh7j-ipython-2.3.1/bin/ipython >> locate profile haskell >> which >> /nix/store/hnmgn0jrpzlp5d91m5xawpr2iqsjxh7j-ipython-2.3.1/bin/ipython >> >> Exception: /root/.ipython/profile_haskell/ipython_config.py: openFile: >> does not exist (No such file or directory) >> >> I poked around a bit and the root/.ipython/profile_haskell/ contains none >> of the expected profile files... Could not find anything by googling >> around except the suggestion to run the ihaskell setup command which seems >> to have been discontinued >> >> Can anyone suggest what I could try? >> >> Thanks >> Riaan >> >> >> _______________________________________________ >> Haskell-Cafe mailing list >> Haskel...(a)haskell.org <javascript:> >> http://www.haskell.org/mailman/listinfo/haskell-cafe >> >

1 0