Parent Directory
| 
Revision Log
| Sticky Revision: |
New external library naming scheme on Windows: 'name_ascend.dll' instead of 'name.dll'. This applies to solvers as well as external methods, extpy, import handlers, etc.
Updated config for CONOPT_LINKED
changed to ASC_ENV_SOLVERS and ASC_ENV_LIBRARY in C code. Rerranged order of Tcl/Tk initialisation so that path env vars are set before solvers are loaded.
Adding support for ASCENDSOLVERS env var, for defining location of the ASCEND solvers. By default this will be /usr/share/ascend/solvers.
More of the TRON stuff
Removed ASC_IDA_NEW_ANALYSE condition from the codebase. Cleaned up some debug output.
Added bindings to UFSPARSE sparse matrix library, which provides matrix algebra operations via the CXSparse library which it includes. Tested with the UFSPARSE that comes with Ubuntu 6.10. Added simple test case (no validation)
Some disgraceful hacking to obtain SIGFPE details in ida.c. Working next to fix the proper ASCEND ascSignal implementation (which I broke at some point).
Added getParameter by name in SolverParameters. Fixed up refences to signal-related header files. Fixed up various warnings from -pedantic compilation. Identified an apparent bug with RelationCalcGradient (unsafe version). Added support for compilation of a standalone C++ program (idatest) in SCons build.
Huge reorganisation -- all mtx and lin files moved to base/generic/linear. Required many #include statements to be changed.
Got paranoid about turning off signal handling... added '#ifdef ASC_SIGNAL_TRAPS' anywhere that signal handling referenced. Signal handling is OFF by default now.
Added Matrix Market export routines (from math.nist.gov). Added necessary build commands for above. Work on Jacobi preconditioner for IDA (ongoing) Set integrator_analye to make a call to slv_block_partition. Not sure if that's a good idea or not. Tidied up comments in linsol.h Moved unnecessary #defines from model_reorder.h into model_reorder.c. Cleaned up codedocs in mtx*.h Added WITH_MMIO config.h flag.
Added SCons tests to check SIGINT and to replace ascresetneeded (need replacement for this in Autoconf as well).
Removed debugging from createinst.c
Typo (text) in evaluate.c
Commented out redundant code in importhandler.c
Added signal handling in ExecuteCASGN.
Added missing ospath_free in ModuleSearchPath.
Exported InitSymbolTable, DestroySymbolTable in symtab (dubious)
Moved FPRESET macro out of ascConfig.h and into ascSignal.h
Added Asc_SignalHandler{Push,Pop}Default.
Added ASC_RESETNEEDED and HAVE_C99FPE macros in config.h.in.
Found the bug causing the SIGFPE in idakryx.a4c (raises a question about int/float division in modelling, I think)
Added system_destroy call in Simulation::~Simulation (dubious).
Added support for dmalloc (http://dmalloc.com) Added Integrator::setEngine(string name) Minor other stuff.
Moved the location of the PATHENVIRONMENTVAR to config.h.in so that it can be shared with other places that want to use that variable. Did some bug fixing of the simple TMY reader.
Fixed for CONOPT under Windows.
Moved 'SearchArchiveLibraryPath' into ascDynaLoad so that it can be shared between compiler/packages and solver/conopt. Changed debug messages in ospath. Fixed conopt to search for consub3 in $LD_LIBRARY_PATH or /usr/lib:/usr/local/lib by default on Linux (or %PATH% or c:\Program Files\ASCEND on Windows). Modified definition of testconopt so that some parameters can be varied via the interface.
First work towards reinstating the CONOPT solver. So far, I've got it building and linking without errors, but haven't tested the resulting runtime.
Merged changes from DAE branch (revisions 702 to 819) back into trunk. This adds the Integration API to the ASCEND solver (in base/generic). Also provides pre-alpha support for 'IDA' from the SUNDIALS suite, a DAE solver. Many other minor code clean-ups, including adoption of new 'ASC_NEW' and friends (to replace 'ascmalloc') Added some very sketchy stuff providing 'DIFF(...)' syntax, although it is anticipated that this will be removed.
Fixed up standard paths stuff to use 'INSTALL_ASCDATA' as the place where tcltk/python/models etc will live. The confusion here came from the fact that I was taking 'INSTALL_SHARE' to be be assumed to mean 'the location of /usr/share' -- hence the resulting '$INSTALL_SHARE/ascend' that was bothering Ben. The 'INSTALL_ASCDATA' fixes this, and behaves as he will like, I hope. Also note that I have switched the default situation in config.h.in to use relative paths by default, which is the preferred CMU way. SCons builds will default to absolute paths on non-Windows systems.
Summary: most of the recent complaints about autotools build fixed, but the buildbot might find a problem with the scons build/install/rpmbuild. Details: configure,Driver.c,utilities/config.h.in: The choice of INSTALL_DATA as a configure variable name has been repaired (renamed INSTALL_SHARE); INSTALL_DATA is the unix canonical name for the program 'install' handling a data file. Both autotools and scons builds have been updated to reflect this, in the process correcting the definition from prefix/share to prefix/share/ascend. autotools build: Added --with-quiet option to make the install of already installed and unchanged files less verbose. Put in a workaround for the '' problem in tkConfig.sh seen by KC/JP. Added --datadir support to autotools configure. If you configure the autotools --datadir=PREFIX/share/ascend you get the same behavior as the Pye packaging; by default you get the CMU packaging convention instead. Added utilities/env.c to build. compiler: Turned off AWAL in anontype.c. Noodling away at blackboxes continues...
Completing missing part of RELATIVE_PATHS configuration for SCons, to address bug #262.
woops
Still working on a way to add options to config.h.in such that it doesn't break autotools.
fixed for scons, can't work out a way to allows @MISSING_SOMETHING@ to fallback safely, unfort.
fixing so that Ben won't need to add this macro to autoconf, hopefully.
Adding configuration option to turn off xterm colours in output.
Fixing some malloc bugs.
This fixes the new Driver.c implementation. Now, default values of the environment variables are specified in the file base/generic/utilities/config.h.in, which should be tooled up with you configuration process. These default values will be subject to environment substiution, so that if you set the default ASCENDTK=$ASCENDDIST/TK, that will be expanded *at runtime*. This is to help with the 'application folder' style of installation off-root.
Working on improving handling of paths.
Created 'scons install' command that also works for Tcl/Tk interface. These changes required some modifications to the default values of the environment variables. Installed Tcl/Tk version now works with only the LD_LIBRARY_PATH env var being required.
This is a patch to allow statements like "IMPORT johnpye/extfn/extfntest". It modifies the methods in packages.c. Plan is to reuse the same code for "REQUIRE xxxxxx" statements as well.
This form allows you to request diffs between any two revisions of this file. For each of the two "sides" of the diff, enter a numeric revision.
| john.pye@anu.edu.au | ViewVC Help |
| Powered by ViewVC 1.1.22 |