Commits on Nov 2, 2012
  1. Updates to build on GHC 7.4.

    rrnewton committed Nov 2, 2012
Commits on Aug 12, 2011
  1. minor: doc change

    rrnewton committed Aug 12, 2011
  2. No, this is the release candidate for I removed the Graphaly…

    …ze dependence, and therefore other random dependences like pandoc.
    rrnewton committed Aug 12, 2011
  3. Release candidate for

    rrnewton committed Aug 12, 2011
Commits on Aug 11, 2011
  1. Minor fix for ghc 7.0.4

    rrnewton committed Aug 11, 2011
  2. MAJOR REPO CHANGE -- I cloned and then pruned the haskell cnc repo to…

    … factor out the CnC Spec Compiler. In this revision I start to crop out irrelevant stuff and make it stand on its own.
    rrnewton committed Aug 11, 2011
Commits on Jun 24, 2011
  1. Broke a cyclic definition.

    cchen15 committed Jun 24, 2011
Commits on May 10, 2011
  1. minor: comment

    rrnewton committed May 10, 2011
  2. Oops, fix to the previous fix.

    rrnewton committed May 10, 2011
  3. Fixed copy paste bug

    rrnewton committed May 10, 2011
Commits on Apr 13, 2011
Commits on Mar 16, 2011
  1. Minor -- comment tweak

    rrnewton committed Mar 16, 2011
Commits on Mar 3, 2011
Commits on Feb 14, 2011
  1. Added function applications to parsed tag-function grammar.

    Tweaked makefiles and added some scripts.
    Went through the legacy tests and categorized them appropriately.
    "make test" should work under tests_spec.
    rrnewton committed Feb 14, 2011
Commits on Feb 10, 2011
Commits on Feb 9, 2011
  1. Added all existing tests on the spec language to tests_spec. Here are

    the findings:
    Failures in supposedly good ones (i.e. the ones that are accepted by the old translator):
    1. The "prescription" relation is bi-directional. (test004.cnc)
    2. Duplicate collection declarations are allowed. (test014.cnc)
    3. Syntax for tag components is not handled. (test016.cnc and many other failures)
    4. ENV is a keyword (same as env). (test033.cnc)
    5. <tag> -> env should be allowed. (testnq_cnc.cnc, testsearchinsortedtree.cnc)
    Failures in supposedly bad ones (i.e. the ones that are not accepted by the old translator):
    1. Multi-line comment enclosed in /*.... */. (test005.cnc)
    2. <- env -> (test034.cnc)
    3. -> env <- (test035.cnc)
    4. env, env (test036.cnc)
    5. Items produced only by env cannot be consumed only by env (test045.cnc)
    I think the problems in the first group should be fixed for backward compatibility. As for the second group, the first 4 are actually not an issue (rather, they are improvements that don't break backward compatibility). The last one does need to be addressed, though.
    One other thing I observed is that the new translator reports only the first error it encounters. If that's the case, then I guess the syntax for tag components should be addressed first so that those tests that currently fail because of that can be tested for real (and not masked by the tag component syntax error).
    cchen15 committed Feb 9, 2011
Commits on Feb 8, 2011
Commits on Jan 26, 2011
  1. Convered from darcs to git. First git checkin. Git is detecting the p…

    …ermissions changes on the executable files. Very nice work, Git.
    newton committed Jan 26, 2011
  2. RN242# Added a by-hand Data.Binary instance for traces. This version …

    …should produce correct files -- BUT there seems to be a space leak when daling with the binary files that doesn't occur with ASCII. committed Jan 26, 2011
  3. RN241# Moved trace parsing over to ByteString. But this BREAKS the bi…

    …nary format because you can't use the generic marshalling for a ByteString, which apparently is a foreign pointe. committed Jan 26, 2011
  4. RN240# Bumped version number for binary traces. committed Jan 26, 2011