Question

This happens to me every once in a while and I always end up solving it the same way and then wishing for a cleaner way.

I start out with a calls to related utility functions, followed by and update call.

SynchA();
SynchB();
UpdateLastTime();

Then later I add check boxes so I have:

if(synchA.Checked)
{
  SynchA();
}
if(synchB.Checked)
{
  SynchB();
}

BUT now I only want to call UpdateLastTime() of ONE OR BOTH the two executed so invariably I end up with:

bool synchHappened = false;
if(synchA.Checked)
{
  SynchA();
  synchHappened = true;
}
if(synchB.Checked)
{
  SynchB();
  synchHappened = true;
}
if(synchHappened)
{
  UpdateLastTime();
}

That final step always bothers me because I’m spreading this one bool around to three branches of logic.

Is there some obvious better approach to the above logic/scenario that I could use?

Answer 1

The main goal would be – each time as logic has changed – code should be affected at least as possible.
So you’ve to structure such things once and then it will work for you.

In your particular case I would suggest to Keep It Simple (no Strategy Pattern, and so on), so
extract and encapsulate a logic of switches into the properties. So each time as requirements will be changed – you’ve to update either logic of particular switch or main logic itself.

Switches with encapsulated rules:

bool IsUpdateLastTime
{
 get
 {
    // logic here even can be fully or partially injected 
    // as Func<bool>
    return this.IsSyncA || this.IsSyncB;
 }
}

bool IsSyncA { get { return synchA.Checked; } }
bool IsSyncB { get { return synchB.Checked; } }

Main logic:

if (this.IsUpdateLastTime)
{
  this.UpdateLastTime();
}