ecell-devel Mailing List for E-Cell System

Hi,

thanks for the info. that makes us relieved.
Gabor

--- efe...@eb... wrote: > Hi
> 
> I reported a crash problem of gecell with PyGTK
> 2.3.93 and it is now
> solved with 2.3.94 (released on FTP mirror). It was
> obviously a PyGTK bug.
> 
> Eric
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic
> Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1
> today.
>
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
> _______________________________________________
> Ecell-devel mailing list
> Ece...@li...
>
https://lists.sourceforge.net/lists/listinfo/ecell-devel
>  


	
	
		
___________________________________________________________ALL-NEW Yahoo! Messenger - sooooo many all-new ways to express yourself http://uk.messenger.yahoo.com

Hi

I reported a crash problem of gecell with PyGTK 2.3.93 and it is now
solved with 2.3.94 (released on FTP mirror). It was obviously a PyGTK bug.

Eric

Hi Gabor
>
> I will try to check it. Have you built pygtk 2.3.93
> from source?

No, I used a package for Mandrake Cooker (dev version). I might try to
compile it myself though. However, it was working with the previous
version 2.3.92 package.

> But I suggest that until we find out what the problem
> was you should try to build a pygtk version somewhat
> behind. Currently we are using stable 2.2 and it is
> working fine.

Yes, I have reverted and it is working. The problem appeared with new
version and I was wondering if it was a PyGTK bug.

Eric

Hi Eric,

I will try to check it. Have you built pygtk 2.3.93
from source?
But I suggest that until we find out what the problem
was you should try to build a pygtk version somewhat
behind. Currently we are using stable 2.2 and it is
working fine.
Gabor


--- efe...@eb... wrote: > Hi everyone
> 
> As announced on the PyGTK project page
> (http://www.pygtk.org/), PyGTK 0.6
> branch is now obsolete and will be dropped. I just
> installed their new
> version PyGTK 2.3.93, which is the frozen state of
> the future 2.4.0
> release, to check compatibility with ecell3. They
> will just do bug fixes
> now.
> 
> With today's ecell3 cvs version, I get an error when
> launching gecell :
> 
> $ gecell -f simple.eml
> Traceback (most recent call last):
>   File "/usr/local/lib/osogo/opengui.py", line 198,
> in ?
>     main()
>   File "/usr/local/lib/osogo/opengui.py", line 138,
> in main
>     aMainWindow = aSession.openWindow('MainWindow')
>   File
>
"/usr/local/lib/python2.3/site-packages/ecell/GtkSessionMonitor.py",
> line 190, in openWindow
>     self.theFundamentalWindows[ aWindowName
> ].openWindow()
>   File "/usr/local/lib/osogo/MainWindow.py", line
> 84, in openWindow
>     OsogoWindow.openWindow(self)
>   File "/usr/local/lib/osogo/OsogoWindow.py", line
> 180, in openWindow
>     Window.openWindow(self)
>   File
>
"/usr/local/lib/python2.3/site-packages/ecell/Window.py",
> line 103,
> in openWindow
>     self.widgets = gtk.glade.XML( self.theGladeFile,
> root=None )
> TypeError: typedict must be a mapping
> 
> 
> Do you think it is a bug in PyGTK or is it a change
> in PyGTK spec ?
> Thanks
> Eric
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic
> Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1
> today.
>
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
> _______________________________________________
> Ecell-devel mailing list
> Ece...@li...
>
https://lists.sourceforge.net/lists/listinfo/ecell-devel
>  


	
	
		
___________________________________________________________ALL-NEW Yahoo! Messenger - sooooo many all-new ways to express yourself http://uk.messenger.yahoo.com

Hi Eric,

I've not yet tried pygtk-2.3.93, but it is working fine with 2.3.92
here.

I'm going to try 2.3.93 when its RPM is available.

