Custom Query (33 matches)
| Ticket | Summary | Priority | Status | Owner | Topic | Difficulty |
|---|---|---|---|---|---|---|
| #25 | GSL Binding | -- | assigned | johan_silver | Bindings | unknown |
| description |
Extend the GSLHaskell library to cover all the GSL functions. Implement (possibly using additional numerical libraries) important Octave functions not available in the GSL. Interested Mentors
Interested Students
|
|||||
| #72 | GStreamer bindings | -- | new | none | Bindings | unknown |
| description |
GStreamer is a multimedia framework. The goal would be to create haskell bindings, to allow easy creation of multimedia applications in haskell. Interested Mentors
Interested Students
|
|||||
| #1103 | COM interop library and IDL compiler for Haskell | -- | new | none | Bindings | 1 person Summer |
| description |
Quite often someone from the mailing lists is asking how to bind Haskell to COM. HaskellDirect? is a tool that allows direct translation from IDL to Haskell. The problem is that currently it is seriously ill. The generated code is often full of both runtime and compile errors. The generated code is using a COM interop library which was developed before the current FFI addendum and doesn't conform to it very well. I have developed an initial replacement for the library that I intend to use in Visual Haskell. The aim of this project would be to implement an IDL to Haskell translator for it. |
|||||
| #28 | Bioinformatics tools | -- | new | none | Bioinformatics | unknown |
| description |
1. Further develop RBR, a tool for masking repeats. This can include a) optimize (using FastPackedString and/or a new internal data structure); b) extend functionality. For more details, click here. 2. Develop a tool for annotation/classification of sequences. This would involve computation on and visualization of graphs (experience with the latter would be really great). For more details, click here. Prior bioinformatics knowledge is not a requirement. Please contact me for details. Interested Mentors
Interested Students
|
|||||
| #1121 | bio library development | -- | new | none | Bioinformatics | unknown |
| description |
The "bio" package is a collection of useful functionality aimed at bioinformatics. Its development has been largely driven by the immediate needs of the applications that use it, and its current contents reflect this. Ideally, this would develop into a general, broadly scoped bioinformatics library (akin to bioperl and biopython). The library can be extended in many directions, and this will to a large part be dictated by student interests and background. Some possibilities are:
Application-driven library development may be a useful route, so library development as part of solving concrete (biological or otherwise) problems is welcome. If this sounds interesting, please e-mail me at <ketil at malde dot org> to discuss the details. |
|||||
| #80 | BSPHlib - A parallel programming library based on BSP model | -- | new | none | Concurrency | unknown |
| description |
Implementation of a different aproach to parallel programming in Haskell, based on BSP model and using the MPI library for message passing, instead of PVM. The great advantage of this aproach is that the BSP model have an easy efficiency prediction. BSP Reference: http://www.bsp-worldwide.org/ MPI Reference: http://www.mpi-forum.org Interested Mentors
Interested Students
|
|||||
| #33 | Port HaskellDB to HList | -- | new | none | Databases | unknown |
| description |
HaskellDB currently uses its own record system, which (I believe) is less powerful than HList. The project is to port HaskellDB to use HList instead, making any necessary changes to the interface to make this possible and to fit HList better. Interested Mentors
Interested Students
|
|||||
| #35 | Add support for optimization features in HaskellDB | -- | new | none | Databases | unknown |
| description |
The projects is to added support for indexes, prepared statements and other optimization features to HaskellDB. Interested Mentors
Interested Students
|
|||||
| #70 | integrate searchpath and ghc | -- | new | none | GHC | unknown |
| description |
Searchpath currently operates as a wrapper around ghc and ghci. This means that a :r in ghci does not attempt a full reload from the internet. It would be nice if :r reloaded needed modules from the internet. For more info on searchpath, see http://searchpath.org. |
|||||
| #55 | Improvements to the INBlobs tool developed at U. Minho | -- | new | none | Graphics | 1 person Summer |
| description |
