[clam-devel] Changes to NetworkEditor

David García Garzón david.garcia at barcelonamedia.org
Mon Oct 18 15:49:37 PDT 2010


Some after session changes: Extracted the NetworkUpgrader class so it can be 
eventually turned into a lib and reused anywhere. Also added some c++/python 
helpers and fixed some memory issues on error handling.

David.


On Monday 18 October 2010 20:27:40 David García Garzón wrote:
> After a pair programming session with Xavier, this feature is already
> working on svn. We moved the migration script from plugin/spatializaiton
> to NetworkEditor keeping a link in the former place.
> 
> We added error reporting functions to the previous patch so now failing
> migration scripts or even errors in the clamRefactor are notified to the
> user. We consider this task finished, so Xavi can move on his next goal,
> but any feedback on upgrading networks is still welcome.
> 
> David.
> 
> On Wednesday 06 October 2010 19:59:34 Xavier Serra Román wrote:
> > Hi guys,
> > I finally completed the integration with clamrefactor. Now it works
> > without setting the PYTHONPATH. The code has en error though, when the
> > function that uses clamrefactor is called this shows: Traceback (most
> > recent call
> > 
> > last):
> >   File "<string>", line 342, in <module>
> >   File "/usr/lib/python2.6/optparse.py", line 1376, in parse_args
> >   
> >     rargs = self._get_args(args)
> >   
> >   File "/usr/lib/python2.6/optparse.py", line 1358, in _get_args
> >   
> >     return sys.argv[1:]
> > 
> > I don't know why or how to change it.
> > 
> > Bye!
> > 
> > On Mon, Oct 4, 2010 at 4:31 PM, Xavier Serra Román
> > <xvr.serra at gmail.com<mailto:xvr.serra at gmail.com>> wrote: Patch after
> > another refactoring.
> > The error messages are shown through Qt message box and not on the
> > console. Also, the migrationScript is now embedded in NetworkEditor, so
> > the only thing left to decide is where to put the migrationScript (right
> > now it's in the parent directory of NetworkEditor) and change the line
> > where it is embedded to the final location.
> > 
> > I'm still working to solve the issue about clamrefactor.py, because right
> > now the PYTHONPATH enviroment variable has to be set up to where
> > clamrefactor.py is for it to fully work
> > 
> > Bye!
> > 
> > 
> > On Fri, Sep 10, 2010 at 2:10 PM, David García Garzón
> > <david.garcia at barcelonamedia.org<mailto:david.garcia at barcelonamedia.org>>
> > wrote: After a refactoring session with Xavi, the patch is commited but
> > the actual upgrading is disabled as it needs the migration script to be
> > at specific locations. When Xavi solves that the functionality will be
> > working.
> > 
> > Important: That commit adds a new dependency on python-dev to compile
> > NetworkEditor.
> > 
> > David.
> > 
> > On Saturday 31 July 2010 02:57:04 Xavier Serra wrote:
> > > On Jul 30, 2010, at 6:19 PM, David García Garzón wrote:
> > > 
> > > On Tuesday 27 July 2010 18:55:43 Xavier Serra Román wrote:
> > > Hi guys,
> > > I finished the integration of clamrefactor with NetworkEditor. The code
> > > is in the patch. I'm sorry for all the warning it throws, I don't know
> > > how to silence them. The code has some limitations:
> > > 
> > > 1. clamrefactor.py and migrationScript need to be in the same directory
> > > from where you launch NetworkEditor 2. NetworkEditor has to be launched
> > > using: PYTHONPATH=`pwd` ./NetworkEditor This is needed for the Python
> > > Interpreter embedded to find the module.
> > > 
> > > The SConstruct file is changed too because it's needed to link with the
> > > Python libraries
> > > 
> > > Bye!
> > > 
> > > We should meet to decide what to do with the scripts and paths.
> > > Embedding would be the perfect solution, at least for the migration
> > > script but also for the clamrefactor module. Is that possible?
> > > 
> > > 
> > > I don't really know. The safest assumption would be that it's possible.
> > > I'll search around to see If I find something. We should decide too how
> > > to treat errors (tell the user the specific error; just tell that
> > > something happened; do nothing and load the network file without any
> > > change...)
> > > 
> > > 
> > > It also looks like the loadFromString and loadFrom file have some code
> > > duplication, and that the python bit looks like an alien we should move
> > > apart to some class.
> > > 
> > > 
> > > I though so too, but I didn't know how to solve the problem. Maybe we
> > > can come up with a solution together.
> > > 
> > > If you are still by Barcelona, drop yourself by my office and let's do
> > > some pair programming.
> > > 
> > > David.
> > > _______________________________________________
> > > clam-devel mailing list
> > > clam-devel at lists.clam-project.org<mailto:clam-devel at lists.clam-project.
> > > or
> > > g><mailto:clam-devel at lists.clam-project.org<mailto:clam-devel at lists.cl
> > > am- project.org>>
> > > http://lists.clam-project.org/listinfo.cgi/clam-devel-clam-project.org
> > 
> > _______________________________________________
> > clam-devel mailing list
> > clam-devel at lists.clam-project.org<mailto:clam-devel at lists.clam-project.or
> > g> http://lists.clam-project.org/listinfo.cgi/clam-devel-clam-project.org
> 
> _______________________________________________
> clam-devel mailing list
> clam-devel at lists.clam-project.org
> http://lists.clam-project.org/listinfo.cgi/clam-devel-clam-project.org



More information about the clam-devel mailing list