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. 4 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. 9 10 # Printing 11 12 The verbs: 13 14 General: 15 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 21 22 Boolean: 23 24 %t the word true or false 25 26 Integer: 27 28 %b base 2 29 %c the character represented by the corresponding Unicode code point 30 %d base 10 31 %o base 8 32 %O base 8 with 0o prefix 33 %q a single-quoted character literal safely escaped with Go syntax. 34 %x base 16, with lower-case letters for a-f 35 %X base 16, with upper-case letters for A-F 36 %U Unicode format: U+1234; same as "U+%04X" 37 38 Floating-point and complex constituents: 39 40 %b decimalless scientific notation with exponent a power of two, 41 in the manner of strconv.FormatFloat with the 'b' format, 42 e.g. -123456p-78 43 %e scientific notation, e.g. -1.234456e+78 44 %E scientific notation, e.g. -1.234456E+78 45 %f decimal point but no exponent, e.g. 123.456 46 %F synonym for %f 47 %g %e for large exponents, %f otherwise. Precision is discussed below. 48 %G %E for large exponents, %F otherwise 49 %x hexadecimal notation (with decimal power of two exponent), e.g. -0x1.23abcp+20 50 %X upper-case hexadecimal notation, e.g. -0X1.23ABCP+20 51 52 String and slice of bytes (treated equivalently with these verbs): 53 54 %s the uninterpreted bytes of the string or slice 55 %q a double-quoted string safely escaped with Go syntax 56 %x base 16, lower-case, two characters per byte 57 %X base 16, upper-case, two characters per byte 58 59 Slice: 60 61 %p address of 0th element in base 16 notation, with leading 0x 62 63 Pointer: 64 65 %p base 16 notation, with leading 0x 66 The %b, %d, %o, %x and %X verbs also work with pointers, 67 formatting the value exactly as if it were an integer. 68 69 The default format for %v is: 70 71 bool: %t 72 int, int8 etc.: %d 73 uint, uint8 etc.: %d, %#x if printed with %#v 74 float32, complex64, etc: %g 75 string: %s 76 chan: %p 77 pointer: %p 78 79 For compound objects, the elements are printed using these rules, recursively, 80 laid out like this: 81 82 struct: {field0 field1 ...} 83 array, slice: [elem0 elem1 ...] 84 maps: map[key1:value1 key2:value2 ...] 85 pointer to above: &{}, &[], &map[] 86 87 Width is specified by an optional decimal number immediately preceding the verb. 88 If absent, the width is whatever is necessary to represent the value. 89 Precision is specified after the (optional) width by a period followed by a 90 decimal number. If no period is present, a default precision is used. 91 A period with no following number specifies a precision of zero. 92 Examples: 93 94 %f default width, default precision 95 %9f width 9, default precision 96 %.2f default width, precision 2 97 %9.2f width 9, precision 2 98 %9.f width 9, precision 0 99 100 Width and precision are measured in units of Unicode code points, 101 that is, runes. (This differs from C's printf where the 102 units are always measured in bytes.) Either or both of the flags 103 may be replaced with the character '*', causing their values to be 104 obtained from the next operand (preceding the one to format), 105 which must be of type int. 106 107 For most values, width is the minimum number of runes to output, 108 padding the formatted form with spaces if necessary. 109 110 For strings, byte slices and byte arrays, however, precision 111 limits the length of the input to be formatted (not the size of 112 the output), truncating if necessary. Normally it is measured in 113 runes, but for these types when formatted with the %x or %X format 114 it is measured in bytes. 115 116 For floating-point values, width sets the minimum width of the field and 117 precision sets the number of places after the decimal, if appropriate, 118 except that for %g/%G precision sets the maximum number of significant 119 digits (trailing zeros are removed). For example, given 12.345 the format 120 %6.3f prints 12.345 while %.3g prints 12.3. The default precision for %e, %f 121 and %#g is 6; for %g it is the smallest number of digits necessary to identify 122 the value uniquely. 123 124 For complex numbers, the width and precision apply to the two 125 components independently and the result is parenthesized, so %f applied 126 to 1.2+3.4i produces (1.200000+3.400000i). 127 128 When formatting a single integer code point or a rune string (type []rune) 129 with %q, invalid Unicode code points are changed to the Unicode replacement 130 character, U+FFFD, as in strconv.QuoteRune. 131 132 Other flags: 133 134 '+' always print a sign for numeric values; 135 guarantee ASCII-only output for %q (%+q) 136 '-' pad with spaces on the right rather than the left (left-justify the field) 137 '#' alternate format: add leading 0b for binary (%#b), 0 for octal (%#o), 138 0x or 0X for hex (%#x or %#X); suppress 0x for %p (%#p); 139 for %q, print a raw (backquoted) string if strconv.CanBackquote 140 returns true; 141 always print a decimal point for %e, %E, %f, %F, %g and %G; 142 do not remove trailing zeros for %g and %G; 143 write e.g. U+0078 'x' if the character is printable for %U (%#U). 