ecell-devel Mailing List for E-Cell System

Dear sha,

       Thank you for valuable opinions.

 > - Since use of MSVC is not a feature, but it is an optional package to use,
 >   I think --with-msvc is appropriate here.

       I agree with the opinion. I change it.

 > - I couldn't find/remember where this ECELL_API macro is defined.
 >   I hope someday we will be able to get rid of it in some way.

       In ecell/libecs/Defs.hpp .
       I seek a way not to define.

 > - How does the build system find ecell_config.h.msvc?   Is the user
 >   supposed to manually make a symbolic link to ecell_config.h?

       I think so, or its exchages.
       I feel MSVC users are easy to understand it.

Best regards,
YUASA Takeshi


> Yuasasan,
> 
> Thanks for the up to date patch for the unified source tree.
> 
> Generally I think it is ready to be committed to svn.
> 
> Small comments/questions;
> 
> - You added --enable-msvc option to configure.    Since use of MSVC
>   is not a feature, but it is an optional package to use, I think
>   --with-msvc is appropriate here.   Even better way is to detect the
>   compiler automatically.  It could also be specified manually with
>   CC or CXX envvar.  Should be simple to implement.
> 
> - I couldn't find/remember where this ECELL_API macro is defined.
>   I understand that it's a price of supporting VC, and we have to
>   live with that for now.   However it is obvious that this makes
>   maintenance of the code complicated than before.   We have to
>   remember and care about what this
>   macro means and does everytime we define new functions or modify
>   existing ones.    I hope someday we will be able to get rid
>   of it in some way.
> 
> - How does the build system find ecell_config.h.msvc?   Is the user
>   supposed to manually make a symbolic link to ecell_config.h?
> 
> 
> sha

Yuasasan,

Thanks for the up to date patch for the unified source tree.

Generally I think it is ready to be committed to svn.

Small comments/questions;

- You added --enable-msvc option to configure.    Since use of MSVC
  is not a feature, but it is an optional package to use, I think
  --with-msvc is appropriate here.   Even better way is to detect the
  compiler automatically.  It could also be specified manually with
  CC or CXX envvar.  Should be simple to implement.

- I couldn't find/remember where this ECELL_API macro is defined.
  I understand that it's a price of supporting VC, and we have to
  live with that for now.   However it is obvious that this makes
  maintenance of the code complicated than before.   We have to
  remember and care about what this
  macro means and does everytime we define new functions or modify
  existing ones.    I hope someday we will be able to get rid
  of it in some way.

- How does the build system find ecell_config.h.msvc?   Is the user
  supposed to manually make a symbolic link to ecell_config.h?


sha



> Hi,
> 
>     I have shared Ecell "Core code" of GCC and VC8.0.
> 
>     Because autoconf is not run on VC8.0,
>       I have prepared ecell/ecell_config.h of VC8.0 in advance.
> 
>     A file that shows differences existed
>       between current and shared code is attached.
> 
>     Please confirm the file.
>     If there is no problem, the code is committed to CVS.
> 
> Best regards,
> YUASA Takeshi
> 
> 
> 
> ------------------------------------------------------------------------
> 
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys - and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Ecell-devel mailing list
> Ece...@li...
> https://lists.sourceforge.net/lists/listinfo/ecell-devel

Hi Sandeep,  (sorry if you get this twice)

What's important to you is only in the last line:


NotFound: [Process:/:Cysthati_formn]: VariableReference [Phser] not
found in
this Process."

It seems you forgot to define Phser variable in /Cysthati_formn.


