The association in the model stores the fact that "" has been used as the container (in language properties of the role). I don't understand why there is an ambiguity. "Andrew McVeigh" wrote in message news:963npn$lqa$1. We'll definitely consider it, and will try to fix the problem ASAP. This is a pretty good idea to use language properties in this situation. "Andrew McVeigh" wrote in message news:95uuc4$aqj$1. #Magicdraw uml repeated attribute softwareOO software design can be an appealing challenge if you have If you have some suggestions, how would you like the MagicDraw to behave, let us know, we'd be glad to hear you opinion. This is not a proper functionality, but we are working on it. Hope this will clear the situation a bit. So you have two members/associations instead of original one. Because you choose to merge model and code, old association is left untouched. It finds a member with Vector type, and creates an appropriate member in the class. Let me tell now how the reverse engine works. So, generating a member which type is vector, so loose all connection between the aggregating and aggregated classes. When you have an association with a multiplicity greater than 1, MagicDraw generates a container member in the java source code. The problem is in the ambiguity of the situation. So, it doesn't seem to recognise that the reversed attribute "rooms" is the same as the current association, even though it has the same type and name, and I have checked "merge model and code"!!! I am using this current option set on the advice of MagicDraw in the first place, and it works well for multiplicities of 1. However, if I try to generate again, I get: #Magicdraw uml repeated attribute codeTo demonstrate this, create a House class which aggregates a number of Rooms: Then, assign to a code engineering set, and generate using the following options: The reversing options are set to: The first version of "House.java" looks great: public class House You are right, this is the problem, which we promise to fix ASAP. "Jolita Guzauskaite" wrote in message news:3A76D763.EF0AC75D. Any idea on when it will be fixed? Any chance of a patched version? I have downloaded the final version of MD4.0, and duplicate reversed attributes is still a problem for 1 to many associations using vector. > place, and it works well for multiplicities of 1. > using this current option set on the advice of MagicDraw in the first > type and name, and I have checked "merge model and code"!!! I am > is the same as the current association, even though it has the same > So, it doesn't seem to recognise that the reversed attribute "rooms" > However, if I try to generate again, I get: > first version of "House.java" looks great: public class House > the following options: The reversing options are set to: The > Rooms: Then, assign to a code engineering set, and generate using > To demonstrate this, create a House class which aggregates a number of OO software design can be appealing challenge if you have You are right, this is the problem, which
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |