Commit | Line | Data |
---|---|---|
a0d0e21e | 1 | =head1 NAME |
d74e8afc | 2 | X<operator> |
a0d0e21e LW |
3 | |
4 | perlop - Perl operators and precedence | |
5 | ||
d042e63d MS |
6 | =head1 DESCRIPTION |
7 | ||
d74e8afc ITB |
8 | =head2 Operator Precedence and Associativity |
9 | X<operator, precedence> X<precedence> X<associativity> | |
d042e63d MS |
10 | |
11 | Operator precedence and associativity work in Perl more or less like | |
12 | they do in mathematics. | |
13 | ||
14 | I<Operator precedence> means some operators are evaluated before | |
15 | others. For example, in C<2 + 4 * 5>, the multiplication has higher | |
16 | precedence so C<4 * 5> is evaluated first yielding C<2 + 20 == | |
17 | 22> and not C<6 * 5 == 30>. | |
18 | ||
19 | I<Operator associativity> defines what happens if a sequence of the | |
20 | same operators is used one after another: whether the evaluator will | |
21 | evaluate the left operations first or the right. For example, in C<8 | |
22 | - 4 - 2>, subtraction is left associative so Perl evaluates the | |
23 | expression left to right. C<8 - 4> is evaluated first making the | |
24 | expression C<4 - 2 == 2> and not C<8 - 2 == 6>. | |
a0d0e21e LW |
25 | |
26 | Perl operators have the following associativity and precedence, | |
19799a22 GS |
27 | listed from highest precedence to lowest. Operators borrowed from |
28 | C keep the same precedence relationship with each other, even where | |
29 | C's precedence is slightly screwy. (This makes learning Perl easier | |
30 | for C folks.) With very few exceptions, these all operate on scalar | |
31 | values only, not array values. | |
a0d0e21e LW |
32 | |
33 | left terms and list operators (leftward) | |
34 | left -> | |
35 | nonassoc ++ -- | |
36 | right ** | |
37 | right ! ~ \ and unary + and - | |
54310121 | 38 | left =~ !~ |
a0d0e21e LW |
39 | left * / % x |
40 | left + - . | |
41 | left << >> | |
42 | nonassoc named unary operators | |
43 | nonassoc < > <= >= lt gt le ge | |
0d863452 | 44 | nonassoc == != <=> eq ne cmp ~~ |
a0d0e21e LW |
45 | left & |
46 | left | ^ | |
47 | left && | |
c963b151 | 48 | left || // |
137443ea | 49 | nonassoc .. ... |
a0d0e21e LW |
50 | right ?: |
51 | right = += -= *= etc. | |
52 | left , => | |
53 | nonassoc list operators (rightward) | |
a5f75d66 | 54 | right not |
a0d0e21e | 55 | left and |
c963b151 | 56 | left or xor err |
a0d0e21e LW |
57 | |
58 | In the following sections, these operators are covered in precedence order. | |
59 | ||
5a964f20 TC |
60 | Many operators can be overloaded for objects. See L<overload>. |
61 | ||
a0d0e21e | 62 | =head2 Terms and List Operators (Leftward) |
d74e8afc | 63 | X<list operator> X<operator, list> X<term> |
a0d0e21e | 64 | |
62c18ce2 | 65 | A TERM has the highest precedence in Perl. They include variables, |
5f05dabc | 66 | quote and quote-like operators, any expression in parentheses, |
a0d0e21e LW |
67 | and any function whose arguments are parenthesized. Actually, there |
68 | aren't really functions in this sense, just list operators and unary | |
69 | operators behaving as functions because you put parentheses around | |
70 | the arguments. These are all documented in L<perlfunc>. | |
71 | ||
72 | If any list operator (print(), etc.) or any unary operator (chdir(), etc.) | |
73 | is followed by a left parenthesis as the next token, the operator and | |
74 | arguments within parentheses are taken to be of highest precedence, | |
75 | just like a normal function call. | |
76 | ||
77 | In the absence of parentheses, the precedence of list operators such as | |
78 | C<print>, C<sort>, or C<chmod> is either very high or very low depending on | |
54310121 | 79 | whether you are looking at the left side or the right side of the operator. |
a0d0e21e LW |
80 | For example, in |
81 | ||
82 | @ary = (1, 3, sort 4, 2); | |
83 | print @ary; # prints 1324 | |
84 | ||
19799a22 GS |
85 | the commas on the right of the sort are evaluated before the sort, |
86 | but the commas on the left are evaluated after. In other words, | |
87 | list operators tend to gobble up all arguments that follow, and | |
a0d0e21e | 88 | then act like a simple TERM with regard to the preceding expression. |
19799a22 | 89 | Be careful with parentheses: |
a0d0e21e LW |
90 | |
91 | # These evaluate exit before doing the print: | |
92 | print($foo, exit); # Obviously not what you want. | |
93 | print $foo, exit; # Nor is this. | |
94 | ||
95 | # These do the print before evaluating exit: | |
96 | (print $foo), exit; # This is what you want. | |
97 | print($foo), exit; # Or this. | |
98 | print ($foo), exit; # Or even this. | |
99 | ||
100 | Also note that | |
101 | ||
102 | print ($foo & 255) + 1, "\n"; | |
103 | ||
d042e63d MS |
104 | probably doesn't do what you expect at first glance. The parentheses |
105 | enclose the argument list for C<print> which is evaluated (printing | |
106 | the result of C<$foo & 255>). Then one is added to the return value | |
107 | of C<print> (usually 1). The result is something like this: | |
108 | ||
109 | 1 + 1, "\n"; # Obviously not what you meant. | |
110 | ||
111 | To do what you meant properly, you must write: | |
112 | ||
113 | print(($foo & 255) + 1, "\n"); | |
114 | ||
115 | See L<Named Unary Operators> for more discussion of this. | |
a0d0e21e LW |
116 | |
117 | Also parsed as terms are the C<do {}> and C<eval {}> constructs, as | |
54310121 | 118 | well as subroutine and method calls, and the anonymous |
a0d0e21e LW |
119 | constructors C<[]> and C<{}>. |
120 | ||
2ae324a7 | 121 | See also L<Quote and Quote-like Operators> toward the end of this section, |
c07a80fd | 122 | as well as L<"I/O Operators">. |
a0d0e21e LW |
123 | |
124 | =head2 The Arrow Operator | |
d74e8afc | 125 | X<arrow> X<dereference> X<< -> >> |
a0d0e21e | 126 | |
35f2feb0 | 127 | "C<< -> >>" is an infix dereference operator, just as it is in C |
19799a22 GS |
128 | and C++. If the right side is either a C<[...]>, C<{...}>, or a |
129 | C<(...)> subscript, then the left side must be either a hard or | |
130 | symbolic reference to an array, a hash, or a subroutine respectively. | |
131 | (Or technically speaking, a location capable of holding a hard | |
132 | reference, if it's an array or hash reference being used for | |
133 | assignment.) See L<perlreftut> and L<perlref>. | |
a0d0e21e | 134 | |
19799a22 GS |
135 | Otherwise, the right side is a method name or a simple scalar |
136 | variable containing either the method name or a subroutine reference, | |
137 | and the left side must be either an object (a blessed reference) | |
138 | or a class name (that is, a package name). See L<perlobj>. | |
a0d0e21e | 139 | |
5f05dabc | 140 | =head2 Auto-increment and Auto-decrement |
d74e8afc | 141 | X<increment> X<auto-increment> X<++> X<decrement> X<auto-decrement> X<--> |
a0d0e21e | 142 | |
d042e63d MS |
143 | "++" and "--" work as in C. That is, if placed before a variable, |
144 | they increment or decrement the variable by one before returning the | |
145 | value, and if placed after, increment or decrement after returning the | |
146 | value. | |
147 | ||
148 | $i = 0; $j = 0; | |
149 | print $i++; # prints 0 | |
150 | print ++$j; # prints 1 | |
a0d0e21e | 151 | |
b033823e A |
152 | Note that just as in C, Perl doesn't define B<when> the variable is |
153 | incremented or decremented. You just know it will be done sometime | |
154 | before or after the value is returned. This also means that modifying | |
155 | a variable twice in the same statement will lead to undefined behaviour. | |
156 | Avoid statements like: | |
157 | ||
158 | $i = $i ++; | |
159 | print ++ $i + $i ++; | |
160 | ||
161 | Perl will not guarantee what the result of the above statements is. | |
162 | ||
54310121 | 163 | The auto-increment operator has a little extra builtin magic to it. If |
a0d0e21e LW |
164 | you increment a variable that is numeric, or that has ever been used in |
165 | a numeric context, you get a normal increment. If, however, the | |
5f05dabc | 166 | variable has been used in only string contexts since it was set, and |
5a964f20 | 167 | has a value that is not the empty string and matches the pattern |
9c0670e1 | 168 | C</^[a-zA-Z]*[0-9]*\z/>, the increment is done as a string, preserving each |
a0d0e21e LW |
169 | character within its range, with carry: |
170 | ||
171 | print ++($foo = '99'); # prints '100' | |
172 | print ++($foo = 'a0'); # prints 'a1' | |
173 | print ++($foo = 'Az'); # prints 'Ba' | |
174 | print ++($foo = 'zz'); # prints 'aaa' | |
175 | ||
6a61d433 HS |
176 | C<undef> is always treated as numeric, and in particular is changed |
177 | to C<0> before incrementing (so that a post-increment of an undef value | |
178 | will return C<0> rather than C<undef>). | |
179 | ||
5f05dabc | 180 | The auto-decrement operator is not magical. |
a0d0e21e LW |
181 | |
182 | =head2 Exponentiation | |
d74e8afc | 183 | X<**> X<exponentiation> X<power> |
a0d0e21e | 184 | |
19799a22 | 185 | Binary "**" is the exponentiation operator. It binds even more |
cb1a09d0 AD |
186 | tightly than unary minus, so -2**4 is -(2**4), not (-2)**4. (This is |
187 | implemented using C's pow(3) function, which actually works on doubles | |
188 | internally.) | |
a0d0e21e LW |
189 | |
190 | =head2 Symbolic Unary Operators | |
d74e8afc | 191 | X<unary operator> X<operator, unary> |
a0d0e21e | 192 | |
5f05dabc | 193 | Unary "!" performs logical negation, i.e., "not". See also C<not> for a lower |
a0d0e21e | 194 | precedence version of this. |
d74e8afc | 195 | X<!> |
a0d0e21e LW |
196 | |
197 | Unary "-" performs arithmetic negation if the operand is numeric. If | |
198 | the operand is an identifier, a string consisting of a minus sign | |
199 | concatenated with the identifier is returned. Otherwise, if the string | |
200 | starts with a plus or minus, a string starting with the opposite sign | |
bff5667c | 201 | is returned. One effect of these rules is that -bareword is equivalent |
8705167b | 202 | to the string "-bareword". If, however, the string begins with a |
06705523 SP |
203 | non-alphabetic character (exluding "+" or "-"), Perl will attempt to convert |
204 | the string to a numeric and the arithmetic negation is performed. If the | |
205 | string cannot be cleanly converted to a numeric, Perl will give the warning | |
206 | B<Argument "the string" isn't numeric in negation (-) at ...>. | |
d74e8afc | 207 | X<-> X<negation, arithmetic> |
a0d0e21e | 208 | |
972b05a9 JH |
209 | Unary "~" performs bitwise negation, i.e., 1's complement. For |
210 | example, C<0666 & ~027> is 0640. (See also L<Integer Arithmetic> and | |
211 | L<Bitwise String Operators>.) Note that the width of the result is | |
212 | platform-dependent: ~0 is 32 bits wide on a 32-bit platform, but 64 | |
213 | bits wide on a 64-bit platform, so if you are expecting a certain bit | |
d042e63d | 214 | width, remember to use the & operator to mask off the excess bits. |
d74e8afc | 215 | X<~> X<negation, binary> |
a0d0e21e LW |
216 | |
217 | Unary "+" has no effect whatsoever, even on strings. It is useful | |
218 | syntactically for separating a function name from a parenthesized expression | |
219 | that would otherwise be interpreted as the complete list of function | |
5ba421f6 | 220 | arguments. (See examples above under L<Terms and List Operators (Leftward)>.) |
d74e8afc | 221 | X<+> |
a0d0e21e | 222 | |
19799a22 GS |
223 | Unary "\" creates a reference to whatever follows it. See L<perlreftut> |
224 | and L<perlref>. Do not confuse this behavior with the behavior of | |
225 | backslash within a string, although both forms do convey the notion | |
226 | of protecting the next thing from interpolation. | |
d74e8afc | 227 | X<\> X<reference> X<backslash> |
a0d0e21e LW |
228 | |
229 | =head2 Binding Operators | |
d74e8afc | 230 | X<binding> X<operator, binding> X<=~> X<!~> |
a0d0e21e | 231 | |
c07a80fd | 232 | Binary "=~" binds a scalar expression to a pattern match. Certain operations |
cb1a09d0 AD |
233 | search or modify the string $_ by default. This operator makes that kind |
234 | of operation work on some other string. The right argument is a search | |
2c268ad5 TP |
235 | pattern, substitution, or transliteration. The left argument is what is |
236 | supposed to be searched, substituted, or transliterated instead of the default | |
f8bab1e9 GS |
237 | $_. When used in scalar context, the return value generally indicates the |
238 | success of the operation. Behavior in list context depends on the particular | |
d7782e69 SP |
239 | operator. See L</"Regexp Quote-Like Operators"> for details and |
240 | L<perlretut> for examples using these operators. | |
f8bab1e9 GS |
241 | |
242 | If the right argument is an expression rather than a search pattern, | |
2c268ad5 | 243 | substitution, or transliteration, it is interpreted as a search pattern at run |
573e01ca | 244 | time. |
a0d0e21e LW |
245 | |
246 | Binary "!~" is just like "=~" except the return value is negated in | |
247 | the logical sense. | |
248 | ||
249 | =head2 Multiplicative Operators | |
d74e8afc | 250 | X<operator, multiplicative> |
a0d0e21e LW |
251 | |
252 | Binary "*" multiplies two numbers. | |
d74e8afc | 253 | X<*> |
a0d0e21e LW |
254 | |
255 | Binary "/" divides two numbers. | |
d74e8afc | 256 | X</> X<slash> |
a0d0e21e | 257 | |
54310121 | 258 | Binary "%" computes the modulus of two numbers. Given integer |
259 | operands C<$a> and C<$b>: If C<$b> is positive, then C<$a % $b> is | |
260 | C<$a> minus the largest multiple of C<$b> that is not greater than | |
261 | C<$a>. If C<$b> is negative, then C<$a % $b> is C<$a> minus the | |
262 | smallest multiple of C<$b> that is not less than C<$a> (i.e. the | |
6bb4e6d4 | 263 | result will be less than or equal to zero). |
0412d526 | 264 | Note that when C<use integer> is in scope, "%" gives you direct access |
55d729e4 GS |
265 | to the modulus operator as implemented by your C compiler. This |
266 | operator is not as well defined for negative operands, but it will | |
267 | execute faster. | |
d74e8afc | 268 | X<%> X<remainder> X<modulus> X<mod> |
55d729e4 | 269 | |
62d10b70 GS |
270 | Binary "x" is the repetition operator. In scalar context or if the left |
271 | operand is not enclosed in parentheses, it returns a string consisting | |
272 | of the left operand repeated the number of times specified by the right | |
273 | operand. In list context, if the left operand is enclosed in | |
3585017f YST |
274 | parentheses or is a list formed by C<qw/STRING/>, it repeats the list. |
275 | If the right operand is zero or negative, it returns an empty string | |
276 | or an empty list, depending on the context. | |
d74e8afc | 277 | X<x> |
a0d0e21e LW |
278 | |
279 | print '-' x 80; # print row of dashes | |
280 | ||
281 | print "\t" x ($tab/8), ' ' x ($tab%8); # tab over | |
282 | ||
283 | @ones = (1) x 80; # a list of 80 1's | |
