A list for the developers of CellML tools
Text archives Help
- From: ak.miller at auckland.ac.nz (Andrew Miller)
- Subject: [cellml-dev] The model of a connection
- Date: Thu, 01 Apr 2010 10:52:06 +1300
Lucian Smith wrote:
> ...anyone? Bug? Misunderstanding? Something else?
Hi Lucian,
If the connection really is in Pandit_Hinch_Niederer, and it is
returning the wrong model, it is a bug.
A few things: are you sure that the call succeeded? Are you checking
your rvs to make sure something didn't fail, leaving you with a stale
value of something?
Can you produce a small test program (preferably not using the XPCOM
bridge) that demonstrates this problem?
Best wishes,
Andrew
>
> -Lucian
>
> * Lucian Smith <lpsmith at spod-central.org> [2010-03-29 19:46] writes:
>> I have what I *think* is a bug in the CellML API, but I want to check to
>> make sure it's not a misunderstanding first.
>>
>> I am reading the model
>> terkildsen_niederer_crampin_hunter_smith_2008/Pandit_Hinch_Niederer.cellml
>>
>> which, in turn, imports Pandit_et_al_2001_endo.cellml
>>
>> The first model imports a bunch of components from the second, and then
>> makes some connections between two components that it imported from that
>> second model:
>>
>> <!-- These components are imported from the Pandit et al endocardial cell
>> model -->
>> <import xmlns:xlink="http://www.w3.org/1999/xlink";
>> xlink:href="Pandit_et_al_2001_endo.cellml">
>> <units name="mV" units_ref="millivolt"/>
>> <units name="uF" units_ref="microF"/>
>> <units name="mM" units_ref="millimolar"/>
>> <units name="mJ_per_mole_K"
>> units_ref="millijoule_per_mole_kelvin"/>
>> <units name="C_per_mole" units_ref="coulomb_per_mole"/>
>>
>> <component component_ref="sodium_current"
>> name="pandit_sodium_current"/>
>> <component
>> component_ref="Ca_independent_transient_outward_K_current"
>> name="pandit_Ca_independent_transient_outward_K_current"/>
>> <component component_ref="steady_state_outward_K_current"
>> name="pandit_steady_state_outward_K_current"/>
>> <component component_ref="inward_rectifier"
>> name="pandit_inward_rectifier"/>
>> <component
>> component_ref="hyperpolarisation_activated_current"
>> name="pandit_hyperpolarisation_activated_current"/>
>> <component component_ref="background_currents"
>> name="pandit_background_currents"/>
>> <component component_ref="standard_ionic_concentrations"
>> name="pandit_standard_ionic_concentrations"/>
>> </import>
>>
>> And, later:
>>
>> <connection>
>> <map_components component_1="pandit_sodium_current"
>> component_2="pandit_background_currents"/>
>> <map_variables variable_1="E_Na" variable_2="E_Na"/>
>> </connection>
>>
>>
>> Now, you'll notice that the imported components have been given new
>> names--what used to be the component 'sodium_current' is now
>> 'pandit_sodium_current'. So when I get to the connection in question, I
>> need to know which model it came from--if it came from the main model
>> (which it does), the components to connect will be 'pandit_sodium_current'
>> and 'pandit_background_currents', while if that connection was present in
>> the submodel, the component names will be 'sodium_current' and
>> 'background_currents'. So, I ask the connection which model it came from:
>>
>> rv = connection->GetModelElement(getter_AddRefs(topmodel));
>>
>> ...and this is where the API fails me, because 'topmodel' is
>> 'pandit_et_al_2001_endo' instead of 'Pandit_Hinch_Niederer_Model'.
>>
>> Presumably, it's lying to me because both of the variables it wants to
>> connect are in the first and not directly in the latter, but since the
>> names have changed, I can no longer find the varibles in question.
>>
>> On the positive side, this is the first time the issue's come up when
>> going through all the models on cellml.org alphabetically, so making it to
>> the p's is pretty good!
>>
>> -Lucian
>> _______________________________________________
>> cellml-tools-developers mailing list
>> cellml-tools-developers at cellml.org
>> http://www.cellml.org/mailman/listinfo/cellml-tools-developers
> _______________________________________________
> cellml-tools-developers mailing list
> cellml-tools-developers at cellml.org
> http://www.cellml.org/mailman/listinfo/cellml-tools-developers
- [cellml-dev] The model of a connection, Lucian Smith, 04/01/2010
- [cellml-dev] The model of a connection, Andrew Miller, 04/01/2010
- [cellml-dev] The model of a connection, Lucian Smith, 04/01/2010
- [cellml-dev] The model of a connection, Andrew Miller, 04/01/2010
Archive powered by MHonArc 2.6.18.
