This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Edits to perlrecharclass.pod
[perl5.git] / pod / perlop.pod
CommitLineData
a0d0e21e 1=head1 NAME
d74e8afc 2X<operator>
a0d0e21e
LW
3
4perlop - Perl operators and precedence
5
d042e63d
MS
6=head1 DESCRIPTION
7
89d205f2 8=head2 Operator Precedence and Associativity
d74e8afc 9X<operator, precedence> X<precedence> X<associativity>
d042e63d
MS
10
11Operator precedence and associativity work in Perl more or less like
12they do in mathematics.
13
14I<Operator precedence> means some operators are evaluated before
15others. For example, in C<2 + 4 * 5>, the multiplication has higher
16precedence so C<4 * 5> is evaluated first yielding C<2 + 20 ==
1722> and not C<6 * 5 == 30>.
18
19I<Operator associativity> defines what happens if a sequence of the
20same operators is used one after another: whether the evaluator will
21evaluate the left operations first or the right. For example, in C<8
22- 4 - 2>, subtraction is left associative so Perl evaluates the
23expression left to right. C<8 - 4> is evaluated first making the
24expression C<4 - 2 == 2> and not C<8 - 2 == 6>.
a0d0e21e
LW
25
26Perl operators have the following associativity and precedence,
19799a22
GS
27listed from highest precedence to lowest. Operators borrowed from
28C keep the same precedence relationship with each other, even where
29C's precedence is slightly screwy. (This makes learning Perl easier
30for C folks.) With very few exceptions, these all operate on scalar
31values 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
f23102e2 56 left or xor
a0d0e21e
LW
57
58In the following sections, these operators are covered in precedence order.
59
5a964f20
TC
60Many operators can be overloaded for objects. See L<overload>.
61
a0d0e21e 62=head2 Terms and List Operators (Leftward)
d74e8afc 63X<list operator> X<operator, list> X<term>
a0d0e21e 64
62c18ce2 65A TERM has the highest precedence in Perl. They include variables,
5f05dabc 66quote and quote-like operators, any expression in parentheses,
a0d0e21e
LW
67and any function whose arguments are parenthesized. Actually, there
68aren't really functions in this sense, just list operators and unary
69operators behaving as functions because you put parentheses around
70the arguments. These are all documented in L<perlfunc>.
71
72If any list operator (print(), etc.) or any unary operator (chdir(), etc.)
73is followed by a left parenthesis as the next token, the operator and
74arguments within parentheses are taken to be of highest precedence,
75just like a normal function call.
76
77In the absence of parentheses, the precedence of list operators such as
78C<print>, C<sort>, or C<chmod> is either very high or very low depending on
54310121 79whether you are looking at the left side or the right side of the operator.
a0d0e21e
LW
80For example, in
81
82 @ary = (1, 3, sort 4, 2);
83 print @ary; # prints 1324
84
19799a22
GS
85the commas on the right of the sort are evaluated before the sort,
86but the commas on the left are evaluated after. In other words,
87list operators tend to gobble up all arguments that follow, and
a0d0e21e 88then act like a simple TERM with regard to the preceding expression.
19799a22 89Be 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 ($foo), exit; # Or even this.
99
100Also note that
101
102 print ($foo & 255) + 1, "\n";
103
d042e63d
MS
104probably doesn't do what you expect at first glance. The parentheses
105enclose the argument list for C<print> which is evaluated (printing
106the result of C<$foo & 255>). Then one is added to the return value
107of C<print> (usually 1). The result is something like this:
108
109 1 + 1, "\n"; # Obviously not what you meant.
110
111To do what you meant properly, you must write:
112
113 print(($foo & 255) + 1, "\n");
114
115See L<Named Unary Operators> for more discussion of this.
a0d0e21e
LW
116
117Also parsed as terms are the C<do {}> and C<eval {}> constructs, as
54310121 118well as subroutine and method calls, and the anonymous
a0d0e21e
LW
119constructors C<[]> and C<{}>.
120
2ae324a7 121See also L<Quote and Quote-like Operators> toward the end of this section,
da87341d 122as well as L</"I/O Operators">.
a0d0e21e
LW
123
124=head2 The Arrow Operator
d74e8afc 125X<arrow> X<dereference> X<< -> >>
a0d0e21e 126
35f2feb0 127"C<< -> >>" is an infix dereference operator, just as it is in C
19799a22
GS
128and C++. If the right side is either a C<[...]>, C<{...}>, or a
129C<(...)> subscript, then the left side must be either a hard or
130symbolic reference to an array, a hash, or a subroutine respectively.