-sha

> As announced on the PyGTK project page (http://www.pygtk.org/), PyGTK 0.6
> branch is now obsolete and will be dropped. I just installed their new
> version PyGTK 2.3.93, which is the frozen state of the future 2.4.0
> release, to check compatibility with ecell3. They will just do bug fixes
> now.
> 
> With today's ecell3 cvs version, I get an error when launching gecell :
> 
> $ gecell -f simple.eml
> Traceback (most recent call last):
>   File "/usr/local/lib/osogo/opengui.py", line 198, in ?
>     main()
>   File "/usr/local/lib/osogo/opengui.py", line 138, in main
>     aMainWindow = aSession.openWindow('MainWindow')
>   File
> "/usr/local/lib/python2.3/site-packages/ecell/GtkSessionMonitor.py",
> line 190, in openWindow
>     self.theFundamentalWindows[ aWindowName ].openWindow()
>   File "/usr/local/lib/osogo/MainWindow.py", line 84, in openWindow
>     OsogoWindow.openWindow(self)
>   File "/usr/local/lib/osogo/OsogoWindow.py", line 180, in openWindow
>     Window.openWindow(self)
>   File "/usr/local/lib/python2.3/site-packages/ecell/Window.py", line 103,
> in openWindow
>     self.widgets = gtk.glade.XML( self.theGladeFile, root=None )
> TypeError: typedict must be a mapping
> 
> 
> Do you think it is a bug in PyGTK or is it a change in PyGTK spec ?
> Thanks
> Eric
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1 today.
> http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
> _______________________________________________
> Ecell-devel mailing list
> Ece...@li...
> https://lists.sourceforge.net/lists/listinfo/ecell-devel

Hi everyone

As announced on the PyGTK project page (http://www.pygtk.org/), PyGTK 0.6
branch is now obsolete and will be dropped. I just installed their new
version PyGTK 2.3.93, which is the frozen state of the future 2.4.0
release, to check compatibility with ecell3. They will just do bug fixes
now.

With today's ecell3 cvs version, I get an error when launching gecell :

$ gecell -f simple.eml
Traceback (most recent call last):
  File "/usr/local/lib/osogo/opengui.py", line 198, in ?
    main()
  File "/usr/local/lib/osogo/opengui.py", line 138, in main
    aMainWindow = aSession.openWindow('MainWindow')
  File
"/usr/local/lib/python2.3/site-packages/ecell/GtkSessionMonitor.py",
line 190, in openWindow
    self.theFundamentalWindows[ aWindowName ].openWindow()
  File "/usr/local/lib/osogo/MainWindow.py", line 84, in openWindow
    OsogoWindow.openWindow(self)
  File "/usr/local/lib/osogo/OsogoWindow.py", line 180, in openWindow
    Window.openWindow(self)
  File "/usr/local/lib/python2.3/site-packages/ecell/Window.py", line 103,
in openWindow
    self.widgets = gtk.glade.XML( self.theGladeFile, root=None )
TypeError: typedict must be a mapping


Do you think it is a bug in PyGTK or is it a change in PyGTK spec ?
Thanks
Eric

Dear all,

Here I share with you our experience of porting Ecell 3.1.102 to Mac  
OSX.

So far on Mac OS X, Ecell can start in both command line mode and GUI  
mode. All the dynamic modules can be loaded without error. However, we  
now come across a boost.python.ArgumentError that breaks ecell model  
loading (ecell-session -f simple.eml). Any suggestions are high  
welcome.

Boost.Python.ArgumentError: Python argument types in
     Simulator.loadEntityProperty(Simulator, str, list)
did not match C++ signature:
     loadEntityProperty(libemc::Simulator {lvalue}, std::string,  
libecs::Polymorph)

Cheers,
--
Chen Peng <che...@tl...>
Senior System Engineer
Temasek Life Sciences Laboratory

------------------------------------------------------------------------ 
---------------------------
Port Ecell 3.1.102 to Mac OSX

Testbed: 	PowerBook G4, 1Ghz, 1GB Memory
		Mac OSX 10.3.4, gcc version 3.3 20030304

------------------------------------------------------------------------ 
---------------------------
Download boost-jam from sourceforge
http://sourceforge.net/project/showfiles.php?group_id=7586

To install Boost.Jam, copy the bjam executable to a location accessible  
in your PATH.
I downloaded macppc binary from sourceforge and put it in  
/opt/local/bin.

