Avoid inheritance for implementation
m (Reverted edits by Ebybymic (Talk); changed back to last version by JaninaVoigt) |
|||
| (3 intermediate revisions by 3 users not shown) | |||
| Line 9: | Line 9: | ||
Instead, the Grapher should contain a DataList object internally. This avoids the above issues, as well as being more semantically correct (a Graph '''has''' a DataList) | Instead, the Grapher should contain a DataList object internally. This avoids the above issues, as well as being more semantically correct (a Graph '''has''' a DataList) | ||
== See Also== | == See Also== | ||
| − | [[Favor composition over inheritance]] | + | *[[Favor composition over inheritance]] |
| + | *[[Don't burn your base class]] | ||
| + | *[[BaseBean]] | ||
Latest revision as of 03:05, 25 November 2010
The core idea here is not to create a subclass merely to access the functionality of the superclass. An example:
class Grapher extends DataList
Now the Graphing class can use all the functionality of the DataList class. But intuitively, the declaration looks wrong, mainly because a Grapher isn't a type of DataList. A graph should have have a DataList, but it isn't strictly a list of data itself.
This will cause problems since Grapher exposes its internal data list to outside classes (bad for encapsulation), we can't extend the class easily if we want a graph that draws two lists of data, rather than a single one, and also, we are creating unnecessary class hierarchy.
Instead, the Grapher should contain a DataList object internally. This avoids the above issues, as well as being more semantically correct (a Graph has a DataList)
