Pablo Pazos wrote: > and more!, may be the validValue() method of CDomainType has to be in > the LeafContraint class, because if you want to validate primitive > data, the CPrimitiveObject class dont have a validValue() method. > If validValue() is in LeafConstraint then both CPrimitiveObject and > CDomainType inherit validValue(). Please note that LeafConstraint class is not part of the RM specification. It is created to facilitate implementation of C_DOMAIN_TYPE and C_PRIMITIVE. In fact, these two classes share nearly identical interface which probably should be extracted as a common interface called C_DATA_TYPE. Also C_PRIMITIVE_OBJECT doesn't seem to be very useful. It can be skipped and instead C_PRIMITVE can be moved into constraint model package and renamed to C_PRIMITIVE_TYPE to be consistent.
Cheers, Rong > > Pablo Pazos > NIB > > ----- Original Message ----- > *From:* Rodrigo Filgueira <mailto:rfilgue at fing.edu.uy> > *To:* openehr-technical at openehr.org > <mailto:openehr-technical at openehr.org> > *Sent:* Tuesday, August 15, 2006 12:19 PM > *Subject:* C_PRIMITIVE and C_DOMAIN_TYPE > > Should C_DOMAIN_TYPE classes be defined based (using) on C_PRIMITIVE? > > and more, > > are any of the C_PRIMITIVE classes used directly as a constraint > to an archetype element? > > these questions refer to the fact that string or integers by > themselves do not have any semantic meaning, they acquire meaning > inside DATA_VALUE instances, do they not? > > So as we advance in using and understanding AM we find that more > classes may be needed in the PROFILE package in order to validate > DATA_VALUES. > > just thoughts, and doubts, we are trying to understand the model, > and how it is used. > If you could tell us how you use it, I believe it would be very > helpful. > thank you > > > ------------------------------------------------------------------------ > > Rodrigo Filgueira > Asistente Docente/Investigador > N?cleo de Ingenier?a Biom?dica, FING - UDELAR >