------------------------------------------------------------------------ 
---------------------------
Compiling boost C++ lib 1.31.0
http://sourceforge.net/project/showfiles.php?group_id=7586

bjam -sPYTHON_VERSION=2.3 -sTOOLS=darwin \
	"-sBUILD=framework bundle-loader" --prefix=/opt/local install

Compiled in 25 minutes and there is no error (but a few warnings)

------------------------------------------------------------------------ 
---------------------------
Python-numeric module 23.1
http://www.pfdubois.com/numpy/
http://sourceforge.net/projects/numpy

Binary installation is available (NOT recommended at all as it does not  
include the headers)
Python Package Manager:  
http://homepages.cwi.nl/~jack/macpython/download.html
Check python module database  
http://undefined.org/python/pimp/darwin-7.2.0-Power_Macintosh.plist

export LDFLAGS=""-framework vecLib -L/opt/local/lib"

I use darwinport to install py-numeric module.

------------------------------------------------------------------------ 
---------------------------
Install GSL 1.4
http://sources.redhat.com/gsl/

GNU Scientific Library are available from darwinport

------------------------------------------------------------------------ 
---------------------------
GUI Requirements
It is important to mention that darwinport might break in the middle  
when one dependent lib fails compilation. In this circumstance, we have  
to go to the work folder and fix the problem. One possible cause is too  
aggressive CFLAGS (try CFLAGS="" sudo install package). And there are  
all kinds of minor problem during linking stage. They are solved at  
case-by-case basis.

Also important that if OSX already has the library that is dependent to  
the library to be installed, we must be extremely careful. Usually  
there are incompatibilities. We need to check if the dependent lib  
really needs to be installed by port, or already exists with OSX, where  
we should hack the Portfile to suit us.

# cheat darwinport that python 2.3 has been installed by port system
sudo ln -s /usr/bin/python /opt/local/bin/python

* py-gtk2 must be installed after libglade2.

Installed Packages (including dependent libraries):
atk-1.6.1
audiofile-0.2.5
cotvnc-1.3.6
dlcompat-20030629
docbook-xml-4.2
eel-2.6.1
esound-0.2.29
expat-1.95.7
fontconfig-2.2.2
freetype-2.1.7
gail-1.6.3
gconf-2.6.1
gettext-0.14.1
glib2-2.2.3
gnome-applets-2.6.1
gnome-desktop-2.6.1
gnome-icon-theme-1.2.1
gnome-keyring-0.2.1
gnome-mime-data-2.4.1
gnome-panel-2.6.1
gnome-session-2.6.1
gnome-terminal-2.6.1
gnome-themes-2.6.1
gnome-vfs-2.6.1.1
gnutls-0.9.96
gsl-1.4
gtk-engines2-2.2.0
gtk2-2.4.1
hicolor-icon-theme-0.4
intltool-0.30
jpeg-6b
lcms-1.12
libart_lgpl-2.3.11
libbonobo-2.6.0
libbonoboui-2.4.1
libcroco-0.5.0
libexif-0.5.12
libgcrypt-1.1.44
libglade2-2.3.6
libgnome-2.6.0
libgnomecanvas-2.6.1.1
libgnomeui-2.6.1.1
libgpg-error-0.5
libgsf-1.9.0
libgtkhtml-2.4.1
libgtop-2.6.0
libidl-0.8.2
libmng-1.0.6
libpng-1.2.5
librsvg-2.6.5
libtasn1-0.2.6
libungif-4.1.0b1
libwmf-0.2.8.2
libwnck-2.6.1
libxklavier-1.02
libxml2-2.6.7
libxslt-1.0.33
libzvt-2.0.1
metacity-2.8.1
nautilus-2.6.1
orbit2-2.10.2
pango-1.2.5
pkgconfig-0.15.0
popt-1.7
py-gtk2-2.3.92
py-orbit-2.0.0
readline-4.3
scrollkeeper-0.3.12
startup-notification-0.5
tiff-3.5.7
vte-0.11.11

