This chapter explains the features, technical details and syntaxes of the C programming language. I assume that you could write some simple programs. Otherwise, read "Introduction to Programming in C for Novices and First-time Programmers".
Introduction to C
C StandardsC is standardized as ISO/IEC 9899.
- K&R C: Pre-standardized C, based on Brian Kernighan and Dennis Ritchie [K&R] "The C Programming Language" 1978 book.
- C90 [ISO/IEC 9899:1990 "Programming Languages. C"]. Also known as ANSI C 89.
- C99 [ISO/IEC 9899:1999 "Programming Languages. C"]
- C11 [ISO/IEC 9899:2011 "Programming Languages. C"]
[TODO]
C Strength and Pitfall[TODO]
Basic Syntaxes
Revision
Below is a simple C program that illustrates the important programming constructs [sequential flow, while-loop, and if-else] and input/output. Read "Introduction to Programming in C for Novices and First-time Programmers" if you need help in understanding this program.
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 | #include int main[] { int sumOdd = 0; int sumEven = 0; int upperbound; int absDiff; printf["Enter the upperbound: "]; scanf["%d", &upperbound]; int number = 1; while [number sumEven] { absDiff = sumOdd - sumEven; } else { absDiff = sumEven - sumOdd; } printf["The sum of odd numbers is %d.\n", sumOdd]; printf["The sum of even numbers is %d.\n", sumEven]; printf["The absolute difference is %d.\n", absDiff]; return 0; } |
Comments
Comments are used to document and explain your codes and program logic. Comments are not programming statements and are ignored by the compiler, but they VERY IMPORTANT for providing documentation and explanation for others to understand your program [and also for yourself three days later].
There are two kinds of comments in C:
- Multi-line Comment: begins with a /* and ends with a */, and can span several lines.
- End-of-line Comment: begins with // and lasts till the end of the current line.
You should use comments liberally to explain and document your codes. During program development, instead of deleting a chunk of statements permanently, you could comment-out these statements so that you could get them back later, if needed.
Statements and Blocks
Statement: A programming statement is the smallest independent unit in a program, just like a sentence in the English language. It performs a piece of programming action. A programming statement must be terminated by a semi-colon [;], just like an English sentence ends with a period. [Why not ends with a period like an english sentence? This is because period crashes with decimal point - it is hard for the dumb computer to differentiate between period and decimal point!]
For examples,
int number1 = 10; int number2, number3 = 99; int product; product = number1 * number2 * number3; printf["Hello\n"];Block: A block [or a compound statement] is a group of statements surrounded by braces { }. All the statements inside the block is treated as one unit. Blocks are used as the body in constructs like function, if-else and loop, which may contain multiple statements but are treated as one unit. There is no need to put a semi-colon after the closing brace to end a complex statement. Empty block [without any statement] is permitted. For examples,
if [mark >= 50] { printf["PASS\n"]; printf["Well Done!\n"]; printf["Keep it Up!\n"]; } if [number == 88] { printf["Got it\n"]; } else { printf["Try Again\n"]; } i = 1; while [i < 8] { printf["%d\n", i]; ++i; } int main[] { ...statements... }White Spaces and Formatting Source Codes
White Spaces: Blank, tab and new-line are collectively called white spaces. C ignores extra white spaces. That is, multiple contiguous white spaces are treated as a single white space.
You need to use a white space to separate two keywords or tokens, e.g.,
int sum=0; double average; average=sum/100.0;Additional white spaces and extra lines are, however, ignored, e.g.,
int sum = 0 ; double average ; average = sum / 100.0;Formatting Source Codes: As mentioned, extra white spaces are ignored and have no computational significance. However, proper indentation [with tabs and blanks] and extra empty lines greatly improves the readability of the program, which is extremely important for others [and yourself three days later] to understand your programs. For example, the following hello-world works, but can you understand the program?
#include int main[]{printf["Hello, world!\n"];return 0;}Braces: Place the beginning brace at the end of the line, and align the ending brace with the start of the statement.
Indentation: Indent the body of a block by an extra 3 [or 4 spaces], according to its level.
For example,
#include int main[] { int mark = 70; if [mark >= 50] { printf["You Pass!\n"]; } else { printf["You Fail!\n"]; } return 0; }Most IDEs [such as CodeBlocks, Eclipse and NetBeans] have a command to re-format your source code automatically.
Note: Traditional C-style formatting places the beginning and ending braces on the same column. For example,
#include int main[] { int mark = 70; if [mark >= 50] { printf["You Pass!\n"]; } else { printf["You Fail!\n"]; } return 0; }Preprocessor Directives
C source code is preprocessed before it is compiled into object code [as illustrated].
A preprocessor directive, which begins with a # sign [such as #include, #define], tells the preprocessor to perform a certain action [such as including a header file, or performing text replacement], before compiling the source code into object code. Preprocessor directives are not programming statements, and therefore should NOT be terminated with a semi-colon. For example,
#include #include #define PI 3.14159265 // DO NOT terminate preprocessor directive with a semi-colonIn almost all of the C programs, we use #include to include the input/output stream library header into our program, so as to use the IO library function to carry out input/output operations [such as printf[] and scanf[]].
More on preprocessor directives later.
Variables and Types
Variables
Computer programs manipulate [or process] data. A variable is used to store a piece of data for processing. It is called variable because you can change the value stored.
