1 of 60

9

C Formatted Input/Output

1

2 of 60

2

All the news that’s fit to print.

Adolph S. Ochs

What mad pursuit? What struggle to escape?

John Keats

Remove not the landmark on the boundary�of the fields.

Amenemope

3 of 60

OBJECTIVES

In this chapter you will learn:

To use input and output streams.
To use all print formatting capabilities.
To use all input formatting capabilities.
To print with field widths and precisions.
To use formatting flags in the printf format control string.
To output literals and escape sequences.
To format input using scanf.

3

4 of 60

9.1 Introduction

9.2 Streams

9.3 Formatting Output with printf

9.4 Printing Integers

9.5 Printing Floating-Point Numbers

9.6 Printing Strings and Characters

9.7 Other Conversion Specifiers

9.8 Printing with Field Widths and Precision

9.9 Using Flags in the printf Format Control String

9.10 Printing Literals and Escape Sequences

9.11 Reading Formatted Input with scanf

4

5 of 60

9.1 Introduction

In this chapter

Presentation of results
scanf and printf
Streams (input and output)

gets, puts, getchar, putchar (in <stdio.h>)

5

6 of 60

9.2 Streams

Streams

Sequences of characters organized into lines

Each line consists of zero or more characters and ends with newline character
ANSI C must support lines of at least 254 characters

Performs all input and output
Can often be redirected

Standard input – keyboard
Standard output – screen
Standard error – screen
More in Chapter 11

6

7 of 60

9.3 Formatting Output with printf

printf

Precise output formatting

Conversion specifications: flags, field widths, precisions, etc.

Can perform rounding, aligning columns, right/left justification, inserting literal characters, exponential format, hexadecimal format, and fixed width and precision

Format

printf( format-control-string, other-arguments );
Format control string: describes output format
Other-arguments: correspond to each conversion specification in format-control-string

Each specification begins with a percent sign(%), ends with conversion specifier

7

8 of 60

Common Programming Error 9.1

Forgetting to enclose a format-control-string in quotation marks is a syntax error.

8

9 of 60

Good Programming Practice 9.1

Format outputs neatly for presentation to make program outputs more readable and reduce user errors.

9

10 of 60

9.4 Printing Integers

Integer

Whole number (no decimal point): 25, 0, -9
Positive, negative, or zero
Only minus sign prints by default (later we will change this)

10

11 of 60

11

Fig. 9.1 | Integer conversion specifiers.

12 of 60

12

Outline

fig09_02.c

(1 of 2 )

d and i specify signed integers

h specifies a short number

l specifies a long number

o specifies an octal integer

u specifies an unsigned integer

13 of 60

13

Outline

fig09_02.c

(2 of 2 )

x and X specify hexadecimal integers

14 of 60

Common Programming Error 9.2

Printing a negative value with a conversion specifier that expects an unsigned value.

14

15 of 60

9.5 Printing Floating-Point Numbers

Floating Point Numbers

Have a decimal point (33.5)
Exponential notation (computer's version of scientific notation)

150.3 is 1.503 x 10² in scientific
150.3 is 1.503E+02 in exponential

15

16 of 60

16

Fig. 9.3 | Floating-point conversion specifiers.

17 of 60

Error-Prevention Tip 9.1

When outputting data, be sure that the user is aware of situations in which data may be imprecise due to formatting (e.g., rounding errors from specifying precisions).

17

18 of 60

18

Outline

fig09_04.c

e and E specify exponential notation

f specifies fixed-point notation

g and G specify either exponential or fixed-point notation depending on the number’s size

19 of 60

9.6 Printing Strings and Characters

c

Prints char argument
Cannot be used to print the first character of a string

s

Requires a pointer to char as an argument
Prints characters until NULL ('\0') encountered
Cannot print a char argument

Remember

Single quotes for character constants ('z')
Double quotes for strings "z" (which actually contains two characters, 'z' and '\0')

19

20 of 60

Common Programming Error 9.3

Using %c to print a string is an error. The conversion specifier %c expects a char argument. A string is a pointer to char (i.e., a char *).

20

21 of 60

Common Programming Error 9.4

Using %s to print a char argument, on�some systems, causes a fatal execution-time error called an access violation. The conversion specifier %s expects an argument of type pointer to char.

21

22 of 60

Common Programming Error 9.5

Using single quotes around character strings is a syntax error. Character strings must be enclosed in double quotes.

22

23 of 60

Common Programming Error 9.6

Using double quotes around a character constant creates a pointer to a string consisting of two characters, the second of which is the terminating null. A character constant is a single character enclosed in single quotes.

23

24 of 60

24

Outline

fig09_05.c

c specifies a character will be printed

s specifies a string will be printed

25 of 60

9.7 Other Conversion Specifiers

p

Displays pointer value (address)

n

Stores number of characters already output by current printf statement
Takes a pointer to an integer as an argument
Nothing printed by a %n specification
Every printf call returns a value

Number of characters output
Negative number if error occurs

%

Prints a percent sign
%%

25

26 of 60

Portability Tip 9.1

The conversion specifier p displays an address in an implementation-defined manner (on many systems, hexadecimal notation is used rather than decimal notation).

