CODEWITHSUNDEEP
c#casting
user393267
user393267

Reputation:

VS tell me that the cast is redundant, but it won't return the correct datatype even without

Bear with me: I have a very simple comparison, between a value, contained in a dictionary, which is an int, and a float.

Before the comparison, I divide the value by 100, so I get a decimal value (the int is 50, divide by 100, get 0.5).

baseline = 0f;
if ((mydict["intvalue"] / 100) > baseline)
    ...do something

0.5 is greater than 0.0, so it should "do someting".

Now, the comparison fail all the times; because an int divided by 100, where the result is between 0 and 1, will end up being an int with value 0, from what I can see. I did cast the value; to solve the issue, but VS tell me that the cast is redundant; and either way, the result is still 0. I don't get what I am doing wrong here.

baseline = 0f;
if ((float)(mydict["intvalue"] / 100) > baseline)
    ...do something

The compiler tell me that the cast is not necessary, but even if I leave the cast there; I still get 0.

Upvotes: 1

Views: 380

Answers (4)

Yeezh
Yeezh

Reputation: 161

(float)(mydict["intvalue"] / 100) > baseline

this code run in this order.

mydict["intvalue"]=50

(float)(mydict["intvalue"] / 100) > baseline => (float)(50 / 100) > baseline

50 / 100 = 0 here is the problem

(float)(50 / 100) > baseline => (float)0 > baseline

(float)0=0.0f

(float)0 > baseline => 0.0f>0.0f =>false

Upvotes: 0

BWA
BWA

Reputation: 5764

Your code 

if ((float)(mydict["intvalue"] / 100) > baseline)

says: Get int divde by int and result also int cast to float. So you need to modify code to operate on floats:

if (((float)mydict["intvalue"] / 100) > baseline)

or

if ((mydict["intvalue"] / 100f) > baseline)

Upvotes: 0

romain-aga
romain-aga

Reputation: 1561

Or implicit cast like this

if (mydict["intvalue"] / 100f > baseline)

Upvotes: 0

Ian
Ian

Reputation: 30813

That is because the float casting should be done in one of the items (either mydict["intvalue"] or 100) and not their result:

//here is one correct way of doing it
if (((float)mydict["intvalue"] / 100) > baseline)

//alternatively
if ((mydict["intvalue"] / 100f) > baseline)

What you do before is having integer division, and then you cast to float and therefore it is redundant:

//note that here you do division first before casting to float
if ((float)(mydict["intvalue"] / 100) > baseline)

Upvotes: 3

Related Questions