We propose the following improvements to the INblobs tool developed at U. Minho: - To fix graphical portability problems of INblobs. This implies a few tricks with wxHaskell and the wx toolkit. - Adding new features to the tool, namely:
Interested Mentors
Interested Students
|
|||||
| #77 | 3D GUI system and widget library | -- | new | none | Graphics | unknown |
| description |
A framework for writing 3D, skinnable GUIs in Haskell (simmilar to some of the projects at http://www.haskell.org/haskellwiki/Libraries_and_tools/GUI_libraries). The system will be intended for use in some games, CAD/CAM applications, 3D art tools, and any other program that needs a heavy-duty GUI in a primarly 3D environment. It could also be potentially used (if only for inspiration) in an OSS response to the current trends in 3D desktop composition engines. The goals for the Summer of Code 2006 will be:
Interested Mentors
Interested Students
|
|||||
| #85 | Haven vector graphics | -- | new | none | Graphics | unknown |
| description |
Haven is a vector grapics library for Haskell, currently implemented via the Java 2D rendering API. While potentially very useful, Haven is also unmaintained, undocumented, and tricky to get working. Problems have also cropped up with the toolchain it depends on. For example, GCJNI hasn't had a release in several years and changes to Cygwin have made it tricky to compile on Windows platforms. Goals for this project would be to:
Implementing the PDF generator would give this library functionality very comparable to ReportLab and PDF::Writer, two libraries for Python and Ruby respectively. These libraries have a very similar API. A major annoyance of mine is that points are passed as two floating-point arguments, preventing the composition of functions. Haven's API is quite a bit different. Much effort was put into making Haven's API compositional, above and beyond this annoyance. Interested Mentors
Interested Students
|
|||||
| #43 | JHC Hacking | -- | new | none | JHC | unknown |
| description |
Implement a new feature in JHC, depending on interests and skills. e.g. MPTC+FD, fast arrays, STM, Template Haskell, open datatypes, nice records, or something that you are interested in. Interested Mentors
Interested Students
|
|||||
| #1108 | .NET CLR back end for jhc | -- | new | none | JHC | unknown |
| description |
jhc currently has front end support for .NET, accepting hugs98.NET compatable foreign declarations and allowing selection of .NET as a target, there is just no .NET back end. one can be written to either convert core or grin to .NET. jhc core is similar enough to ghc core that there might be relevant literature, but grin -> imperitive code would be a much shorter hop. Interested mentors* John Meacham Interested students* Aaron Tomb <atomb@…> |
|||||
| #30 | Haskellnet | -- | new | none | Networking | unknown |
| description |
We have got cgi, ftp, http, and irc. Get them into shape in the hierarchical libraries as well as adding a number of other protocols, like nntp, smtp and pop3, imap4, ... much like the Ocamlnet project. Interested Mentors
Interested Students
|
|||||
| #1117 | SNMP MIB compiler using Parsec | -- | new | none | Networking | 2 people, 2 weeks |
| description |
Write a SMIv1, SMIv2 SNMP MIB compiler using the Parsec combinator. |
|||||
| #61 | Pugs - Fast Mutable Collection Types. | -- | new | none | Pugs | 2 people, 2 weeks |
| description |
Currently, Haskell lacks production-grade bindings for mutable data structures, and has to rely on immutable data structures for storage. For example, a sparse mutable array may be emulated by (IORef IntMap), and a mutable set may be emulated by (IORef Set). However, these data structures have to be reconstructed almost from scratch for each insertion/deletions, resulting in low performance. The Pugs project implements Perl 6 on top of Haskell; we would need fast implementations for mutable sparse array, mutable ordered mappings, mutable resizable byte-buffers, as well as mutable string-indexed Hash tables. One approach is to write a binding to Judy (judy.sf.net) or other C-based libraries, and present an API conformant to the existing Haskell libraries (e.g. MArray). The result of this effort will be reusable to other Haskell projects as well. Interested Mentors
Interested Students
|
|||||
| #62 | Pugs - Embedding Interface | -- | new | none | Pugs | 2 people, 2 weeks |
| description |
