Source file src/fmt/doc.go

Documentation: fmt

     1  // Copyright 2009 The Go Authors. All rights reserved.
     2  // Use of this source code is governed by a BSD-style
     3  // license that can be found in the LICENSE file.
     5  /*
     6  	Package fmt implements formatted I/O with functions analogous
     7  	to C's printf and scanf.  The format 'verbs' are derived from C's but
     8  	are simpler.
    11  	Printing
    13  	The verbs:
    15  	General:
    16  		%v	the value in a default format
    17  			when printing structs, the plus flag (%+v) adds field names
    18  		%#v	a Go-syntax representation of the value
    19  		%T	a Go-syntax representation of the type of the value
    20  		%%	a literal percent sign; consumes no value
    22  	Boolean:
    23  		%t	the word true or false
    24  	Integer:
    25  		%b	base 2
    26  		%c	the character represented by the corresponding Unicode code point
    27  		%d	base 10
    28  		%o	base 8
    29  		%q	a single-quoted character literal safely escaped with Go syntax.
    30  		%x	base 16, with lower-case letters for a-f
    31  		%X	base 16, with upper-case letters for A-F
    32  		%U	Unicode format: U+1234; same as "U+%04X"
    33  	Floating-point and complex constituents:
    34  		%b	decimalless scientific notation with exponent a power of two,
    35  			in the manner of strconv.FormatFloat with the 'b' format,
    36  			e.g. -123456p-78
    37  		%e	scientific notation, e.g. -1.234456e+78
    38  		%E	scientific notation, e.g. -1.234456E+78
    39  		%f	decimal point but no exponent, e.g. 123.456
    40  		%F	synonym for %f
    41  		%g	%e for large exponents, %f otherwise. Precision is discussed below.
    42  		%G	%E for large exponents, %F otherwise
    43  	String and slice of bytes (treated equivalently with these verbs):
    44  		%s	the uninterpreted bytes of the string or slice
    45  		%q	a double-quoted string safely escaped with Go syntax
    46  		%x	base 16, lower-case, two characters per byte
    47  		%X	base 16, upper-case, two characters per byte
    48  	Slice:
    49  		%p	address of 0th element in base 16 notation, with leading 0x
    50  	Pointer:
    51  		%p	base 16 notation, with leading 0x
    52  		The %b, %d, %o, %x and %X verbs also work with pointers,
    53  		formatting the value exactly as if it were an integer.
    55  	The default format for %v is:
    56  		bool:                    %t
    57  		int, int8 etc.:          %d
    58  		uint, uint8 etc.:        %d, %#x if printed with %#v
    59  		float32, complex64, etc: %g
    60  		string:                  %s
    61  		chan:                    %p
    62  		pointer:                 %p
    63  	For compound objects, the elements are printed using these rules, recursively,
    64  	laid out like this:
    65  		struct:             {field0 field1 ...}
    66  		array, slice:       [elem0 elem1 ...]
    67  		maps:               map[key1:value1 key2:value2 ...]
    68  		pointer to above:   &{}, &[], &map[]
    70  	Width is specified by an optional decimal number immediately preceding the verb.
    71  	If absent, the width is whatever is necessary to represent the value.
    72  	Precision is specified after the (optional) width by a period followed by a
    73  	decimal number. If no period is present, a default precision is used.
    74  	A period with no following number specifies a precision of zero.
    75  	Examples:
    76  		%f     default width, default precision
    77  		%9f    width 9, default precision
    78  		%.2f   default width, precision 2
    79  		%9.2f  width 9, precision 2
    80  		%9.f   width 9, precision 0
    82  	Width and precision are measured in units of Unicode code points,
    83  	that is, runes. (This differs from C's printf where the
    84  	units are always measured in bytes.) Either or both of the flags
    85  	may be replaced with the character '*', causing their values to be
    86  	obtained from the next operand (preceding the one to format),
    87  	which must be of type int.