284 | @ones = (5) x @ones; # set all elements to 5 | |
285 | ||
286 | ||
287 | =head2 Additive Operators | |
d74e8afc | 288 | X<operator, additive> |
a0d0e21e LW |
289 | |
290 | Binary "+" returns the sum of two numbers. | |
d74e8afc | 291 | X<+> |
a0d0e21e LW |
292 | |
293 | Binary "-" returns the difference of two numbers. | |
d74e8afc | 294 | X<-> |
a0d0e21e LW |
295 | |
296 | Binary "." concatenates two strings. | |
d74e8afc ITB |
297 | X<string, concatenation> X<concatenation> |
298 | X<cat> X<concat> X<concatenate> X<.> | |
a0d0e21e LW |
299 | |
300 | =head2 Shift Operators | |
d74e8afc ITB |
301 | X<shift operator> X<operator, shift> X<<< << >>> |
302 | X<<< >> >>> X<right shift> X<left shift> X<bitwise shift> | |
303 | X<shl> X<shr> X<shift, right> X<shift, left> | |
a0d0e21e | 304 | |
55497cff | 305 | Binary "<<" returns the value of its left argument shifted left by the |
306 | number of bits specified by the right argument. Arguments should be | |
982ce180 | 307 | integers. (See also L<Integer Arithmetic>.) |
a0d0e21e | 308 | |
55497cff | 309 | Binary ">>" returns the value of its left argument shifted right by |
310 | the number of bits specified by the right argument. Arguments should | |
982ce180 | 311 | be integers. (See also L<Integer Arithmetic>.) |
a0d0e21e | 312 | |
b16cf6df JH |
313 | Note that both "<<" and ">>" in Perl are implemented directly using |
314 | "<<" and ">>" in C. If C<use integer> (see L<Integer Arithmetic>) is | |
315 | in force then signed C integers are used, else unsigned C integers are | |
316 | used. Either way, the implementation isn't going to generate results | |
317 | larger than the size of the integer type Perl was built with (32 bits | |
318 | or 64 bits). | |
319 | ||
320 | The result of overflowing the range of the integers is undefined | |
321 | because it is undefined also in C. In other words, using 32-bit | |
322 | integers, C<< 1 << 32 >> is undefined. Shifting by a negative number | |
323 | of bits is also undefined. | |
324 | ||
a0d0e21e | 325 | =head2 Named Unary Operators |
d74e8afc | 326 | X<operator, named unary> |
a0d0e21e LW |
327 | |
328 | The various named unary operators are treated as functions with one | |
568e6d8b | 329 | argument, with optional parentheses. |
a0d0e21e LW |
330 | |
331 | If any list operator (print(), etc.) or any unary operator (chdir(), etc.) | |
332 | is followed by a left parenthesis as the next token, the operator and | |
333 | arguments within parentheses are taken to be of highest precedence, | |
3981b0eb JA |
334 | just like a normal function call. For example, |
335 | because named unary operators are higher precedence than ||: | |
a0d0e21e LW |
336 | |
337 | chdir $foo || die; # (chdir $foo) || die | |
338 | chdir($foo) || die; # (chdir $foo) || die | |
339 | chdir ($foo) || die; # (chdir $foo) || die | |
340 | chdir +($foo) || die; # (chdir $foo) || die | |
341 | ||
3981b0eb | 342 | but, because * is higher precedence than named operators: |
a0d0e21e LW |
343 | |
344 | chdir $foo * 20; # chdir ($foo * 20) | |
345 | chdir($foo) * 20; # (chdir $foo) * 20 | |
346 | chdir ($foo) * 20; # (chdir $foo) * 20 | |
347 | chdir +($foo) * 20; # chdir ($foo * 20) | |
348 | ||
349 | rand 10 * 20; # rand (10 * 20) | |
350 | rand(10) * 20; # (rand 10) * 20 | |
351 | rand (10) * 20; # (rand 10) * 20 | |
352 | rand +(10) * 20; # rand (10 * 20) | |
353 | ||
568e6d8b RGS |
354 | Regarding precedence, the filetest operators, like C<-f>, C<-M>, etc. are |
355 | treated like named unary operators, but they don't follow this functional | |
356 | parenthesis rule. That means, for example, that C<-f($file).".bak"> is | |
357 | equivalent to C<-f "$file.bak">. | |
d74e8afc | 358 | X<-X> X<filetest> X<operator, filetest> |
568e6d8b | 359 | |
5ba421f6 | 360 | See also L<"Terms and List Operators (Leftward)">. |
a0d0e21e LW |
361 | |
362 | =head2 Relational Operators | |
d74e8afc | 363 | X<relational operator> X<operator, relational> |
a0d0e21e | 364 | |
35f2feb0 | 365 | Binary "<" returns true if the left argument is numerically less than |
a0d0e21e | 366 | the right argument. |
d74e8afc | 367 | X<< < >> |
a0d0e21e | 368 | |
35f2feb0 | 369 | Binary ">" returns true if the left argument is numerically greater |
a0d0e21e | 370 | than the right argument. |
d74e8afc | 371 | X<< > >> |
a0d0e21e | 372 | |
35f2feb0 | 373 | Binary "<=" returns true if the left argument is numerically less than |
a0d0e21e | 374 | or equal to the right argument. |
d74e8afc | 375 | X<< <= >> |
a0d0e21e | 376 | |
35f2feb0 | 377 | Binary ">=" returns true if the left argument is numerically greater |
a0d0e21e | 378 | than or equal to the right argument. |
d74e8afc | 379 | X<< >= >> |
a0d0e21e LW |
380 | |
381 | Binary "lt" returns true if the left argument is stringwise less than | |
382 | the right argument. | |
d74e8afc | 383 | X<< lt >> |
a0d0e21e LW |
384 | |
385 | Binary "gt" returns true if the left argument is stringwise greater | |
386 | than the right argument. | |
d74e8afc | 387 | X<< gt >> |
a0d0e21e LW |
388 | |
389 | Binary "le" returns true if the left argument is stringwise less than | |
390 | or equal to the right argument. | |
d74e8afc | 391 | X<< le >> |
a0d0e21e LW |
392 | |
393 | Binary "ge" returns true if the left argument is stringwise greater | |
394 | than or equal to the right argument. | |
d74e8afc | 395 | X<< ge >> |
a0d0e21e LW |
396 | |
397 | =head2 Equality Operators | |
d74e8afc | 398 | X<equality> X<equal> X<equals> X<operator, equality> |
a0d0e21e LW |
399 | |
400 | Binary "==" returns true if the left argument is numerically equal to | |
401 | the right argument. | |
d74e8afc | 402 | X<==> |
a0d0e21e LW |
403 | |
404 | Binary "!=" returns true if the left argument is numerically not equal | |
405 | to the right argument. | |
d74e8afc | 406 | X<!=> |
a0d0e21e | 407 | |
35f2feb0 | 408 | Binary "<=>" returns -1, 0, or 1 depending on whether the left |
6ee5d4e7 | 409 | argument is numerically less than, equal to, or greater than the right |
d4ad863d | 410 | argument. If your platform supports NaNs (not-a-numbers) as numeric |
7d3a9d88 NC |
411 | values, using them with "<=>" returns undef. NaN is not "<", "==", ">", |
412 | "<=" or ">=" anything (even NaN), so those 5 return false. NaN != NaN | |
413 | returns true, as does NaN != anything else. If your platform doesn't | |
414 | support NaNs then NaN is just a string with numeric value 0. | |
d74e8afc | 415 | X<< <=> >> X<spaceship> |
7d3a9d88 | 416 | |
2b54f59f YST |
417 | perl -le '$a = "NaN"; print "No NaN support here" if $a == $a' |
418 | perl -le '$a = "NaN"; print "NaN support here" if $a != $a' | |
a0d0e21e LW |
419 | |
420 | Binary "eq" returns true if the left argument is stringwise equal to | |
421 | the right argument. | |
d74e8afc | 422 | X<eq> |
a0d0e21e LW |
423 | |
424 | Binary "ne" returns true if the left argument is stringwise not equal | |
425 | to the right argument. | |
d74e8afc | 426 | X<ne> |
a0d0e21e | 427 | |
d4ad863d JH |
428 | Binary "cmp" returns -1, 0, or 1 depending on whether the left |
429 | argument is stringwise less than, equal to, or greater than the right | |
430 | argument. | |
d74e8afc | 431 | X<cmp> |
a0d0e21e | 432 | |
0d863452 RH |
433 | Binary "~~" does a smart match between its arguments. Smart matching |
434 | is described in L<perlsyn/"Smart Matching in Detail">. | |
435 | This operator is only available if you enable the "~~" feature: | |
436 | see L<feature> for more information. | |
437 | X<~~> | |
438 | ||
a034a98d DD |
439 | "lt", "le", "ge", "gt" and "cmp" use the collation (sort) order specified |
440 | by the current locale if C<use locale> is in effect. See L<perllocale>. | |
441 | ||
a0d0e21e | 442 | =head2 Bitwise And |
d74e8afc | 443 | X<operator, bitwise, and> X<bitwise and> X<&> |
a0d0e21e | 444 | |
2cdc098b | 445 | Binary "&" returns its operands ANDed together bit by bit. |
2c268ad5 | 446 | (See also L<Integer Arithmetic> and L<Bitwise String Operators>.) |
a0d0e21e | 447 | |
2cdc098b MG |
448 | Note that "&" has lower priority than relational operators, so for example |
449 | the brackets are essential in a test like | |
450 | ||
451 | print "Even\n" if ($x & 1) == 0; | |
452 | ||
a0d0e21e | 453 | =head2 Bitwise Or and Exclusive Or |
d74e8afc ITB |
454 | X<operator, bitwise, or> X<bitwise or> X<|> X<operator, bitwise, xor> |
455 | X<bitwise xor> X<^> | |
a0d0e21e | 456 | |
2cdc098b | 457 | Binary "|" returns its operands ORed together bit by bit. |
2c268ad5 | 458 | (See also L<Integer Arithmetic> and L<Bitwise String Operators>.) |
a0d0e21e | 459 | |
2cdc098b | 460 | Binary "^" returns its operands XORed together bit by bit. |
2c268ad5 | 461 | (See also L<Integer Arithmetic> and L<Bitwise String Operators>.) |
a0d0e21e | 462 | |
2cdc098b MG |
463 | Note that "|" and "^" have lower priority than relational operators, so |
464 | for example the brackets are essential in a test like | |
465 | ||
466 | print "false\n" if (8 | 2) != 10; | |
467 | ||
a0d0e21e | 468 | =head2 C-style Logical And |
d74e8afc | 469 | X<&&> X<logical and> X<operator, logical, and> |
a0d0e21e LW |
470 | |
471 | Binary "&&" performs a short-circuit logical AND operation. That is, | |
472 | if the left operand is false, the right operand is not even evaluated. | |
473 | Scalar or list context propagates down to the right operand if it | |
474 | is evaluated. | |
475 | ||
476 | =head2 C-style Logical Or | |
d74e8afc | 477 | X<||> X<operator, logical, or> |
a0d0e21e LW |
478 | |
479 | Binary "||" performs a short-circuit logical OR operation. That is, | |
480 | if the left operand is true, the right operand is not even evaluated. | |
481 | Scalar or list context propagates down to the right operand if it | |
482 | is evaluated. | |
483 | ||
c963b151 | 484 | =head2 C-style Logical Defined-Or |
d74e8afc | 485 | X<//> X<operator, logical, defined-or> |
c963b151 BD |
486 | |
487 | Although it has no direct equivalent in C, Perl's C<//> operator is related | |
488 | to its C-style or. In fact, it's exactly the same as C<||>, except that it | |
489 | tests the left hand side's definedness instead of its truth. Thus, C<$a // $b> | |
490 | is similar to C<defined($a) || $b> (except that it returns the value of C<$a> | |
491 | rather than the value of C<defined($a)>) and is exactly equivalent to | |
492 | C<defined($a) ? $a : $b>. This is very useful for providing default values | |
d042e63d MS |
493 | for variables. If you actually want to test if at least one of C<$a> and |
494 | C<$b> is defined, use C<defined($a // $b)>. | |
c963b151 | 495 | |
d042e63d MS |
496 | The C<||>, C<//> and C<&&> operators return the last value evaluated |
497 | (unlike C's C<||> and C<&&>, which return 0 or 1). Thus, a reasonably | |
498 | portable way to find out the home directory might be: | |
a0d0e21e | 499 | |
c963b151 BD |
500 | $home = $ENV{'HOME'} // $ENV{'LOGDIR'} // |
501 | (getpwuid($<))[7] // die "You're homeless!\n"; | |
a0d0e21e | 502 | |
5a964f20 TC |
503 | In particular, this means that you shouldn't use this |
504 | for selecting between two aggregates for assignment: | |
505 | ||
506 | @a = @b || @c; # this is wrong | |
507 | @a = scalar(@b) || @c; # really meant this | |
508 | @a = @b ? @b : @c; # this works fine, though | |
509 | ||
c963b151 BD |
510 | As more readable alternatives to C<&&>, C<//> and C<||> when used for |
511 | control flow, Perl provides C<and>, C<err> and C<or> operators (see below). | |
512 | The short-circuit behavior is identical. The precedence of "and", "err" | |
513 | and "or" is much lower, however, so that you can safely use them after a | |
5a964f20 | 514 | list operator without the need for parentheses: |
a0d0e21e LW |
515 | |
516 | unlink "alpha", "beta", "gamma" | |
517 | or gripe(), next LINE; | |
518 | ||
519 | With the C-style operators that would have been written like this: | |
520 | ||
521 | unlink("alpha", "beta", "gamma") | |
522 | || (gripe(), next LINE); | |
523 | ||
eeb6a2c9 | 524 | Using "or" for assignment is unlikely to do what you want; see below. |
5a964f20 TC |
525 | |
526 | =head2 Range Operators | |
d74e8afc | 527 | X<operator, range> X<range> X<..> X<...> |
a0d0e21e LW |
528 | |
529 | Binary ".." is the range operator, which is really two different | |
fb53bbb2 | 530 | operators depending on the context. In list context, it returns a |
54ae734e | 531 | list of values counting (up by ones) from the left value to the right |
2cdbc966 | 532 | value. If the left value is greater than the right value then it |
fb53bbb2 | 533 | returns the empty list. The range operator is useful for writing |
54ae734e | 534 | C<foreach (1..10)> loops and for doing slice operations on arrays. In |
2cdbc966 JD |
535 | the current implementation, no temporary array is created when the |
536 | range operator is used as the expression in C<foreach> loops, but older | |
537 | versions of Perl might burn a lot of memory when you write something | |
538 | like this: | |
a0d0e21e LW |
539 | |
540 | for (1 .. 1_000_000) { | |
541 | # code | |
54310121 | 542 | } |
a0d0e21e | 543 | |
54ae734e MG |
544 | The range operator also works on strings, using the magical auto-increment, |
545 | see below. | |
546 | ||
5a964f20 | 547 | In scalar context, ".." returns a boolean value. The operator is |
a0d0e21e LW |
548 | bistable, like a flip-flop, and emulates the line-range (comma) operator |
549 | of B<sed>, B<awk>, and various editors. Each ".." operator maintains its | |
550 | own boolean state. It is false as long as its left operand is false. | |
551 | Once the left operand is true, the range operator stays true until the | |
552 | right operand is true, I<AFTER> which the range operator becomes false | |
19799a22 | 553 | again. It doesn't become false till the next time the range operator is |
a0d0e21e LW |
554 | evaluated. It can test the right operand and become false on the same |
555 | evaluation it became true (as in B<awk>), but it still returns true once. | |
19799a22 GS |
556 | If you don't want it to test the right operand till the next |
557 | evaluation, as in B<sed>, just use three dots ("...") instead of | |
558 | two. In all other regards, "..." behaves just like ".." does. | |
559 | ||
560 | The right operand is not evaluated while the operator is in the | |
561 | "false" state, and the left operand is not evaluated while the | |
562 | operator is in the "true" state. The precedence is a little lower | |
563 | than || and &&. The value returned is either the empty string for | |
564 | false, or a sequence number (beginning with 1) for true. The | |
565 | sequence number is reset for each range encountered. The final | |
566 | sequence number in a range has the string "E0" appended to it, which | |
567 | doesn't affect its numeric value, but gives you something to search | |
568 | for if you want to exclude the endpoint. You can exclude the | |
