Getting things right

So let's get out of the illusion of comparing code one character at a time and try to find an abstraction that can be applied to updating all of these data types. There are two key points of this abstraction that have already been mentioned in the previous section:

• A change contains the information that can transform the value of an item from one to another

• Multiple changes could be generated or applied to one data item during a single synchronization

Now, starting from changes, let's think about what happens when an update method of a client is called.

Finding abstraction

Take a closer look to the method update of client:

• For data of the 'value' type, first we create the change, including a new value, and then update the change list to make the newly created change the only one. After that, we update the value of data item.

• For data of the 'increment' type, we add a change including the increment in the change list; or if a change that has not be synchronized already exists...