144 ' ' (space) leave a space for elided sign in numbers (% d); 145 put spaces between bytes printing strings or slices in hex (% x, % X) 146 '0' pad with leading zeros rather than spaces; 147 for numbers, this moves the padding after the sign; 148 ignored for strings, byte slices and byte arrays 149 150 Flags are ignored by verbs that do not expect them. 151 For example there is no alternate decimal format, so %#d and %d 152 behave identically. 153 154 For each Printf-like function, there is also a Print function 155 that takes no format and is equivalent to saying %v for every 156 operand. Another variant Println inserts blanks between 157 operands and appends a newline. 158 159 Regardless of the verb, if an operand is an interface value, 160 the internal concrete value is used, not the interface itself. 161 Thus: 162 163 var i interface{} = 23 164 fmt.Printf("%v\n", i) 165 166 will print 23. 167 168 Except when printed using the verbs %T and %p, special 169 formatting considerations apply for operands that implement 170 certain interfaces. In order of application: 171 172 1. If the operand is a reflect.Value, the operand is replaced by the 173 concrete value that it holds, and printing continues with the next rule. 174 175 2. If an operand implements the Formatter interface, it will 176 be invoked. In this case the interpretation of verbs and flags is 177 controlled by that implementation. 178 179 3. If the %v verb is used with the # flag (%#v) and the operand 180 implements the GoStringer interface, that will be invoked. 181 182 If the format (which is implicitly %v for Println etc.) is valid 183 for a string (%s %q %v %x %X), the following two rules apply: 184 185 4. If an operand implements the error interface, the Error method 186 will be invoked to convert the object to a string, which will then 187 be formatted as required by the verb (if any). 188 189 5. If an operand implements method String() string, that method 190 will be invoked to convert the object to a string, which will then 191 be formatted as required by the verb (if any). 192 193 For compound operands such as slices and structs, the format 194 applies to the elements of each operand, recursively, not to the 195 operand as a whole. Thus %q will quote each element of a slice 196 of strings, and %6.2f will control formatting for each element 197 of a floating-point array. 198 199 However, when printing a byte slice with a string-like verb 200 (%s %q %x %X), it is treated identically to a string, as a single item. 201 202 To avoid recursion in cases such as 203 204 type X string 205 func (x X) String() string { return Sprintf("<%s>", x) } 206 207 convert the value before recurring: 208 209 func (x X) String() string { return Sprintf("<%s>", string(x)) } 210 211 Infinite recursion can also be triggered by self-referential data 212 structures, such as a slice that contains itself as an element, if 213 that type has a String method. Such pathologies are rare, however, 214 and the package does not protect against them. 215 216 When printing a struct, fmt cannot and therefore does not invoke 217 formatting methods such as Error or String on unexported fields. 218 219 # Explicit argument indexes 220 221 In Printf, Sprintf, and Fprintf, the default behavior is for each 222 formatting verb to format successive arguments passed in the call. 223 However, the notation [n] immediately before the verb indicates that the 224 nth one-indexed argument is to be formatted instead. The same notation 225 before a '*' for a width or precision selects the argument index holding 226 the value. After processing a bracketed expression [n], subsequent verbs 227 will use arguments n+1, n+2, etc. unless otherwise directed. 228 229 For example, 230 231 fmt.Sprintf("%[2]d %[1]d\n", 11, 22) 232 233 will yield "22 11", while 234 235 fmt.Sprintf("%[3]*.[2]*[1]f", 12.0, 2, 6) 236 237 equivalent to 238 239 fmt.Sprintf("%6.2f", 12.0) 240 241 will yield " 12.00". Because an explicit index affects subsequent verbs, 242 this notation can be used to print the same values multiple times 243 by resetting the index for the first argument to be repeated: 244 245 fmt.Sprintf("%d %d %#[1]x %#x", 16, 17) 246 247 will yield "16 17 0x10 0x11". 248 249 # Format errors 250 251 If an invalid argument is given for a verb, such as providing 252 a string to %d, the generated string will contain a 253 description of the problem, as in these examples: 254 255 Wrong type or unknown verb: %!verb(type=value) 256 Printf("%d", "hi"): %!d(string=hi) 257 Too many arguments: %!(EXTRA type=value) 258 Printf("hi", "guys"): hi%!(EXTRA string=guys) 259 Too few arguments: %!verb(MISSING) 260 Printf("hi%d"): hi%!d(MISSING) 261 Non-int for width or precision: %!(BADWIDTH) or %!(BADPREC) 262 Printf("%*s", 4.5, "hi"): %!(BADWIDTH)hi 263 Printf("%.*s", 4.5, "hi"): %!(BADPREC)hi 264 Invalid or invalid use of argument index: %!(BADINDEX) 265 Printf("%*[2]d", 7): %!d(BADINDEX) 266 Printf("%.[2]d", 7): %!d(BADINDEX) 267 268 All errors begin with the string "%!" followed sometimes 269 by a single character (the verb) and end with a parenthesized 270 description. 