------------------------------------------------------------------------ 
---------------------------
Gnome-Python Hack
# Ecell uses gnome.canvas.path_def_new to check gnome.canvas version
# but this function is only defined in gnome-python on 2004-03-15
# according to gnome-python cvs ChangeLog
#  
http://cvs.gnome.org/viewcvs/gnome-python/gnome-python/ChangeLog? 
rev=1.217&view=markup

# py-gnome from darwinports install only ver 2.0.0 which was released  
in 2003
# gnome-python2-python23 from fink is verion 2.0.0.11
# Therefore, this function is not available for gnome python in OSX
#
# the solution is to install gnome-python 2.0.2 (released on 8th April  
2004)
# or to install from gnome-python cvs (July 05 2004), which I believe  
is much better
#
# note that the latest gnome-python cvs requires automake 1.7 or higher
# I install 1.7.9 via darwinport
#
# the CVS version has fixed most of the OSX linking bugs
# still have a few bugs on vfs bundle
#
$ cvs -d:pserver:ano...@an...:/cvs/gnome login
$ cvs -z9 -d:pserver:ano...@an...:/cvs/gnome co  
gnome-python/gnome-python
#
# vi gnome/vfs-aysnc-handle.c
# add #define NO_IMPORT_PYGOBJECT to line 2
#
# vi gtkhtml2/gtkhtml2.override
#define NO_IMPORT_PYGOBJECT
#include "pygobject.h"
#
# autogen.sh needs to be fixed for OSX
$ ./autogen.sh  --prefix=/opt/local
$ make && sudo make install
#
# this fixed gnome.canvas version check bug in OSX
#
# To test
$ python
Python 2.3 (#1, Sep 13 2003, 00:49:11)
[GCC 3.3 20030304 (Apple Computer, Inc. build 1495)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
 >>> import pygtk
 >>> pygtk.require('2.0')
 >>> import gtk
 >>> import gnome.canvas
 >>> import gnome.ui
# TEST OK, important!!

------------------------------------------------------------------------ 
---------------------------
Some missing python libraries on Mac OSX
DLFCN.py should be a default module from python linux distribution  
(/usr/lib/python2.3/plat-linux2/DLFCN.py
). It is interesting that it is not available on Macpython, which is  
the default python for OSX.

How to get DLFCN.py on Mac OS X.
1. download h2py from http://freshmeat.net/projects/h2py/
2. The current version should be 0.11, modify h2py
# line 68
   output=command("cpp -dM "+pathname)
3. cp h2py.py /opt/local/bin
4. h2py.py /usr/include/dlfcn.h
5. cp dlfcn.py /opt/local/lib/python2.3/site-packages/DLFCN.py

Unfortunately, Apple could not ship the readline module with Python  
because that particular module is based on GPL'd code. Of course, code  
under the GPL isn't actually free and, as such, the entire core  
operating system cannot contain any non-free or non-Apple owned code.

so we need to install readline from darwinport and install py-readline  
(readline-0.0.0.0.tar.gz), which is not a official patch for python on  
OSX, as the reason explained above.