More precisely, a variable is a named storage location, that stores a value of a particular data type. In other words, a variable has a name, a type and stores a value.
- A variable has a name [or identifier], e.g., radius, area, age, height. The name is needed to uniquely identify each variable, so as to assign a value to the variable [e.g., radius=1.2], and retrieve the value stored [e.g., area = radius*radius*3.1416].
- A variable has a type. Examples of type are,
- int: for integers [whole numbers] such as 123 and -456;
- double: for floating-point or real numbers such as 3.1416, -55.66, having a decimal point and fractional part.
- A variable can store a value of that particular type. It is important to take note that a variable in most programming languages is associated with a type, and can only store value of the particular type. For example, a int variable can store an integer value such as 123, but NOT real number such as 12.34, nor texts such as "Hello".
- The concept of type was introduced into the early programming languages to simplify interpretation of data made up of 0s and 1s. The type determines the size and layout of the data, the range of its values, and the set of operations that can be applied.
The following diagram illustrates two types of variables: int and double. An int variable stores an integer [whole number]. A double variable stores a real number.
Identifiers
An identifier is needed to name a variable [or any other entity such as a function or a class]. C imposes the following rules on identifiers:
- An identifier is a sequence of characters, of up to a certain length [compiler-dependent, typically 255 characters], comprising uppercase and lowercase letters [a-z, A-Z], digits [0-9], and underscore "_".
- White space [blank, tab, new-line] and other special characters [such as +, -, *, /, @, &, commas, etc.] are not allowed.
- An identifier must begin with a letter or underscore. It cannot begin with a digit. Identifiers beginning with an underscore are typically reserved for system use.
- An identifier cannot be a reserved keyword or a reserved literal [e.g.,int, double, if, else, for].
- Identifiers are case-sensitive. A rose is NOT a Rose, and is NOT a ROSE.
Caution: Programmers don't use blank character in names. It is either not supported, or will pose you more challenges.
Variable Naming ConventionA variable name is a noun, or a noun phrase made up of several words. The first word is in lowercase, while the remaining words are initial-capitalized, with no spaces between words. For example, thefontSize, roomNumber, xMax, yMin, xTopLeft and thisIsAVeryLongVariableName. This convention is also known as camel-case.
- It is important to choose a name that is self-descriptive and closely reflects the meaning of the variable, e.g., numberOfStudents or numStudents.
- Do not use meaningless names like a, b, c, d, i, j, k, i1, j99.
- Avoid single-alphabet names, which is easier to type but often meaningless, unless they are common names like x, y, z for coordinates, i for index.
- It is perfectly okay to use long names of says 30 characters to make sure that the name accurately reflects its meaning!
- Use singular and plural nouns prudently to differentiate between singular and plural variables. For example, you may use the variable row to refer to a single row number and the variable rows to refer to many rows [such as an array of rows - to be discussed later].
Variable Declaration
To use a variable in your program, you need to first "introduce" it by declaring its name and type, in one of the following syntaxes:
| // Declare a variable of a specified type type identifier; // Declare multiple variables of the same type, separated by commas type identifier-1, identifier-2, ..., identifier-n; // Declare a variable and assign an initial value type identifier = value; // Declare multiple variables with initial values type identifier-1 = value-1, ..., identifier-n = value-n; | int option; double sum, difference, product, quotient; int magicNumber = 88; double sum = 0.0, product = 1.0; |
Example,
int mark1; mark1 = 76; int mark2; mark2 = mark1 + 10; double average; average = [mark1 + mark2] / 2.0; int mark1; mark2 = "Hello";Take note that:
- In C, you need to declare the name of a variable before it can be used.
- C is a "strongly-type" language. A variable takes on a type. Once the type of a variable is declared, it can only store a value belonging to this particular type. For example, an int variable can hold only integer such as 123, and NOT floating-point number such as -2.17 or text string such as "Hello". The concept of type was introduced into the early programming languages to simplify interpretation of data made up of 0s and 1s. Knowing the type of a piece of data greatly simplifies its interpretation and processing.
- Each variable can only be declared once.
- In C, you can declare a variable anywhere inside the program, as long as it is declared before used. [In C prior to C99, all the variables must be declared at the beginning of functions.] It is recommended that your declare a variable just before it is first used.
- The type of a variable cannot be changed inside the program.
When a variable is declared, it contains garbage until you assign an initial value. It is important to take note that C does not issue any warning/error if you use a variable before initialize it - which certainly leads to some unexpected results. For example,
| 1 2 3 4 5 6 7 8 | #include int main[] { int number; printf["%d\n", number]; return 0; } |
Constants [const]
Constants are non-modifiable variables, declared with keyword const. Their values cannot be changed during program execution. Also, const must be initialized during declaration. For examples:
const double PI = 3.1415926;Constant Naming Convention: Use uppercase words, joined with underscore. For example, MIN_VALUE, MAX_SIZE.
Expressions
An expression is a combination of operators [such as addition '+', subtraction '-', multiplication '*', division '/'] and operands [variables or literal values], that can be evaluated to yield a single value of a certain type. For example,
1 + 2 * 3 int sum, number; sum + number double principal, interestRate; principal * [1 + interestRate]Assignment [=]
An assignment statement:
- assigns a literal value [of the RHS] to a variable [of the LHS]; or
- evaluates an expression [of the RHS] and assign the resultant value to a variable [of the LHS].