Pugs is a Perl 6 implementation; as such, it can be used as a dynamic scripting language for Haskell- and C-based applications, for example Yi or hIDE. Currently, Pugs has support for "Opaque" objects, using existential types that allows the embedder to define coercion functions to convert from/to Perl 6 values. We also have a C-based embedding API that provides "eval" and "apply" of Perl 6 code, based on FFI for foreign export. However, the embedding API for running and introspecting the interpreter is currently ill-defined; coming up with a sane API for them will allow Perl 6 be easily embedded as a Haskell-native scripting language, as well as reusable for C-based applications such as Emacs or Vim. Interested Mentors
Interested Students
|
|||||
| #63 | Pugs - Debugger and Profiler | -- | new | none | Pugs | 2 people, 2 weeks |
| description |
Pugs is a Perl6 implementation in Haskell, with a reference interpreting environment for the Perl 6 AST. However, although Pugs internals support pluggable evaluators, only one exists to date: it performs straightforward tree-reduction rules on the Perl 6 AST, optionally with tracing information. Implementing an alternate evaluator with a minimal small-step debugger can simplify Perl 6 programming a lot; the same infrastructure can be used to write a simple line- or function-based profiler as well. Interested Mentors
Interested Students
|
|||||
| #68 | Pugs - Foreign Function Interface | -- | new | none | Pugs | 2 people, 2 weeks |
| description |
Pugs is a Perl 6 implementation written in Haskell. Currently Pugs has support for using Perl 5 and Haskell libraries, but not for other foreign languages like C. Implementing the ability to use C libraries with Pugs would open up new possibilities in Perl 6 programming. Interested Mentors
Interested Students
|
|||||
| #76 | Darcs project management web application | -- | new | none | Systems | unknown |
| description |
A web based project management application for Darcs that would honor it's distributed and patch based nature. There exist projects to make Trac work with Darcs, but the suggester of the project (and the first interested student) feels a system that supports every aspect of Darcs (unrecord and unpull) in addition to designing the system to match Darcs nature will be beneficial. Implementation language would not necessarily be Haskell. Ruby was suggested. Interested Mentors* ? Interested Students* Eivind Uggedal (rfsu) <uggedal@…> * Juan José Olivera Rodríguez <jotajota@…> |
|||||
| #78 | Graphical type analysis | -- | new | none | Systems | 1 person Summer |
| description |
Type errors can be frustrating to beginner and intermediate programmers in haskell. Indeed, with its powerful type system and baffling compiler output, finding type errors for programs compiled in ghc can be a task which can be extremely tedious and time consuming. As a teaching assistant, I have seen first hand the exasperation of students when trying to find their errors. I propose to let the programmer see which types have been assigned to what parts of the code graphically in Yi. A user would simply have to point to the piece of code he wants to scrutinise and all the type information would be instantly displayed. Furthermore, the exact location of the typing incongruities would be quickly displayed in red. This would involve:
Interested Mentors
Interested Students
|
|||||
| #57 | A lojbanic parser for haskell | -- | new | none | Tools | unknown |
| description |
Haskell Lojban tools. Interested Mentors
Interested Students
|
|||||
| #1104 | GuiHaskell, to superceed WinHugs | -- | new | none | Tools | 1 person Summer |
| description |
I started a GuiHaskell? project some time ago ( http://www-users.cs.york.ac.uk/~ndm/projects/guihaskell.php), the goal of this project is to write a replacement for WinHugs? but with several important enhancements:
The project was blocked on a missing Gtk2Hs feature, which is now present. The initial prototype is capable of executing code as a proof of concept, but the vast majority of the code has not yet been written. The project would aim to achieve at the very least:
And it would be cool if the student looked towards adding the following features (at least a few would be expected, all would be great!):
The aim is NOT to write a Haskell text editing widget that you can use to edit your code, but to do everything else that might be handy in an IDE. Neil Mitchell offers to mentor (and wrote the current version of WinHugs?, and the GuiHaskell? prototype) Interested Students
|
|||||
| #19 | Continuations-based DSL on top of HAppS | -- | new | none | Web Development | unknown |
| description |