569 | beginning point by waiting for the sequence number to be greater | |
df5f8116 CW |
570 | than 1. |
571 | ||
572 | If either operand of scalar ".." is a constant expression, | |
573 | that operand is considered true if it is equal (C<==>) to the current | |
574 | input line number (the C<$.> variable). | |
575 | ||
576 | To be pedantic, the comparison is actually C<int(EXPR) == int(EXPR)>, | |
577 | but that is only an issue if you use a floating point expression; when | |
578 | implicitly using C<$.> as described in the previous paragraph, the | |
579 | comparison is C<int(EXPR) == int($.)> which is only an issue when C<$.> | |
580 | is set to a floating point value and you are not reading from a file. | |
581 | Furthermore, C<"span" .. "spat"> or C<2.18 .. 3.14> will not do what | |
582 | you want in scalar context because each of the operands are evaluated | |
583 | using their integer representation. | |
584 | ||
585 | Examples: | |
a0d0e21e LW |
586 | |
587 | As a scalar operator: | |
588 | ||
df5f8116 CW |
589 | if (101 .. 200) { print; } # print 2nd hundred lines, short for |
590 | # if ($. == 101 .. $. == 200) ... | |
9f10b797 RGS |
591 | |
592 | next LINE if (1 .. /^$/); # skip header lines, short for | |
df5f8116 | 593 | # ... if ($. == 1 .. /^$/); |
9f10b797 RGS |
594 | # (typically in a loop labeled LINE) |
595 | ||
596 | s/^/> / if (/^$/ .. eof()); # quote body | |
a0d0e21e | 597 | |
5a964f20 TC |
598 | # parse mail messages |
599 | while (<>) { | |
600 | $in_header = 1 .. /^$/; | |
df5f8116 CW |
601 | $in_body = /^$/ .. eof; |
602 | if ($in_header) { | |
603 | # ... | |
604 | } else { # in body | |
605 | # ... | |
606 | } | |
5a964f20 | 607 | } continue { |
df5f8116 | 608 | close ARGV if eof; # reset $. each file |
5a964f20 TC |
609 | } |
610 | ||
acf31ca5 SF |
611 | Here's a simple example to illustrate the difference between |
612 | the two range operators: | |
613 | ||
614 | @lines = (" - Foo", | |
615 | "01 - Bar", | |
616 | "1 - Baz", | |
617 | " - Quux"); | |
618 | ||
9f10b797 RGS |
619 | foreach (@lines) { |
620 | if (/0/ .. /1/) { | |
acf31ca5 SF |
621 | print "$_\n"; |
622 | } | |
623 | } | |
624 | ||
9f10b797 RGS |
625 | This program will print only the line containing "Bar". If |
626 | the range operator is changed to C<...>, it will also print the | |
acf31ca5 SF |
627 | "Baz" line. |
628 | ||
629 | And now some examples as a list operator: | |
a0d0e21e LW |
630 | |
631 | for (101 .. 200) { print; } # print $_ 100 times | |
3e3baf6d | 632 | @foo = @foo[0 .. $#foo]; # an expensive no-op |
a0d0e21e LW |
633 | @foo = @foo[$#foo-4 .. $#foo]; # slice last 5 items |
634 | ||
5a964f20 | 635 | The range operator (in list context) makes use of the magical |
5f05dabc | 636 | auto-increment algorithm if the operands are strings. You |
a0d0e21e LW |
637 | can say |
638 | ||
639 | @alphabet = ('A' .. 'Z'); | |
640 | ||
54ae734e | 641 | to get all normal letters of the English alphabet, or |
a0d0e21e LW |
642 | |
643 | $hexdigit = (0 .. 9, 'a' .. 'f')[$num & 15]; | |
644 | ||
645 | to get a hexadecimal digit, or | |
646 | ||
647 | @z2 = ('01' .. '31'); print $z2[$mday]; | |
648 | ||
649 | to get dates with leading zeros. If the final value specified is not | |
650 | in the sequence that the magical increment would produce, the sequence | |
651 | goes until the next value would be longer than the final value | |
652 | specified. | |
653 | ||
df5f8116 CW |
654 | Because each operand is evaluated in integer form, C<2.18 .. 3.14> will |
655 | return two elements in list context. | |
656 | ||
657 | @list = (2.18 .. 3.14); # same as @list = (2 .. 3); | |
658 | ||
a0d0e21e | 659 | =head2 Conditional Operator |
d74e8afc | 660 | X<operator, conditional> X<operator, ternary> X<ternary> X<?:> |
a0d0e21e LW |
661 | |
662 | Ternary "?:" is the conditional operator, just as in C. It works much | |
663 | like an if-then-else. If the argument before the ? is true, the | |
664 | argument before the : is returned, otherwise the argument after the : | |
cb1a09d0 AD |
665 | is returned. For example: |
666 | ||
54310121 | 667 | printf "I have %d dog%s.\n", $n, |
cb1a09d0 AD |
668 | ($n == 1) ? '' : "s"; |
669 | ||
670 | Scalar or list context propagates downward into the 2nd | |
54310121 | 671 | or 3rd argument, whichever is selected. |
cb1a09d0 AD |
672 | |
673 | $a = $ok ? $b : $c; # get a scalar | |
674 | @a = $ok ? @b : @c; # get an array | |
675 | $a = $ok ? @b : @c; # oops, that's just a count! | |
676 | ||
677 | The operator may be assigned to if both the 2nd and 3rd arguments are | |
678 | legal lvalues (meaning that you can assign to them): | |
a0d0e21e LW |
679 | |
680 | ($a_or_b ? $a : $b) = $c; | |
681 | ||
5a964f20 TC |
682 | Because this operator produces an assignable result, using assignments |
683 | without parentheses will get you in trouble. For example, this: | |
684 | ||
685 | $a % 2 ? $a += 10 : $a += 2 | |
686 | ||
687 | Really means this: | |
688 | ||
689 | (($a % 2) ? ($a += 10) : $a) += 2 | |
690 | ||
691 | Rather than this: | |
692 | ||
693 | ($a % 2) ? ($a += 10) : ($a += 2) | |
694 | ||
19799a22 GS |
695 | That should probably be written more simply as: |
696 | ||
697 | $a += ($a % 2) ? 10 : 2; | |
698 | ||
4633a7c4 | 699 | =head2 Assignment Operators |
d74e8afc | 700 | X<assignment> X<operator, assignment> X<=> X<**=> X<+=> X<*=> X<&=> |
5ac3b81c | 701 | X<<< <<= >>> X<&&=> X<-=> X</=> X<|=> X<<< >>= >>> X<||=> X<//=> X<.=> |
d74e8afc | 702 | X<%=> X<^=> X<x=> |
a0d0e21e LW |
703 | |
704 | "=" is the ordinary assignment operator. | |
705 | ||
706 | Assignment operators work as in C. That is, | |
707 | ||
708 | $a += 2; | |
709 | ||
710 | is equivalent to | |
711 | ||
712 | $a = $a + 2; | |
713 | ||
714 | although without duplicating any side effects that dereferencing the lvalue | |
54310121 | 715 | might trigger, such as from tie(). Other assignment operators work similarly. |
716 | The following are recognized: | |
a0d0e21e LW |
717 | |
718 | **= += *= &= <<= &&= | |
9f10b797 RGS |
719 | -= /= |= >>= ||= |
720 | .= %= ^= //= | |
721 | x= | |
a0d0e21e | 722 | |
19799a22 | 723 | Although these are grouped by family, they all have the precedence |
a0d0e21e LW |
724 | of assignment. |
725 | ||
b350dd2f GS |
726 | Unlike in C, the scalar assignment operator produces a valid lvalue. |
727 | Modifying an assignment is equivalent to doing the assignment and | |
728 | then modifying the variable that was assigned to. This is useful | |
729 | for modifying a copy of something, like this: | |
a0d0e21e LW |
730 | |
731 | ($tmp = $global) =~ tr [A-Z] [a-z]; | |
732 | ||
733 | Likewise, | |
734 | ||
735 | ($a += 2) *= 3; | |
736 | ||
737 | is equivalent to | |
738 | ||
739 | $a += 2; | |
740 | $a *= 3; | |
741 | ||
b350dd2f GS |
742 | Similarly, a list assignment in list context produces the list of |
743 | lvalues assigned to, and a list assignment in scalar context returns | |
744 | the number of elements produced by the expression on the right hand | |
745 | side of the assignment. | |
746 | ||
748a9306 | 747 | =head2 Comma Operator |
d74e8afc | 748 | X<comma> X<operator, comma> X<,> |
a0d0e21e | 749 | |
5a964f20 | 750 | Binary "," is the comma operator. In scalar context it evaluates |
a0d0e21e LW |
751 | its left argument, throws that value away, then evaluates its right |
752 | argument and returns that value. This is just like C's comma operator. | |
753 | ||
5a964f20 | 754 | In list context, it's just the list argument separator, and inserts |
a0d0e21e LW |
755 | both its arguments into the list. |
756 | ||
d042e63d | 757 | The C<< => >> operator is a synonym for the comma, but forces any word |
719b43e8 | 758 | (consisting entirely of word characters) to its left to be interpreted |
a44e5664 MS |
759 | as a string (as of 5.001). This includes words that might otherwise be |
760 | considered a constant or function call. | |
761 | ||
762 | use constant FOO => "something"; | |
763 | ||
764 | my %h = ( FOO => 23 ); | |
765 | ||
766 | is equivalent to: | |
767 | ||
768 | my %h = ("FOO", 23); | |
769 | ||
770 | It is I<NOT>: | |
771 | ||
772 | my %h = ("something", 23); | |
773 | ||
774 | If the argument on the left is not a word, it is first interpreted as | |
775 | an expression, and then the string value of that is used. | |
719b43e8 RGS |
776 | |
777 | The C<< => >> operator is helpful in documenting the correspondence | |
778 | between keys and values in hashes, and other paired elements in lists. | |
748a9306 | 779 | |
a44e5664 MS |
780 | %hash = ( $key => $value ); |
781 | login( $username => $password ); | |
782 | ||
a0d0e21e | 783 | =head2 List Operators (Rightward) |
d74e8afc | 784 | X<operator, list, rightward> X<list operator> |
a0d0e21e LW |
785 | |
786 | On the right side of a list operator, it has very low precedence, | |
787 | such that it controls all comma-separated expressions found there. | |
788 | The only operators with lower precedence are the logical operators | |
789 | "and", "or", and "not", which may be used to evaluate calls to list | |
790 | operators without the need for extra parentheses: | |
791 | ||
792 | open HANDLE, "filename" | |
793 | or die "Can't open: $!\n"; | |
794 | ||
5ba421f6 | 795 | See also discussion of list operators in L<Terms and List Operators (Leftward)>. |
a0d0e21e LW |
796 | |
797 | =head2 Logical Not | |
d74e8afc | 798 | X<operator, logical, not> X<not> |
a0d0e21e LW |
799 | |
800 | Unary "not" returns the logical negation of the expression to its right. | |
801 | It's the equivalent of "!" except for the very low precedence. | |
802 | ||
803 | =head2 Logical And | |
d74e8afc | 804 | X<operator, logical, and> X<and> |
a0d0e21e LW |
805 | |
806 | Binary "and" returns the logical conjunction of the two surrounding | |
807 | expressions. It's equivalent to && except for the very low | |
5f05dabc | 808 | precedence. This means that it short-circuits: i.e., the right |
a0d0e21e LW |
809 | expression is evaluated only if the left expression is true. |
810 | ||
c963b151 | 811 | =head2 Logical or, Defined or, and Exclusive Or |
d74e8afc ITB |
812 | X<operator, logical, or> X<operator, logical, xor> X<operator, logical, err> |
813 | X<operator, logical, defined or> X<operator, logical, exclusive or> | |
814 | X<or> X<xor> X<err> | |
a0d0e21e LW |
815 | |
816 | Binary "or" returns the logical disjunction of the two surrounding | |
5a964f20 TC |
817 | expressions. It's equivalent to || except for the very low precedence. |
818 | This makes it useful for control flow | |
819 | ||
820 | print FH $data or die "Can't write to FH: $!"; | |
821 | ||
822 | This means that it short-circuits: i.e., the right expression is evaluated | |
823 | only if the left expression is false. Due to its precedence, you should | |
824 | probably avoid using this for assignment, only for control flow. | |
825 | ||
826 | $a = $b or $c; # bug: this is wrong | |
827 | ($a = $b) or $c; # really means this | |
828 | $a = $b || $c; # better written this way | |
829 | ||
19799a22 | 830 | However, when it's a list-context assignment and you're trying to use |
5a964f20 TC |
831 | "||" for control flow, you probably need "or" so that the assignment |
832 | takes higher precedence. | |
833 | ||
834 | @info = stat($file) || die; # oops, scalar sense of stat! | |
835 | @info = stat($file) or die; # better, now @info gets its due | |
836 | ||
c963b151 BD |
837 | Then again, you could always use parentheses. |
838 | ||
9f10b797 RGS |
839 | Binary "err" is equivalent to C<//>--it's just like binary "or", except it |
840 | tests its left argument's definedness instead of its truth. There are two | |
841 | ways to remember "err": either because many functions return C<undef> on | |
842 | an B<err>or, or as a sort of correction: C<$a = ($b err 'default')>. This | |
843 | keyword is only available when the 'err' feature is enabled: see | |
844 | L<feature> for more information. | |
a0d0e21e LW |
845 | |
846 | Binary "xor" returns the exclusive-OR of the two surrounding expressions. | |
847 | It cannot short circuit, of course. | |
848 | ||
849 | =head2 C Operators Missing From Perl | |
d74e8afc ITB |
850 | X<operator, missing from perl> X<&> X<*> |
851 | X<typecasting> X<(TYPE)> | |
a0d0e21e LW |
852 | |
853 | Here is what C has that Perl doesn't: | |
854 | ||
855 | =over 8 | |
856 | ||
857 | =item unary & | |
858 | ||
859 | Address-of operator. (But see the "\" operator for taking a reference.) | |
860 | ||
861 | =item unary * | |
862 | ||
54310121 | 863 | Dereference-address operator. (Perl's prefix dereferencing |
a0d0e21e LW |
864 | operators are typed: $, @, %, and &.) |
865 | ||
866 | =item (TYPE) | |
867 | ||
19799a22 | 868 | Type-casting operator. |
a0d0e21e LW |
869 | |
870 | =back | |
871 | ||
5f05dabc | 872 | =head2 Quote and Quote-like Operators |
d74e8afc ITB |
873 | X<operator, quote> X<operator, quote-like> X<q> X<qq> X<qx> X<qw> X<m> |
874 | X<qr> X<s> X<tr> X<'> X<''> X<"> X<""> X<//> X<`> X<``> X<<< << >>> | |
875 | X<escape sequence> X<escape> | |
876 | ||
a0d0e21e LW |
877 | |
878 | While we usually think of quotes as literal values, in Perl they | |
879 | function as operators, providing various kinds of interpolating and | |
880 | pattern matching capabilities. Perl provides customary quote characters | |
881 | for these behaviors, but also provides a way for you to choose your | |
882 | quote character for any of them. In the following table, a C<{}> represents | |
9f10b797 | 883 | any pair of delimiters you choose. |
a0d0e21e | 884 | |
2c268ad5 TP |
885 | Customary Generic Meaning Interpolates |
886 | '' q{} Literal no | |
887 | "" qq{} Literal yes | |
af9219ee | 888 | `` qx{} Command yes* |
2c268ad5 | 889 | qw{} Word list no |
af9219ee MG |
890 | // m{} Pattern match yes* |
891 | qr{} Pattern yes* | |
892 | s{}{} Substitution yes* | |
2c268ad5 | 893 | tr{}{} Transliteration no (but see below) |
7e3b091d | 894 | <<EOF here-doc yes* |
a0d0e21e | 895 | |
af9219ee MG |
896 | * unless the delimiter is ''. |
897 | ||
87275199 GS |
898 | Non-bracketing delimiters use the same character fore and aft, but the four |
899 | sorts of brackets (round, angle, square, curly) will all nest, which means | |
9f10b797 | 900 | that |
87275199 | 901 | |
9f10b797 | 902 | q{foo{bar}baz} |
35f2feb0 | 903 | |
9f10b797 | 904 | is the same as |
87275199 GS |
905 | |
906 | 'foo{bar}baz' | |
907 | ||
908 | Note, however, that this does not always work for quoting Perl code: | |
909 | ||
910 | $s = q{ if($a eq "}") ... }; # WRONG | |
911 | ||
83df6a1d JH |
912 | is a syntax error. The C<Text::Balanced> module (from CPAN, and |
913 | starting from Perl 5.8 part of the standard distribution) is able | |
914 | to do this properly. | |
87275199 | 915 | |
19799a22 | 916 | There can be whitespace between the operator and the quoting |
fb73857a | 917 | characters, except when C<#> is being used as the quoting character. |
19799a22 GS |