26

27 of 60

Common Programming Error 9.7

Trying to print a literal percent character using % rather than %% in the format control string. When % appears in a format control string, it must be followed by a conversion specifier.

27

28 of 60

28

Fig. 9.6 | Other conversion specifiers.

29 of 60

29

Outline

fig09_07.c

(1 of 2 )

p specifies a memory address will be printed

n stores the number of characters printed on a line

% prints a percent sign

30 of 60

30

Outline

fig09_07.c

(2 of 2 )

31 of 60

9.8 Printing with Field Widths and Precision

Field width

Size of field in which data is printed
If width larger than data, default right justified

If field width too small, increases to fit data
Minus sign uses one character position in field

Integer width inserted between % and conversion specifier
%4d – field width of 4

31

32 of 60

32

Outline

fig09_08.c

(1 of 2 )

A field width of 4 will make C attempt to print the number in a 4-character space

Note that C considers the minus sign a character

33 of 60

33

Outline

fig09_08.c

(2 of 2 )

34 of 60

Common Programming Error 9.8

Not providing a sufficiently large field�width to handle a value to be printed can offset other data being printed and can produce confusing outputs. Know your data!

34

35 of 60

9.8 Printing with Field Widths and Precision

Precision

Meaning varies depending on data type
Integers (default 1)

Minimum number of digits to print

If data too small, prefixed with zeros

Floating point

Number of digits to appear after decimal (e and f)

For g – maximum number of significant digits

Strings

Maximum number of characters to be written from string

Format

Use a dot (.) then precision number after %

%.3f

35

36 of 60

36

Outline

fig09_09.c

(1 of 2 )

Precision for integers specifies the minimum number of characters to be printed

Precision for f and e specifiers controls the number of digits after the decimal point

Precision for the g specifier controls the maximum number of significant digits printed

37 of 60

37

Outline

fig09_09.c

(2 of 2 )

Precision for strings specifies the maximum number of characters to be printed

38 of 60

9.9 Using Flags in the printf Format Control String

Flags

Supplement formatting capabilities
Place flag immediately to the right of percent sign
Several flags may be combined

38

39 of 60

39

Fig. 9.10 | Format control string flags.

40 of 60

40

Outline

fig09_11.c

- flag left justifies characters in a field

41 of 60

41

Outline

fig09_12.c

+ flag forces a plus sign on positive numbers

42 of 60

42

Outline

fig09_13.c

Space flag forces a space on positive numbers

43 of 60

43

Outline

fig09_14.c

# flag prefixes a 0 before octal integers

# flag prefixes a 0x before hexadecimal integers

# flag forces a decimal point on floating-point numbers with no fractional part

44 of 60

44

Outline

fig09_15.c

0 flag fills empty spaces with zeros

45 of 60

9.10 Printing Literals and Escape Sequences

Printing Literals

Most characters can be printed
Certain "problem" characters, such as the quotation mark "
Must be represented by escape sequences

Represented by a backslash \ followed by an escape character

45

46 of 60

Common Programming Error 9.9

Attempting to print as literal data in a printf statement a single quote, double quote or backslash character without preceding that character with a backslash to form a proper escape sequence is an error.

46

47 of 60

47

Fig. 9.16 | Escape sequences.

48 of 60

9.11 Formatting Input with scanf

scanf

Input can be formatted much like output can
scanf conversion specifiers are slightly different from those used with printf

48

49 of 60

Good Programming Practice 9.2

When inputting data, prompt the user for one data item or a few data items at a time. Avoid asking the user to enter many data items in response to a single prompt.

49

50 of 60

Good Programming Practice 9.3

Always consider what the user and your program will do when (not if) incorrect data is entered—for example, a value for an integer that is nonsensical in a program’s context, or a string with missing punctuation or spaces.

50

51 of 60

51

Fig. 9.17 | Conversion specifiers for scanf. (Part 1 of 3.)

52 of 60

52

Fig. 9.17 | Conversion specifiers for scanf. (Part 2 of 3.)

53 of 60

53

Fig. 9.17 | Conversion specifiers for scanf. (Part 3 of 3.)

54 of 60

54

Outline

fig09_18.c

d specifies a decimal integer will be input

i specifies an integer will be input

o specifies an octal integer will be input

u specifies an unsigned decimal integer will be input

x specifies a hexadecimal integer will be input

55 of 60

55

Outline

fig09_19.c

e, f, and g specify a floating-point number will be input

l specifies a double or long double will be input

56 of 60

56

Outline

fig09_20.c

c specifies a character will be input

s specifies a string will be input

57 of 60

57

Outline

fig09_21.c

[] specifies only the initial segment of a string that contains the characters in brackets will be read

58 of 60

58

Outline

fig09_22.c

[] and ^ specify only the initial segment of a string that does not contain the characters in brackets will be read

59 of 60

59

Outline

fig09_23.c

A field width of 2 tells C to only read the first 2 characters of that input

60 of 60

60

Outline

fig09_24.c

* is a wildcard—scanf will disregard anything between the two inputs on either side of it