The RHS shall be a value; and the LHS shall be a variable [or memory address].
The syntax for assignment statement is:
| // Assign the literal value [of the RHS] to the variable [of the LHS] variable = literal-value; // Evaluate the expression [RHS] and assign the result to the variable [LHS] variable = expression; | number = 88; sum = sum + number; |
The assignment statement should be interpreted this way: The expression on the right-hand-side [RHS] is first evaluated to produce a resultant value [called rvalue or right-value]. The rvalue is then assigned to the variable on the left-hand-side [LHS] [or lvalue, which is a location that can hold a rvalue]. Take note that you have to first evaluate the RHS, before assigning the resultant value to the LHS. For examples,
number = 8; number = number + 1;The symbol "=" is known as the assignment operator. The meaning of "=" in programming is different from Mathematics. It denotes assignment instead of equality. The RHS is a literal value; or an expression that evaluates to a value; while the LHS must be a variable. Note that x = x + 1 is valid [and often used] in programming. It evaluates x + 1 and assign the resultant value to the variable x. x = x + 1 illegal in Mathematics. While x + y = 1 is allowed in Mathematics, it is invalid in programming [because the LHS of an assignment statement must be a variable]. Some programming languages use symbol ":=", "←", "->", or "→" as the assignment operator to avoid confusion with equality.
Fundamental Types
Integers: C supports these integer types: char, short, int, long, long long [in C11] in a non-decreasing order of size. The actual size depends on the implementation. The integers [except char] are signed number [which can hold zero, positive and negative numbers]. You could use the keyword unsigned [char|short|int|long|long long] to declare an unsigned integers [which can hold zero and positive numbers]. There are a total 10 types of integers - signed|unsigned combined with char|short|int|long|long long.
Characters: Characters [e.g., 'a', 'Z', '0', '9'] are encoded in ASCII into integers, and kept in type char. For example, character '0' is 48 [decimal] or 30H [hexadecimal]; character 'A' is 65 [decimal] or 41H [hexadecimal]; character 'a' is 97 [decimal] or 61H [hexadecimal]. Take note that the type char can be interpreted as character in ASCII code, or an 8-bit integer. Unlike int or long, which is signed, char could be signed or unsigned, depending on the implementation. You can use signed char or unsigned char to explicitly declare signed or unsigned char.
Floating-point Numbers: There are 3 floating point types: float, double and long double, for single, double and long double precision floating point numbers. float and double are represented as specified by IEEE 754 standard. A float can represent a number between ±1.40239846×10^-45 and ±3.40282347×10^38, approximated. A double can represented a number between ±4.94065645841246544×10^-324 and ±1.79769313486231570×10^308, approximated. Take note that not all real numbers can be represented by float and double, because there are infinite real numbers. Most of the values are approximated.
The table below shows the typical size, minimum, maximum for the primitive types. Again, take note that the sizes are implementation dependent.
| Integers | int [or signed int] | Signed integer [of at least 16 bits] | 4 [2] | -2147483648 | 2147483647 |
| unsigned int | Unsigned integer [of at least 16 bits] | 4 [2] | 0 | 4294967295 | |
| char | Character [can be either signed or unsigned depends on implementation] | 1 | |||
| signed char | Character or signed tiny integer [guarantee to be signed] | 1 | -128 | 127 | |
| unsigned char | Character or unsigned tiny integer [guarantee to be unsigned] | 1 | 0 | 255 | |
| short [or short int] [or signed short] [or signed short int] | Short signed integer [of at least 16 bits] | 2 | -32768 | 32767 | |
| unsigned short [or unsigned shot int] | Unsigned short integer [of at least 16 bits] | 2 | 0 | 65535 | |
| long [or long int] [or signed long] [or signed long int] | Long signed integer [of at least 32 bits] | 4 [8] | -2147483648 | 2147483647 | |
| unsigned long [or unsigned long int] | Unsigned long integer [of at least 32 bits] | 4 [8] | 0 | same as above | |
| long long [or long long int] [or signed long long] [or signed long long int] | Very long signed integer [of at least 64 bits] | 8 | -263 | 263-1 | |
| unsigned long long [or unsigned long long int] | Unsigned very long integer [of at least 64 bits] | 8 | 0 | 264-1 | |
| Real Numbers | float | Floating-point number, ≈7 digits [IEEE 754 single-precision floating point format] | 4 | 3.4e38 | 3.4e-38 |
| double | Double precision floating-point number, ≈15 digits [IEEE 754 double-precision floating point format] | 8 | 1.7e308 | 1.7e-308 | |
| long double | Long double precision floating-point number, ≈19 digits [IEEE 754 quadruple-precision floating point format] | 12 [8] | |||
| Wide Characters | wchar_t | Wide [double-byte] character | 2 [4] |
In addition, many C library functions use a type called size_t, which is equivalent [typedef] to a unsigned int, meant for counting, size or length, with 0 and positive integers.
*The sizeof OperatorC provides an unary sizeof operator to get the size of the operand [in bytes]. The following program uses sizeof operator to print the size of the fundamental types.
