Using Else with blank If

[Lost User]

The bottom one is definitely better. Anyone who writes code like the first example you posted is confused and either needs a vacation or a book thrown at his face.

█▒▒▒▒▒██▒█▒██ █▒█████▒▒▒▒▒█ █▒██████▒█▒██ █▒█████▒▒▒▒▒█ █▒▒▒▒▒██▒█▒██

[pbraun]

Uhm yeah, I think most developers are guilty of doing something like this at one time or another. It certainly isn't the end of the world. I've seen far worse than this in production code that is currently running around the world. Phil

[AncientCodeslinger]

Both examples make me shudder. Why would you compare a boolean to True or False? It IS true or false.

[Roger Bamforth]

Absolutely. Along the same lines, my pet peeve is

bool b;

if (condition)
{
b = true;
}
else
{
b = false;
}

what on earth is wrong with

bool b = condition;

Regards - Roger

[Marcus J Smith]

I think it is just easier to read when you say say what you want it to equal.

If boolean = True

instead of just saying

If boolean

I just couldnt think of a good example but what I was getting at is really saying something like

If strType = "GOOD" OrElse strType = "COOL" Then
'Good Record
Else
blnError = True
End If

instead of saying

If strType <> "GOOD" AndAlso strType <> "COOL" Then
blnError = True
End If

---

CleaKO

"I think you'll be okay here, they have a thin candy shell. 'Surprised you didn't know that." - Tommy Boy
"Fill it up again! Fill it up again! Once it hits your lips, it's so good!" - Frank the Tank (Old School)

[jhwurmbach]

CleaKO wrote:

think it is just easier to read when you say say what you want it to equal. If boolean = True instead of just saying If boolean

Funny. For me its just the opposite: The superfluous = True imposes the nagging feel in me, I have missed something while reading the code. But then, the whole VB code gives me a screaming fit anyway. :rolleyes:

---

Failure is not an option - it's built right in.

[Dave Kreskowiak]

Yeah, it's just the opposite for me too. Especially if you name your boolean properties or methods correctly.

If UserIsInRole(user, role) Then

Just the name of the function implies that it returns a True/False value.

Dave Kreskowiak Microsoft MVP - Visual Basic

[Shog9 0]

I prefer

If ( isFlag ) Then Continue

Succinct. :cool:

----

It appears that everybody is under the impression that I approve of the documentation. You probably also blame Ken Burns for supporting slavery.

--Raymond Chen on MSDN

[Josh Smith]

Roger Bamforth wrote:

what on earth is wrong with bool b = condition;

In general I completely agree with you. The only time I might use the "bad" way is for debugging purposes. If I only want a breakpoint to hit when 'condition' is false, then I will create an explicit if/else branch. Of course, after debugging I revert the code back to bool b = condition; (I'm usually too lazy to setup debugger conditions!)

:josh: My WPF Blog[^]

[PIEBALDconsult]

CleaKO wrote:

If boolean = True

But then you may have trouble when the code is ported to C, so you really need: If True = boolean

--| "Every tool is a hammer." |--

[Hal Angseesing]

Yep, done this myself. And debugger conditional breakpoints are actually quite a lot slower than IP traps.

[John R Shaw]

It is a good idea to place the constant value to the left of the equality symbol in any language that supports them, to catch syntax errors. It just feels so unnatural to type it that way. ;)

INTP "Program testing can be used to show the presence of bugs, but never to show their absence."Edsger Dijkstra

[John R Shaw]

jhwurmbach wrote:

But then, the whole VB code gives me a screaming fit anyway.

I know the feeling. ;)

INTP "Program testing can be used to show the presence of bugs, but never to show their absence."Edsger Dijkstra

[Shog9 0]

You might still have trouble in C, if you run up against code that uses other non-zero values for "true"...

----

It appears that everybody is under the impression that I approve of the documentation. You probably also blame Ken Burns for supporting slavery.

--Raymond Chen on MSDN

[AncientCodeslinger]

If that's how you use booleans, you might as well just use an integer set to 1 or 0 or a string set to "Y" or "N" (or "True" or "False"). How about: somethingIsWrong = (type <> "GOOD" AndAlso type <> "Cool") With a variable name that clearly expresses the condition represented by the boolean, the comparison to True or False becomes clearly redundant.

[Graham Bradshaw]

Well, if the condition is to test the equality of two variables,

bool b;

if (x = y)
{
b = true;
}
else
{
b = false;
}

will give you a compiler warning, and

bool b = (x = y);

will not. (Note the assignment vs equality test bug)

[Kevin McFarlane]

It's also quite common. I often find myself starting with the first one, then after testing, stepping through, etc., I refactor to the second.

Kevin

[Kevin McFarlane]

The reason for the first is that some people have a fixation about negative conditionals, such that they will try and avoid them at all costs - such as the cost you describe!

Kevin

[Kevin McFarlane]

jhwurmbach wrote:

But then, the whole VB code gives me a screaming fit anyway.

But you see such style in the C-family languages too, though it's probably less common.

Kevin

[Roger Bamforth]

That's a good point and is actually something that had never occurred to me. However, whether or not you get a warning depends upon the types of x, y and b is not as simple as it seems. It is probably also language and compiler dependant. e.g in Visual C++

bool x = true;
bool y = true;
bool b = (x = y);

gives no warning, as you say, but

int x = true;
int y = true;
bool b = (x = y);

does generate a warning (forcing an int to be a bool) and

int x = true;
int y = true;
BOOL b = (x = y);

doesn't.

Regards - Roger