Changes between Version 7 and Version 8 of Status/Oct13


Ignore:
Timestamp:
Oct 10, 2013 5:25:09 PM (2 years ago)
Author:
thoughtpolice
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • Status/Oct13

    v7 v8  
    2727
    2828 * **New Template Haskell** - Geoffrey Mainland implemented support for New Template Haskell, fixing a lot of long-standing bugs with the TH implementation, while making it significantly more expressive, including support for typed quotes, pattern splices and more. This allows us to write for example a typed, staged power function. For details and examples, you can see Geoff's blog[NewTH1] and the GHC wiki pages describing the design [NewTH2].
     29
     30 * Joachim Brietner spent time earlier this year implementing a new, fancy **Coercible** design in GHC, which helps eliminate the cases in which newtypes are 'not free'. This occurs for example, when we say something like `newtype Age = MkAge Int` and then `map MkAge [0..10]` - while `newtype`s are supposed to be free, this particular example has a runtime cost!
     31 While Coercible is still in flux, we hope it will be useful as a 'preview technology' in the 7.8 release.
    2932
    3033 * A small menagerie of various other language improvements and extensions, including:
     
    7982 * **Explicit Type Application** - Stephanie Weirich, Richard Eisenburg and Hamidhasan Ahmed have been working on adding explicit type applications to GHC. This allows the programmer to specify the *types* that should be instantiated for arguments to a function application, where normally they would be inferred. While this capability already exists in FC-pro (indeed, every FC-pro program has function application with explicitly applied types,) it has not been available in Haskell itself. While a lot of the syntax and design is not quite final, there are some details about the design available on the wiki[TA].
    8083
     84 * **Git management changes** - For a long time, GHC has used a loosely coupled set of repositories during development. However, as we've added more contributors, this practice has become increasingly problematic, preventing us from using useful tools like `git bisect` to track down bugs. Our plans after the 7.8 release are to sort this out, and hopefully have stable, reproducible GHC builds for all.
     85
    8186But we're not sure what else might happen. It's a great time to step up to the plate and do something fun!
    8287
     
    8792Amongst those who have rolled up their sleeves and helped us recently:
    8893
    89  * Herbert Valerio Riedel
    90  * Takano Akio
    91  * Reid Barton
    92  * Kazu Yamamoto
    93  * Joachim Breitner
    94  * Carter Schonwald
    95  * Krzysztof Gogolewski
    96  * Jan Stolarek
    97  * Muhaimin Ahsan
     94 * **Herbert Valerio Riedel** has stepped up to the plate to help maintain http://ghc.haskell.org, all of our git repositories, and has become increasingly involved in library maintenance and development.
     95 * **Takano Akio** has submitted many great bug reports and patches recently, and has a keen eye for reproducing challenging or tricky bugs.
     96 * **Reid Barton** has submitted some excellent patches to fix some awkward bugs, and help us as we move towards Android cross compiler support.
     97 * **Kazu Yamamoto** helped Andreas Voellmy in the improvements to the I/O manager, and has been extremely helpful in pointing out and reproducing bugs, and helping us find solutions.
     98 * **Joachim Breitner** implemented Coercible, but has also stepped up to merge patches, fix more bugs, and help set up some Continuous-Integration for GHC.
     99 * **Carter Schonwald** has been involved in talking with many who wish to get involved and inspiring them to get their hands dirty!
     100 * **Krzysztof Gogolewski** has fixed many bugs, and like Takano, has helped reproduce and track down issues with fantastic test cases.
     101 * **Jan Stolarek** has helped improve the documentation of the low-level parts of the compiler, as well as implementing optimizations for us too.
     102 * **Muhaimin Ahsan** has been helping with infrastructure, writing patches, and generally offering help to us who are already busy.
    98103
    99104As ever, there is a ton of stuff in the future for us to do. If you want something done - don't wait, it might take a while. You should join us instead!