Parent Directory | Revision Log
Sticky Revision: |
In slv9, made changes to keep CONOPT from complaining. CMSlv still not working as expected though. Decreased required version of Bison to 2.0. Switched search path for CONOPT on Linux to the CONOPT_PATH env var (LD_LIBRARY_PATH was being interfered with when ASC_DEV=1). Some other debug message changes.
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.
Few small changes to fix for Linux after the recent MSVC++ oriented changes.
Fixes for MSVC++
Added general-purpose text-information dialog to PyGTK interface. Added reporting of clique of a variable to PyGTK GUI. Changed 'fixable' and 'new bounds' to use the new dialog. Removed 'set<...>' from ascpy.i (it wasn't working), changed to vector<...>. Exported a couple of var incidences fns to the DLL/SO.
Added '--models' option to ascend-config. Fixed a bug with packages.c wrt generating absolute paths to library files. Added SCons directives to install header files in INSTALL_INCLUDE subdirs. Changed rel.c so that init function is only run if provided in the package. Added 'ospath_getabs' to convert relative paths into absolute paths by adding prefix of fully-resolved '.'
Fixed up ExtRel_Evaluate_LHS (was not accessing the arglist correctly), it seems pretty inefficient at the moment, but we'll persist.
Monster commit! Lots of recommenting and reorganising of external relations-related stuff. Replaced a lot of ascmalloc and asccalloc calls with the new ASC_NEW* macros. Fixed (?) the problem Art is having with icons in PyGTK. Turned on -Wall in SConstruct and fixed up a stack of warnings. Removed the redundant exit(2) from after Asc_Panic calls and added __attribute__((noreturn)). Set doxygen to create callgraphs to level 2, updated doxyfile to version 1.4.7. Fixed up building of extfntest.c.
Removed 'instance.c' (it said we should). A little bit more work on blackbox_evaluate_residual.
Temporarily removed extrel.* from disused dir (to avoid spurious grep) Commenting and tidying up of some other files related to extrels.
Fixed a problem with external packages when loaded from PyGTK
Fixed up behaviour of default ASCENDLIBRARY in python interface (for the case where you're using ascpy without using the PyGTK GUI). Removed some debug output when loading external libraries. Fixed up the 'extfntest' example & SConscript file.
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.
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...
Fix windows behaviour for 'IMPORT "johnpye/extfn/extfntest";'
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.
Working on platform-independent pathnames for the IMPORT command. Added 'ospath.c' to base/generic/general for this purpose. Patched kvalues and sensitivity to use the 'IMPORT "libname";' syntax instead of 'IMPORT registerfn FROM libname;'. Fixed pathnames in create.nsi. Added GPL header to coupla files. Added quoting to 'IMPORT' syntax in ascParse.y. Removed 'PackageOption' from Tcl/Tk related Scons options so that missing Tcl/Tk doesn't cause breakage.
Working of debugging the mysterious PyGTK bug. Seems to be coming from a wrong char *filename in an error_reporter call from the DEBUG_RELS sections of the compiler code. Also renamed the WSEM and WSNM macros for clarity, removed unused file pointer.
display ASC_SHLIBSUFFIX error only if platform is unknow
More fixes for compiling and running on ubuntu 5.10 with GCC: gcc version 4.0.2 20050808 (prerelease) (Ubuntu 4.0.1-4ubuntu9) Added ASC_SHLIBSUFFIX and ASC_SHLIBPREFIX which are used in packages.c in preference to platform-specific #ifdefs.
Changed the syntax of import from IMPORT id FROM id; to IMPORT id FROM "quoted/text/lib"; Writing external METHOD calls via the EXTERNAL methname(arglist); now works. Against my better judgement in a few cases, cleaned up the old EXT statement code from Abbott to work in a typesafe manner. A better solution is to get the CALL syntax working as defined in the grammar, but this takes a good deal more work and meanwhile people want to get their phds... Still some known bugs in relation_util. I need to commit the test models that go with the EXT statements. Fixed a bunch of lintish compiler complaints. Don't try to solve external models yet.
tentative: Disabled ascDynaload functions when DYNAMIC_PACKAGES is not defined. Altered error.[ch] so that MS VC++ variadic macros are used when available. Disabled Windows MessageBox in ascPanic, so that compilation of base engine is possible with the Platform SDK. Added SConscript files to build FORTRAN components. Removed some autoconf-related stuff from the pygtk/interface directory.
- Fixed missing header and casting problems in initialize.c around JP FIX implementation. - Repaired extfunc header declaration-- someone deleted array pointers without asking. - Repaired casting errors in extfunc.c, probably ala kirk. - Put in a #define for zz_lval in scanner.h to work around brokenness in flex 2.5.4; if this breaks something for others, we need a better solution.
Normalised all #include statements so that files in the current directory are included as "localfile.h" and files in other directories are included as <directory/file.h> This is in accordance with the spec at http://gcc.gnu.org/onlinedocs/gcc-4.1.0/cpp/Include-Syntax.html#Include-Syntax
move the function call after the variable define, a compile bug fix
Fixing 'auto package registration' with 'libstemname_register'. Fixed bug #247.
Fixed up dynamic loading so that external libraries can be used on Windows -- no unresolved symbols in the DLL.
Added some debug output related to external relations. Added ./configure flags for relation debugging and compiler timing. Removed some <!-- --> comments.
Debugging output for loading external packages. Export error_reporter to windows DLL (needs testing).
Documentation, indentation and debugging-message changes.
Minor rework of jam system to stop relinking of libraries when sources not modified. Minor tweaks to stem the tide of compiler warnings.
Documentation and CONSOLE_DEBUGging.
- Simplifying many 'error_reporter' calls with new macros: ERROR_REPORTER_NOLINE and ERROR_REPORTER_HERE - Adding some cygwin cases in the PyGTK build.
Removed support for Ultrix, OSF, which seem to be dead. This makes the ascDynaload.c file much smaller. Also removed unused DynamicLoad function (use Asc_DynamicLoad instead). Please let me know if these changes break anything for you.
Reworked jam build system: - integrated Jamrules.in and Jamrules_tcltk98.in with autoconf - now support single-point building from trunk/jam - old jam files still in place but deprecated - may be buggy, please report back bugs for fixing Compiled on both Windows and Linux.
A little more progress killing compiler warnings.
Small changes to eliminate GCC warnings
My first commit: Added 'error_reporter' for standardised error reporting Working on getting external functions running again Also: issue 130, issue 68, issue 131, issue 89, issue 127, issue 138, issue 94.
Manual rework of doxygen comments in all headers. - Added @file comment to all headers. - Added parameter names to all function declarations in headers. - Corrected comment referencing where necessary. - Split some comments which documented blocks of declarations. - Converted notes about required work into @todo comments so doxygen can generate a todo list. Minor bug fixes.
moving things to base/generic
Setting up web subdirectory in repository
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 |