131(Or technically speaking, a location capable of holding a hard
132reference, if it's an array or hash reference being used for
133assignment.) See L<perlreftut> and L<perlref>.
a0d0e21e 134
19799a22
GS
135Otherwise, the right side is a method name or a simple scalar
136variable containing either the method name or a subroutine reference,
137and the left side must be either an object (a blessed reference)
138or a class name (that is, a package name). See L<perlobj>.
a0d0e21e 139
5f05dabc 140=head2 Auto-increment and Auto-decrement
d74e8afc 141X<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,
144they increment or decrement the variable by one before returning the
145value, and if placed after, increment or decrement after returning the
146value.
147
148 $i = 0; $j = 0;
149 print $i++; # prints 0
150 print ++$j; # prints 1
a0d0e21e 151
b033823e 152Note that just as in C, Perl doesn't define B<when> the variable is
89d205f2 153incremented or decremented. You just know it will be done sometime
b033823e
A
154before or after the value is returned. This also means that modifying
155a variable twice in the same statement will lead to undefined behaviour.
156Avoid statements like:
157
158 $i = $i ++;
159 print ++ $i + $i ++;
160
161Perl will not guarantee what the result of the above statements is.
162
54310121 163The auto-increment operator has a little extra builtin magic to it. If
a0d0e21e
LW
164you increment a variable that is numeric, or that has ever been used in
165a numeric context, you get a normal increment. If, however, the
5f05dabc 166variable has been used in only string contexts since it was set, and
5a964f20 167has a value that is not the empty string and matches the pattern
9c0670e1 168C</^[a-zA-Z]*[0-9]*\z/>, the increment is done as a string, preserving each
a0d0e21e
LW
169character within its range, with carry:
170
171 print ++($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
176C<undef> is always treated as numeric, and in particular is changed
177to C<0> before incrementing (so that a post-increment of an undef value
178will return C<0> rather than C<undef>).
179
5f05dabc 180The auto-decrement operator is not magical.
a0d0e21e
LW
181
182=head2 Exponentiation
d74e8afc 183X<**> X<exponentiation> X<power>
a0d0e21e 184
19799a22 185Binary "**" is the exponentiation operator. It binds even more
cb1a09d0
AD
186tightly than unary minus, so -2**4 is -(2**4), not (-2)**4. (This is
187implemented using C's pow(3) function, which actually works on doubles
188internally.)
a0d0e21e
LW
189
190=head2 Symbolic Unary Operators
d74e8afc 191X<unary operator> X<operator, unary>
a0d0e21e 192
5f05dabc 193Unary "!" performs logical negation, i.e., "not". See also C<not> for a lower
a0d0e21e 194precedence version of this.
d74e8afc 195X<!>
a0d0e21e
LW
196
197Unary "-" performs arithmetic negation if the operand is numeric. If
198the operand is an identifier, a string consisting of a minus sign
199concatenated with the identifier is returned. Otherwise, if the string
200starts with a plus or minus, a string starting with the opposite sign
bff5667c 201is returned. One effect of these rules is that -bareword is equivalent
8705167b 202to the string "-bareword". If, however, the string begins with a
353c6505 203non-alphabetic character (excluding "+" or "-"), Perl will attempt to convert
06705523
SP
204the string to a numeric and the arithmetic negation is performed. If the
205string cannot be cleanly converted to a numeric, Perl will give the warning
206B<Argument "the string" isn't numeric in negation (-) at ...>.
d74e8afc 207X<-> X<negation, arithmetic>
a0d0e21e 208
972b05a9
JH
209Unary "~" performs bitwise negation, i.e., 1's complement. For
210example, C<0666 & ~027> is 0640. (See also L<Integer Arithmetic> and
211L<Bitwise String Operators>.) Note that the width of the result is
212platform-dependent: ~0 is 32 bits wide on a 32-bit platform, but 64
213bits wide on a 64-bit platform, so if you are expecting a certain bit
d042e63d 214width, remember to use the & operator to mask off the excess bits.
d74e8afc 215X<~> X<negation, binary>
a0d0e21e
LW
216
217Unary "+" has no effect whatsoever, even on strings. It is useful
218syntactically for separating a function name from a parenthesized expression
219that would otherwise be interpreted as the complete list of function
5ba421f6 220arguments. (See examples above under L<Terms and List Operators (Leftward)>.)
d74e8afc 221X<+>
a0d0e21e 222
19799a22
GS
223Unary "\" creates a reference to whatever follows it. See L<perlreftut>
224and L<perlref>. Do not confuse this behavior with the behavior of
225backslash within a string, although both forms do convey the notion
226of protecting the next thing from interpolation.