Ok we need to suppress that verbose backtraces except in
debug builds.   Anyone interested in this hack?


sha






> Dear E-cell Users,
> 
> I am getting the following error message when I load the model into E-cell 
> session monitor. Kindly tell me what this error message indicates. The error 
> message obtained is as follows:
> 
> "Loading Model file C:\Program Files\E-Cell 3\share\sucess.eml
> 
> Can't load [C:\Program Files\E-Cell 3\share\sucess.eml]
> Traceback (most recent call last):
> 
>   File "C:\Program Files\E-Cell 3\lib\osogo\MainWindow.py", line 631, in 
> __loadData
>     self.theSession.updateFundamentalWindows()
> 
>   File "C:\Program Files\E-Cell 
> 3\lib\python2.4\site-packages\ecell\GtkSessionMonitor.py", line 430, in 
> updateFundamentalWindows
>     anEntityListWindow.update()
> 
>   File "C:\Program Files\E-Cell 3\lib\osogo\EntityListWindow.py", line 543, 
> in update
>     self.thePropertyWindow.update()
> 
>   File "C:\Program Files\E-Cell 3\lib\osogo\plugins\PropertyWindow.py", line 
> 314, in update
>     self.thePrePropertyMap[str(aProperty)][0] =\
> 
>   File "C:\Program Files\E-Cell 
> 3\lib\python2.4\site-packages\ecell\EntityStub.py", line 180, in getProperty
>     return self.theSimulator.getEntityProperty( aFullPN )
> 
> RuntimeError: const libecs::VariableReference& 
> libecs::Process::getVariableReference(const libecs::String&):
> NotFound: [Process:/:Cysthati_formn]: VariableReference [Phser] not found in 
> this Process."
> 
> 
> Hope to get an early reply. Thanks in advance.
> 
> Regards,
> Sandeep Dixit.
> 
> _________________________________________________________________
> Shah Rukh fan? Know all about the Baadshah of Bollywood. On MSN Search 
> http://server1.msn.co.in/profile/shahrukh.asp
> 
> _______________________________________________
> e-cell-users mailing list
> e-c...@e-...
> http://mail.e-cell.org/mailman/listinfo/e-cell-users
> 

Hi,

    I have shared Ecell "Core code" of GCC and VC8.0.

    Because autoconf is not run on VC8.0,
      I have prepared ecell/ecell_config.h of VC8.0 in advance.

    A file that shows differences existed
      between current and shared code is attached.

    Please confirm the file.
    If there is no problem, the code is committed to CVS.

Best regards,
YUASA Takeshi

Good idea.
'make test' doubles the benefit of automatic nightly
builds.   If we have to prioritize things I would suggest
implementing the test make target first, and then proceed
to nightly build.   While doing so we would be able to find
and setup some machine which we can setup a cron job on.

sha

> Hei,
> 
> Could you consider running a test build nightly from the trunk source  
> tree? The benefit is that any build breaks would not go unnoticed too  
> long.  It ought to be enough to put the checkout and compile commands  
> to cron job and redirect output to a file. If the compilation fails,  
> users gets an email. It would be even better if there was even very  
> simple test suite. The sample models should suffice to begin with.  
> Please consider implementing this, because fixing compile and  
> regression issues can be very frustrating thing to do.
> 
> --
> Petteri
> 
> 
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Ecell-devel mailing list
> Ece...@li...
> https://lists.sourceforge.net/lists/listinfo/ecell-devel

Hei,

Could you consider running a test build nightly from the trunk source  
tree? The benefit is that any build breaks would not go unnoticed too  
long.  It ought to be enough to put the checkout and compile commands  
to cron job and redirect output to a file. If the compilation fails,  
users gets an email. It would be even better if there was even very  
simple test suite. The sample models should suffice to begin with.  
Please consider implementing this, because fixing compile and  
regression issues can be very frustrating thing to do.

--
Petteri

Hi,

Have anyone checked the SBML compatibility recently? I could not run  
convert my model from eml to SBML. I tried the SBML files in the src/ 
test-data directory of SBML 2.3.4 distribution package, and only some  
files can be converted to EML. For example, converting 1803- 
pass00-01.xml will bring a segmentation fault. Can anyone look into  
this? Thanks.

Bin