Do you have a vision how to do better than WASH? Integrate continuation based interaction with client or use something like Functional Forms for the interaction. How to best to interact with XML etc. Other HAppS related projects also possible. Interested Mentors
Interested Students
|
|||||
| #53 | Tie HAppS with SQL databases | -- | new | none | Web Development | 1 person Summer |
| description |
HAppS is a web framework offering transactional semantics. Tying it to SQL transactions is an interesting exercise requiring some knowledge in haskell database bindings (implementing two phased commit for different databases) and extending HAppS MACID to handle the two-phased commit. Interested Mentors
Interested Students
|
|||||
| #1112 | A lightweight web framework | -- | new | none | Web Development | unknown |
| description |
It'd be great to have a lightweight web framework like Rails that makes creating web apps simple. My suggestion would be to fork HAppS (since that's already an established platform, which has been fine-tuned for speed), get rid of most of the useless stuff in HAppS.Protocols.*, and add some generics for dealing with deletion of objects and so on. Interested Students
|
|||||
| #51 | XMPP (aka Jabber) bindings for Haskell | -- | new | none | misc | unknown |
| description |
Support for the XMPP ( http://www.xmpp.org, http://www.jabber.org) will enable us to write IM client, message and file transfer tools, notifiers, IM bots and otherwise join an XMPP frenzly that seems to engulf the whole world :) Interested Mentors
Interested Students
|
|||||
| #69 | Implement a model checker | -- | new | none | misc | unknown |
| description |
jhypernets is a simulator of a formal model of mobile computations called "Petri hypernets", with a core implemented in Haskell. The model supports model checking (automatic checking of satisfiability of logical formulas). At the moment there is a very simple model checker implemented in Haskell which is too slow for using it for non-trivial models. The idea is to implement a fast model checker in Haskell using Symbolic Model Checking. It requires implementing or creating bindings to a BDD (Binary Decision Diagrams) library, implementing the model checker itself (which is a non-standard one) and possibly strong profiling of code. The project could be a proof of a hypothesis "Haskell is a great tool for implementing systems specified formally". The hypothesis is partially proved at the moment (Haskell component is a very successful part of the project), but creating a model checker is a very implementation-speed-sensitive and we do not know whether we can have the high-levelness of Haskell and speed of imperative languages at the same time (in the domain of model checkers). Interested Mentors
Interested Students
|
|||||
| #1113 | New I/O library (async I/O+unicode filenames+filesystem manipulations+...) | -- | new | Bulat | misc | unknown |
| description |
Existing ghc i/o library is hard to extend, closely coupled with ghc rts, can't be ported to other haskell compilers. it will be great to detach new i/o library, based on code of current ghc i/o, streams, SSC, network-alt, fps, filepath and other libs. detailed explanation at http://haskell.org/haskellwiki/Library/IO Interested Mentors
Interested Students
|
|||||
| #1118 | Debian package autobuilder for hackage packages | -- | new | none | misc | unknown |
| description |
Creating an infrastructure for automatically generating Debian source packages for Hackage packages, building them, and publishing the resulting binaries in a publicly-accessible APT archive. Interested Mentors
Interested Students
|
|||||
| #1126 | MIME library | -- | new | none | misc | unknown |
| description |
The goal is to create the one, true MIME library for Haskell. Currently, there are a lot of partial MIME libraries, but nothing really complete. For more information, see http://www.haskell.org/pipermail/haskell-cafe/2007-March/023598.html and http://www.haskell.org/haskellwiki/Libraries_and_tools/MIMEStrikeForce. |
|||||
| #1540 | Debugger for Attribute Grammar using Haskell | -- | new | Yogesh Mali | misc | unknown |
| description |
I intended to design a debugger for attribute grammar specification language.The plan is to finish this project in summer, but i give myself some flexibility of time till summer plus two more months.My current work focuses on attribute grammar specification language,silver which has been designed using Haskell. Silver is used for language design and extend them with new domain specific features. |
|||||