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | #include int main[] { printf["sizeof[char] is %d bytes.\n", sizeof[char]]; printf["sizeof[short] is %d bytes.\n", sizeof[short]]; printf["sizeof[int] is %d bytes.\n", sizeof[int]]; printf["sizeof[long] is %d bytes.\n", sizeof[long]]; printf["sizeof[long long] is %d bytes.\n", sizeof[long long]]; printf["sizeof[float] is %d bytes.\n", sizeof[float]]; printf["sizeof[double] is %d bytes.\n", sizeof[double]]; printf["sizeof[long double] is %d bytes.\n", sizeof[long double]]; return 0; } |
The results may vary among different systems.
*HeaderThe limits.h header contains information about limits of integer type. For example,
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 | #include #include int main[] { printf["int max = %d\n", INT_MAX]; printf["int min = %d\n", INT_MIN]; printf["unsigned int max = %u\n", UINT_MAX]; printf["long max = %ld\n", LONG_MAX]; printf["long min = %ld\n", LONG_MIN]; printf["unsigned long max = %lu\n", ULONG_MAX]; printf["long long max = %lld\n", LLONG_MAX]; printf["long long min = %lld\n", LLONG_MIN]; printf["unsigned long long max = %llu\n", ULLONG_MAX]; printf["Bits in char = %d\n", CHAR_BIT]; printf["char max = %d\n", CHAR_MAX]; printf["char min = %d\n", CHAR_MIN]; printf["signed char max = %d\n", SCHAR_MAX]; printf["signed char min = %d\n", SCHAR_MIN]; printf["unsigned char max = %u\n", UCHAR_MAX]; return 0; } |
Again, the outputs depend on the system.
The minimum of unsigned integer is always 0. The other constants are SHRT_MAX, SHRT_MIN, USHRT_MAX, LONG_MIN, LONG_MAX, ULONG_MAX. Try inspecting this header [search for limits.h under your compiler].
*HeaderSimilarly, the float.h header contain information on limits for floating point numbers, such as minimum number of significant digits [FLT_DIG, DBL_DIG, LDBL_DIG for float, double and long double], number of bits for mantissa [FLT_MANT_DIG, DBL_MANT_DIG, LDBL_MANT_DIG], maximum and minimum exponent values, etc. Try inspecting this header [search for cfloat under your compiler].
Choosing TypesAs a programmer, you need to choose variables and decide on the type of the variables to be used in your programs. Most of the times, the decision is intuitive. For example, use an integer type for counting and whole number; a floating-point type for number with fractional part, char for a single character, and boolean for binary outcome.
Rule of Thumb- Use int for integer and double for floating point numbers. Use byte, short, long and float only if you have a good reason to choose that specific precision.
- Use int [or unsigned int] for counting and indexing, NOT floating-point type [float or double]. This is because integer type are precise and more efficient in operations.
- Use an integer type if possible. Use a floating-point type only if the number contains a fractional part.
Read my article on "Data Representation" if you wish to understand how the numbers and characters are represented inside the computer memory. In brief, It is important to take note that char '1' is different from int 1, short 1, float 1.0, double 1.0, and String "1". They are represented differently in the computer memory, with different precision and interpretation. For example, short 1 is "00000000 00000001", int 1 is "00000000 00000000 00000000 00000001", long long 1 is "00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000001", float 1.0 is "0 01111111 0000000 00000000 00000000", double 1.0 is "0 01111111111 0000 00000000 00000000 00000000 00000000 00000000 00000000", char '1' is "00110001".
There is a subtle difference between int 0 and double 0.0.
Furthermore, you MUST know the type of a value before you can interpret a value. For example, this value "00000000 00000000 00000000 00000001" cannot be interpreted unless you know the type.
*The typedef StatementTyping "unsigned int" many time can get annoying. The typedef statement can be used to create a new name for an existing type. For example, you can create a new type called "uint" for "unsigned int" as follow. You should place the typedef immediately after #include. Use typedef with care because it makes the program hard to read and understand.
Many C compilers define a type called size_t, which is a typedef of unsigned int.
typedef unsigned int size_t;Output via printf[] Function
C programs use function printf[] of library stdio to print output to the console. You need to issue a so-called preprocessor directive "#include " to use printf[].
To print a string literal such as "Hello, world", simply place it inside the parentheses, as follow:
printf[aStringLiteral];For example,
The \n represents the newline character. Printing a newline advances the cursor [denoted by _ in the above example] to the beginning of next line. printf[], by default, places the cursor after the printed string, and does not advance the cursor to the next line. For example,
printf["Hello"]; printf[", "]; printf["world!"]; printf["\n"]; printf["Hello\nworld\nagain\n"];Hello, world! Hello world again _Formatted Output via printf[]The "f" in printf[] stands for "formatted" printing. To do formatted printing, you need to use the following syntax:
printf[formattingString, variable1, variable2, ...]The formattingString is a string composing of normal texts and conversion specifiers. Normal texts will be printed as they are. A conversion specifier begins with a percent sign [%], followed by a code to specify the type of variable and format of the output [such as the field width and number of decimal places]. For example, %d denotes an int; %3d for an int with field-width of 3. The conversion specifiers are used as placeholders, which will be substituted by the variables given after the formatting string in a sequential manner. For example,
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 | #include int main[] { int number1 = 12345, number2 = 678; printf["Hello, number1 is %d.\n", number1]; printf["number1=%d, number2=%d.\n", number1, number2]; printf["number1=%8d, number2=%5d.\n", number1, number2]; printf["number1=%08d, number2=%05d.\n", number1, number2]; printf["number1=%-8d, number2=%-5d.\n", number1, number2]; return 0; } |
The commonly-used type conversion codes are:
| Integers | %d [or %i] | [signed] int |
| %u | unsigned int | |
| %o | int in octal | |
| %x, %X | int in hexadecimal [%X uses uppercase A-F] | |
| %hd, %hu | short, unsigned short | |
| %ld, %lu | long, unsigned long | |
| %lld, %llu | long long, unsigned long long | |
| Floating-point | %f | float in fixed notation |
| %e, %E | float in scientific notation | |
| %g, %G | float in fixed/scientific notation depending on its value | |
| %f, %lf [printf], %lf [scanf] | double: Use %f or %lf in printf[], but %lf in scanf[]. | |
| %Lf, %Le, %LE, %Lg, %LG | long double | |
| Character | %c | char |
| String | %s | string |
Notes:
- For double, you must use %lf [for long float] in scanf[] [or %le, %lE, %lg, %lG], but you can use either %f or %lf in printf[] [or %e, %E, %g, %G, %le, %lE, %lg, %lG].