    89  	For most values, width is the minimum number of runes to output,
    90  	padding the formatted form with spaces if necessary.
    92  	For strings, byte slices and byte arrays, however, precision
    93  	limits the length of the input to be formatted (not the size of
    94  	the output), truncating if necessary. Normally it is measured in
    95  	runes, but for these types when formatted with the %x or %X format
    96  	it is measured in bytes.
    98  	For floating-point values, width sets the minimum width of the field and
    99  	precision sets the number of places after the decimal, if appropriate,
   100  	except that for %g/%G precision sets the maximum number of significant
   101  	digits (trailing zeros are removed). For example, given 12.345 the format
   102  	%6.3f prints 12.345 while %.3g prints 12.3. The default precision for %e, %f
   103  	and %#g is 6; for %g it is the smallest number of digits necessary to identify
   104  	the value uniquely.
   106  	For complex numbers, the width and precision apply to the two
   107  	components independently and the result is parenthesized, so %f applied
   108  	to 1.2+3.4i produces (1.200000+3.400000i).
   110  	Other flags:
   111  		+	always print a sign for numeric values;
   112  			guarantee ASCII-only output for %q (%+q)
   113  		-	pad with spaces on the right rather than the left (left-justify the field)
   114  		#	alternate format: add leading 0 for octal (%#o), 0x for hex (%#x);
   115  			0X for hex (%#X); suppress 0x for %p (%#p);
   116  			for %q, print a raw (backquoted) string if strconv.CanBackquote
   117  			returns true;
   118  			always print a decimal point for %e, %E, %f, %F, %g and %G;
   119  			do not remove trailing zeros for %g and %G;
   120  			write e.g. U+0078 'x' if the character is printable for %U (%#U).
   121  		' '	(space) leave a space for elided sign in numbers (% d);
   122  			put spaces between bytes printing strings or slices in hex (% x, % X)
   123  		0	pad with leading zeros rather than spaces;
   124  			for numbers, this moves the padding after the sign
   126  	Flags are ignored by verbs that do not expect them.
   127  	For example there is no alternate decimal format, so %#d and %d
   128  	behave identically.
   130  	For each Printf-like function, there is also a Print function
   131  	that takes no format and is equivalent to saying %v for every
   132  	operand.  Another variant Println inserts blanks between
   133  	operands and appends a newline.
   135  	Regardless of the verb, if an operand is an interface value,
   136  	the internal concrete value is used, not the interface itself.
   137  	Thus:
   138  		var i interface{} = 23
   139  		fmt.Printf("%v\n", i)
   140  	will print 23.
   142  	Except when printed using the verbs %T and %p, special
   143  	formatting considerations apply for operands that implement
   144  	certain interfaces. In order of application:
   146  	1. If the operand is a reflect.Value, the operand is replaced by the
   147  	concrete value that it holds, and printing continues with the next rule.
   149  	2. If an operand implements the Formatter interface, it will
   150  	be invoked. Formatter provides fine control of formatting.
   152  	3. If the %v verb is used with the # flag (%#v) and the operand
   153  	implements the GoStringer interface, that will be invoked.
   155  	If the format (which is implicitly %v for Println etc.) is valid
   156  	for a string (%s %q %v %x %X), the following two rules apply:
   158  	4. If an operand implements the error interface, the Error method
   159  	will be invoked to convert the object to a string, which will then
   160  	be formatted as required by the verb (if any).
   162  	5. If an operand implements method String() string, that method
   163  	will be invoked to convert the object to a string, which will then
   164  	be formatted as required by the verb (if any).
   166  	For compound operands such as slices and structs, the format
   167  	applies to the elements of each operand, recursively, not to the
   168  	operand as a whole. Thus %q will quote each element of a slice
   169  	of strings, and %6.2f will control formatting for each element
   170  	of a floating-point array.
   172  	However, when printing a byte slice with a string-like verb
   173  	(%s %q %x %X), it is treated identically to a string, as a single item.