d74e8afc 227X<\> X<reference> X<backslash>
a0d0e21e
LW
228
229=head2 Binding Operators
d74e8afc 230X<binding> X<operator, binding> X<=~> X<!~>
a0d0e21e 231
c07a80fd 232Binary "=~" binds a scalar expression to a pattern match. Certain operations
cb1a09d0
AD
233search or modify the string $_ by default. This operator makes that kind
234of operation work on some other string. The right argument is a search
2c268ad5
TP
235pattern, substitution, or transliteration. The left argument is what is
236supposed to be searched, substituted, or transliterated instead of the default
f8bab1e9
GS
237$_. When used in scalar context, the return value generally indicates the
238success of the operation. Behavior in list context depends on the particular
89d205f2 239operator. See L</"Regexp Quote-Like Operators"> for details and
d7782e69 240L<perlretut> for examples using these operators.
f8bab1e9
GS
241
242If the right argument is an expression rather than a search pattern,
2c268ad5 243substitution, or transliteration, it is interpreted as a search pattern at run
89d205f2
YO
244time. Note that this means that its contents will be interpolated twice, so
245
246 '\\' =~ q'\\';
247
248is not ok, as the regex engine will end up trying to compile the
249pattern C<\>, which it will consider a syntax error.
a0d0e21e
LW
250
251Binary "!~" is just like "=~" except the return value is negated in
252the logical sense.
253
254=head2 Multiplicative Operators
d74e8afc 255X<operator, multiplicative>
a0d0e21e
LW
256
257Binary "*" multiplies two numbers.
d74e8afc 258X<*>
a0d0e21e
LW
259
260Binary "/" divides two numbers.
d74e8afc 261X</> X<slash>
a0d0e21e 262
f7918450
KW
263Binary "%" is the modulo operator, which computes the division
264remainder of its first argument with respect to its second argument.
265Given integer
54310121 266operands C<$a> and C<$b>: If C<$b> is positive, then C<$a % $b> is
f7918450 267C<$a> minus the largest multiple of C<$b> less than or equal to
54310121
PP
268C<$a>. If C<$b> is negative, then C<$a % $b> is C<$a> minus the
269smallest multiple of C<$b> that is not less than C<$a> (i.e. the
89b4f0ad 270result will be less than or equal to zero). If the operands
4848a83b
ST
271C<$a> and C<$b> are floating point values and the absolute value of
272C<$b> (that is C<abs($b)>) is less than C<(UV_MAX + 1)>, only
273the integer portion of C<$a> and C<$b> will be used in the operation
274(Note: here C<UV_MAX> means the maximum of the unsigned integer type).
275If the absolute value of the right operand (C<abs($b)>) is greater than
276or equal to C<(UV_MAX + 1)>, "%" computes the floating-point remainder
277C<$r> in the equation C<($r = $a - $i*$b)> where C<$i> is a certain
f7918450 278integer that makes C<$r> have the same sign as the right operand
4848a83b
ST
279C<$b> (B<not> as the left operand C<$a> like C function C<fmod()>)
280and the absolute value less than that of C<$b>.
0412d526 281Note that when C<use integer> is in scope, "%" gives you direct access
f7918450 282to the modulo operator as implemented by your C compiler. This
55d729e4
GS
283operator is not as well defined for negative operands, but it will
284execute faster.
f7918450 285X<%> X<remainder> X<modulo> X<mod>
55d729e4 286
62d10b70
GS
287Binary "x" is the repetition operator. In scalar context or if the left
288operand is not enclosed in parentheses, it returns a string consisting
289of the left operand repeated the number of times specified by the right
290operand. In list context, if the left operand is enclosed in
3585017f
YST
291parentheses or is a list formed by C<qw/STRING/>, it repeats the list.
292If the right operand is zero or negative, it returns an empty string
293or an empty list, depending on the context.
d74e8afc 294X<x>
a0d0e21e
LW
295
296 print '-' x 80; # print row of dashes
297
298 print "\t" x ($tab/8), ' ' x ($tab%8); # tab over
299
300 @ones = (1) x 80; # a list of 80 1's
301 @ones = (5) x @ones; # set all elements to 5
302
303
304=head2 Additive Operators
d74e8afc 305X<operator, additive>
a0d0e21e
LW
306
307Binary "+" returns the sum of two numbers.
d74e8afc 308X<+>
a0d0e21e
LW
309
310Binary "-" returns the difference of two numbers.
d74e8afc 311X<->
a0d0e21e
LW
312
313Binary "." concatenates two strings.
d74e8afc
ITB
314X<string, concatenation> X<concatenation>
315X<cat> X<concat> X<concatenate> X<.>
a0d0e21e
LW
316
317=head2 Shift Operators
d74e8afc
ITB
318X<shift operator> X<operator, shift> X<<< << >>>
319X<<< >> >>> X<right shift> X<left shift> X<bitwise shift>
320X<shl> X<shr> X<shift, right> X<shift, left>
a0d0e21e 321
55497cff
PP
322Binary "<<" returns the value of its left argument shifted left by the
323number of bits specified by the right argument. Arguments should be
982ce180 324integers. (See also L<Integer Arithmetic>.)