- Use %% to print a % in the formatting string.
For example,
int anInt = 12345; float aFloat = 55.6677; double aDouble = 11.2233; char aChar = 'a'; char aStr[] = "Hello"; printf["The int is %d.\n", anInt]; printf["The float is %f.\n", aFloat]; printf["The double is %lf.\n", aDouble]; printf["The char is %c.\n", aChar]; printf["The string is %s.\n", aStr]; printf["The int [in hex] is %x.\n", anInt]; printf["The double [in scientific] is %le.\n", aDouble]; printf["The float [in scientific] is %E.\n", aFloat];Using the wrong type conversion code usually produces garbage.
Field WidthYou can optionally specify a field-width before the type conversion code, e.g., %3d, %6f, %20s. If the value to be formatted is shorter than the field width, it will be padded with spaces [by default]. Otherwise, the field-width will be ignored. For example,
int number = 123456; printf["number=%d.\n", number]; printf["number=%8d.\n", number]; printf["number=%3d.\n", number]; Precision [Decimal Places] for Floating-point NumbersFor floating-point numbers, you can optionally specify the number of decimal places to be printed, e.g., %6.2f, %8.3f. For example,
double value = 123.14159265; printf["value=%lf;\n", value]; printf["value=%6.2lf;\n", value]; printf["value=%9.4lf;\n", value]; printf["value=%3.2lf;\n", value]; AlignmentThe output are right-aligned by default. You could include a "-" flag [before the field width] to ask for left-aligned. For example,
int i1 = 12345, i2 = 678; printf["Hello, first int is %d, second int is %5d.\n", i1, i2]; printf["Hello, first int is %d, second int is %-5d.\n", i1, i2]; char msg[] = "Hello"; printf["xx%20sxx\n", msg]; printf["xx%-20sxx\n", msg]; Others- + [plus sign]: display plus or minus sign preceding the number.
- # or 0: Pad with leading # or 0.
C11 introduces more secure version of printf[]/scanf[] called printf_s[]/scanf_s[] to deal with mismatched conversion specifiers. Microsoft Visual C implemented its own versions of printf_s[]/scanf_s[] before C11, and issues a deprecated warning for using printf[]/scanf[].
Input via scanf[] Function
In C, you can use scanf[] function of to read inputs from keyboard. scanf[] uses the type-conversion code like printf[]. For example,
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 | /* * TestScanf.c */ #include int main[] { int anInt; float aFloat; double aDouble; printf["Enter an int: "]; // Prompting message scanf["%d", &anInt]; // Read an int from keyboard and assign to variable anInt. printf["The value entered is %d.\n", anInt]; printf["Enter a floating-point number: "]; // Prompting message scanf["%f", &aFloat]; // Read a double from keyboard and assign to variable aFloat. printf["The value entered is %f.\n", aFloat]; printf["Enter a floating-point number: "]; // Prompting message scanf["%lf", &aDouble]; // Read a double from keyboard and assign to variable aDouble. printf["The value entered is %lf.\n", aDouble]; return 0; } |
Notes:
- To place the input into a variable in scanf[], you need to prefix the variable name by an ampersand sign [&]. The ampersand [&] is called address-of operator, which will be explained later. However, it is important to stress that missing ampersand [&] is a common error.
- For double, you must use type conversion code %lf for scanf[]. You could use %f or %lf for printf[].
The scanf[] returns an int indicating the number of values read.
For example,
int number1 = 55, number2 = 66; int rcode = scanf["%d", &number]; printf["return code is %d\n", rcode]; printf["number1 is %d\n", number1]; printf["number2 is %d\n", number2];The scanf[] returns 1 if user enters an integer which is read into the variable number. It returns 0 if user enters a non-integer [such as "hello"], and variable number is not assigned.
int number1 = 55, number2 = 66; int rcode = scanf["%d%d", &number1, &number2]; printf["return code is %d\n", rcode]; printf["number1 is %d\n", number1]; printf["number2 is %d\n", number2];The scanf[] returns 2 if user enters two integers that are read into number1 and number2. It returns 1 if user enters an integer followed by a non-integer, and number2 will not be affected. It returns 0 if user enters a non-integer, and both number1 and number2 will not be affected.
Checking the return code of scanf[] is recommended for secure coding.