readline-0.0.0.0.tar.gz is made for python 2.2, to make it work for  
python 2.3, we need to copy readline.c from python 2.3 source code  
(src/Modules) to replace to the readline.c in its distribution. During  
the compilation, error may come as

gcc -Wl,-F. -Wl,-F. -bundle -framework Python -L/opt/local/lib -mcpu=G4  
-mtune=G4 -faltivec -O3 -I/opt/local/include  
-I/opt/local/include/python2.3  
build/temp.darwin-7.4.0-Power_Macintosh-2.3/readline.o -o  
build/lib.darwin-7.4.0-Power_Macintosh-2.3/readline.so -g  
-F/System/Library/PrivateFrameworks/ -framework readline
ld: can't locate framework for: -framework readline

Copy the command and run it as

gcc -Wl,-F. -Wl,-F. -bundle -framework Python -L/opt/local/lib -mcpu=G4  
-mtune=G4 -faltivec -O3 -I/opt/local/include  
-I/opt/local/include/python2.3  
build/temp.darwin-7.4.0-Power_Macintosh-2.3/readline.o -o  
build/lib.darwin-7.4.0-Power_Macintosh-2.3/readline.so -g  
-F/System/Library/PrivateFrameworks/ -lreadline

Then python setup.py install. Readline for python is OK.

------------------------------------------------------------------------ 
---------------------------
ECELL Code Hack Summary (updated July 07 2004)
------------------------------------------------------------------------ 
---------------------------
Create a autogen.sh.osx from autogen.sh, modify autogen.sh.osx for  
aclocal path. OSX's libtool is not GNU libtool, we should modify to use  
glibtool and glibtoolize in OSX.

# remove previous libltdl folder, MUST do this
# note that Apple's libltdl determines dynamic modules goes with suffix  
.so
# the original libltdl sets suffix to empty
rm -rf libltdl
# copy OS X hacked libltdl for ecell
cp -r /usr/share/libtool/libltdl .

vi dmtools/dmcompile.in
# line 15
SHLEXT='.so'
# line 43
def link( CXX, OBJ, SOBJ, LDFLAGS ):
     return execute( CXX + ' -bundle ' + OBJ + ' -o ' + SOBJ + ' ' +  
LDFLAGS )
# line 81
     LDFLAGS = '-framework python ' + LDFLAGS

vi ecell/dm/Makefile.am
# line 2
SUFFIXES = .cpp .so
# line 4
LTDL_SHLIB_EXT = .so
# line 19, fix for your additional python headers
PYTHON_INCLUDES=-I/opt/local/include @PYTHON_INCLUDES@
# line 117
SOS = ${DMS:.cpp=.so}
# line 119
%$(LTDL_SHLIB_EXT) : %.cpp
     ${ECELL3_DMC} --verbose --dmcompile="../../dmtool" --no-stdinclude  
--no-stdlibdir --ldflags="-L${top_srcdir}/libecs/.libs ${BOOST_LIB}"  
--cxxflags='-I${top_srcdir}/libecs/ -I${top_srcdir}/.. -I${top_srcdir}  
${PYTHON_INCLUDES} ${BOOST_INCLUDE}' $<

vi ecell/libecs/Makefile.am
#line 31, remove -lstdc++
libecs_la_LIBADD=${LIBLTDL} ${GSL_LIBADD}

vi ecell/pyecs/Makefile.am
# line 66
ECS_SO=_ecs.so
EMC_SO=_emc.so

vi ecell/libecs/Makefile.am
# line 4
if OS_WIN32
   no_undefined = -no-undefined
   SHLEXT = .dll.a
else
   SHLEXT = .so
endif
#line 31, to remove -lstdc++
libecs_la_LIBADD=${LIBLTDL} ${GSL_LIBADD}

# !!! finally understand
vi modeleditor/acinclude.m4
vi osogo/acinclude.m4
vi toollancher/acinclude.m4
# line 29 "import $1", the prefix should be 2 tabs, but it is one tab  
and some spaces
# this confuses python with wrong indent
# fix it with two tabs
# this fixed gnome.canvas and gnome.ui check bug in OSX

vi modeleditor/configure.in
# line 15
dnl check for pygtk>=2.0.0
prog="
import pygtk,sys
pygtk.require('2.0')
import gtk
req=(2,0,0)
inst=gtk.pygtk_version
if inst>=req:
     sys.exit(0)
sys.exit(1)
"
# this fixed PyGtk version check bug in OSX

vi ecell/libecs/osif.cpp
# line 123, replace sys/vfs.h
include "sys/mount.h".

