Hi Pariya, You are right; if you want to use a composition you would have to use slots. "Using of a slot" is a simple statement on its own, but it has implications in the implementation space. A slot allows you to define a set of archetypes that can be used at that point, not necessarily a single one. The archetype author may leave a slot with a slightly generic constraint, allowing someone to insert one of the multiple archetypes matching the set defined by the constrained, or the author may simply define the constraint so that only a single archetype can be used at that node. If the author chooses the first option, then there has to be another phase in the modelling process where someone makes a choice for that node, or in adl speak, constraints the definition of that node further. At this point you now need a mechanism to express further constraints on archetypes, which leads to templates in the openEHR specification. I'll get back to this in a second.. If the author choses the second option and declares the archetype to be used in the slot from the beginning, an implementer still has to process an archetype nested into another.
The first case requires a) tooling to further narrow down the constraint, or in other words to select one of the allowed archetypes, b) a method of expressing this selection, saying that this node in this archetype will be filled with this other archetype. You can skip tooling for this if you have to, but one has to have a method of expressing the slot filling archetype. This formalism is being developed in the context of template specifications, which seems to be going in a good direction IMHO. In Opereffa, I did not implement this yet, but it is quite likely that I will not wait for the template specs to be finalized and archetype parser to catch up. I'll probably use my own temporary tooling and method for allowing design phase filling of archetype slots. The second case on the other hand, is simpler to implement, because all you have to do is to handle slot filling as you read archetype descriptions, there are no choices to make. So the logic is, when a wrapper is wrapping an archetype description node by node, slots would be handled recursively. I've actually implemented this on my working copy of Opereffa, so that I can load a composition archetype which specifies single slot filling archetypes. I've also added other capabilities, so this is good news. Bad news is: I've changed lots of code, and now I have to update the rest of the code that makes Opereffa tick. UI generation, UI layer to back end communication, persistence, reporting etc all have to be updated to work with this new set of wrappers which know how to consume nested archetypes. I would like to think that it will be over in about a month, but I can't give any guarantees. Given this situation, you have two choices: you can wait till I release the next version which can at least process the second use case, or you can go ahead and implement these using the existing source code. Of course, if you think you have a third option, I would be more than happy to hear it. Kind regards Seref On Tue, Mar 23, 2010 at 2:27 PM, Pariya Kashfi <hajar.kashfi at chalmers.se>wrote: > Hi, > yes I've seen the link, should read it more carefully though. Actually, I'm > trying to find out how one should connect a composition / section archetype > to the GUI. > In the example on opereffa the archetype is an observation which includes > items/values, but for case that we have nested archetypes I have no idea how > to do this... > > Sincerely > Pariya > > On Mar 19, 2010, at 4:52 PM, Seref Arikan wrote: > > Hi Pariya, > Did you see http://www.openehr.org/wiki/display/projects/Opereffa+Project > ? Any particular questions? > > Kind regards > Seref > > > On Thu, Mar 18, 2010 at 12:19 PM, Pariya Kashfi <hajar.kashfi at > chalmers.se>wrote: > >> Hi there, >> Does anyone know if there exist any documentation for opereffa framework? >> The only thing I could find was the short tutorial on the website >> >> Regards >> Pariya >> >> MSc; PhD Candidate >> Department of Computing Science and Engineering >> Chalmers University of Technology >> http://www.chalmers.se/cse/EN/people/kashfi-hajar >> >> >> >> >> >> >> >> _______________________________________________ >> openEHR-technical mailing list >> openEHR-technical at openehr.org >> http://lists.chime.ucl.ac.uk/mailman/listinfo/openehr-technical >> >> > <ATT00001..txt> > > > Regards > Pariya > > MSc; PhD Candidate > Department of Computing Science and Engineering > Chalmers University of Technology > http://www.chalmers.se/cse/EN/people/kashfi-hajar > > > > > > > > _______________________________________________ > openEHR-technical mailing list > openEHR-technical at openehr.org > http://lists.chime.ucl.ac.uk/mailman/listinfo/openehr-technical > > -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.openehr.org/mailman/private/openehr-technical_lists.openehr.org/attachments/20100323/6b7d2525/attachment.html>