271 272 If an Error or String method triggers a panic when called by a 273 print routine, the fmt package reformats the error message 274 from the panic, decorating it with an indication that it came 275 through the fmt package. For example, if a String method 276 calls panic("bad"), the resulting formatted message will look 277 like 278 279 %!s(PANIC=bad) 280 281 The %!s just shows the print verb in use when the failure 282 occurred. If the panic is caused by a nil receiver to an Error 283 or String method, however, the output is the undecorated 284 string, "<nil>". 285 286 # Scanning 287 288 An analogous set of functions scans formatted text to yield 289 values. Scan, Scanf and Scanln read from os.Stdin; Fscan, 290 Fscanf and Fscanln read from a specified io.Reader; Sscan, 291 Sscanf and Sscanln read from an argument string. 292 293 Scan, Fscan, Sscan treat newlines in the input as spaces. 294 295 Scanln, Fscanln and Sscanln stop scanning at a newline and 296 require that the items be followed by a newline or EOF. 297 298 Scanf, Fscanf, and Sscanf parse the arguments according to a 299 format string, analogous to that of Printf. In the text that 300 follows, 'space' means any Unicode whitespace character 301 except newline. 302 303 In the format string, a verb introduced by the % character 304 consumes and parses input; these verbs are described in more 305 detail below. A character other than %, space, or newline in 306 the format consumes exactly that input character, which must 307 be present. A newline with zero or more spaces before it in 308 the format string consumes zero or more spaces in the input 309 followed by a single newline or the end of the input. A space 310 following a newline in the format string consumes zero or more 311 spaces in the input. Otherwise, any run of one or more spaces 312 in the format string consumes as many spaces as possible in 313 the input. Unless the run of spaces in the format string 314 appears adjacent to a newline, the run must consume at least 315 one space from the input or find the end of the input. 316 317 The handling of spaces and newlines differs from that of C's 318 scanf family: in C, newlines are treated as any other space, 319 and it is never an error when a run of spaces in the format 320 string finds no spaces to consume in the input. 321 322 The verbs behave analogously to those of Printf. 323 For example, %x will scan an integer as a hexadecimal number, 324 and %v will scan the default representation format for the value. 325 The Printf verbs %p and %T and the flags # and + are not implemented. 326 For floating-point and complex values, all valid formatting verbs 327 (%b %e %E %f %F %g %G %x %X and %v) are equivalent and accept 328 both decimal and hexadecimal notation (for example: "2.3e+7", "0x4.5p-8") 329 and digit-separating underscores (for example: "3.14159_26535_89793"). 330 331 Input processed by verbs is implicitly space-delimited: the 332 implementation of every verb except %c starts by discarding 333 leading spaces from the remaining input, and the %s verb 334 (and %v reading into a string) stops consuming input at the first 335 space or newline character. 336 337 The familiar base-setting prefixes 0b (binary), 0o and 0 (octal), 338 and 0x (hexadecimal) are accepted when scanning integers 339 without a format or with the %v verb, as are digit-separating 340 underscores. 341 342 Width is interpreted in the input text but there is no 343 syntax for scanning with a precision (no %5.2f, just %5f). 344 If width is provided, it applies after leading spaces are 345 trimmed and specifies the maximum number of runes to read 346 to satisfy the verb. For example, 347 348 Sscanf(" 1234567 ", "%5s%d", &s, &i) 349 350 will set s to "12345" and i to 67 while 351 352 Sscanf(" 12 34 567 ", "%5s%d", &s, &i) 353 354 will set s to "12" and i to 34. 355 356 In all the scanning functions, a carriage return followed 357 immediately by a newline is treated as a plain newline 358 (\r\n means the same as \n). 359 360 In all the scanning functions, if an operand implements method 361 Scan (that is, it implements the Scanner interface) that 362 method will be used to scan the text for that operand. Also, 363 if the number of arguments scanned is less than the number of 364 arguments provided, an error is returned. 365 366 All arguments to be scanned must be either pointers to basic 367 types or implementations of the Scanner interface. 368 369 Like Scanf and Fscanf, Sscanf need not consume its entire input. 370 There is no way to recover how much of the input string Sscanf used. 371 372 Note: Fscan etc. can read one character (rune) past the input 373 they return, which means that a loop calling a scan routine 374 may skip some of the input. This is usually a problem only 375 when there is no space between input values. If the reader 376 provided to Fscan implements ReadRune, that method will be used 377 to read characters. If the reader also implements UnreadRune, 378 that method will be used to save the character and successive 379 calls will not lose data. To attach ReadRune and UnreadRune 380 methods to a reader without that capability, use 381 bufio.NewReader. 382 */ 383 package fmt 384