# Now start compiling ECell
# LDFLAGS can NOT include -Wno-long-double
export CFLAGS="-O3 -faltivec"
export CPPFLAGS="-I/opt/local/include -I/opt/local/include/python2.3"
export  
PYTHONPATH="/opt/local/lib/python2.3/site-packages:/opt/local/lib/ 
python2.3/site-packages/Numeric"

./autogen.sh.osx --prefix=/opt/biotools/ecell-3.1.102  
--with-boost-root-dir=/opt/local

make -j2
# The make process is almost perfect. Without any complains... ;)

sudo make install

# Can run and test ecell with some commands
export PATH=/opt/biotools/ecell-3.1.102:$PATH
$ ecell3-session
ecell3-session [ E-Cell SE Version 3.1.102, on Python Version 2.3.0 ]
Copyright (C) 1996-2004 Keio University.
Send feedback to Kouichi Takahashi <sh...@e-...>
ecell3-session>>> import ecell
ecell3-session>>> dir( ecell );
ecell3-session>>>
$
$ gecell &

#
# Now the remaining problem is Boost.Python.ArgumentError, which does  
not appear in Linux/Win
#
# END OF DOCUMENT
------------------------------------------------------------------------ 
---------------------------

macgregor curiosity classificatory pappy burma doppler embedded 
ester anaplasmosis gardner marjorie float cameron 
cholesterol stoichiometry consequential loot progression regular gulp 
asher taper cantaloupe through component 
thermionic argus resuming catnip 
conn flam beatrice innocuous landscape 
bowline abolish stake anxious cloak elope dour 

On Fri, 2004-07-02 at 17:00, Kouichi Takahashi wrote:
> Hi eric,
> 
> ok, this is fixed on the cvs.
> 
> thanks for the report.
> 
> -sha

Thanks a lot. It works flawlessly under Mandrake Cooker now (the new
stable gcc has fixed all issues, including the memory allocation bug
with Pyecs). I shall be pleased to create a package for Mandrake soon
and to make it available for everyone.

Cheers
Eric

Hi eric,

ok, this is fixed on the cvs.

thanks for the report.

-sha


> Finally gcc-3.4.1 has been released (maybe not publicly but the freeze has
> been officially announced).
> I get an error when compiling ecell3 cvs, due to a new gcc-3.4 constraint :
> 
> make[4]: Entering directory `/usr/local/src/ecell3/ecell/dm'
> ../pyecs/ecell3-python ../../bin/ecell3-dmc --verbose
> --dmcompile="../../dmtool" --no-stdinclude --no-stdlibdir
> --ldflags="-L../libecs -L/usr/lib" --cxxflags='-I../libecs -I../.. -I..
> -I/usr/include/python2.3 -I/usr/include' ExpressionFluxProcess.cpp
> "/usr/bin/python" /usr/local/src/ecell3/dmtool/dmcompile -v
> "ExpressionFluxProcess.cpp" "-D_ECELL3_DM_CLASSNAME=ExpressionFluxProcess"
> -I../libecs -I../.. -I.. -I/usr/include/python2.3 -I/usr/include
> g++ -g -O2 ""-D_ECELL3_DM_CLASSNAME=ExpressionFluxProcess"" "-I../libecs"
> "-I../.." "-I.." "-I/usr/include/python2.3" "-I/usr/include" -fPIC -c
> ""ExpressionFluxProcess.cpp""
> In file included from ExpressionFluxProcess.cpp:34:
> ExpressionProcessBase.hpp: In member function `const libecs::Real
> libecs::ExpressionProcessBase::VirtualMachine::execute(const
> std::vector<char, std::allocator<char> >&)':
> ExpressionProcessBase.hpp:492: error: ISO C++ forbids cast to
> non-reference type used as lvalue
> ExpressionProcessBase.hpp:504: error: ISO C++ forbids cast to
> non-reference type used as lvalue
> E-Cell3 dmcompile: Error: The compiler failed to create
> ExpressionFluxProcess.o
> 
> Eric
> 
> 
> -------------------------------------------------------
> This SF.Net email sponsored by Black Hat Briefings & Training.
> Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
> digital self defense, top technical experts, no vendor pitches, 
> unmatched networking opportunities. Visit www.blackhat.com
> _______________________________________________
> Ecell-devel mailing list
> Ece...@li...
> https://lists.sourceforge.net/lists/listinfo/ecell-devel
-- 
Kouichi Takahashi, PhD,
E-Cell Project,
Institute for Advanced Biosciences,
Keio University, Japan.
http://www.sfc.keio.ac.jp/~shafi