918 | C<q#foo#> is parsed as the string C<foo>, while C<q #foo#> is the |
919 | operator C<q> followed by a comment. Its argument will be taken | |
920 | from the next line. This allows you to write: | |
fb73857a | 921 | |
922 | s {foo} # Replace foo | |
923 | {bar} # with bar. | |
924 | ||
904501ec MG |
925 | The following escape sequences are available in constructs that interpolate |
926 | and in transliterations. | |
d74e8afc | 927 | X<\t> X<\n> X<\r> X<\f> X<\b> X<\a> X<\e> X<\x> X<\0> X<\c> X<\N> |
a0d0e21e | 928 | |
6ee5d4e7 | 929 | \t tab (HT, TAB) |
5a964f20 | 930 | \n newline (NL) |
6ee5d4e7 | 931 | \r return (CR) |
932 | \f form feed (FF) | |
933 | \b backspace (BS) | |
934 | \a alarm (bell) (BEL) | |
935 | \e escape (ESC) | |
a0ed51b3 LW |
936 | \033 octal char (ESC) |
937 | \x1b hex char (ESC) | |
938 | \x{263a} wide hex char (SMILEY) | |
19799a22 | 939 | \c[ control char (ESC) |
95cc3e0c | 940 | \N{name} named Unicode character |
2c268ad5 | 941 | |
4c77eaa2 AE |
942 | B<NOTE>: Unlike C and other languages, Perl has no \v escape sequence for |
943 | the vertical tab (VT - ASCII 11). | |
944 | ||
904501ec MG |
945 | The following escape sequences are available in constructs that interpolate |
946 | but not in transliterations. | |
d74e8afc | 947 | X<\l> X<\u> X<\L> X<\U> X<\E> X<\Q> |
904501ec | 948 | |
a0d0e21e LW |
949 | \l lowercase next char |
950 | \u uppercase next char | |
951 | \L lowercase till \E | |
952 | \U uppercase till \E | |
953 | \E end case modification | |
1d2dff63 | 954 | \Q quote non-word characters till \E |
a0d0e21e | 955 | |
95cc3e0c JH |
956 | If C<use locale> is in effect, the case map used by C<\l>, C<\L>, |
957 | C<\u> and C<\U> is taken from the current locale. See L<perllocale>. | |
958 | If Unicode (for example, C<\N{}> or wide hex characters of 0x100 or | |
959 | beyond) is being used, the case map used by C<\l>, C<\L>, C<\u> and | |
960 | C<\U> is as defined by Unicode. For documentation of C<\N{name}>, | |
961 | see L<charnames>. | |
a034a98d | 962 | |
5a964f20 TC |
963 | All systems use the virtual C<"\n"> to represent a line terminator, |
964 | called a "newline". There is no such thing as an unvarying, physical | |
19799a22 | 965 | newline character. It is only an illusion that the operating system, |
5a964f20 TC |
966 | device drivers, C libraries, and Perl all conspire to preserve. Not all |
967 | systems read C<"\r"> as ASCII CR and C<"\n"> as ASCII LF. For example, | |
968 | on a Mac, these are reversed, and on systems without line terminator, | |
969 | printing C<"\n"> may emit no actual data. In general, use C<"\n"> when | |
970 | you mean a "newline" for your system, but use the literal ASCII when you | |
971 | need an exact character. For example, most networking protocols expect | |
2a380090 | 972 | and prefer a CR+LF (C<"\015\012"> or C<"\cM\cJ">) for line terminators, |
5a964f20 TC |
973 | and although they often accept just C<"\012">, they seldom tolerate just |
974 | C<"\015">. If you get in the habit of using C<"\n"> for networking, | |
975 | you may be burned some day. | |
d74e8afc ITB |
976 | X<newline> X<line terminator> X<eol> X<end of line> |
977 | X<\n> X<\r> X<\r\n> | |
5a964f20 | 978 | |
904501ec MG |
979 | For constructs that do interpolate, variables beginning with "C<$>" |
980 | or "C<@>" are interpolated. Subscripted variables such as C<$a[3]> or | |
ad0f383a A |
981 | C<< $href->{key}[0] >> are also interpolated, as are array and hash slices. |
982 | But method calls such as C<< $obj->meth >> are not. | |
af9219ee MG |
983 | |
984 | Interpolating an array or slice interpolates the elements in order, | |
985 | separated by the value of C<$">, so is equivalent to interpolating | |
904501ec MG |
986 | C<join $", @array>. "Punctuation" arrays such as C<@+> are only |
987 | interpolated if the name is enclosed in braces C<@{+}>. | |
af9219ee | 988 | |
1d2dff63 GS |
989 | You cannot include a literal C<$> or C<@> within a C<\Q> sequence. |
990 | An unescaped C<$> or C<@> interpolates the corresponding variable, | |
991 | while escaping will cause the literal string C<\$> to be inserted. | |
992 | You'll need to write something like C<m/\Quser\E\@\Qhost/>. | |
993 | ||
a0d0e21e LW |
994 | Patterns are subject to an additional level of interpretation as a |
995 | regular expression. This is done as a second pass, after variables are | |
996 | interpolated, so that regular expressions may be incorporated into the | |
997 | pattern from the variables. If this is not what you want, use C<\Q> to | |
998 | interpolate a variable literally. | |
999 | ||
19799a22 GS |
1000 | Apart from the behavior described above, Perl does not expand |
1001 | multiple levels of interpolation. In particular, contrary to the | |
1002 | expectations of shell programmers, back-quotes do I<NOT> interpolate | |
1003 | within double quotes, nor do single quotes impede evaluation of | |
1004 | variables when used within double quotes. | |
a0d0e21e | 1005 | |
5f05dabc | 1006 | =head2 Regexp Quote-Like Operators |
d74e8afc | 1007 | X<operator, regexp> |
cb1a09d0 | 1008 | |
5f05dabc | 1009 | Here are the quote-like operators that apply to pattern |
cb1a09d0 AD |
1010 | matching and related activities. |
1011 | ||
a0d0e21e LW |
1012 | =over 8 |
1013 | ||
1014 | =item ?PATTERN? | |
d74e8afc | 1015 | X<?> |
a0d0e21e LW |
1016 | |
1017 | This is just like the C</pattern/> search, except that it matches only | |
1018 | once between calls to the reset() operator. This is a useful | |
5f05dabc | 1019 | optimization when you want to see only the first occurrence of |
a0d0e21e LW |
1020 | something in each file of a set of files, for instance. Only C<??> |
1021 | patterns local to the current package are reset. | |
1022 | ||
5a964f20 TC |
1023 | while (<>) { |
1024 | if (?^$?) { | |
1025 | # blank line between header and body | |
1026 | } | |
1027 | } continue { | |
1028 | reset if eof; # clear ?? status for next file | |
1029 | } | |
1030 | ||
483b4840 | 1031 | This usage is vaguely deprecated, which means it just might possibly |
19799a22 GS |
1032 | be removed in some distant future version of Perl, perhaps somewhere |
1033 | around the year 2168. | |
a0d0e21e | 1034 | |
fb73857a | 1035 | =item m/PATTERN/cgimosx |
d74e8afc ITB |
1036 | X<m> X<operator, match> |
1037 | X<regexp, options> X<regexp> X<regex, options> X<regex> | |
1038 | X</c> X</i> X</m> X</o> X</s> X</x> | |
a0d0e21e | 1039 | |
fb73857a | 1040 | =item /PATTERN/cgimosx |
a0d0e21e | 1041 | |
5a964f20 | 1042 | Searches a string for a pattern match, and in scalar context returns |
19799a22 GS |
1043 | true if it succeeds, false if it fails. If no string is specified |
1044 | via the C<=~> or C<!~> operator, the $_ string is searched. (The | |
1045 | string specified with C<=~> need not be an lvalue--it may be the | |
1046 | result of an expression evaluation, but remember the C<=~> binds | |
1047 | rather tightly.) See also L<perlre>. See L<perllocale> for | |
1048 | discussion of additional considerations that apply when C<use locale> | |
1049 | is in effect. | |
a0d0e21e LW |
1050 | |
1051 | Options are: | |
1052 | ||
fb73857a | 1053 | c Do not reset search position on a failed match when /g is in effect. |
5f05dabc | 1054 | g Match globally, i.e., find all occurrences. |
a0d0e21e LW |
1055 | i Do case-insensitive pattern matching. |
1056 | m Treat string as multiple lines. | |
5f05dabc | 1057 | o Compile pattern only once. |
a0d0e21e LW |
1058 | s Treat string as single line. |
1059 | x Use extended regular expressions. | |
1060 | ||
1061 | If "/" is the delimiter then the initial C<m> is optional. With the C<m> | |
01ae956f | 1062 | you can use any pair of non-alphanumeric, non-whitespace characters |
19799a22 GS |
1063 | as delimiters. This is particularly useful for matching path names |
1064 | that contain "/", to avoid LTS (leaning toothpick syndrome). If "?" is | |
7bac28a0 | 1065 | the delimiter, then the match-only-once rule of C<?PATTERN?> applies. |
19799a22 | 1066 | If "'" is the delimiter, no interpolation is performed on the PATTERN. |
a0d0e21e LW |
1067 | |
1068 | PATTERN may contain variables, which will be interpolated (and the | |
f70b4f9c | 1069 | pattern recompiled) every time the pattern search is evaluated, except |
1f247705 GS |
1070 | for when the delimiter is a single quote. (Note that C<$(>, C<$)>, and |
1071 | C<$|> are not interpolated because they look like end-of-string tests.) | |
f70b4f9c AB |
1072 | If you want such a pattern to be compiled only once, add a C</o> after |
1073 | the trailing delimiter. This avoids expensive run-time recompilations, | |
1074 | and is useful when the value you are interpolating won't change over | |
1075 | the life of the script. However, mentioning C</o> constitutes a promise | |
1076 | that you won't change the variables in the pattern. If you change them, | |
13a2d996 | 1077 | Perl won't even notice. See also L<"qr/STRING/imosx">. |
a0d0e21e | 1078 | |
5a964f20 | 1079 | If the PATTERN evaluates to the empty string, the last |
d65afb4b HS |
1080 | I<successfully> matched regular expression is used instead. In this |
1081 | case, only the C<g> and C<c> flags on the empty pattern is honoured - | |
1082 | the other flags are taken from the original pattern. If no match has | |
1083 | previously succeeded, this will (silently) act instead as a genuine | |
1084 | empty pattern (which will always match). | |
a0d0e21e | 1085 | |
c963b151 BD |
1086 | Note that it's possible to confuse Perl into thinking C<//> (the empty |
1087 | regex) is really C<//> (the defined-or operator). Perl is usually pretty | |
1088 | good about this, but some pathological cases might trigger this, such as | |
1089 | C<$a///> (is that C<($a) / (//)> or C<$a // />?) and C<print $fh //> | |
1090 | (C<print $fh(//> or C<print($fh //>?). In all of these examples, Perl | |
1091 | will assume you meant defined-or. If you meant the empty regex, just | |
1092 | use parentheses or spaces to disambiguate, or even prefix the empty | |
1093 | regex with an C<m> (so C<//> becomes C<m//>). | |
1094 | ||
19799a22 | 1095 | If the C</g> option is not used, C<m//> in list context returns a |
a0d0e21e | 1096 | list consisting of the subexpressions matched by the parentheses in the |
f7e33566 GS |
1097 | pattern, i.e., (C<$1>, C<$2>, C<$3>...). (Note that here C<$1> etc. are |
1098 | also set, and that this differs from Perl 4's behavior.) When there are | |
1099 | no parentheses in the pattern, the return value is the list C<(1)> for | |
1100 | success. With or without parentheses, an empty list is returned upon | |
1101 | failure. | |
a0d0e21e LW |
1102 | |
1103 | Examples: | |
1104 | ||
1105 | open(TTY, '/dev/tty'); | |
1106 | <TTY> =~ /^y/i && foo(); # do foo if desired | |
1107 | ||
1108 | if (/Version: *([0-9.]*)/) { $version = $1; } | |
1109 | ||
1110 | next if m#^/usr/spool/uucp#; | |
1111 | ||
1112 | # poor man's grep | |
1113 | $arg = shift; | |
1114 | while (<>) { | |
1115 | print if /$arg/o; # compile only once | |
1116 | } | |
1117 | ||
1118 | if (($F1, $F2, $Etc) = ($foo =~ /^(\S+)\s+(\S+)\s*(.*)/)) | |
1119 | ||
1120 | This last example splits $foo into the first two words and the | |
5f05dabc | 1121 | remainder of the line, and assigns those three fields to $F1, $F2, and |
1122 | $Etc. The conditional is true if any variables were assigned, i.e., if | |
a0d0e21e LW |
1123 | the pattern matched. |
1124 | ||
19799a22 GS |
1125 | The C</g> modifier specifies global pattern matching--that is, |
1126 | matching as many times as possible within the string. How it behaves | |
1127 | depends on the context. In list context, it returns a list of the | |
1128 | substrings matched by any capturing parentheses in the regular | |
1129 | expression. If there are no parentheses, it returns a list of all | |
1130 | the matched strings, as if there were parentheses around the whole | |
1131 | pattern. | |
a0d0e21e | 1132 | |
7e86de3e | 1133 | In scalar context, each execution of C<m//g> finds the next match, |
19799a22 | 1134 | returning true if it matches, and false if there is no further match. |
7e86de3e MG |
1135 | The position after the last match can be read or set using the pos() |
1136 | function; see L<perlfunc/pos>. A failed match normally resets the | |
1137 | search position to the beginning of the string, but you can avoid that | |
1138 | by adding the C</c> modifier (e.g. C<m//gc>). Modifying the target | |
1139 | string also resets the search position. | |
c90c0ff4 | 1140 | |
1141 | You can intermix C<m//g> matches with C<m/\G.../g>, where C<\G> is a | |
1142 | zero-width assertion that matches the exact position where the previous | |
5d43e42d DC |
1143 | C<m//g>, if any, left off. Without the C</g> modifier, the C<\G> assertion |
1144 | still anchors at pos(), but the match is of course only attempted once. | |
1145 | Using C<\G> without C</g> on a target string that has not previously had a | |
1146 | C</g> match applied to it is the same as using the C<\A> assertion to match | |
fe4b3f22 RGS |
1147 | the beginning of the string. Note also that, currently, C<\G> is only |
1148 | properly supported when anchored at the very beginning of the pattern. | |
c90c0ff4 | 1149 | |
1150 | Examples: | |
a0d0e21e LW |
1151 | |
1152 | # list context | |
1153 | ($one,$five,$fifteen) = (`uptime` =~ /(\d+\.\d+)/g); | |
1154 | ||
1155 | # scalar context | |
5d43e42d | 1156 | $/ = ""; |
19799a22 GS |
1157 | while (defined($paragraph = <>)) { |
1158 | while ($paragraph =~ /[a-z]['")]*[.!?]+['")]*\s/g) { | |
1159 | $sentences++; | |
a0d0e21e LW |
1160 | } |
1161 | } | |
1162 | print "$sentences\n"; | |
1163 | ||
c90c0ff4 | 1164 | # using m//gc with \G |
137443ea | 1165 | $_ = "ppooqppqq"; |
44a8e56a | 1166 | while ($i++ < 2) { |
1167 | print "1: '"; | |
c90c0ff4 | 1168 | print $1 while /(o)/gc; print "', pos=", pos, "\n"; |
44a8e56a | 1169 | print "2: '"; |
c90c0ff4 | 1170 | print $1 if /\G(q)/gc; print "', pos=", pos, "\n"; |
44a8e56a | 1171 | print "3: '"; |
c90c0ff4 | 1172 | print $1 while /(p)/gc; print "', pos=", pos, "\n"; |
44a8e56a | 1173 | } |
5d43e42d | 1174 | print "Final: '$1', pos=",pos,"\n" if /\G(.)/; |
44a8e56a | 1175 | |
1176 | The last example should print: | |
1177 | ||
1178 | 1: 'oo', pos=4 | |
137443ea | 1179 | 2: 'q', pos=5 |
44a8e56a | 1180 | 3: 'pp', pos=7 |
1181 | 1: '', pos=7 | |
137443ea | 1182 | 2: 'q', pos=8 |
1183 | 3: '', pos=8 | |
5d43e42d DC |
1184 | Final: 'q', pos=8 |
1185 | ||
1186 | Notice that the final match matched C<q> instead of C<p>, which a match | |
1187 | without the C<\G> anchor would have done. Also note that the final match | |
1188 | did not update C<pos> -- C<pos> is only updated on a C</g> match. If the | |
1189 | final match did indeed match C<p>, it's a good bet that you're running an | |
1190 | older (pre-5.6.0) Perl. | |
44a8e56a | 1191 | |
c90c0ff4 | 1192 | A useful idiom for C<lex>-like scanners is C</\G.../gc>. You can |
