Commits on Jul 18, 2009
  1. Remove web-UI

    Hopefully switches between master will be easier
    committed Jul 18, 2009
  2. Fix remaining bug in slot mapping.

    Return from c_indyStudeHowIndy is entirely
    canonical equation slots.
    committed Jul 17, 2009
Commits on Jul 17, 2009
  1. Only map equation slots for student equations.

    This fixes a bug in 2e4943c
    committed Jul 16, 2009
Commits on Jul 15, 2009
  1. merge web-UI from master

    committed Jul 14, 2009
Commits on Jul 7, 2009
  1. Map from equation id's to solver slots.

    In Andes2, equations were labeled by integers.  In Andes3, objects
    are labeled by a hash string.
    In solver, remove separate category for student equations.
    committed Jul 7, 2009
Commits on Jul 3, 2009
  1. web-UI merged from master

    committed Jul 3, 2009
Commits on Jul 1, 2009
  1. Remove special declaration for symbols already declared constant.

    **Premature-subst-help** **premature-entry-help**
    **Dead-Path-Help** **Forbidden-Help** **Nogood-Help**
    (cherry picked from commit f9cd927)
    committed Jun 11, 2009
Commits on Apr 7, 2009
Commits on Apr 6, 2009
  1. changes to match asdf in master

    committed Apr 6, 2009
  2. merge changes to asd

    committed Apr 6, 2009
Commits on Mar 20, 2009
  1. update KB from master

    committed Mar 19, 2009
  2. update Algebra/src from master

    committed Mar 19, 2009
Commits on Mar 9, 2009
  1. Looks like it is working for the short example log file.

    Ran Fall2008-USNA.dat but still need "control condition" to
    compare it to.  (running on gideon).  Tweaked print formatting
    for c_indy... and commented out debug prints.
    committed Mar 9, 2009
Commits on Mar 7, 2009
Commits on Mar 6, 2009
  1. Assume positions are always continuous: always use

    closed intervals when checking for conditions on
    (cherry picked from commit 731ee6e)
    committed Mar 6, 2009
  2. check that both bodies have same coordinates when applying

    relative-position-displacement.  Unfortunately, this didn't seem
    to help dt6a much.
    (cherry picked from commit 72db95f)
    committed Mar 6, 2009
Commits on Mar 4, 2009
  1. Better check for running process and work on print

    formatting for expressions sent to solver.  Not
    very sure why it behaves as it does.
    committed Mar 4, 2009
  2. solver-load only loads if not currently loaded.

    This is the behavior that the old system had.
    The help system explicitly unloads solver at the
    end of every session.  Also, turn off pretty printing
    when adding new equations.
    committed Mar 4, 2009
  3. remove cruft SGG/

    committed Mar 4, 2009
  4. remove force-reload

    committed Mar 3, 2009
  5. working for kt1a after removing leading space and putting

    in absolute path name.  Still need to test what happens when
    there is a solver error and test over generation of all problems
    and test over a semester of help.  Also need to test that the
    process is actually shut down properly, even on an error.
    committed Mar 3, 2009
  6. was mis-using :search in run-program.

    Switch to using full path name.
    committed Mar 3, 2009
Commits on Mar 3, 2009
  1. almost working for kt1a

    committed Mar 3, 2009
  2. remove cruft: interact.cpp

    committed Mar 3, 2009
Commits on Mar 1, 2009
  1. wrote but not tested Andes side of solver interface

    need to fix asd files (import changes from main branch) so
    that helpsystem loads ok
    committed Mar 1, 2009