Objective-C Fundamentals

The basic data types
35
while growing to a 64-bit integer when compiling the same source code on a 64-bit
platform. This allows code to take advantage of the increased range provided by 64-bit
integers while not using excessive memory when targeting 32-bit systems.
Eagled-eyed developers may wonder why new data types such as NSInteger were
introduced when existing data types such as long int would appear to already fit the
desired role. NSInteger exists for use in APIs that should be 64-bit integers on 64-bit
platforms but for one reason or another must be typed as int instead of long int on
32-bit platforms.
Good habits learned now mean less hassle in the future
Declaring your variables using data types such as NSInteger and NSUInteger can
be considered a form of future-proofing your source code. Although they’re identical
to int and unsigned int when compiling for the iPhone today, who knows what’s
around the corner? Perhaps a future iPhone or iPad will be a 64-bit device, or you’ll
want to reuse some of your source code in a matching desktop application.
It’s much easier to establish the habit of using portable data types such as NSInteger
or NSUInteger now (even if you don’t get much immediate benefit from it) than
to have to correct such portability issues in the future.
2.2.2 Filling in the gaps—floating-point numbers
When modeling the real world, it’s common to come across numbers that contain a
fractional part, such as 0.25 or 1234.56. An integer variable can’t store such values, so
Objective-C provides alternative data types called float and double for storing this
kind of data. As an example, you can declare a variable f of type float and assign it
the value 1.4 as follows:
float f = 1.4;
Floating-point constants can also be expressed in scientific or exponential notation by
using the character e to separate the mantissa from the exponent. The following two
variable declarations are both initialized to the same value:
float a = 0.0001;
float b = 1e-4;
The first variable is assigned the value 0.0001 via a familiar decimal constant. The second
variable is assigned the same value but this time via a constant expressed in scientific
notation. The constant 1e-4 is shorthand for 1 × 10 -4 , which, once calculated, produces
the result 0.0001. The e can be thought of as representing “times 10 to the power of.”
In Objective-C, floating-point variables are available via two main data types, which
trade off the range of possible values they can represent and the amount of memory
they use, as shown in table 2.3.
You may wonder how a variable of type float or double can store such a wide
range of values when a similar sized int can only store a much smaller range of values.
The answer lies in how floating-point variables store their values.