Why is casting double to char allowed in Java?

Question

Why is char c = (char)65.8; allowed in Java?

Shouldn't it throw an error since 65.8 is not an exact Unicode value? I understand that the double is truncated to an integer, in this case, 65, but it seems like bad design to me to allow the programmer to make such a cast.

Answer 1

That is called Narrowing type casting. From oracle docs:

22 specific conversions on primitive types are called the narrowing primitive conversions:

short to byte or char

char to byte or short

int to byte, short, or char

long to byte, short, char, or int

float to byte, short, char, int, or long

double to byte, short, char, int, long, or float

A narrowing primitive conversion may lose information about the overall magnitude of a numeric value and may also lose precision and range.

In Java, there are two basic types of type conversions: widening and narrowing.

A widening conversion occurs when you convert from a type with smaller (or narrower) to type with larger (or wider) range. Because of this, there is no chance for data loss and the conversion is considered "safe."

A narrowing conversion occurs when you convert from a type with larger (or wider) to type with smaller (or narrower) range. Since we are shrinking the range, there is a chance of data loss so this conversion is considered "unsafe"

The conversion from byte to char is a special case and represents widening and narrowing at the same time. The conversion starts by converting the byte to an int and then the int gets converted to the char.

One reason I can think of why narrowing type casting doesn't result in an error/exception is to allow for a convenient/easy/quick type conversion in the cases when no data will be loss. Compiler leaves it up to us to make sure converted data will be able to fit in the smaller range. It is also useful if we want to quickly truncate values such as rounding the value of a double (by type-casting it to an int).

Answer 2

it doesn't happen automatically on assignment: that would be a compilation error.

The fact that the programmer makes a conscious choice (e.g. the type cast) means she is taking into consideration the possibility of, and responsibility for, possible truncation.

Answer 3

You may have code such as cipher algorithms that may find useful to cast a double or float to char. Also, char is an unsigned type, which means (char)200.5 yields something different than (char)(byte)200.5.