What is the proper control structure to impose here?

[SlurrerOfSpeech]

I often encounter situations in my professional development like

if ( condition1 ) 
{
    // do this
}
if ( condition2 ) 
{
   // do that
}
if ( !condition1 && !condition2 ) 
{
   // do something else
}

and I handle it in exactly that manner. (Note: this is not an if-elseif-else situation. condition1 and condition2 are not mutually exclusive.)

However, my programming intuition says there's something wrong when it sees that the truthiness of condition1 and condition2 are each checked twice. The alternative

if ( !condition1 && !condition2 )
{
    // do something else
}
else
{
    if ( condition1 ) 
   {
       // do this
   }
   if ( condition2 ) 
   {
      // do that
    }
}

is even more like spaghetti code. I have yet a third way that gets rid of the redundant conditions by doing

int condval = 1 * (condition1 ? 1 : 0 ) + 2 * (condition2 ? 1 : 0);
if ( condval & 1 ) 
{
    // do this
}
if ( condval & 2 ) 
{
   // do that
}
if ( !condval  ) 
{
   // do something else
}

Yes, I've actually used the above thingy in real life. I feel bad for anyone who has to read my code and figure out what it does.

What is the proper way of handling this?

 

[Jarvis323]

Not necessarilly better, but you could also do this.

bool x=c1;
if (x) {
  Do first thing
}
if (c2) {
  Do second thing
}
else if (!x) {
  Do third thing
}

 

[SlurrerOfSpeech]

Not necessarilly better, but you could also do this.

bool x=c1;
if (x) {
  Do first thing
}
if (c2) {
  Do second thing
}
else if (!x) {
  Do third thing
}

That is not equivalent. If you have !c1 && c2 then the third thing is executed.

 

[Mark44]

Not necessarilly better, but you could also do this.

bool x=c1;
if (x) {
  Do first thing
}
if (c2) {
  Do second thing
}
else if (!x) {
  Do third thing
}

That is not equivalent. If you have !c1 && c2 then the third thing is executed.

I don't think so, if I understand what you're saying here.
If c1 is false (which means that x is false), the first if statement doesn't execute. If c2 is true, the second if statement executes, independent of whether x is true or false, but the third doesn't.

The third if clause executes if c2 is false and x is false.

The first two if statements both execute when x is true and c2 is true.

In case it's not completely obvious, the else if clause is associated with the second if clause.

In cases where the logic is contorted like this, due to conditions that are not mutually exclusive, it's really essential to add comments that explain why the logic is as it is.

It might be simpler to do this:

if (c1 && c2) {
   // Handle case where both conditions are true
}
else (if c1) {
   // Handle case where only c1 is true
}
else if (c2) {
   // Handle case where only c2 is true
}
else {
   // Handle case where neither c1 nor c2 is true
}

 

[Ibix]

In cases where the logic is contorted like this, due to conditions that are not mutually exclusive, it's really essential to add comments that explain why the logic is as it is.

PF needs two like buttons for this.

 

[SlurrerOfSpeech]

In case it's not completely obvious, the else if clause is associated with the second if clause.

Ah, I see now! Thanks.

 

[jtbell]

PF needs two like buttons for this.

I added mine to help you out.

 

[FactChecker]

I like this scheme:

if( condition1 ){
do one;
skip3=true;
}
if( condition2 ){
do two;
skip3=true;
}
if( ! skip3 ){
do three;
}

 

[SlurrerOfSpeech]

I like this scheme:

if( condition1 ){
do one;
skip3=true;
}
if( condition2 ){
do two;
skip3=true;
}
if( ! skip3 ){
do three;
}

^ I'm stealing this from you.

 

[Svein]

I have yet a third way that gets rid of the redundant conditions by doing

A slight adjustment:

int condval = (condition1 ? 1 : 0 ) + (condition2 ? 2 : 0);
switch (condval) {
  case 3: // condition1 AND condition2
  ...
     break;
  case 2: // condition2 AND (not condition1)
  ...
    break;
  case 1: // condition1 AND (not condition2)
  ...
    break;
  case 0: // (not condition1) AND (not condition2)
  ...
    break;
  default: // How the hell did you get here?
  ...
    break;
    };

 

[Jarvis323]

One thing to consider is adherence to the DRY principle.

 

[FactChecker]

^ I'm stealing this from you.

I was sloppy and forgot to initialize skip3 to false at the top.