a0d0e21e 325
55497cff
PP
326Binary ">>" returns the value of its left argument shifted right by
327the number of bits specified by the right argument. Arguments should
982ce180 328be integers. (See also L<Integer Arithmetic>.)
a0d0e21e 329
b16cf6df
JH
330Note that both "<<" and ">>" in Perl are implemented directly using
331"<<" and ">>" in C. If C<use integer> (see L<Integer Arithmetic>) is
332in force then signed C integers are used, else unsigned C integers are
333used. Either way, the implementation isn't going to generate results
334larger than the size of the integer type Perl was built with (32 bits
335or 64 bits).
336
337The result of overflowing the range of the integers is undefined
338because it is undefined also in C. In other words, using 32-bit
339integers, C<< 1 << 32 >> is undefined. Shifting by a negative number
340of bits is also undefined.
341
a0d0e21e 342=head2 Named Unary Operators
d74e8afc 343X<operator, named unary>
a0d0e21e
LW
344
345The various named unary operators are treated as functions with one
568e6d8b 346argument, with optional parentheses.
a0d0e21e
LW
347
348If any list operator (print(), etc.) or any unary operator (chdir(), etc.)
349is followed by a left parenthesis as the next token, the operator and
350arguments within parentheses are taken to be of highest precedence,
3981b0eb 351just like a normal function call. For example,
352because named unary operators are higher precedence than ||:
a0d0e21e
LW
353
354 chdir $foo || die; # (chdir $foo) || die
355 chdir($foo) || die; # (chdir $foo) || die
356 chdir ($foo) || die; # (chdir $foo) || die
357 chdir +($foo) || die; # (chdir $foo) || die
358
3981b0eb 359but, because * is higher precedence than named operators:
a0d0e21e
LW
360
361 chdir $foo * 20; # chdir ($foo * 20)
362 chdir($foo) * 20; # (chdir $foo) * 20
363 chdir ($foo) * 20; # (chdir $foo) * 20
364 chdir +($foo) * 20; # chdir ($foo * 20)
365
366 rand 10 * 20; # rand (10 * 20)
367 rand(10) * 20; # (rand 10) * 20
368 rand (10) * 20; # (rand 10) * 20
369 rand +(10) * 20; # rand (10 * 20)
370
568e6d8b
RGS
371Regarding precedence, the filetest operators, like C<-f>, C<-M>, etc. are
372treated like named unary operators, but they don't follow this functional
373parenthesis rule. That means, for example, that C<-f($file).".bak"> is
374equivalent to C<-f "$file.bak">.
d74e8afc 375X<-X> X<filetest> X<operator, filetest>
568e6d8b 376
5ba421f6 377See also L<"Terms and List Operators (Leftward)">.
a0d0e21e
LW
378
379=head2 Relational Operators
d74e8afc 380X<relational operator> X<operator, relational>
a0d0e21e 381
35f2feb0 382Binary "<" returns true if the left argument is numerically less than
a0d0e21e 383the right argument.
d74e8afc 384X<< < >>
a0d0e21e 385
35f2feb0 386Binary ">" returns true if the left argument is numerically greater
a0d0e21e 387than the right argument.
d74e8afc 388X<< > >>
a0d0e21e 389
35f2feb0 390Binary "<=" returns true if the left argument is numerically less than
a0d0e21e 391or equal to the right argument.
d74e8afc 392X<< <= >>
a0d0e21e 393
35f2feb0 394Binary ">=" returns true if the left argument is numerically greater
a0d0e21e 395than or equal to the right argument.
d74e8afc 396X<< >= >>
a0d0e21e
LW
397
398Binary "lt" returns true if the left argument is stringwise less than
399the right argument.
d74e8afc 400X<< lt >>
a0d0e21e
LW
401
402Binary "gt" returns true if the left argument is stringwise greater
403than the right argument.
d74e8afc 404X<< gt >>
a0d0e21e
LW
405
406Binary "le" returns true if the left argument is stringwise less than
407or equal to the right argument.
d74e8afc 408X<< le >>
a0d0e21e
LW
409
410Binary "ge" returns true if the left argument is stringwise greater
411than or equal to the right argument.
d74e8afc 412X<< ge >>
a0d0e21e
LW
413
414=head2 Equality Operators
d74e8afc 415X<equality> X<equal> X<equals> X<operator, equality>
a0d0e21e
LW
416
417Binary "==" returns true if the left argument is numerically equal to
418the right argument.
d74e8afc 419X<==>
a0d0e21e
LW
420
421Binary "!=" returns true if the left argument is numerically not equal
422to the right argument.