e7ea3e70 | 1193 | combine several regexps like this to process a string part-by-part, |
c90c0ff4 | 1194 | doing different actions depending on which regexp matched. Each |
1195 | regexp tries to match where the previous one leaves off. | |
e7ea3e70 | 1196 | |
3fe9a6f1 | 1197 | $_ = <<'EOL'; |
e7ea3e70 | 1198 | $url = new URI::URL "http://www/"; die if $url eq "xXx"; |
3fe9a6f1 | 1199 | EOL |
1200 | LOOP: | |
e7ea3e70 | 1201 | { |
c90c0ff4 | 1202 | print(" digits"), redo LOOP if /\G\d+\b[,.;]?\s*/gc; |
1203 | print(" lowercase"), redo LOOP if /\G[a-z]+\b[,.;]?\s*/gc; | |
1204 | print(" UPPERCASE"), redo LOOP if /\G[A-Z]+\b[,.;]?\s*/gc; | |
1205 | print(" Capitalized"), redo LOOP if /\G[A-Z][a-z]+\b[,.;]?\s*/gc; | |
1206 | print(" MiXeD"), redo LOOP if /\G[A-Za-z]+\b[,.;]?\s*/gc; | |
1207 | print(" alphanumeric"), redo LOOP if /\G[A-Za-z0-9]+\b[,.;]?\s*/gc; | |
1208 | print(" line-noise"), redo LOOP if /\G[^A-Za-z0-9]+/gc; | |
e7ea3e70 IZ |
1209 | print ". That's all!\n"; |
1210 | } | |
1211 | ||
1212 | Here is the output (split into several lines): | |
1213 | ||
1214 | line-noise lowercase line-noise lowercase UPPERCASE line-noise | |
1215 | UPPERCASE line-noise lowercase line-noise lowercase line-noise | |
1216 | lowercase lowercase line-noise lowercase lowercase line-noise | |
1217 | MiXeD line-noise. That's all! | |
44a8e56a | 1218 | |
a0d0e21e | 1219 | =item q/STRING/ |
d74e8afc | 1220 | X<q> X<quote, double> X<'> X<''> |
a0d0e21e LW |
1221 | |
1222 | =item C<'STRING'> | |
1223 | ||
19799a22 | 1224 | A single-quoted, literal string. A backslash represents a backslash |
68dc0745 | 1225 | unless followed by the delimiter or another backslash, in which case |
1226 | the delimiter or backslash is interpolated. | |
a0d0e21e LW |
1227 | |
1228 | $foo = q!I said, "You said, 'She said it.'"!; | |
1229 | $bar = q('This is it.'); | |
68dc0745 | 1230 | $baz = '\n'; # a two-character string |
a0d0e21e LW |
1231 | |
1232 | =item qq/STRING/ | |
d74e8afc | 1233 | X<qq> X<quote, double> X<"> X<""> |
a0d0e21e LW |
1234 | |
1235 | =item "STRING" | |
1236 | ||
1237 | A double-quoted, interpolated string. | |
1238 | ||
1239 | $_ .= qq | |
1240 | (*** The previous line contains the naughty word "$1".\n) | |
19799a22 | 1241 | if /\b(tcl|java|python)\b/i; # :-) |
68dc0745 | 1242 | $baz = "\n"; # a one-character string |
a0d0e21e | 1243 | |
eec2d3df | 1244 | =item qr/STRING/imosx |
d74e8afc | 1245 | X<qr> X</i> X</m> X</o> X</s> X</x> |
eec2d3df | 1246 | |
322edccd | 1247 | This operator quotes (and possibly compiles) its I<STRING> as a regular |
19799a22 GS |
1248 | expression. I<STRING> is interpolated the same way as I<PATTERN> |
1249 | in C<m/PATTERN/>. If "'" is used as the delimiter, no interpolation | |
1250 | is done. Returns a Perl value which may be used instead of the | |
1251 | corresponding C</STRING/imosx> expression. | |
4b6a7270 IZ |
1252 | |
1253 | For example, | |
1254 | ||
1255 | $rex = qr/my.STRING/is; | |
1256 | s/$rex/foo/; | |
1257 | ||
1258 | is equivalent to | |
1259 | ||
1260 | s/my.STRING/foo/is; | |
1261 | ||
1262 | The result may be used as a subpattern in a match: | |
eec2d3df GS |
1263 | |
1264 | $re = qr/$pattern/; | |
0a92e3a8 GS |
1265 | $string =~ /foo${re}bar/; # can be interpolated in other patterns |
1266 | $string =~ $re; # or used standalone | |
4b6a7270 IZ |
1267 | $string =~ /$re/; # or this way |
1268 | ||
1269 | Since Perl may compile the pattern at the moment of execution of qr() | |
19799a22 | 1270 | operator, using qr() may have speed advantages in some situations, |
4b6a7270 IZ |
1271 | notably if the result of qr() is used standalone: |
1272 | ||
1273 | sub match { | |
1274 | my $patterns = shift; | |
1275 | my @compiled = map qr/$_/i, @$patterns; | |
1276 | grep { | |
1277 | my $success = 0; | |
a7665c5e | 1278 | foreach my $pat (@compiled) { |
4b6a7270 IZ |
1279 | $success = 1, last if /$pat/; |
1280 | } | |
1281 | $success; | |
1282 | } @_; | |
1283 | } | |
1284 | ||
19799a22 GS |
1285 | Precompilation of the pattern into an internal representation at |
1286 | the moment of qr() avoids a need to recompile the pattern every | |
1287 | time a match C</$pat/> is attempted. (Perl has many other internal | |
1288 | optimizations, but none would be triggered in the above example if | |
1289 | we did not use qr() operator.) | |
eec2d3df GS |
1290 | |
1291 | Options are: | |
1292 | ||
1293 | i Do case-insensitive pattern matching. | |
1294 | m Treat string as multiple lines. | |
1295 | o Compile pattern only once. | |
1296 | s Treat string as single line. | |
1297 | x Use extended regular expressions. | |
1298 | ||
0a92e3a8 GS |
1299 | See L<perlre> for additional information on valid syntax for STRING, and |
1300 | for a detailed look at the semantics of regular expressions. | |
1301 | ||
a0d0e21e | 1302 | =item qx/STRING/ |
d74e8afc | 1303 | X<qx> X<`> X<``> X<backtick> |
a0d0e21e LW |
1304 | |
1305 | =item `STRING` | |
1306 | ||
43dd4d21 JH |
1307 | A string which is (possibly) interpolated and then executed as a |
1308 | system command with C</bin/sh> or its equivalent. Shell wildcards, | |
1309 | pipes, and redirections will be honored. The collected standard | |
1310 | output of the command is returned; standard error is unaffected. In | |
1311 | scalar context, it comes back as a single (potentially multi-line) | |
1312 | string, or undef if the command failed. In list context, returns a | |
1313 | list of lines (however you've defined lines with $/ or | |
1314 | $INPUT_RECORD_SEPARATOR), or an empty list if the command failed. | |
5a964f20 TC |
1315 | |
1316 | Because backticks do not affect standard error, use shell file descriptor | |
1317 | syntax (assuming the shell supports this) if you care to address this. | |
1318 | To capture a command's STDERR and STDOUT together: | |
a0d0e21e | 1319 | |
5a964f20 TC |
1320 | $output = `cmd 2>&1`; |
1321 | ||
1322 | To capture a command's STDOUT but discard its STDERR: | |
1323 | ||
1324 | $output = `cmd 2>/dev/null`; | |
1325 | ||
1326 | To capture a command's STDERR but discard its STDOUT (ordering is | |
1327 | important here): | |
1328 | ||
1329 | $output = `cmd 2>&1 1>/dev/null`; | |
1330 | ||
1331 | To exchange a command's STDOUT and STDERR in order to capture the STDERR | |
1332 | but leave its STDOUT to come out the old STDERR: | |
1333 | ||
1334 | $output = `cmd 3>&1 1>&2 2>&3 3>&-`; | |
1335 | ||
1336 | To read both a command's STDOUT and its STDERR separately, it's easiest | |
2359510d SD |
1337 | to redirect them separately to files, and then read from those files |
1338 | when the program is done: | |
5a964f20 | 1339 | |
2359510d | 1340 | system("program args 1>program.stdout 2>program.stderr"); |
5a964f20 TC |
1341 | |
1342 | Using single-quote as a delimiter protects the command from Perl's | |
1343 | double-quote interpolation, passing it on to the shell instead: | |
1344 | ||
1345 | $perl_info = qx(ps $$); # that's Perl's $$ | |
1346 | $shell_info = qx'ps $$'; # that's the new shell's $$ | |
1347 | ||
19799a22 | 1348 | How that string gets evaluated is entirely subject to the command |
5a964f20 TC |
1349 | interpreter on your system. On most platforms, you will have to protect |
1350 | shell metacharacters if you want them treated literally. This is in | |
1351 | practice difficult to do, as it's unclear how to escape which characters. | |
1352 | See L<perlsec> for a clean and safe example of a manual fork() and exec() | |
1353 | to emulate backticks safely. | |
a0d0e21e | 1354 | |
bb32b41a GS |
1355 | On some platforms (notably DOS-like ones), the shell may not be |
1356 | capable of dealing with multiline commands, so putting newlines in | |
1357 | the string may not get you what you want. You may be able to evaluate | |
1358 | multiple commands in a single line by separating them with the command | |
1359 | separator character, if your shell supports that (e.g. C<;> on many Unix | |
1360 | shells; C<&> on the Windows NT C<cmd> shell). | |
1361 | ||
0f897271 GS |
1362 | Beginning with v5.6.0, Perl will attempt to flush all files opened for |
1363 | output before starting the child process, but this may not be supported | |
1364 | on some platforms (see L<perlport>). To be safe, you may need to set | |
1365 | C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method of | |
1366 | C<IO::Handle> on any open handles. | |
1367 | ||
bb32b41a GS |
1368 | Beware that some command shells may place restrictions on the length |
1369 | of the command line. You must ensure your strings don't exceed this | |
1370 | limit after any necessary interpolations. See the platform-specific | |
1371 | release notes for more details about your particular environment. | |
1372 | ||
5a964f20 TC |
1373 | Using this operator can lead to programs that are difficult to port, |
1374 | because the shell commands called vary between systems, and may in | |
1375 | fact not be present at all. As one example, the C<type> command under | |
1376 | the POSIX shell is very different from the C<type> command under DOS. | |
1377 | That doesn't mean you should go out of your way to avoid backticks | |
1378 | when they're the right way to get something done. Perl was made to be | |
1379 | a glue language, and one of the things it glues together is commands. | |
1380 | Just understand what you're getting yourself into. | |
bb32b41a | 1381 | |
dc848c6f | 1382 | See L<"I/O Operators"> for more discussion. |
a0d0e21e | 1383 | |
945c54fd | 1384 | =item qw/STRING/ |
d74e8afc | 1385 | X<qw> X<quote, list> X<quote, words> |
945c54fd JH |
1386 | |
1387 | Evaluates to a list of the words extracted out of STRING, using embedded | |
1388 | whitespace as the word delimiters. It can be understood as being roughly | |
1389 | equivalent to: | |
1390 | ||
1391 | split(' ', q/STRING/); | |
1392 | ||
efb1e162 CW |
1393 | the differences being that it generates a real list at compile time, and |
1394 | in scalar context it returns the last element in the list. So | |
945c54fd JH |
1395 | this expression: |
1396 | ||
1397 | qw(foo bar baz) | |
1398 | ||
1399 | is semantically equivalent to the list: | |
1400 | ||
1401 | 'foo', 'bar', 'baz' | |
1402 | ||
1403 | Some frequently seen examples: | |
1404 | ||
1405 | use POSIX qw( setlocale localeconv ) | |
1406 | @EXPORT = qw( foo bar baz ); | |
1407 | ||
1408 | A common mistake is to try to separate the words with comma or to | |
1409 | put comments into a multi-line C<qw>-string. For this reason, the | |
1410 | C<use warnings> pragma and the B<-w> switch (that is, the C<$^W> variable) | |
1411 | produces warnings if the STRING contains the "," or the "#" character. | |
1412 | ||
a0d0e21e | 1413 | =item s/PATTERN/REPLACEMENT/egimosx |
d74e8afc ITB |
1414 | X<substitute> X<substitution> X<replace> X<regexp, replace> |
1415 | X<regexp, substitute> X</e> X</g> X</i> X</m> X</o> X</s> X</x> | |
a0d0e21e LW |
1416 | |
1417 | Searches a string for a pattern, and if found, replaces that pattern | |
1418 | with the replacement text and returns the number of substitutions | |
e37d713d | 1419 | made. Otherwise it returns false (specifically, the empty string). |
a0d0e21e LW |
1420 | |
1421 | If no string is specified via the C<=~> or C<!~> operator, the C<$_> | |
1422 | variable is searched and modified. (The string specified with C<=~> must | |
5a964f20 | 1423 | be scalar variable, an array element, a hash element, or an assignment |
5f05dabc | 1424 | to one of those, i.e., an lvalue.) |
a0d0e21e | 1425 | |
19799a22 | 1426 | If the delimiter chosen is a single quote, no interpolation is |
a0d0e21e LW |
1427 | done on either the PATTERN or the REPLACEMENT. Otherwise, if the |
1428 | PATTERN contains a $ that looks like a variable rather than an | |
1429 | end-of-string test, the variable will be interpolated into the pattern | |
5f05dabc | 1430 | at run-time. If you want the pattern compiled only once the first time |
a0d0e21e | 1431 | the variable is interpolated, use the C</o> option. If the pattern |
5a964f20 | 1432 | evaluates to the empty string, the last successfully executed regular |
a0d0e21e | 1433 | expression is used instead. See L<perlre> for further explanation on these. |
5a964f20 | 1434 | See L<perllocale> for discussion of additional considerations that apply |
a034a98d | 1435 | when C<use locale> is in effect. |
a0d0e21e LW |
1436 | |
1437 | Options are: | |
1438 | ||
1439 | e Evaluate the right side as an expression. | |
5f05dabc | 1440 | g Replace globally, i.e., all occurrences. |
a0d0e21e LW |
1441 | i Do case-insensitive pattern matching. |
1442 | m Treat string as multiple lines. | |
5f05dabc | 1443 | o Compile pattern only once. |
a0d0e21e LW |
1444 | s Treat string as single line. |
1445 | x Use extended regular expressions. | |
1446 | ||
1447 | Any non-alphanumeric, non-whitespace delimiter may replace the | |
1448 | slashes. If single quotes are used, no interpretation is done on the | |
e37d713d | 1449 | replacement string (the C</e> modifier overrides this, however). Unlike |
54310121 | 1450 | Perl 4, Perl 5 treats backticks as normal delimiters; the replacement |
e37d713d | 1451 | text is not evaluated as a command. If the |
a0d0e21e | 1452 | PATTERN is delimited by bracketing quotes, the REPLACEMENT has its own |
5f05dabc | 1453 | pair of quotes, which may or may not be bracketing quotes, e.g., |
35f2feb0 | 1454 | C<s(foo)(bar)> or C<< s<foo>/bar/ >>. A C</e> will cause the |
cec88af6 GS |
1455 | replacement portion to be treated as a full-fledged Perl expression |
1456 | and evaluated right then and there. It is, however, syntax checked at | |
1457 | compile-time. A second C<e> modifier will cause the replacement portion | |
1458 | to be C<eval>ed before being run as a Perl expression. | |
a0d0e21e LW |
1459 | |
1460 | Examples: | |
1461 | ||
1462 | s/\bgreen\b/mauve/g; # don't change wintergreen | |
1463 | ||
1464 | $path =~ s|/usr/bin|/usr/local/bin|; | |
1465 | ||
1466 | s/Login: $foo/Login: $bar/; # run-time pattern | |
1467 | ||
5a964f20 | 1468 | ($foo = $bar) =~ s/this/that/; # copy first, then change |
a0d0e21e | 1469 | |
5a964f20 | 1470 | $count = ($paragraph =~ s/Mister\b/Mr./g); # get change-count |
a0d0e21e LW |
1471 | |
1472 | $_ = 'abc123xyz'; | |
1473 | s/\d+/$&*2/e; # yields 'abc246xyz' | |
1474 | s/\d+/sprintf("%5d",$&)/e; # yields 'abc 246xyz' | |
1475 | s/\w/$& x 2/eg; # yields 'aabbcc 224466xxyyzz' | |
1476 | ||
1477 | s/%(.)/$percent{$1}/g; # change percent escapes; no /e | |
1478 | s/%(.)/$percent{$1} || $&/ge; # expr now, so /e | |
1479 | s/^=(\w+)/&pod($1)/ge; # use function call | |
1480 | ||
5a964f20 TC |
1481 | # expand variables in $_, but dynamics only, using |
1482 | # symbolic dereferencing | |
1483 | s/\$(\w+)/${$1}/g; | |
1484 | ||
cec88af6 GS |
1485 | # Add one to the value of any numbers in the string |
1486 | s/(\d+)/1 + $1/eg; | |
1487 | ||
1488 | # This will expand any embedded scalar variable | |