Literals for Fundamental Types and String
A literal is a specific constant value, such as 123, -456, 3.14, 'a', "Hello", that can be assigned directly to a variable; or used as part of an expression. They are called literals because they literally and explicitly identify their values.
Integer LiteralsA whole number, such as 123 and -456, is treated as an int, by default. For example,
int number = -123; int sum = 4567; int bigSum = 8234567890;An int literal may precede with a plus [+] or minus [-] sign, followed by digits. No commas or special symbols [e.g., $ or space] is allowed [e.g., 1,234 and $123 are invalid]. No preceding 0 is allowed too [e.g., 007 is invalid].
Besides the default base 10 integers, you can use a prefix '0' [zero] to denote a value in octal, prefix '0x' for a value in hexadecimal, and prefix '0b' for binary value [in some compilers], e.g.,
int number1 = 1234; int number2 = 01234; int number3 = 0x1abc; int number4 = 0b10001001;A long literal is identified by a suffix 'L' or 'l' [avoid lowercase, which can be confused with the number one]. A long long int is identified by a suffix 'LL'. You can also use suffix 'U' for unsigned int, 'UL' for unsigned long, and 'ULL' for unsigned long long int. For example,
long number = 12345678L; long sum = 123; long long bigNumber = 987654321LL;No suffix is needed for short literals. But you can only use integer values in the permitted range. For example,
short smallNumber = 1234567890; short midSizeNumber = -12345;Floating-point LiteralsA number with a decimal point, such as 55.66 and -33.44, is treated as a double, by default. You can also express them in scientific notation, e.g., 1.2e3, -5.5E-6, where e or E denotes the exponent in power of 10. You could precede the fractional part or exponent with a plus [+] or minus [-] sign. Exponent shall be an integer. There should be no space or other characters [e.g., space] in the number.
You MUST use a suffix of 'f' or 'F' for float literals, e.g., -1.2345F. For example,
float average = 55.66; float average = 55.66f;Use suffix 'L' [or 'l'] for long double.
Character Literals and Escape SequencesA printable char literal is written by enclosing the character with a pair of single quotes, e.g., 'z', '$', and '9'. In C, characters are represented using 8-bit ASCII code, and can be treated as a 8-bit signed integers in arithmetic operations. In other words, char and 8-bit signed integer are interchangeable. You can also assign an integer in the range of [-128, 127] to a char variable; and [0, 255] to an unsigned char.
You can find the ASCII code table HERE.
For example,
char letter = 'a'; char anotherLetter = 98; printf["%c\n", letter]; printf["%c\n", anotherLetter]; anotherLetter += 2; printf["%c\n", anotherLetter]; printf["%d\n", anotherLetter];Non-printable and control characters can be represented by so-called escape sequences, which begins with a back-slash [\] followed by a code. The commonly-used escape sequences are:
| \n | New-line [or Line-feed] | 0AH [10D] |
| \r | Carriage-return | 0DH [13D] |
| \t | Tab | 09H [9D] |
| \" | Double-quote [needed to include " in double-quoted string] | 22H [34D] |
| \' | Single-quote | 27H [39D] |
| \\ | Back-slash [to resolve ambiguity] | 5CH [92D] |
Notes:
- New-line [0AH] and carriage return [0dH], represented by \n, and \r respectively, are used as line delimiter [or end-of-line, or EOL]. However, take note that UNIX/Linux/Mac use \n as EOL, Windows use \r\n.
- Horizontal Tab [09H] is represented as \t.
- To resolve ambiguity, characters back-slash [\], single-quote ['] and double-quote ["] are represented using escape sequences \\, \' and \", respectively. This is because a single back-slash begins an escape sequence, while single-quotes and double-quotes are used to enclose character and string.
- Other less commonly-used escape sequences are: \? or ?, \a for alert or bell, \b for backspace, \f for form-feed, \v for vertical tab. These may not be supported in some consoles.
The ctype.h header provides functions such as isalpha[], isdigit[], isspace[], ispunct[], isalnum[], isupper[], islower[] to determine the type of character; and toupper[], tolower[] for case conversion.
String LiteralsA String literal is composed of zero of more characters surrounded by a pair of double quotes, e.g., "Hello, world!", "The sum is ", "".
String literals may contains escape sequences. Inside a String, you need to use \" for double-quote to distinguish it from the ending double-quote, e.g. "\"quoted\"". Single quote inside a String does not require escape sequence. For example,
printf["Use \\\" to place\n a \" within\ta\tstring\n"];Use \" to place a " within a stringTRY: Write a program to print the following picture. Take note that you need to use escape sequences to print special characters.