d74e8afc 423X<!=>
a0d0e21e 424
35f2feb0 425Binary "<=>" returns -1, 0, or 1 depending on whether the left
6ee5d4e7 426argument is numerically less than, equal to, or greater than the right
d4ad863d 427argument. If your platform supports NaNs (not-a-numbers) as numeric
7d3a9d88
NC
428values, using them with "<=>" returns undef. NaN is not "<", "==", ">",
429"<=" or ">=" anything (even NaN), so those 5 return false. NaN != NaN
430returns true, as does NaN != anything else. If your platform doesn't
431support NaNs then NaN is just a string with numeric value 0.
d74e8afc 432X<< <=> >> X<spaceship>
7d3a9d88 433
2b54f59f
YST
434 perl -le '$a = "NaN"; print "No NaN support here" if $a == $a'
435 perl -le '$a = "NaN"; print "NaN support here" if $a != $a'
a0d0e21e
LW
436
437Binary "eq" returns true if the left argument is stringwise equal to
438the right argument.
d74e8afc 439X<eq>
a0d0e21e
LW
440
441Binary "ne" returns true if the left argument is stringwise not equal
442to the right argument.
d74e8afc 443X<ne>
a0d0e21e 444
d4ad863d
JH
445Binary "cmp" returns -1, 0, or 1 depending on whether the left
446argument is stringwise less than, equal to, or greater than the right
447argument.
d74e8afc 448X<cmp>
a0d0e21e 449
0d863452 450Binary "~~" does a smart match between its arguments. Smart matching
0f7107a0 451is described in L<perlsyn/"Smart matching in detail">.
0d863452
RH
452X<~~>
453
a034a98d
DD
454"lt", "le", "ge", "gt" and "cmp" use the collation (sort) order specified
455by the current locale if C<use locale> is in effect. See L<perllocale>.
456
a0d0e21e 457=head2 Bitwise And
d74e8afc 458X<operator, bitwise, and> X<bitwise and> X<&>
a0d0e21e 459
2cdc098b 460Binary "&" returns its operands ANDed together bit by bit.
2c268ad5 461(See also L<Integer Arithmetic> and L<Bitwise String Operators>.)
a0d0e21e 462
2cdc098b
MG
463Note that "&" has lower priority than relational operators, so for example
464the brackets are essential in a test like
465
466 print "Even\n" if ($x & 1) == 0;
467
a0d0e21e 468=head2 Bitwise Or and Exclusive Or
d74e8afc
ITB
469X<operator, bitwise, or> X<bitwise or> X<|> X<operator, bitwise, xor>
470X<bitwise xor> X<^>
a0d0e21e 471
2cdc098b 472Binary "|" returns its operands ORed together bit by bit.
2c268ad5 473(See also L<Integer Arithmetic> and L<Bitwise String Operators>.)
a0d0e21e 474
2cdc098b 475Binary "^" returns its operands XORed together bit by bit.
2c268ad5 476(See also L<Integer Arithmetic> and L<Bitwise String Operators>.)
a0d0e21e 477
2cdc098b
MG
478Note that "|" and "^" have lower priority than relational operators, so
479for example the brackets are essential in a test like
480
481 print "false\n" if (8 | 2) != 10;
482
a0d0e21e 483=head2 C-style Logical And
d74e8afc 484X<&&> X<logical and> X<operator, logical, and>
a0d0e21e
LW
485
486Binary "&&" performs a short-circuit logical AND operation. That is,
487if the left operand is false, the right operand is not even evaluated.
488Scalar or list context propagates down to the right operand if it
489is evaluated.
490
491=head2 C-style Logical Or
d74e8afc 492X<||> X<operator, logical, or>
a0d0e21e
LW
493
494Binary "||" performs a short-circuit logical OR operation. That is,
495if the left operand is true, the right operand is not even evaluated.
496Scalar or list context propagates down to the right operand if it
497is evaluated.
498
c963b151 499=head2 C-style Logical Defined-Or
d74e8afc 500X<//> X<operator, logical, defined-or>
c963b151
BD
501
502Although it has no direct equivalent in C, Perl's C<//> operator is related
89d205f2 503to its C-style or. In fact, it's exactly the same as C<||>, except that it
c963b151 504tests the left hand side's definedness instead of its truth. Thus, C<$a // $b>
89d205f2
YO
505is similar to C<defined($a) || $b> (except that it returns the value of C<$a>
506rather than the value of C<defined($a)>) and is exactly equivalent to
c963b151 507C<defined($a) ? $a : $b>. This is very useful for providing default values
89d205f2 508for variables. If you actually want to test if at least one of C<$a> and
d042e63d 509C<$b> is defined, use C<defined($a // $b)>.
