Permalink
Commits on Jun 5, 2009
  1. Merge 1630:1632 from trunk on OneBigGrin-branch.

    (Last merge was 1610:1630)
    jleeuwes committed Jun 5, 2009
Commits on Jun 4, 2009
  1. Merge trunk on OneBigGrin branch.

    Small chunk-fix to make variant 8 compile.
    jleeuwes committed Jun 4, 2009
Commits on Jun 3, 2009
  1. Some cleaning up.

    jleeuwes committed Jun 3, 2009
Commits on Jun 2, 2009
  1. Keep the arity map in hi-files.

    When merging separate Grin modules, absence of arity information caused (overloaded) functions to be eval'ed, which doesn't make sense.
    
    Fix suggested by Atze.
    jleeuwes committed Jun 2, 2009
  2. Correctly parse and pretty-print calls.

    Correctly pretty-print underscores in DICT-related stuff.
    jleeuwes committed Jun 2, 2009
Commits on May 26, 2009
  1. Postpone cleaning of Grin when doing full program analysis.

    Remove duplicate globals/types when merging Grin modules.
    jleeuwes committed May 26, 2009
Commits on May 18, 2009
  1. Parse BindAnns. Horrible problems (syntax clashes with block comment …

    …syntax) ahead.
    jleeuwes committed May 18, 2009
  2. Make some FFI-related words keywords in Grin.

    jleeuwes committed May 18, 2009
  3. Simplify OneBigGrin: closer to trunk version.

    Next problem: GRIN-parser seems a bit broken (works on some ehclib modules, fails on others).
    jleeuwes committed May 18, 2009
Commits on May 15, 2009
  1. Cached Grin is stored and, in a later compile run, parsed again. Howe…

    …ver, the
    
    compiler sometimes also reuses Grin when the Haskell source has changed.
    jleeuwes committed May 15, 2009
Commits on May 14, 2009
  1. Branch for moving module merging further down the pipeline ('one big …

    …grin'
    
    instead of 'one big core').
    jleeuwes committed May 14, 2009
  2. Branch for moving module merging further down the pipeline ('one big …

    …grin'
    
    instead of 'one big core').
    jleeuwes committed May 14, 2009
  3. Removed dependency on UU.DData.Seq and UU.DData.Scc.

    (UU.DData does not exist anymore)
    
    Partially regress-tested (some regress-tests fail on my system because 
    I'm using a 64-bit architecture).
    amiddelk committed May 14, 2009
Commits on May 13, 2009
  1. Fix for issue 9: correcter check for allowed combi's of duplicate dat…

    …atype field names.
    
    This a bit involved:
    - not ok: duplicate names for different datatypes
    - ok: duplicate names for 1 datatype and different constructors, except when having a different type (so must be postponed until after type checking)
    - not ok: duplicate names for 1 datatype and 1 constructor
    atzedijkstra committed May 13, 2009
Commits on May 8, 2009
  1. Work on HS paper: final version (I think). Did a last spell check (wi…

    …th a lot of hits!), and added a new paragraph in the introduction praising the many-transformations-approach. As a bonus, the new paragraph forces better page breaks on the other pages.
    JeroenFokker committed May 8, 2009
  2. Work on HS paper: processed Doaitse's an my own last-minute correctio…

    …ns (including the correct characterization of the VSS:ICFP09 paper).
    JeroenFokker committed May 8, 2009
  3. hs09

    atzedijkstra committed May 8, 2009
  4. hs09 small remarks

    atzedijkstra committed May 8, 2009
  5. hs09 conclusion

    atzedijkstra committed May 8, 2009
  6. More HS paper: abstract

    JeroenFokker committed May 8, 2009
  7. hs: Core, Ruler

    atzedijkstra committed May 8, 2009
Commits on May 7, 2009
  1. hs09 stuff

    atzedijkstra committed May 7, 2009
  2. hs09 stuff

    atzedijkstra committed May 7, 2009