| Creating a macroscopic difffrom a collection of unit diffs [message #1820944] |
Sun, 02 February 2020 23:47 |
 Steve Hickman
Messages: 13
Registered: January 2017 |
Junior Member |
|
|
The Developer Guide description of 'Refinements' is 'spartan'. Between that and the 1 topic on this forum that mentions macroscopic diffs, there seems to be an implication that creating means picking one of the diffs as the macroscopic diff and setting the
'refines' and 'refinedBy' pointers to the other diffs.
In my application, all the 'unit' diffs are 'equal' - none of the objects referenced in any of the diffs 'contains' any of the other referenced objects. Further, there are several different kinds of 'macroscopic' diffs that could occur - and I want to distinguish them for the end user. My initial thought is to simply define additional classes that implement the 'Diff' interface - 1 for each kind of macroscopic diff - and the create an instance of the appropriate type and have it refinedBy the relevant set of unit diffs.
Before I start implementing this approach, I thought I would check to see if this is the preferred approach or if there is some other approach that I should consider.
My questions:
1) Is there a preferred approach for defining macroscopic diffs?
2) If there are multiple approaches, are there any specific issues I should keep in mind when deciding between approaches?
|
|
|
Powered by
FUDForum. Page generated in 0.06845 seconds
] 
Search
Help
Register
Login
Home