'__' [oo] +========\/ / || %%% || * ||-----|| "" ""Example [Literals]| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 | #include int main[] { char gender = 'm'; unsigned short numChildren = 8; short yearOfBirth = 1945; unsigned int salary = 88000; double weight = 88.88; float gpa = 3.88f; printf["Gender is %c.\n", gender]; printf["Number of children is %u.\n", numChildren]; printf["Year of birth is %d.\n", yearOfBirth]; printf["Salary is %u.\n", salary]; printf["Weight is %.2lf.\n", weight]; printf["GPA is %.2f.\n", gpa]; return 0; } |
Operations
Arithmetic Operators
C supports the following arithmetic operators for numbers: short, int, long, long long, char [treated as 8-bit signed integer], unsigned short, unsigned int, unsigned long, unsigned long long, unsigned char, float, double and long double.
| * | Multiplication | expr1 * expr2 | 2 * 3 → 6; 3.3 * 1.0 → 3.3 |
| / | Division | expr1 / expr2 | 1 / 2 → 0; 1.0 / 2.0 → 0.5 |
| % | Remainder [Modulus] | expr1 % expr2 | 5 % 2 → 1; -5 % 2 → -1 |
| + | Addition | expr1 + expr2 | 1 + 2 → 3; 1.1 + 2.2 → 3.3 |
| - | Subtraction | expr1 - expr2 | 1 - 2 → -1; 1.1 - 2.2 → -1.1 |
All the above operators are binary operators, i.e., they take two operands. The multiplication, division and remainder take precedence over addition and subtraction. Within the same precedence level [e.g., addition and subtraction], the expression is evaluated from left to right. For example, 1+2+3-4 is evaluated as [[1+2]+3]-4.
It is important to take note that int/int produces an int, with the result truncated, e.g., 1/2 → 0 [instead of 0.5].
Take note that C does not have an exponent [power] operator ['^' is exclusive-or, not exponent].
Arithmetic Expressions
In programming, the following arithmetic expression:
must be written as [1+2*a]/3 + [4*[b+c]*[5-d-e]]/f - 6*[7/g+h]. You cannot omit the multiplication symbol '*' [as in Mathematics].
Like Mathematics, the multiplication '*' and division '/' take precedence over addition '+' and subtraction '-'. Parentheses [] have higher precedence. The operators '+', '-', '*', and '/' are left-associative. That is, 1 + 2 + 3 + 4 is treated as [[[1+2] + 3] + 4].
Mixed-Type Operations
If both the operands of an arithmetic operation belong to the same type, the operation is carried out in that type, and the result belongs to that type. For example, int/int → int; double/double → double.
However, if the two operands belong to different types, the compiler promotes the value of the smaller type to the larger type [known as implicit type-casting]. The operation is then carried out in the larger type. For example, int/double → double/double → double. Hence, 1/2 → 0, 1.0/2.0 → 0.5, 1.0/2 → 0.5, 1/2.0 → 0.5.
For example,
| int | 2 + 3 | int 2 + int 3 → int 5 |
| double | 2.2 + 3.3 | double 2.2 + double 3.3 → double 5.5 |
| mix | 2 + 3.3 | int 2 + double 3.3 → double 2.0 + double 3.3 → double 5.3 |
| int | 1 / 2 | int 1 / int 2 → int 0 |
| double | 1.0 / 2.0 | double 1.0 / double 2.0 → double 0.5 |
| mix | 1 / 2.0 | int 1 / double 2.0 → double 1.0 / double 2.0 → double 0.5 |
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | #include int main[] { int i1 = 2, i2 = 4; double d1 = 2.5, d2 = 5.2; printf["%d + %d = %d\n", i1, i2, i1+i2]; printf["%.1lf + %.1lf = %.1lf\n", d1, d2, d1+d2]; printf["%d + %.1lf = %.1lf\n", i1, d2, i1+d2]; printf["%d / %d = %d\n", i1, i2, i1/i2]; printf["%.1lf / %.1lf = %.2lf\n", d1, d2, d1/d2]; printf["%d / %.1lf = %.2lf\n", i1, d2, i1/d2]; return 0; } |
Overflow/UnderFlow
Study the output of the following program:
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | #include int main[] { int i1 = 2147483647; printf["%d\n", i1 + 1]; printf["%d\n", i1 + 2]; printf["%d\n", i1 * i1]; int i2 = -2147483648; printf["%d\n", i2 - 1]; printf["%d\n", i2 - 2]; printf["%d\n", i2 * i2]; return 0; } |
In arithmetic operations, the resultant value wraps around if it exceeds its range [i.e., overflow or underflow]. C runtime does not issue an error/warning message but produces incorrect result.
It is important to take note that checking of overflow/underflow is the programmer's responsibility, i.e., your job!
This feature is an legacy design, where processors were slow. Checking for overflow/underflow consumes computation power and reduces performance.
To check for arithmetic overflow [known as secure coding] is tedious. Google for "INT32-C. Ensure that operations on signed integers do not result in overflow" @ www.securecoding.cert.org.
Compound Assignment Operators
Besides the usual simple assignment operator '=' described earlier, C also provides the so-called compound assignment operators as listed:
| = | var = expr | Assign the value of the LHS to the variable at the RHS | x = 5; |
| += | var += expr | same as var = var + expr | x += 5; same as x = x + 5 |
| -= | var -= expr | same as var = var - expr | x -= 5; same as x = x - 5 |
| *= | var *= expr | same as var = var * expr | x *= 5; same as x = x * 5 |
| /= | var /= expr | same as var = var / expr | x /= 5; same as x = x / 5 |
| %= | var %= expr | same as var = var % expr | x %= 5; same as x = x % 5 |
Increment/Decrement Operators
C supports these unary arithmetic operators: increment '++' and decrement '--'.