c963b151 510
d042e63d
MS
511The C<||>, C<//> and C<&&> operators return the last value evaluated
512(unlike C's C<||> and C<&&>, which return 0 or 1). Thus, a reasonably
513portable way to find out the home directory might be:
a0d0e21e 514
c963b151
BD
515 $home = $ENV{'HOME'} // $ENV{'LOGDIR'} //
516 (getpwuid($<))[7] // die "You're homeless!\n";
a0d0e21e 517
5a964f20
TC
518In particular, this means that you shouldn't use this
519for selecting between two aggregates for assignment:
520
521 @a = @b || @c; # this is wrong
522 @a = scalar(@b) || @c; # really meant this
523 @a = @b ? @b : @c; # this works fine, though
524
f23102e2
RGS
525As more readable alternatives to C<&&> and C<||> when used for
526control flow, Perl provides the C<and> and C<or> operators (see below).
527The short-circuit behavior is identical. The precedence of "and"
c963b151 528and "or" is much lower, however, so that you can safely use them after a
5a964f20 529list operator without the need for parentheses:
a0d0e21e
LW
530
531 unlink "alpha", "beta", "gamma"
532 or gripe(), next LINE;
533
534With the C-style operators that would have been written like this:
535
536 unlink("alpha", "beta", "gamma")
537 || (gripe(), next LINE);
538
eeb6a2c9 539Using "or" for assignment is unlikely to do what you want; see below.
5a964f20
TC
540
541=head2 Range Operators
d74e8afc 542X<operator, range> X<range> X<..> X<...>
a0d0e21e
LW
543
544Binary ".." is the range operator, which is really two different
fb53bbb2 545operators depending on the context. In list context, it returns a
54ae734e 546list of values counting (up by ones) from the left value to the right
2cdbc966 547value. If the left value is greater than the right value then it
fb53bbb2 548returns the empty list. The range operator is useful for writing
54ae734e 549C<foreach (1..10)> loops and for doing slice operations on arrays. In
2cdbc966
JD
550the current implementation, no temporary array is created when the
551range operator is used as the expression in C<foreach> loops, but older
552versions of Perl might burn a lot of memory when you write something
553like this:
a0d0e21e
LW
554
555 for (1 .. 1_000_000) {
556 # code
54310121 557 }
a0d0e21e 558
8f0f46f8 559The range operator also works on strings, using the magical
560auto-increment, see below.
54ae734e 561
5a964f20 562In scalar context, ".." returns a boolean value. The operator is
8f0f46f8 563bistable, like a flip-flop, and emulates the line-range (comma)
564operator of B<sed>, B<awk>, and various editors. Each ".." operator
565maintains its own boolean state, even across calls to a subroutine
566that contains it. It is false as long as its left operand is false.
a0d0e21e
LW
567Once the left operand is true, the range operator stays true until the
568right operand is true, I<AFTER> which the range operator becomes false
8f0f46f8 569again. It doesn't become false till the next time the range operator
570is evaluated. It can test the right operand and become false on the
571same evaluation it became true (as in B<awk>), but it still returns
572true once. If you don't want it to test the right operand until the
573next evaluation, as in B<sed>, just use three dots ("...") instead of
19799a22
GS
574two. In all other regards, "..." behaves just like ".." does.
575
576The right operand is not evaluated while the operator is in the
577"false" state, and the left operand is not evaluated while the
578operator is in the "true" state. The precedence is a little lower
579than || and &&. The value returned is either the empty string for
8f0f46f8 580false, or a sequence number (beginning with 1) for true. The sequence
581number is reset for each range encountered. The final sequence number
582in a range has the string "E0" appended to it, which doesn't affect
583its numeric value, but gives you something to search for if you want
584to exclude the endpoint. You can exclude the beginning point by
585waiting for the sequence number to be greater than 1.
df5f8116
CW
586
587If either operand of scalar ".." is a constant expression,
588that operand is considered true if it is equal (C<==>) to the current
589input line number (the C<$.> variable).
590
591To be pedantic, the comparison is actually C<int(EXPR) == int(EXPR)>,
592but that is only an issue if you use a floating point expression; when
593implicitly using C<$.> as described in the previous paragraph, the
594comparison is C<int(EXPR) == int($.)> which is only an issue when C<$.>
595is set to a floating point value and you are not reading from a file.
596Furthermore, C<"span" .. "spat"> or C<2.18 .. 3.14> will not do what
597you want in scalar context because each of the operands are evaluated
598using their integer representation.