1489 | # (including lexicals) in $_ : First $1 is interpolated | |
1490 | # to the variable name, and then evaluated | |
a0d0e21e LW |
1491 | s/(\$\w+)/$1/eeg; |
1492 | ||
5a964f20 | 1493 | # Delete (most) C comments. |
a0d0e21e | 1494 | $program =~ s { |
4633a7c4 LW |
1495 | /\* # Match the opening delimiter. |
1496 | .*? # Match a minimal number of characters. | |
1497 | \*/ # Match the closing delimiter. | |
a0d0e21e LW |
1498 | } []gsx; |
1499 | ||
6b0ac556 | 1500 | s/^\s*(.*?)\s*$/$1/; # trim whitespace in $_, expensively |
5a964f20 | 1501 | |
6b0ac556 | 1502 | for ($variable) { # trim whitespace in $variable, cheap |
5a964f20 TC |
1503 | s/^\s+//; |
1504 | s/\s+$//; | |
1505 | } | |
a0d0e21e LW |
1506 | |
1507 | s/([^ ]*) *([^ ]*)/$2 $1/; # reverse 1st two fields | |
1508 | ||
54310121 | 1509 | Note the use of $ instead of \ in the last example. Unlike |
35f2feb0 GS |
1510 | B<sed>, we use the \<I<digit>> form in only the left hand side. |
1511 | Anywhere else it's $<I<digit>>. | |
a0d0e21e | 1512 | |
5f05dabc | 1513 | Occasionally, you can't use just a C</g> to get all the changes |
19799a22 | 1514 | to occur that you might want. Here are two common cases: |
a0d0e21e LW |
1515 | |
1516 | # put commas in the right places in an integer | |
19799a22 | 1517 | 1 while s/(\d)(\d\d\d)(?!\d)/$1,$2/g; |
a0d0e21e LW |
1518 | |
1519 | # expand tabs to 8-column spacing | |
1520 | 1 while s/\t+/' ' x (length($&)*8 - length($`)%8)/e; | |
1521 | ||
6940069f | 1522 | =item tr/SEARCHLIST/REPLACEMENTLIST/cds |
d74e8afc | 1523 | X<tr> X<y> X<transliterate> X</c> X</d> X</s> |
a0d0e21e | 1524 | |
6940069f | 1525 | =item y/SEARCHLIST/REPLACEMENTLIST/cds |
a0d0e21e | 1526 | |
2c268ad5 | 1527 | Transliterates all occurrences of the characters found in the search list |
a0d0e21e LW |
1528 | with the corresponding character in the replacement list. It returns |
1529 | the number of characters replaced or deleted. If no string is | |
2c268ad5 | 1530 | specified via the =~ or !~ operator, the $_ string is transliterated. (The |
54310121 | 1531 | string specified with =~ must be a scalar variable, an array element, a |
1532 | hash element, or an assignment to one of those, i.e., an lvalue.) | |
8ada0baa | 1533 | |
2c268ad5 TP |
1534 | A character range may be specified with a hyphen, so C<tr/A-J/0-9/> |
1535 | does the same replacement as C<tr/ACEGIBDFHJ/0246813579/>. | |
54310121 | 1536 | For B<sed> devotees, C<y> is provided as a synonym for C<tr>. If the |
1537 | SEARCHLIST is delimited by bracketing quotes, the REPLACEMENTLIST has | |
1538 | its own pair of quotes, which may or may not be bracketing quotes, | |
2c268ad5 | 1539 | e.g., C<tr[A-Z][a-z]> or C<tr(+\-*/)/ABCD/>. |
a0d0e21e | 1540 | |
cc255d5f JH |
1541 | Note that C<tr> does B<not> do regular expression character classes |
1542 | such as C<\d> or C<[:lower:]>. The <tr> operator is not equivalent to | |
1543 | the tr(1) utility. If you want to map strings between lower/upper | |
1544 | cases, see L<perlfunc/lc> and L<perlfunc/uc>, and in general consider | |
1545 | using the C<s> operator if you need regular expressions. | |
1546 | ||
8ada0baa JH |
1547 | Note also that the whole range idea is rather unportable between |
1548 | character sets--and even within character sets they may cause results | |
1549 | you probably didn't expect. A sound principle is to use only ranges | |
1550 | that begin from and end at either alphabets of equal case (a-e, A-E), | |
1551 | or digits (0-4). Anything else is unsafe. If in doubt, spell out the | |
1552 | character sets in full. | |
1553 | ||
a0d0e21e LW |
1554 | Options: |
1555 | ||
1556 | c Complement the SEARCHLIST. | |
1557 | d Delete found but unreplaced characters. | |
1558 | s Squash duplicate replaced characters. | |
1559 | ||
19799a22 GS |
1560 | If the C</c> modifier is specified, the SEARCHLIST character set |
1561 | is complemented. If the C</d> modifier is specified, any characters | |
1562 | specified by SEARCHLIST not found in REPLACEMENTLIST are deleted. | |
1563 | (Note that this is slightly more flexible than the behavior of some | |
1564 | B<tr> programs, which delete anything they find in the SEARCHLIST, | |
1565 | period.) If the C</s> modifier is specified, sequences of characters | |
1566 | that were transliterated to the same character are squashed down | |
1567 | to a single instance of the character. | |
a0d0e21e LW |
1568 | |
1569 | If the C</d> modifier is used, the REPLACEMENTLIST is always interpreted | |
1570 | exactly as specified. Otherwise, if the REPLACEMENTLIST is shorter | |
1571 | than the SEARCHLIST, the final character is replicated till it is long | |
5a964f20 | 1572 | enough. If the REPLACEMENTLIST is empty, the SEARCHLIST is replicated. |
a0d0e21e LW |
1573 | This latter is useful for counting characters in a class or for |
1574 | squashing character sequences in a class. | |
1575 | ||
1576 | Examples: | |
1577 | ||
1578 | $ARGV[1] =~ tr/A-Z/a-z/; # canonicalize to lower case | |
1579 | ||
1580 | $cnt = tr/*/*/; # count the stars in $_ | |
1581 | ||
1582 | $cnt = $sky =~ tr/*/*/; # count the stars in $sky | |
1583 | ||
1584 | $cnt = tr/0-9//; # count the digits in $_ | |
1585 | ||
1586 | tr/a-zA-Z//s; # bookkeeper -> bokeper | |
1587 | ||
1588 | ($HOST = $host) =~ tr/a-z/A-Z/; | |
1589 | ||
1590 | tr/a-zA-Z/ /cs; # change non-alphas to single space | |
1591 | ||
1592 | tr [\200-\377] | |
1593 | [\000-\177]; # delete 8th bit | |
1594 | ||
19799a22 GS |
1595 | If multiple transliterations are given for a character, only the |
1596 | first one is used: | |
748a9306 LW |
1597 | |
1598 | tr/AAA/XYZ/ | |
1599 | ||
2c268ad5 | 1600 | will transliterate any A to X. |
748a9306 | 1601 | |
19799a22 | 1602 | Because the transliteration table is built at compile time, neither |
a0d0e21e | 1603 | the SEARCHLIST nor the REPLACEMENTLIST are subjected to double quote |
19799a22 GS |
1604 | interpolation. That means that if you want to use variables, you |
1605 | must use an eval(): | |
a0d0e21e LW |
1606 | |
1607 | eval "tr/$oldlist/$newlist/"; | |
1608 | die $@ if $@; | |
1609 | ||
1610 | eval "tr/$oldlist/$newlist/, 1" or die $@; | |
1611 | ||
7e3b091d | 1612 | =item <<EOF |
d74e8afc | 1613 | X<here-doc> X<heredoc> X<here-document> X<<< << >>> |
7e3b091d DA |
1614 | |
1615 | A line-oriented form of quoting is based on the shell "here-document" | |
1616 | syntax. Following a C<< << >> you specify a string to terminate | |
1617 | the quoted material, and all lines following the current line down to | |
1618 | the terminating string are the value of the item. The terminating | |
1619 | string may be either an identifier (a word), or some quoted text. If | |
1620 | quoted, the type of quotes you use determines the treatment of the | |
1621 | text, just as in regular quoting. An unquoted identifier works like | |
1622 | double quotes. There must be no space between the C<< << >> and | |
1623 | the identifier, unless the identifier is quoted. (If you put a space it | |
1624 | will be treated as a null identifier, which is valid, and matches the first | |
1625 | empty line.) The terminating string must appear by itself (unquoted and | |
1626 | with no surrounding whitespace) on the terminating line. | |
1627 | ||
1628 | print <<EOF; | |
1629 | The price is $Price. | |
1630 | EOF | |
1631 | ||
1632 | print << "EOF"; # same as above | |
1633 | The price is $Price. | |
1634 | EOF | |
1635 | ||
1636 | print << `EOC`; # execute commands | |
1637 | echo hi there | |
1638 | echo lo there | |
1639 | EOC | |
1640 | ||
1641 | print <<"foo", <<"bar"; # you can stack them | |
1642 | I said foo. | |
1643 | foo | |
1644 | I said bar. | |
1645 | bar | |
1646 | ||
1647 | myfunc(<< "THIS", 23, <<'THAT'); | |
1648 | Here's a line | |
1649 | or two. | |
1650 | THIS | |
1651 | and here's another. | |
1652 | THAT | |
1653 | ||
1654 | Just don't forget that you have to put a semicolon on the end | |
1655 | to finish the statement, as Perl doesn't know you're not going to | |
1656 | try to do this: | |
1657 | ||
1658 | print <<ABC | |
1659 | 179231 | |
1660 | ABC | |
1661 | + 20; | |
1662 | ||
1663 | If you want your here-docs to be indented with the | |
1664 | rest of the code, you'll need to remove leading whitespace | |
1665 | from each line manually: | |
1666 | ||
1667 | ($quote = <<'FINIS') =~ s/^\s+//gm; | |
1668 | The Road goes ever on and on, | |
1669 | down from the door where it began. | |
1670 | FINIS | |
1671 | ||
1672 | If you use a here-doc within a delimited construct, such as in C<s///eg>, | |
1673 | the quoted material must come on the lines following the final delimiter. | |
1674 | So instead of | |
1675 | ||
1676 | s/this/<<E . 'that' | |
1677 | the other | |
1678 | E | |
1679 | . 'more '/eg; | |
1680 | ||
1681 | you have to write | |
1682 | ||
1683 | s/this/<<E . 'that' | |
1684 | . 'more '/eg; | |
1685 | the other | |
1686 | E | |
1687 | ||
1688 | If the terminating identifier is on the last line of the program, you | |
1689 | must be sure there is a newline after it; otherwise, Perl will give the | |
1690 | warning B<Can't find string terminator "END" anywhere before EOF...>. | |
1691 | ||
1692 | Additionally, the quoting rules for the identifier are not related to | |
1693 | Perl's quoting rules -- C<q()>, C<qq()>, and the like are not supported | |
1694 | in place of C<''> and C<"">, and the only interpolation is for backslashing | |
1695 | the quoting character: | |
1696 | ||
1697 | print << "abc\"def"; | |
1698 | testing... | |
1699 | abc"def | |
1700 | ||
1701 | Finally, quoted strings cannot span multiple lines. The general rule is | |
1702 | that the identifier must be a string literal. Stick with that, and you | |
1703 | should be safe. | |
1704 | ||
a0d0e21e LW |
1705 | =back |
1706 | ||
75e14d17 | 1707 | =head2 Gory details of parsing quoted constructs |
d74e8afc | 1708 | X<quote, gory details> |
75e14d17 | 1709 | |
19799a22 GS |
1710 | When presented with something that might have several different |
1711 | interpretations, Perl uses the B<DWIM> (that's "Do What I Mean") | |
1712 | principle to pick the most probable interpretation. This strategy | |
1713 | is so successful that Perl programmers often do not suspect the | |
1714 | ambivalence of what they write. But from time to time, Perl's | |
1715 | notions differ substantially from what the author honestly meant. | |
1716 | ||
1717 | This section hopes to clarify how Perl handles quoted constructs. | |
1718 | Although the most common reason to learn this is to unravel labyrinthine | |
1719 | regular expressions, because the initial steps of parsing are the | |
1720 | same for all quoting operators, they are all discussed together. | |
1721 | ||
1722 | The most important Perl parsing rule is the first one discussed | |
1723 | below: when processing a quoted construct, Perl first finds the end | |
1724 | of that construct, then interprets its contents. If you understand | |
1725 | this rule, you may skip the rest of this section on the first | |
1726 | reading. The other rules are likely to contradict the user's | |
1727 | expectations much less frequently than this first one. | |
1728 | ||
1729 | Some passes discussed below are performed concurrently, but because | |
1730 | their results are the same, we consider them individually. For different | |
1731 | quoting constructs, Perl performs different numbers of passes, from | |
1732 | one to five, but these passes are always performed in the same order. | |
75e14d17 | 1733 | |
13a2d996 | 1734 | =over 4 |
75e14d17 IZ |
1735 | |
1736 | =item Finding the end | |
1737 | ||
19799a22 GS |
1738 | The first pass is finding the end of the quoted construct, whether |
1739 | it be a multicharacter delimiter C<"\nEOF\n"> in the C<<<EOF> | |
1740 | construct, a C</> that terminates a C<qq//> construct, a C<]> which | |
35f2feb0 GS |
1741 | terminates C<qq[]> construct, or a C<< > >> which terminates a |
1742 | fileglob started with C<< < >>. | |
75e14d17 | 1743 | |
19799a22 GS |
1744 | When searching for single-character non-pairing delimiters, such |
1745 | as C</>, combinations of C<\\> and C<\/> are skipped. However, | |
1746 | when searching for single-character pairing delimiter like C<[>, | |
1747 | combinations of C<\\>, C<\]>, and C<\[> are all skipped, and nested | |
1748 | C<[>, C<]> are skipped as well. When searching for multicharacter | |
1749 | delimiters, nothing is skipped. | |
75e14d17 | 1750 | |
19799a22 GS |
1751 | For constructs with three-part delimiters (C<s///>, C<y///>, and |
1752 | C<tr///>), the search is repeated once more. | |
75e14d17 | 1753 | |
19799a22 GS |
1754 | During this search no attention is paid to the semantics of the construct. |
1755 | Thus: | |
75e14d17 IZ |
1756 | |
1757 | "$hash{"$foo/$bar"}" | |
1758 | ||
2a94b7ce | 1759 | or: |
75e14d17 IZ |
1760 | |
1761 | m/ | |
2a94b7ce | 1762 | bar # NOT a comment, this slash / terminated m//! |
75e14d17 IZ |
1763 | /x |
1764 | ||
19799a22 GS |
1765 | do not form legal quoted expressions. The quoted part ends on the |
1766 | first C<"> and C</>, and the rest happens to be a syntax error. | |
1767 | Because the slash that terminated C<m//> was followed by a C<SPACE>, | |
1768 | the example above is not C<m//x>, but rather C<m//> with no C</x> | |
1769 | modifier. So the embedded C<#> is interpreted as a literal C<#>. | |
75e14d17 | 1770 | |
0d594e51 TS |
1771 | Also no attention is paid to C<\c\> during this search. |
1772 | Thus the second C<\> in C<qq/\c\/> is interpreted as a part of C<\/>, | |
1773 | and the following C</> is not recognized as a delimiter. | |
1774 | Instead, use C<\034> or C<\x1c> at the end of quoted constructs. | |
1775 | ||
75e14d17 IZ |
1776 | =item Removal of backslashes before delimiters |
1777 | ||
19799a22 GS |
1778 | During the second pass, text between the starting and ending |
1779 | delimiters is copied to a safe location, and the C<\> is removed | |
1780 | from combinations consisting of C<\> and delimiter--or delimiters, | |
1781 | meaning both starting and ending delimiters will should these differ. | |
1782 | This removal does not happen for multi-character delimiters. | |
1783 | Note that the combination C<\\> is left intact, just as it was. | |
75e14d17 | 1784 | |
19799a22 GS |
1785 | Starting from this step no information about the delimiters is |
1786 | used in parsing. | |
75e14d17 IZ |
1787 | |
1788 | =item Interpolation | |
d74e8afc | 1789 | X<interpolation> |
75e14d17 | 1790 | |
19799a22 GS |
1791 | The next step is interpolation in the text obtained, which is now |
1792 | delimiter-independent. There are four different cases. | |
75e14d17 | 1793 | |
13a2d996 | 1794 | =over 4 |
75e14d17 IZ |
1795 | |
1796 | =item C<<<'EOF'>, C<m''>, C<s'''>, C<tr///>, C<y///> | |
1797 | ||
1798 | No interpolation is performed. | |
1799 | ||
1800 | =item C<''>, C<q//> | |
1801 | ||