   175  	To avoid recursion in cases such as
   176  		type X string
   177  		func (x X) String() string { return Sprintf("<%s>", x) }
   178  	convert the value before recurring:
   179  		func (x X) String() string { return Sprintf("<%s>", string(x)) }
   180  	Infinite recursion can also be triggered by self-referential data
   181  	structures, such as a slice that contains itself as an element, if
   182  	that type has a String method. Such pathologies are rare, however,
   183  	and the package does not protect against them.
   185  	When printing a struct, fmt cannot and therefore does not invoke
   186  	formatting methods such as Error or String on unexported fields.
   188  	Explicit argument indexes:
   190  	In Printf, Sprintf, and Fprintf, the default behavior is for each
   191  	formatting verb to format successive arguments passed in the call.
   192  	However, the notation [n] immediately before the verb indicates that the
   193  	nth one-indexed argument is to be formatted instead. The same notation
   194  	before a '*' for a width or precision selects the argument index holding
   195  	the value. After processing a bracketed expression [n], subsequent verbs
   196  	will use arguments n+1, n+2, etc. unless otherwise directed.
   198  	For example,
   199  		fmt.Sprintf("%[2]d %[1]d\n", 11, 22)
   200  	will yield "22 11", while
   201  		fmt.Sprintf("%[3]*.[2]*[1]f", 12.0, 2, 6)
   202  	equivalent to
   203  		fmt.Sprintf("%6.2f", 12.0)
   204  	will yield " 12.00". Because an explicit index affects subsequent verbs,
   205  	this notation can be used to print the same values multiple times
   206  	by resetting the index for the first argument to be repeated:
   207  		fmt.Sprintf("%d %d %#[1]x %#x", 16, 17)
   208  	will yield "16 17 0x10 0x11".
   210  	Format errors:
   212  	If an invalid argument is given for a verb, such as providing
   213  	a string to %d, the generated string will contain a
   214  	description of the problem, as in these examples:
   216  		Wrong type or unknown verb: %!verb(type=value)
   217  			Printf("%d", hi):          %!d(string=hi)
   218  		Too many arguments: %!(EXTRA type=value)
   219  			Printf("hi", "guys"):      hi%!(EXTRA string=guys)
   220  		Too few arguments: %!verb(MISSING)
   221  			Printf("hi%d"):            hi%!d(MISSING)
   222  		Non-int for width or precision: %!(BADWIDTH) or %!(BADPREC)
   223  			Printf("%*s", 4.5, "hi"):  %!(BADWIDTH)hi
   224  			Printf("%.*s", 4.5, "hi"): %!(BADPREC)hi
   225  		Invalid or invalid use of argument index: %!(BADINDEX)
   226  			Printf("%*[2]d", 7):       %!d(BADINDEX)
   227  			Printf("%.[2]d", 7):       %!d(BADINDEX)
   229  	All errors begin with the string "%!" followed sometimes
   230  	by a single character (the verb) and end with a parenthesized
   231  	description.
   233  	If an Error or String method triggers a panic when called by a
   234  	print routine, the fmt package reformats the error message
   235  	from the panic, decorating it with an indication that it came
   236  	through the fmt package.  For example, if a String method
   237  	calls panic("bad"), the resulting formatted message will look
   238  	like
   239  		%!s(PANIC=bad)
   241  	The %!s just shows the print verb in use when the failure
   242  	occurred. If the panic is caused by a nil receiver to an Error
   243  	or String method, however, the output is the undecorated
   244  	string, "<nil>".
   246  	Scanning
   248  	An analogous set of functions scans formatted text to yield
   249  	values.  Scan, Scanf and Scanln read from os.Stdin; Fscan,
   250  	Fscanf and Fscanln read from a specified io.Reader; Sscan,
   251  	Sscanf and Sscanln read from an argument string.
   253  	Scan, Fscan, Sscan treat newlines in the input as spaces.
   255  	Scanln, Fscanln and Sscanln stop scanning at a newline and
   256  	require that the items be followed by a newline or EOF.
