GSoC/GCI Archive
Google Summer of Code 2013

haskell.org

Web Page: http://hackage.haskell.org/trac/summer-of-code/report/1

Mailing List: mailto:haskell-cafe@haskell.org

Haskell.org is the community focus for supporting and promoting the open-source programming language Haskell.

 

Projects

  • Communicating with mobile devices The aim of this project is to develop a server-side library in Haskell for sending push notifications to devices running different OS, such as Android, iOS, Windows Phone, BlackBerry, and so on. And, once we have this libraries, investigate the possibility of mainting a "back and forth" communication between a server and mobile devices and develop a library to handle this.
  • Enhancing Darcsden The goal of this project is to increase the functionality of Darcsden. Darcsden is an open source repository hosting platform for darcs, written in Haskell. The main features are authentication from Github/OpenID, Password Recovery, Editing repository files online, and comparison between a repository and its forks.
  • Extending GHC to support building modules in parallel The aim of this project is to implement native support for building modules in parallel in GHC. This entails making the compilation pipeline thread-safe, and writing a parallel compilation driver next to the existing sequential driver. Focus will be placed on correctness and deterministic output, with speed a latent concern. Nothing in the user's end should change other than there being a new "-j" flag that specifies the number of modules to build in parallel. The cabal-install project should be augmented with the ability to use this new -j flag to speed up builds, alongside its existing package-level parallelization.
  • Haddock extension for Pandoc compatibility Project aiming to extend Haddock to a point where writing reader and writer modules for Pandoc is possible. The general goal is to allow for documentation writing in different formats, including the ever popular Markdown.
  • Haskell Qt Binding Generator This project aims to provide a generation tool that creating Qt bindings automatically for Haskell and make generation as much automation as possible. It parses Qt header files and generates corresponding Qt-style Haskell interfaces (type classes, data types).
  • Improve the feedback of the cabal-install dependency solver The dependency solver can be a mysterious piece of the installation process. I want to give the user the possibility to see what is happening and also give them a better chance to understand what happened if something does not just work. Further, I would like to enable them to use that information to fix the installation in some cases.
  • interactive-diagrams and a paste site with the ability for dynamic rendering of diagrams I want to build an active-diagrams library for compiling diagrams code into active HTML + JS widgets. The diagrams are active in the sense that user can interact with them: for example, a result of type (Bool -> Diagram) should be compiled to a widget that renders a diagram depending on the state of the checkbox. In addition, a pastebin site should be built, that can be used as an interactive scratchpad, where diagrams code can be automatically compiled and the graphical output shown along. This is useful for sharing graphical experiments, teaching beginners and so on. The rendered diagram, together with its interactive capabilities, should be easily embedded in third-party blogs, websites.
  • Overloaded record fields for GHC Haskell's record system lacks support for overloaded field names. This leads to unnecessarily cluttered code and inhibits code reuse. I propose to implement support for overloaded field names and polymorphic record projection as a GHC extension, with the aim to ultimately add them to the language standard in a future revision. This relatively straightforward change would remove a significant source of frustration for Haskell programmers.
  • Parallelise 'cabal build' This project aims to add support for module-level parallel builds to Cabal, a system for building and packaging Haskell libraries and programs. I suggest a novel solution for this problem in the context of GHC/Cabal that, unlike existing approaches, will make a more effective use of multi-core CPUs.
  • Port Charts to use Diagrams Right now the Charts library uses Cairo as backend. Cairo can be difficult to build on platforms other then Linux. Goal of the project would be to port the Charts library to be independent of the used backend. Like that it can support to use Diagrams as backend. Diagrams supports a variety of backends (SVG, Postscript, Cairo (Optional!), Tikz, Gtk), some of which are written platform indepent. As Diagrams is easy to install on every platform supported by Haskell, so will the Charts Library built upon Diagrams. This will make it easy for Haskellers to add charting capabilities to their own applications and libraries in a manner portable across all platforms without any pain.