1802 | The only interpolation is removal of C<\> from pairs C<\\>. | |
1803 | ||
35f2feb0 | 1804 | =item C<"">, C<``>, C<qq//>, C<qx//>, C<< <file*glob> >> |
75e14d17 | 1805 | |
19799a22 GS |
1806 | C<\Q>, C<\U>, C<\u>, C<\L>, C<\l> (possibly paired with C<\E>) are |
1807 | converted to corresponding Perl constructs. Thus, C<"$foo\Qbaz$bar"> | |
1808 | is converted to C<$foo . (quotemeta("baz" . $bar))> internally. | |
1809 | The other combinations are replaced with appropriate expansions. | |
2a94b7ce | 1810 | |
19799a22 GS |
1811 | Let it be stressed that I<whatever falls between C<\Q> and C<\E>> |
1812 | is interpolated in the usual way. Something like C<"\Q\\E"> has | |
1813 | no C<\E> inside. instead, it has C<\Q>, C<\\>, and C<E>, so the | |
1814 | result is the same as for C<"\\\\E">. As a general rule, backslashes | |
1815 | between C<\Q> and C<\E> may lead to counterintuitive results. So, | |
1816 | C<"\Q\t\E"> is converted to C<quotemeta("\t")>, which is the same | |
1817 | as C<"\\\t"> (since TAB is not alphanumeric). Note also that: | |
2a94b7ce IZ |
1818 | |
1819 | $str = '\t'; | |
1820 | return "\Q$str"; | |
1821 | ||
1822 | may be closer to the conjectural I<intention> of the writer of C<"\Q\t\E">. | |
1823 | ||
19799a22 | 1824 | Interpolated scalars and arrays are converted internally to the C<join> and |
92d29cee | 1825 | C<.> catenation operations. Thus, C<"$foo XXX '@arr'"> becomes: |
75e14d17 | 1826 | |
19799a22 | 1827 | $foo . " XXX '" . (join $", @arr) . "'"; |
75e14d17 | 1828 | |
19799a22 | 1829 | All operations above are performed simultaneously, left to right. |
75e14d17 | 1830 | |
19799a22 GS |
1831 | Because the result of C<"\Q STRING \E"> has all metacharacters |
1832 | quoted, there is no way to insert a literal C<$> or C<@> inside a | |
1833 | C<\Q\E> pair. If protected by C<\>, C<$> will be quoted to became | |
1834 | C<"\\\$">; if not, it is interpreted as the start of an interpolated | |
1835 | scalar. | |
75e14d17 | 1836 | |
19799a22 GS |
1837 | Note also that the interpolation code needs to make a decision on |
1838 | where the interpolated scalar ends. For instance, whether | |
35f2feb0 | 1839 | C<< "a $b -> {c}" >> really means: |
75e14d17 IZ |
1840 | |
1841 | "a " . $b . " -> {c}"; | |
1842 | ||
2a94b7ce | 1843 | or: |
75e14d17 IZ |
1844 | |
1845 | "a " . $b -> {c}; | |
1846 | ||
19799a22 GS |
1847 | Most of the time, the longest possible text that does not include |
1848 | spaces between components and which contains matching braces or | |
1849 | brackets. because the outcome may be determined by voting based | |
1850 | on heuristic estimators, the result is not strictly predictable. | |
1851 | Fortunately, it's usually correct for ambiguous cases. | |
75e14d17 IZ |
1852 | |
1853 | =item C<?RE?>, C</RE/>, C<m/RE/>, C<s/RE/foo/>, | |
1854 | ||
19799a22 GS |
1855 | Processing of C<\Q>, C<\U>, C<\u>, C<\L>, C<\l>, and interpolation |
1856 | happens (almost) as with C<qq//> constructs, but the substitution | |
1857 | of C<\> followed by RE-special chars (including C<\>) is not | |
1858 | performed. Moreover, inside C<(?{BLOCK})>, C<(?# comment )>, and | |
1859 | a C<#>-comment in a C<//x>-regular expression, no processing is | |
1860 | performed whatsoever. This is the first step at which the presence | |
1861 | of the C<//x> modifier is relevant. | |
1862 | ||
1863 | Interpolation has several quirks: C<$|>, C<$(>, and C<$)> are not | |
1864 | interpolated, and constructs C<$var[SOMETHING]> are voted (by several | |
1865 | different estimators) to be either an array element or C<$var> | |
1866 | followed by an RE alternative. This is where the notation | |
1867 | C<${arr[$bar]}> comes handy: C</${arr[0-9]}/> is interpreted as | |
1868 | array element C<-9>, not as a regular expression from the variable | |
1869 | C<$arr> followed by a digit, which would be the interpretation of | |
1870 | C</$arr[0-9]/>. Since voting among different estimators may occur, | |
1871 | the result is not predictable. | |
1872 | ||
1873 | It is at this step that C<\1> is begrudgingly converted to C<$1> in | |
1874 | the replacement text of C<s///> to correct the incorrigible | |
1875 | I<sed> hackers who haven't picked up the saner idiom yet. A warning | |
9f1b1f2d GS |
1876 | is emitted if the C<use warnings> pragma or the B<-w> command-line flag |
1877 | (that is, the C<$^W> variable) was set. | |
19799a22 GS |
1878 | |
1879 | The lack of processing of C<\\> creates specific restrictions on | |
1880 | the post-processed text. If the delimiter is C</>, one cannot get | |
1881 | the combination C<\/> into the result of this step. C</> will | |
1882 | finish the regular expression, C<\/> will be stripped to C</> on | |
1883 | the previous step, and C<\\/> will be left as is. Because C</> is | |
1884 | equivalent to C<\/> inside a regular expression, this does not | |
1885 | matter unless the delimiter happens to be character special to the | |
1886 | RE engine, such as in C<s*foo*bar*>, C<m[foo]>, or C<?foo?>; or an | |
1887 | alphanumeric char, as in: | |
2a94b7ce IZ |
1888 | |
1889 | m m ^ a \s* b mmx; | |
1890 | ||
19799a22 | 1891 | In the RE above, which is intentionally obfuscated for illustration, the |
2a94b7ce | 1892 | delimiter is C<m>, the modifier is C<mx>, and after backslash-removal the |
aa863641 | 1893 | RE is the same as for C<m/ ^ a \s* b /mx>. There's more than one |
19799a22 GS |
1894 | reason you're encouraged to restrict your delimiters to non-alphanumeric, |
1895 | non-whitespace choices. | |
75e14d17 IZ |
1896 | |
1897 | =back | |
1898 | ||
19799a22 | 1899 | This step is the last one for all constructs except regular expressions, |
75e14d17 IZ |
1900 | which are processed further. |
1901 | ||
1902 | =item Interpolation of regular expressions | |
d74e8afc | 1903 | X<regexp, interpolation> |
75e14d17 | 1904 | |
19799a22 GS |
1905 | Previous steps were performed during the compilation of Perl code, |
1906 | but this one happens at run time--although it may be optimized to | |
1907 | be calculated at compile time if appropriate. After preprocessing | |
1908 | described above, and possibly after evaluation if catenation, | |
1909 | joining, casing translation, or metaquoting are involved, the | |
1910 | resulting I<string> is passed to the RE engine for compilation. | |
1911 | ||
1912 | Whatever happens in the RE engine might be better discussed in L<perlre>, | |
1913 | but for the sake of continuity, we shall do so here. | |
1914 | ||
1915 | This is another step where the presence of the C<//x> modifier is | |
1916 | relevant. The RE engine scans the string from left to right and | |
1917 | converts it to a finite automaton. | |
1918 | ||
1919 | Backslashed characters are either replaced with corresponding | |
1920 | literal strings (as with C<\{>), or else they generate special nodes | |
1921 | in the finite automaton (as with C<\b>). Characters special to the | |
1922 | RE engine (such as C<|>) generate corresponding nodes or groups of | |
1923 | nodes. C<(?#...)> comments are ignored. All the rest is either | |
1924 | converted to literal strings to match, or else is ignored (as is | |
1925 | whitespace and C<#>-style comments if C<//x> is present). | |
1926 | ||
1927 | Parsing of the bracketed character class construct, C<[...]>, is | |
1928 | rather different than the rule used for the rest of the pattern. | |
1929 | The terminator of this construct is found using the same rules as | |
1930 | for finding the terminator of a C<{}>-delimited construct, the only | |
1931 | exception being that C<]> immediately following C<[> is treated as | |
1932 | though preceded by a backslash. Similarly, the terminator of | |
1933 | C<(?{...})> is found using the same rules as for finding the | |
1934 | terminator of a C<{}>-delimited construct. | |
1935 | ||
1936 | It is possible to inspect both the string given to RE engine and the | |
1937 | resulting finite automaton. See the arguments C<debug>/C<debugcolor> | |
1938 | in the C<use L<re>> pragma, as well as Perl's B<-Dr> command-line | |
4a4eefd0 | 1939 | switch documented in L<perlrun/"Command Switches">. |
75e14d17 IZ |
1940 | |
1941 | =item Optimization of regular expressions | |
d74e8afc | 1942 | X<regexp, optimization> |
75e14d17 | 1943 | |
7522fed5 | 1944 | This step is listed for completeness only. Since it does not change |
75e14d17 | 1945 | semantics, details of this step are not documented and are subject |
19799a22 GS |
1946 | to change without notice. This step is performed over the finite |
1947 | automaton that was generated during the previous pass. | |
2a94b7ce | 1948 | |
19799a22 GS |
1949 | It is at this stage that C<split()> silently optimizes C</^/> to |
1950 | mean C</^/m>. | |
75e14d17 IZ |
1951 | |
1952 | =back | |
1953 | ||
a0d0e21e | 1954 | =head2 I/O Operators |
d74e8afc ITB |
1955 | X<operator, i/o> X<operator, io> X<io> X<while> X<filehandle> |
1956 | X<< <> >> X<@ARGV> | |
a0d0e21e | 1957 | |
54310121 | 1958 | There are several I/O operators you should know about. |
fbad3eb5 | 1959 | |
7b8d334a | 1960 | A string enclosed by backticks (grave accents) first undergoes |
19799a22 GS |
1961 | double-quote interpolation. It is then interpreted as an external |
1962 | command, and the output of that command is the value of the | |
e9c56f9b JH |
1963 | backtick string, like in a shell. In scalar context, a single string |
1964 | consisting of all output is returned. In list context, a list of | |
1965 | values is returned, one per line of output. (You can set C<$/> to use | |
1966 | a different line terminator.) The command is executed each time the | |
1967 | pseudo-literal is evaluated. The status value of the command is | |
1968 | returned in C<$?> (see L<perlvar> for the interpretation of C<$?>). | |
1969 | Unlike in B<csh>, no translation is done on the return data--newlines | |
1970 | remain newlines. Unlike in any of the shells, single quotes do not | |
1971 | hide variable names in the command from interpretation. To pass a | |
1972 | literal dollar-sign through to the shell you need to hide it with a | |
1973 | backslash. The generalized form of backticks is C<qx//>. (Because | |
1974 | backticks always undergo shell expansion as well, see L<perlsec> for | |
1975 | security concerns.) | |
d74e8afc | 1976 | X<qx> X<`> X<``> X<backtick> X<glob> |
19799a22 GS |
1977 | |
1978 | In scalar context, evaluating a filehandle in angle brackets yields | |
1979 | the next line from that file (the newline, if any, included), or | |
1980 | C<undef> at end-of-file or on error. When C<$/> is set to C<undef> | |
1981 | (sometimes known as file-slurp mode) and the file is empty, it | |
1982 | returns C<''> the first time, followed by C<undef> subsequently. | |
1983 | ||
1984 | Ordinarily you must assign the returned value to a variable, but | |
1985 | there is one situation where an automatic assignment happens. If | |
1986 | and only if the input symbol is the only thing inside the conditional | |
1987 | of a C<while> statement (even if disguised as a C<for(;;)> loop), | |
1988 | the value is automatically assigned to the global variable $_, | |
1989 | destroying whatever was there previously. (This may seem like an | |
1990 | odd thing to you, but you'll use the construct in almost every Perl | |
17b829fa | 1991 | script you write.) The $_ variable is not implicitly localized. |
19799a22 GS |
1992 | You'll have to put a C<local $_;> before the loop if you want that |
1993 | to happen. | |
1994 | ||
1995 | The following lines are equivalent: | |
a0d0e21e | 1996 | |
748a9306 | 1997 | while (defined($_ = <STDIN>)) { print; } |
7b8d334a | 1998 | while ($_ = <STDIN>) { print; } |
a0d0e21e LW |
1999 | while (<STDIN>) { print; } |
2000 | for (;<STDIN>;) { print; } | |
748a9306 | 2001 | print while defined($_ = <STDIN>); |
7b8d334a | 2002 | print while ($_ = <STDIN>); |
a0d0e21e LW |
2003 | print while <STDIN>; |
2004 | ||
19799a22 | 2005 | This also behaves similarly, but avoids $_ : |
7b8d334a GS |
2006 | |
2007 | while (my $line = <STDIN>) { print $line } | |
2008 | ||
19799a22 GS |
2009 | In these loop constructs, the assigned value (whether assignment |
2010 | is automatic or explicit) is then tested to see whether it is | |
2011 | defined. The defined test avoids problems where line has a string | |
2012 | value that would be treated as false by Perl, for example a "" or | |
2013 | a "0" with no trailing newline. If you really mean for such values | |
2014 | to terminate the loop, they should be tested for explicitly: | |
7b8d334a GS |
2015 | |
2016 | while (($_ = <STDIN>) ne '0') { ... } | |
2017 | while (<STDIN>) { last unless $_; ... } | |
2018 | ||
35f2feb0 | 2019 | In other boolean contexts, C<< <I<filehandle>> >> without an |
9f1b1f2d GS |
2020 | explicit C<defined> test or comparison elicit a warning if the |
2021 | C<use warnings> pragma or the B<-w> | |
19799a22 | 2022 | command-line switch (the C<$^W> variable) is in effect. |
7b8d334a | 2023 | |
5f05dabc | 2024 | The filehandles STDIN, STDOUT, and STDERR are predefined. (The |
19799a22 GS |
2025 | filehandles C<stdin>, C<stdout>, and C<stderr> will also work except |
2026 | in packages, where they would be interpreted as local identifiers | |
2027 | rather than global.) Additional filehandles may be created with | |
2028 | the open() function, amongst others. See L<perlopentut> and | |
2029 | L<perlfunc/open> for details on this. | |
d74e8afc | 2030 | X<stdin> X<stdout> X<sterr> |
a0d0e21e | 2031 | |
35f2feb0 | 2032 | If a <FILEHANDLE> is used in a context that is looking for |
19799a22 GS |
2033 | a list, a list comprising all input lines is returned, one line per |
2034 | list element. It's easy to grow to a rather large data space this | |
2035 | way, so use with care. | |
a0d0e21e | 2036 | |
35f2feb0 | 2037 | <FILEHANDLE> may also be spelled C<readline(*FILEHANDLE)>. |
19799a22 | 2038 | See L<perlfunc/readline>. |
fbad3eb5 | 2039 | |
35f2feb0 GS |
2040 | The null filehandle <> is special: it can be used to emulate the |
2041 | behavior of B<sed> and B<awk>. Input from <> comes either from | |
a0d0e21e | 2042 | standard input, or from each file listed on the command line. Here's |
35f2feb0 | 2043 | how it works: the first time <> is evaluated, the @ARGV array is |
5a964f20 | 2044 | checked, and if it is empty, C<$ARGV[0]> is set to "-", which when opened |
a0d0e21e LW |
2045 | gives you standard input. The @ARGV array is then processed as a list |
2046 | of filenames. The loop | |
2047 | ||
2048 | while (<>) { | |
2049 | ... # code for each line | |
2050 | } | |
2051 | ||
2052 | is equivalent to the following Perl-like pseudo code: | |
2053 | ||
3e3baf6d | 2054 | unshift(@ARGV, '-') unless @ARGV; |
a0d0e21e LW |
2055 | while ($ARGV = shift) { |
2056 | open(ARGV, $ARGV); | |
2057 | while (<ARGV>) { | |
2058 | ... # code for each line | |
2059 | } | |
2060 | } | |
2061 | ||
19799a22 GS |