| ++ | x++; ++x | Increment by 1, same as x += 1 |
| -- | x--; --x | Decrement by 1, same as x -= 1 |
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 | #include int main[] { int mark = 76; printf["%d\n", mark]; mark++; printf["%d\n", mark]; ++mark; printf["%d\n", mark]; mark = mark + 1; printf["%d\n", mark]; mark--; printf["%d\n", mark]; --mark; printf["%d\n", mark]; mark = mark - 1; printf["%d\n", mark]; return 0; } |
The increment/decrement unary operator can be placed before the operand [prefix operator], or after the operands [postfix operator]. They take on different meaning in operations.
| ++var | Pre-Increment Increment var, then use the new value of var | y = ++x; | same as x=x+1; y=x; |
| var++ | Post-Increment Use the old value of var, then increment var | y = x++; | same as oldX=x; x=x+1; y=oldX; |
| --var | Pre-Decrement | y = --x; | same as x=x-1; y=x; |
| var-- | Post-Decrement | y = x--; | same as oldX=x; x=x-1; y=oldX; |
If '++' or '--' involves another operation, then pre- or post-order is important to specify the order of the two operations. For examples,
x = 5; printf["%d\n", x++]; x = 5; printf["%d\n", ++x];Prefix operator [e.g, ++i] could be more efficient than postfix operator [e.g., i++] in some situations.
Implicit Type-Conversion vs. Explicit Type-Casting
Converting a value from one type to another type is called type casting [or type conversion]. There are two kinds of type casting:
- Implicit type-conversion performed by the compiler automatically, and
- Explicit type-casting via an unary type-casting operator in the form of [new-type]operand.
When you assign a value of a fundamental [built-in] type to a variable of another fundamental type, C automatically converts the value to the receiving type, if the two types are compatible. For examples,
- If you assign an int value to a double variable, the compiler automatically casts the int value to a double double [e.g., from 1 to 1.0] and assigns it to the double variable.
- if you assign a double value of to an int variable, the compiler automatically casts the double value to an int value [e.g., from 1.2 to 1] and assigns it to the int variable. The fractional part would be truncated and lost. Some compilers issue a warning/error "possible loss in precision"; others do not.
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 | #include int main[] { int i; double d; i = 3; d = i; printf["d = %lf\n", d]; d = 5.5; i = d; printf["i = %d\n", i]; i = 6.6; printf["i = %d\n", i]; } |
C will not perform automatic type conversion, if the two types are not compatible.
Explicit Type-CastingYou can explicitly perform type-casting via the so-called unary type-casting operator in the form of [new-type]operand. The type-casting operator takes one operand in the particular type, and returns an equivalent value in the new type. Take note that it is an operation that yields a resultant value, similar to an addition operation although addition involves two operands. For example,
printf["%lf\n", [double]5]; printf["%d\n", [int]5.5]; double aDouble = 5.6; int anInt = [int]aDouble;Example: Suppose that you want to find the average [in double] of the integers between 1 and 100. Study the following codes:
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | #include int main[] { int sum = 0; double average; int number = 1; while [number "]; print[a, size]; printf["\n"]; } } } } void print[const int a[], int size] { int i; printf["{"]; for [i = 0; i < size; ++i] { printf["%d", a[i]]; if [i < size - 1] printf[","]; } printf["} "]; } |
Program Notes:
- [TODO]
Wiki "Insertion Sort" for the algorithm and illustration. In brief, pass thru the list. For each element, compare with all previous elements and insert it at the correct position by shifting the other elements. For example,
{8,4,5,3,2,9,4,1} {8} {4,5,3,2,9,4,1} {4,8} {5,3,2,9,4,1} {4,5,8} {3,2,9,4,1} {3,4,5,8} {2,9,4,1} {2,3,4,5,8} {9,4,1} {2,3,4,5,8,9} {4,1} {2,3,4,4,5,8,9} {1} {1,2,3,4,4,5,8,9}Insertion sort is also not efficient, with complexity of O[n2].
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 | #include
void insertionSort[int a[], int size];
void print[const int a[], int iMin, int iMax];
int main[] {
const int SIZE = 8;
int a[] = {8, 4, 5, 3, 2, 9, 4, 1};
print[a, 0, SIZE - 1];
printf["\n"];
insertionSort[a, SIZE];
print[a, 0, SIZE - 1];
printf["\n"];
}
void insertionSort[int a[], int size] {
int temp;
int i, prev, shift;
for [i = 1; i < size; ++i] {
print[a, 0, i - 1];
print[a, i, size - 1];
printf["\n"];
for [prev = 0; prev < i; ++prev] {
if [a[i] < a[prev]] {
temp = a[i];
for [shift = i; shift > prev; --shift] {
a[shift] = a[shift-1];
}
a[prev] = temp;
break;
}
}
}
}
void print[const int a[], int iMin, int iMax] {
int i;
printf["{"];
for [i = iMin; i {} {1,4,5,3,2,9,4,8}
{1} {4,5,3,2,9,4,8} => {1} {2,5,3,4,9,4,8}
{1,2} {5,3,4,9,4,8} => {1,2} {3,5,4,9,4,8}
{1,2,3} {5,4,9,4,8} => {1,2,3} {4,5,9,4,8}
{1,2,3,4} {5,9,4,8} => {1,2,3,4} {4,9,5,8}
{1,2,3,4,4} {9,5,8} => {1,2,3,4,4} {5,9,8}
{1,2,3,4,4,5} {9,8} => {1,2,3,4,4,5} {8,9}
{1,2,3,4,4,5,8,9} Selection sort is also not efficient, with complexity of O[n2].
|