Hi everyone

Finally gcc-3.4.1 has been released (maybe not publicly but the freeze has
been officially announced).
I get an error when compiling ecell3 cvs, due to a new gcc-3.4 constraint :

make[4]: Entering directory `/usr/local/src/ecell3/ecell/dm'
../pyecs/ecell3-python ../../bin/ecell3-dmc --verbose
--dmcompile="../../dmtool" --no-stdinclude --no-stdlibdir
--ldflags="-L../libecs -L/usr/lib" --cxxflags='-I../libecs -I../.. -I..
-I/usr/include/python2.3 -I/usr/include' ExpressionFluxProcess.cpp
"/usr/bin/python" /usr/local/src/ecell3/dmtool/dmcompile -v
"ExpressionFluxProcess.cpp" "-D_ECELL3_DM_CLASSNAME=ExpressionFluxProcess"
-I../libecs -I../.. -I.. -I/usr/include/python2.3 -I/usr/include
g++ -g -O2 ""-D_ECELL3_DM_CLASSNAME=ExpressionFluxProcess"" "-I../libecs"
"-I../.." "-I.." "-I/usr/include/python2.3" "-I/usr/include" -fPIC -c
""ExpressionFluxProcess.cpp""
In file included from ExpressionFluxProcess.cpp:34:
ExpressionProcessBase.hpp: In member function `const libecs::Real
libecs::ExpressionProcessBase::VirtualMachine::execute(const
std::vector<char, std::allocator<char> >&)':
ExpressionProcessBase.hpp:492: error: ISO C++ forbids cast to
non-reference type used as lvalue
ExpressionProcessBase.hpp:504: error: ISO C++ forbids cast to
non-reference type used as lvalue
E-Cell3 dmcompile: Error: The compiler failed to create
ExpressionFluxProcess.o

Eric

Never give anyone the advice to buy or sell shares, because the most benevolent price of advice can turn out badly.
Empty pockets never held anyone back. Only empty heads and empty hearts can do that.
Man can acquire accomplishments or he can become an animal, whichever he wants. God makes the animals, man makes himself.
In any country there must be people who have to die. They are the Sacrifices any nation has to make to achieve law and order.
Discussion in America means dissent.
Once the state has been founded, there can no longer be any heroes. They come on the scene only in uncivilized conditions.
God does not ask about our ability, but our availability.
A good swordsman is not given to quarrel.
Never give advice in a crowd.
Divorce is a declaration of independence with only two signers.
Unless a capacity for thinking be accompanied by a capacity for action, a superior mind exists in torture.
Dig within. Within is the wellspring of Good and it is always ready to bubble up, if you just dig.
The way to get things done is not to mind who gets the credit for doing them.
Sometimes I need what only you can provide, your absence.