2062 | except that it isn't so cumbersome to say, and will actually work. |
2063 | It really does shift the @ARGV array and put the current filename | |
2064 | into the $ARGV variable. It also uses filehandle I<ARGV> | |
35f2feb0 | 2065 | internally--<> is just a synonym for <ARGV>, which |
19799a22 | 2066 | is magical. (The pseudo code above doesn't work because it treats |
35f2feb0 | 2067 | <ARGV> as non-magical.) |
a0d0e21e | 2068 | |
35f2feb0 | 2069 | You can modify @ARGV before the first <> as long as the array ends up |
a0d0e21e | 2070 | containing the list of filenames you really want. Line numbers (C<$.>) |
19799a22 GS |
2071 | continue as though the input were one big happy file. See the example |
2072 | in L<perlfunc/eof> for how to reset line numbers on each file. | |
5a964f20 TC |
2073 | |
2074 | If you want to set @ARGV to your own list of files, go right ahead. | |
2075 | This sets @ARGV to all plain text files if no @ARGV was given: | |
2076 | ||
2077 | @ARGV = grep { -f && -T } glob('*') unless @ARGV; | |
a0d0e21e | 2078 | |
5a964f20 TC |
2079 | You can even set them to pipe commands. For example, this automatically |
2080 | filters compressed arguments through B<gzip>: | |
2081 | ||
2082 | @ARGV = map { /\.(gz|Z)$/ ? "gzip -dc < $_ |" : $_ } @ARGV; | |
2083 | ||
2084 | If you want to pass switches into your script, you can use one of the | |
a0d0e21e LW |
2085 | Getopts modules or put a loop on the front like this: |
2086 | ||
2087 | while ($_ = $ARGV[0], /^-/) { | |
2088 | shift; | |
2089 | last if /^--$/; | |
2090 | if (/^-D(.*)/) { $debug = $1 } | |
2091 | if (/^-v/) { $verbose++ } | |
5a964f20 | 2092 | # ... # other switches |
a0d0e21e | 2093 | } |
5a964f20 | 2094 | |
a0d0e21e | 2095 | while (<>) { |
5a964f20 | 2096 | # ... # code for each line |
a0d0e21e LW |
2097 | } |
2098 | ||
35f2feb0 | 2099 | The <> symbol will return C<undef> for end-of-file only once. |
19799a22 GS |
2100 | If you call it again after this, it will assume you are processing another |
2101 | @ARGV list, and if you haven't set @ARGV, will read input from STDIN. | |
a0d0e21e | 2102 | |
b159ebd3 | 2103 | If what the angle brackets contain is a simple scalar variable (e.g., |
35f2feb0 | 2104 | <$foo>), then that variable contains the name of the |
19799a22 GS |
2105 | filehandle to input from, or its typeglob, or a reference to the |
2106 | same. For example: | |
cb1a09d0 AD |
2107 | |
2108 | $fh = \*STDIN; | |
2109 | $line = <$fh>; | |
a0d0e21e | 2110 | |
5a964f20 TC |
2111 | If what's within the angle brackets is neither a filehandle nor a simple |
2112 | scalar variable containing a filehandle name, typeglob, or typeglob | |
2113 | reference, it is interpreted as a filename pattern to be globbed, and | |
2114 | either a list of filenames or the next filename in the list is returned, | |
19799a22 | 2115 | depending on context. This distinction is determined on syntactic |
35f2feb0 GS |
2116 | grounds alone. That means C<< <$x> >> is always a readline() from |
2117 | an indirect handle, but C<< <$hash{key}> >> is always a glob(). | |
5a964f20 | 2118 | That's because $x is a simple scalar variable, but C<$hash{key}> is |
ef191992 YST |
2119 | not--it's a hash element. Even C<< <$x > >> (note the extra space) |
2120 | is treated as C<glob("$x ")>, not C<readline($x)>. | |
5a964f20 TC |
2121 | |
2122 | One level of double-quote interpretation is done first, but you can't | |
35f2feb0 | 2123 | say C<< <$foo> >> because that's an indirect filehandle as explained |
5a964f20 TC |
2124 | in the previous paragraph. (In older versions of Perl, programmers |
2125 | would insert curly brackets to force interpretation as a filename glob: | |
35f2feb0 | 2126 | C<< <${foo}> >>. These days, it's considered cleaner to call the |
5a964f20 | 2127 | internal function directly as C<glob($foo)>, which is probably the right |
19799a22 | 2128 | way to have done it in the first place.) For example: |
a0d0e21e LW |
2129 | |
2130 | while (<*.c>) { | |
2131 | chmod 0644, $_; | |
2132 | } | |
2133 | ||
3a4b19e4 | 2134 | is roughly equivalent to: |
a0d0e21e LW |
2135 | |
2136 | open(FOO, "echo *.c | tr -s ' \t\r\f' '\\012\\012\\012\\012'|"); | |
2137 | while (<FOO>) { | |
5b3eff12 | 2138 | chomp; |
a0d0e21e LW |
2139 | chmod 0644, $_; |
2140 | } | |
2141 | ||
3a4b19e4 GS |
2142 | except that the globbing is actually done internally using the standard |
2143 | C<File::Glob> extension. Of course, the shortest way to do the above is: | |
a0d0e21e LW |
2144 | |
2145 | chmod 0644, <*.c>; | |
2146 | ||
19799a22 GS |
2147 | A (file)glob evaluates its (embedded) argument only when it is |
2148 | starting a new list. All values must be read before it will start | |
2149 | over. In list context, this isn't important because you automatically | |
2150 | get them all anyway. However, in scalar context the operator returns | |
069e01df | 2151 | the next value each time it's called, or C<undef> when the list has |
19799a22 GS |
2152 | run out. As with filehandle reads, an automatic C<defined> is |
2153 | generated when the glob occurs in the test part of a C<while>, | |
2154 | because legal glob returns (e.g. a file called F<0>) would otherwise | |
2155 | terminate the loop. Again, C<undef> is returned only once. So if | |
2156 | you're expecting a single value from a glob, it is much better to | |
2157 | say | |
4633a7c4 LW |
2158 | |
2159 | ($file) = <blurch*>; | |
2160 | ||
2161 | than | |
2162 | ||
2163 | $file = <blurch*>; | |
2164 | ||
2165 | because the latter will alternate between returning a filename and | |
19799a22 | 2166 | returning false. |
4633a7c4 | 2167 | |
b159ebd3 | 2168 | If you're trying to do variable interpolation, it's definitely better |
4633a7c4 | 2169 | to use the glob() function, because the older notation can cause people |
e37d713d | 2170 | to become confused with the indirect filehandle notation. |
4633a7c4 LW |
2171 | |
2172 | @files = glob("$dir/*.[ch]"); | |
2173 | @files = glob($files[$i]); | |
2174 | ||
a0d0e21e | 2175 | =head2 Constant Folding |
d74e8afc | 2176 | X<constant folding> X<folding> |
a0d0e21e LW |
2177 | |
2178 | Like C, Perl does a certain amount of expression evaluation at | |
19799a22 | 2179 | compile time whenever it determines that all arguments to an |
a0d0e21e LW |
2180 | operator are static and have no side effects. In particular, string |
2181 | concatenation happens at compile time between literals that don't do | |
19799a22 | 2182 | variable substitution. Backslash interpolation also happens at |
a0d0e21e LW |
2183 | compile time. You can say |
2184 | ||
2185 | 'Now is the time for all' . "\n" . | |
2186 | 'good men to come to.' | |
2187 | ||
54310121 | 2188 | and this all reduces to one string internally. Likewise, if |
a0d0e21e LW |
2189 | you say |
2190 | ||
2191 | foreach $file (@filenames) { | |
5a964f20 | 2192 | if (-s $file > 5 + 100 * 2**16) { } |
54310121 | 2193 | } |
a0d0e21e | 2194 | |
19799a22 GS |
2195 | the compiler will precompute the number which that expression |
2196 | represents so that the interpreter won't have to. | |
a0d0e21e | 2197 | |
fd1abbef | 2198 | =head2 No-ops |
d74e8afc | 2199 | X<no-op> X<nop> |
fd1abbef DN |
2200 | |
2201 | Perl doesn't officially have a no-op operator, but the bare constants | |
2202 | C<0> and C<1> are special-cased to not produce a warning in a void | |
2203 | context, so you can for example safely do | |
2204 | ||
2205 | 1 while foo(); | |
2206 | ||
2c268ad5 | 2207 | =head2 Bitwise String Operators |
d74e8afc | 2208 | X<operator, bitwise, string> |
2c268ad5 TP |
2209 | |
2210 | Bitstrings of any size may be manipulated by the bitwise operators | |
2211 | (C<~ | & ^>). | |
2212 | ||
19799a22 GS |
2213 | If the operands to a binary bitwise op are strings of different |
2214 | sizes, B<|> and B<^> ops act as though the shorter operand had | |
2215 | additional zero bits on the right, while the B<&> op acts as though | |
2216 | the longer operand were truncated to the length of the shorter. | |
2217 | The granularity for such extension or truncation is one or more | |
2218 | bytes. | |
2c268ad5 TP |
2219 | |
2220 | # ASCII-based examples | |
2221 | print "j p \n" ^ " a h"; # prints "JAPH\n" | |
2222 | print "JA" | " ph\n"; # prints "japh\n" | |
2223 | print "japh\nJunk" & '_____'; # prints "JAPH\n"; | |
2224 | print 'p N$' ^ " E<H\n"; # prints "Perl\n"; | |
2225 | ||
19799a22 | 2226 | If you are intending to manipulate bitstrings, be certain that |
2c268ad5 | 2227 | you're supplying bitstrings: If an operand is a number, that will imply |
19799a22 | 2228 | a B<numeric> bitwise operation. You may explicitly show which type of |
2c268ad5 TP |
2229 | operation you intend by using C<""> or C<0+>, as in the examples below. |
2230 | ||
4358a253 SS |
2231 | $foo = 150 | 105; # yields 255 (0x96 | 0x69 is 0xFF) |
2232 | $foo = '150' | 105; # yields 255 | |
2c268ad5 TP |
2233 | $foo = 150 | '105'; # yields 255 |
2234 | $foo = '150' | '105'; # yields string '155' (under ASCII) | |
2235 | ||
2236 | $baz = 0+$foo & 0+$bar; # both ops explicitly numeric | |
2237 | $biz = "$foo" ^ "$bar"; # both ops explicitly stringy | |
a0d0e21e | 2238 | |
1ae175c8 GS |
2239 | See L<perlfunc/vec> for information on how to manipulate individual bits |
2240 | in a bit vector. | |
2241 | ||
55497cff | 2242 | =head2 Integer Arithmetic |
d74e8afc | 2243 | X<integer> |
a0d0e21e | 2244 | |
19799a22 | 2245 | By default, Perl assumes that it must do most of its arithmetic in |
a0d0e21e LW |
2246 | floating point. But by saying |
2247 | ||
2248 | use integer; | |
2249 | ||
2250 | you may tell the compiler that it's okay to use integer operations | |
19799a22 GS |
2251 | (if it feels like it) from here to the end of the enclosing BLOCK. |
2252 | An inner BLOCK may countermand this by saying | |
a0d0e21e LW |
2253 | |
2254 | no integer; | |
2255 | ||
19799a22 GS |
2256 | which lasts until the end of that BLOCK. Note that this doesn't |
2257 | mean everything is only an integer, merely that Perl may use integer | |
2258 | operations if it is so inclined. For example, even under C<use | |
2259 | integer>, if you take the C<sqrt(2)>, you'll still get C<1.4142135623731> | |
2260 | or so. | |
2261 | ||
2262 | Used on numbers, the bitwise operators ("&", "|", "^", "~", "<<", | |
13a2d996 SP |
2263 | and ">>") always produce integral results. (But see also |
2264 | L<Bitwise String Operators>.) However, C<use integer> still has meaning for | |
19799a22 GS |
2265 | them. By default, their results are interpreted as unsigned integers, but |
2266 | if C<use integer> is in effect, their results are interpreted | |
2267 | as signed integers. For example, C<~0> usually evaluates to a large | |
0be96356 | 2268 | integral value. However, C<use integer; ~0> is C<-1> on two's-complement |
19799a22 | 2269 | machines. |
68dc0745 | 2270 | |
2271 | =head2 Floating-point Arithmetic | |
d74e8afc | 2272 | X<floating-point> X<floating point> X<float> X<real> |
68dc0745 | 2273 | |
2274 | While C<use integer> provides integer-only arithmetic, there is no | |
19799a22 GS |
2275 | analogous mechanism to provide automatic rounding or truncation to a |
2276 | certain number of decimal places. For rounding to a certain number | |
2277 | of digits, sprintf() or printf() is usually the easiest route. | |
2278 | See L<perlfaq4>. | |
68dc0745 | 2279 | |
5a964f20 TC |
2280 | Floating-point numbers are only approximations to what a mathematician |
2281 | would call real numbers. There are infinitely more reals than floats, | |
2282 | so some corners must be cut. For example: | |
2283 | ||
2284 | printf "%.20g\n", 123456789123456789; | |
2285 | # produces 123456789123456784 | |
2286 | ||
2287 | Testing for exact equality of floating-point equality or inequality is | |
2288 | not a good idea. Here's a (relatively expensive) work-around to compare | |
2289 | whether two floating-point numbers are equal to a particular number of | |
2290 | decimal places. See Knuth, volume II, for a more robust treatment of | |
2291 | this topic. | |
2292 | ||
2293 | sub fp_equal { | |
2294 | my ($X, $Y, $POINTS) = @_; | |
2295 | my ($tX, $tY); | |
2296 | $tX = sprintf("%.${POINTS}g", $X); | |
2297 | $tY = sprintf("%.${POINTS}g", $Y); | |
2298 | return $tX eq $tY; | |
2299 | } | |
2300 | ||
68dc0745 | 2301 | The POSIX module (part of the standard perl distribution) implements |
19799a22 GS |
2302 | ceil(), floor(), and other mathematical and trigonometric functions. |
2303 | The Math::Complex module (part of the standard perl distribution) | |
2304 | defines mathematical functions that work on both the reals and the | |
2305 | imaginary numbers. Math::Complex not as efficient as POSIX, but | |
68dc0745 | 2306 | POSIX can't work with complex numbers. |
2307 | ||
2308 | Rounding in financial applications can have serious implications, and | |
2309 | the rounding method used should be specified precisely. In these | |
2310 | cases, it probably pays not to trust whichever system rounding is | |
2311 | being used by Perl, but to instead implement the rounding function you | |
2312 | need yourself. | |
5a964f20 TC |
2313 | |
2314 | =head2 Bigger Numbers | |
d74e8afc | 2315 | X<number, arbitrary precision> |
5a964f20 TC |
2316 | |
2317 | The standard Math::BigInt and Math::BigFloat modules provide | |
19799a22 | 2318 | variable-precision arithmetic and overloaded operators, although |
cd5c4fce | 2319 | they're currently pretty slow. At the cost of some space and |
19799a22 GS |
2320 | considerable speed, they avoid the normal pitfalls associated with |
2321 | limited-precision representations. | |
5a964f20 TC |
2322 | |
2323 | use Math::BigInt; | |
2324 | $x = Math::BigInt->new('123456789123456789'); | |
2325 | print $x * $x; | |
2326 | ||
2327 | # prints +15241578780673678515622620750190521 | |
19799a22 | 2328 | |
cd5c4fce T |
2329 | There are several modules that let you calculate with (bound only by |
2330 | memory and cpu-time) unlimited or fixed precision. There are also | |
2331 | some non-standard modules that provide faster implementations via | |
2332 | external C libraries. | |
2333 | ||
2334 | Here is a short, but incomplete summary: | |
2335 | ||
2336 | Math::Fraction big, unlimited fractions like 9973 / 12967 | |
2337 | Math::String treat string sequences like numbers | |
2338 | Math::FixedPrecision calculate with a fixed precision | |
2339 | Math::Currency for currency calculations | |
2340 | Bit::Vector manipulate bit vectors fast (uses C) | |
2341 | Math::BigIntFast Bit::Vector wrapper for big numbers | |
2342 | Math::Pari provides access to the Pari C library | |
2343 | Math::BigInteger uses an external C library | |
2344 | Math::Cephes uses external Cephes C library (no big numbers) | |
2345 | Math::Cephes::Fraction fractions via the Cephes library | |
2346 | Math::GMP another one using an external C library | |
2347 | ||
2348 | Choose wisely. | |
16070b82 GS |
2349 | |
2350 | =cut |