Chronological Thread 

< Chronological >      < Thread >

---

Hi

I’ve updated the document at
https://github.com/codecurve/libcellml/blob/uml01/docs/source/design.rst
to reflect some of the items that it seems will be the conclusion of the
feedback.
- Just libcellml namespace
- “Units” class rather than “Unit”
- Removed “addImport” and “getXML” from Model
- Link to Doxygen (also updated where appropriate)

I still think that Manager will be needed as a support class for handling
some of the nuances that will crop up when serialising a group of models
that are linked via imports. I think that some code for this would shed
light on the issues.

Regards,
Randall


On 3/02/15 4:03 pm, "Randall Britten"
<r.britten AT auckland.ac.nz>
wrote:

>OK, happy to follow this policy: align LibCellML¹s entities¹ (classes,
>members) names to serialisation¹s entities¹ (elements, attributes) names,
>when an obvious mapping exists. The alternative would have been
>consistency with regards to the policy of keeping the library untainted by
>serialisation issues, but I agree that this would just be confusing in
>this case.
>
>I¹ve also added https://tracker.physiomeproject.org/show_bug.cgi?id=3859
>for the record.
>
>Regards,
>Randall
>
>On 3/02/15 1:32 pm, "David Nickerson"
><david.nickerson AT gmail.com>
> wrote:
>
>>> With regards to Unit/Units, see for example
>>> https://en.wikipedia.org/wiki/SI_derived_unit which consistently uses
>>> singular and plural in the way that I described. Given that we are
>>>going
>>> to finalise the CellML 1.2 spec in light of what we learn from
>>>LibCellML,
>>> there is an opportunity to fix the spec too.
>>
>>CellML 1.2 will have a "units" element which may contain a collection
>>of "unit" elements (there is no proposal to change this under
>>https://tracker.physiomeproject.org/show_bug.cgi?id=55). Application
>>developers are free to present data from a CellML model in whatever
>>manner is considered most appropriate for their users. As mentioned
>>many times, it seems sensible to me that objects in the libCellML API
>>which have a 1:1 correspondence to info-items in the specifications
>>use the same name.
>>
>>The primary goal with libCellML in regard to the finalising of the
>>CellML 1.2 specifications is to ensure the objects and rules interact
>>in the manner that is expected based on conceptual discussions. Not to
>>worry about changing the names of elements.
>>
>>
>>Cheers,
>>Andre.
>>
>>--
>>
>>
>>David Nickerson
>>about.me/david.nickerson
>

---

• Re: [[cellml-dev] ] Feedback on libCellML design document, Randall Britten, 02/03/2015
  • <Possible follow-up(s)>
  • RE: [[cellml-dev] ] Feedback on libCellML design document, Richard Christie, 02/03/2015
    • Re: [[cellml-dev] ] Feedback on libCellML design document, Randall Britten, 02/03/2015
      • Re: [[cellml-dev] ] Feedback on libCellML design document, David Nickerson, 02/03/2015
        • Re: [[cellml-dev] ] Feedback on libCellML design document, Randall Britten, 02/03/2015
          • Re: [[cellml-dev] ] Feedback on libCellML design document, Randall Britten, 02/04/2015
            • Re: [[cellml-dev] ] Feedback on libCellML design document, David Nickerson, 02/04/2015
              • Re: [[cellml-dev] ] Feedback on libCellML design document, Randall Britten, 02/09/2015
              • Re: [[cellml-dev] ] Feedback on libCellML design document, David Nickerson, 02/10/2015
              • Re: [[cellml-dev] ] Feedback on libCellML design document, Randall Britten, 02/12/2015

---

Archive powered by MHonArc 2.6.18.