   258  	Scanf, Fscanf, and Sscanf parse the arguments according to a
   259  	format string, analogous to that of Printf. In the text that
   260  	follows, 'space' means any Unicode whitespace character
   261  	except newline.
   263  	In the format string, a verb introduced by the % character
   264  	consumes and parses input; these verbs are described in more
   265  	detail below. A character other than %, space, or newline in
   266  	the format consumes exactly that input character, which must
   267  	be present. A newline with zero or more spaces before it in
   268  	the format string consumes zero or more spaces in the input
   269  	followed by a single newline or the end of the input. A space
   270  	following a newline in the format string consumes zero or more
   271  	spaces in the input. Otherwise, any run of one or more spaces
   272  	in the format string consumes as many spaces as possible in
   273  	the input. Unless the run of spaces in the format string
   274  	appears adjacent to a newline, the run must consume at least
   275  	one space from the input or find the end of the input.
   277  	The handling of spaces and newlines differs from that of C's
   278  	scanf family: in C, newlines are treated as any other space,
   279  	and it is never an error when a run of spaces in the format
   280  	string finds no spaces to consume in the input.
   282  	The verbs behave analogously to those of Printf.
   283  	For example, %x will scan an integer as a hexadecimal number,
   284  	and %v will scan the default representation format for the value.
   285  	The Printf verbs %p and %T and the flags # and + are not implemented.
   286  	The verbs %e %E %f %F %g and %G are all equivalent and scan any
   287  	floating-point or complex value. For float and complex literals in
   288  	scientific notation, both the decimal (e) and binary (p) exponent
   289  	formats are supported (for example: "2.3e+7" and "4.5p-8").
   291  	Input processed by verbs is implicitly space-delimited: the
   292  	implementation of every verb except %c starts by discarding
   293  	leading spaces from the remaining input, and the %s verb
   294  	(and %v reading into a string) stops consuming input at the first
   295  	space or newline character.
   297  	The familiar base-setting prefixes 0 (octal) and 0x
   298  	(hexadecimal) are accepted when scanning integers without
   299  	a format or with the %v verb.
   301  	Width is interpreted in the input text but there is no
   302  	syntax for scanning with a precision (no %5.2f, just %5f).
   303  	If width is provided, it applies after leading spaces are
   304  	trimmed and specifies the maximum number of runes to read
   305  	to satisfy the verb. For example,
   306  	   Sscanf(" 1234567 ", "%5s%d", &s, &i)
   307  	will set s to "12345" and i to 67 while
   308  	   Sscanf(" 12 34 567 ", "%5s%d", &s, &i)
   309  	will set s to "12" and i to 34.
   311  	In all the scanning functions, a carriage return followed
   312  	immediately by a newline is treated as a plain newline
   313  	(\r\n means the same as \n).
   315  	In all the scanning functions, if an operand implements method
   316  	Scan (that is, it implements the Scanner interface) that
   317  	method will be used to scan the text for that operand.  Also,
   318  	if the number of arguments scanned is less than the number of
   319  	arguments provided, an error is returned.
   321  	All arguments to be scanned must be either pointers to basic
   322  	types or implementations of the Scanner interface.
   324  	Like Scanf and Fscanf, Sscanf need not consume its entire input.
   325  	There is no way to recover how much of the input string Sscanf used.
   327  	Note: Fscan etc. can read one character (rune) past the input
   328  	they return, which means that a loop calling a scan routine
   329  	may skip some of the input.  This is usually a problem only
   330  	when there is no space between input values.  If the reader
   331  	provided to Fscan implements ReadRune, that method will be used
   332  	to read characters.  If the reader also implements UnreadRune,
   333  	that method will be used to save the character and successive
   334  	calls will not lose data.  To attach ReadRune and UnreadRune
   335  	methods to a reader without that capability, use
   336  	bufio.NewReader.
   337  */
   338  package fmt

View as plain text