599
600Examples:
a0d0e21e
LW
601
602As a scalar operator:
603
df5f8116 604 if (101 .. 200) { print; } # print 2nd hundred lines, short for
f343f960 605 # if ($. == 101 .. $. == 200) { print; }
9f10b797
RGS
606
607 next LINE if (1 .. /^$/); # skip header lines, short for
f343f960 608 # next LINE if ($. == 1 .. /^$/);
9f10b797
RGS
609 # (typically in a loop labeled LINE)
610
611 s/^/> / if (/^$/ .. eof()); # quote body
a0d0e21e 612
5a964f20
TC
613 # parse mail messages
614 while (<>) {
615 $in_header = 1 .. /^$/;
df5f8116
CW
616 $in_body = /^$/ .. eof;
617 if ($in_header) {
f343f960 618 # do something
df5f8116 619 } else { # in body
f343f960 620 # do something else
df5f8116 621 }
5a964f20 622 } continue {
df5f8116 623 close ARGV if eof; # reset $. each file
5a964f20
TC
624 }
625
acf31ca5
SF
626Here's a simple example to illustrate the difference between
627the two range operators:
628
629 @lines = (" - Foo",
630 "01 - Bar",
631 "1 - Baz",
632 " - Quux");
633
9f10b797
RGS
634 foreach (@lines) {
635 if (/0/ .. /1/) {
acf31ca5
SF
636 print "$_\n";
637 }
638 }
639
9f10b797
RGS
640This program will print only the line containing "Bar". If
641the range operator is changed to C<...>, it will also print the
acf31ca5
SF
642"Baz" line.
643
644And now some examples as a list operator:
a0d0e21e
LW
645
646 for (101 .. 200) { print; } # print $_ 100 times
3e3baf6d 647 @foo = @foo[0 .. $#foo]; # an expensive no-op
a0d0e21e
LW
648 @foo = @foo[$#foo-4 .. $#foo]; # slice last 5 items
649
5a964f20 650The range operator (in list context) makes use of the magical
5f05dabc 651auto-increment algorithm if the operands are strings. You
a0d0e21e
LW
652can say
653
654 @alphabet = ('A' .. 'Z');
655
54ae734e 656to get all normal letters of the English alphabet, or
a0d0e21e
LW
657
658 $hexdigit = (0 .. 9, 'a' .. 'f')[$num & 15];
659
660to get a hexadecimal digit, or
661
662 @z2 = ('01' .. '31'); print $z2[$mday];
663
ea4f5703
YST
664to get dates with leading zeros.
665
666If the final value specified is not in the sequence that the magical
667increment would produce, the sequence goes until the next value would
668be longer than the final value specified.
669
670If the initial value specified isn't part of a magical increment
671sequence (that is, a non-empty string matching "/^[a-zA-Z]*[0-9]*\z/"),
672only the initial value will be returned. So the following will only
673return an alpha:
674
675 use charnames 'greek';
676 my @greek_small = ("\N{alpha}" .. "\N{omega}");
677
678To get lower-case greek letters, use this instead:
679
680 my @greek_small = map { chr } ( ord("\N{alpha}") .. ord("\N{omega}") );
a0d0e21e 681
df5f8116
CW
682Because each operand is evaluated in integer form, C<2.18 .. 3.14> will
683return two elements in list context.
684
685 @list = (2.18 .. 3.14); # same as @list = (2 .. 3);
686
a0d0e21e 687=head2 Conditional Operator
d74e8afc 688X<operator, conditional> X<operator, ternary> X<ternary> X<?:>
a0d0e21e
LW
689
690Ternary "?:" is the conditional operator, just as in C. It works much
691like an if-then-else. If the argument before the ? is true, the
692argument before the : is returned, otherwise the argument after the :
cb1a09d0
AD
693is returned. For example:
694
54310121 695 printf "I have %d dog%s.\n", $n,
cb1a09d0
AD
696 ($n == 1) ? '' : "s";
697
698Scalar or list context propagates downward into the 2nd
54310121 699or 3rd argument, whichever is selected.
cb1a09d0
AD
700
701 $a = $ok ? $b : $c; # get a scalar
702 @a = $ok ? @b : @c; # get an array
703 $a = $ok ? @b : @c; # oops, that's just a count!
704
705The operator may be assigned to if both the 2nd and 3rd arguments are
706legal lvalues (meaning that you can assign to them):
a0d0e21e
LW
707
708 ($a_or_b ? $a : $b) = $c;
709
5a964f20
TC
710Because this operator produces an assignable result, using assignments
711without parentheses will get you in trouble. For example, this:
712
713 $a % 2 ? $a += 10 : $a += 2
714
715Really means this:
716
717 (($a % 2) ? ($a += 10) : $a) += 2
718
719Rather than this:
720
721 ($a % 2) ? ($a += 10) : ($a += 2)
722
19799a22
GS
723That should probably be written more simply as:
724
725 $a += ($a % 2) ? 10 : 2;
726
4633a7c4 727=head2 Assignment Operators
d74e8afc 728X<assignment> X<operator, assignment> X<=> X<**=> X<+=> X<*=> X<&=>
5ac3b81c 729X<<< <<= >>> X<&&=> X<-=> X</=> X<|=> X<<< >>= >>> X<||=> X<//=> X<.=>
d74e8afc 730X<%=> X<^=> X<x=>
a0d0e21e
LW
731
732"=" is the ordinary assignment operator.
