Project: StateCharts cOmPilEr (SCOPE)
Release: scope-0-15_final
     
     Copyright (C) 2001-2003 Andrzej Wasowski
     
     This program is free software; you can redistribute it and/or modify
     it under the terms of the GNU General Public License as published by
     the Free Software Foundation; either version 2 of the License, or
     (at your option) any later version.
     
     This program is distributed in the hope that it will be useful,
     but WITHOUT ANY WARRANTY; without even the implied warranty of
     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
     GNU General Public License for more details.
     
     You should have received a copy of the GNU General Public License
     along with this program; if not, write to the Free Software
     Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.

release-0-11.txt,v 1.8 2004/01/06 21:28:32 wasowski Exp

TODO:

1. opisz zmiany w metodzie kompilacji (tu i w MANUAL)


See MANUAL for installation, usage instructions and credits.
See LICENSE for usage license.

RELEASE NOTES

TBD.

NEW: visualization back-ends (ADot and Dot)
     hsm front-end

     specialization stuff (UNFINISHED)

     flattening code generator

     generated code compiles in --ansi -pedantic mode of GCC

NEW: muddy/buddy included in sources
TODO: write in manuals that SCOPEROOT needs to be configured in the makefile

MAIN BUG FIXES AND NEW FEATURES

TBD
	1. TBD: recompilation manager changed from make to mosmake+make. No
	more possibility of cross compilation.


 	Dodalem dualna implementacje biezacych konfiguracji (dla wszystkich
	ORsts)

	Dodalem back-end do ACFGs (ale jeszcze nie dziala)

	Serious bug in design of internal model representation was corrected.
	The bug could possible (and likely) cause generation of arbitrarily
	incorrect code, by wrong use of persistance in SML. Luckily the bug
	did not affect any published results.
	
LIST OF KNOWN PROBLEMS

	TBD Zakladamy o cichu, ze nazwy stanow sa rozne od nazw zdarzen (no
	error message?)

	TBD Uzywamy rozszerzen jezyka modulow MoscowML, co oznacza, ze trzeba
	wiecej sie narobic, zeby przeniesc to na inne kompialtory. Zdaje sie
	ze trzeba to dez dopisac w MANUALu. A czy nadal uzywamy?

	1. Only models containing a single system are supported (no multi-system
        models at the moment).
        
        2. Internal rules are ignored (not even a warning is issued) 

	3. Similarly Do reactions are ignored (Termination transition causes a
	parse error).
        
        4. Types with absolute values longer than 31 bits are not handled
        (VS_UINT32, VS_INT64, VS_UINT64).

	5. Variable ranges are not checked at runtime (even in debug version
	code).

	6. Transition conflicts are not resolved (the model is assumed to be
	conflict-less which you can verify with original visualSTATE
	verifier).

	7. Signals (aka internal events) may not have parameters.

	8. No debug (test) drivers for VS compatibility mode

	9. VS compatibility mode drivers do not integrate smoothly (you still
	need to do a lot of handwork to compile VS model, though less than
	before).

	10. No parse error messages (at least for the moment).  

	11. In hierarchical code generation (-cC1) Model reinitialization does
	not properly work for history states, i.e. history states are
	reinitialized to recent value of history instead of to the initially
	indicated state. Actually I am not sure what UML semantics expects,
	but probably it does not consider reinitialization, which  means it
	expects full reinitialization. Only models with history which need to
	be reinitialized at runtime (not restarted, which is a different
	thing) suffer from this bug.