733
734Assignment operators work as in C. That is,
735
736 $a += 2;
737
738is equivalent to
739
740 $a = $a + 2;
741
742although without duplicating any side effects that dereferencing the lvalue
54310121
PP
743might trigger, such as from tie(). Other assignment operators work similarly.
744The following are recognized:
a0d0e21e
LW
745
746 **= += *= &= <<= &&=
9f10b797
RGS
747 -= /= |= >>= ||=
748 .= %= ^= //=
749 x=
a0d0e21e 750
19799a22 751Although these are grouped by family, they all have the precedence
a0d0e21e
LW
752of assignment.
753
b350dd2f
GS
754Unlike in C, the scalar assignment operator produces a valid lvalue.
755Modifying an assignment is equivalent to doing the assignment and
756then modifying the variable that was assigned to. This is useful
757for modifying a copy of something, like this:
a0d0e21e
LW
758
759 ($tmp = $global) =~ tr [A-Z] [a-z];
760
761Likewise,
762
763 ($a += 2) *= 3;
764
765is equivalent to
766
767 $a += 2;
768 $a *= 3;
769
b350dd2f
GS
770Similarly, a list assignment in list context produces the list of
771lvalues assigned to, and a list assignment in scalar context returns
772the number of elements produced by the expression on the right hand
773side of the assignment.
774
748a9306 775=head2 Comma Operator
d74e8afc 776X<comma> X<operator, comma> X<,>
a0d0e21e 777
5a964f20 778Binary "," is the comma operator. In scalar context it evaluates
a0d0e21e
LW
779its left argument, throws that value away, then evaluates its right
780argument and returns that value. This is just like C's comma operator.
781
5a964f20 782In list context, it's just the list argument separator, and inserts
ed5c6d31
PJ
783both its arguments into the list. These arguments are also evaluated
784from left to right.
a0d0e21e 785
344f2c40
IG
786The C<< => >> operator is a synonym for the comma except that it causes
787its left operand to be interpreted as a string if it begins with a letter
788or underscore and is composed only of letters, digits and underscores.
789This includes operands that might otherwise be interpreted as operators,
790constants, single number v-strings or function calls. If in doubt about
791this behaviour, the left operand can be quoted explicitly.
792
793Otherwise, the C<< => >> operator behaves exactly as the comma operator
794or list argument separator, according to context.
795
796For example:
a44e5664
MS
797
798 use constant FOO => "something";
799
800 my %h = ( FOO => 23 );
801
802is equivalent to:
803
804 my %h = ("FOO", 23);
805
806It is I<NOT>:
807
808 my %h = ("something", 23);
809
719b43e8
RGS
810The C<< => >> operator is helpful in documenting the correspondence
811between keys and values in hashes, and other paired elements in lists.
748a9306 812
a44e5664
MS
813 %hash = ( $key => $value );
814 login( $username => $password );
815
678ae90b
RGS
816=head2 Yada Yada Operator
817X<...> X<... operator> X<yada yada operator>
be25f609 818
e8163f9b 819The yada yada operator (noted C<...>) is a placeholder for code. Perl
820parses it without error, but when you try to execute a yada yada, it
821throws an exception with the text C<Unimplemented>:
822
823 sub unimplemented { ... }
824
825 eval { unimplemented() };
826 if( $@ eq 'Unimplemented' ) {
827 print "I found the yada yada!\n";
828 }
829
830You can only use the yada yada to stand in for a complete statement.
831These examples of the yada yada work:
832
833 { ... }
834
835 sub foo { ... }
836
837 ...;
838
839 eval { ... };
840
841 sub foo {
842 my( $self ) = shift;
843
844 ...;
845 }
846
847 do { my $n; ...; print 'Hurrah!' };
848
849The yada yada cannot stand in for an expression that is part of a
850larger statement since the C<...> is also the three-dot version of the
851range operator (see L<Range Operators>). These examples of the yada
852yada are still syntax errors:
853
854 print ...;
855
856 open my($fh), '>', '/dev/passwd' or ...;
857
858 if( $condition && ... ) { print "Hello\n" };
859
860There are some cases where Perl can't immediately tell the difference
861between an expression and a statement. For instance, the syntax for a
862block and an anonymous hash reference constructor look the same unless
863there's something in the braces that give Perl a hint. The yada yada
864is a syntax error if Perl doesn't guess that the C<{ ... }> is a
865block. In that case, it doesn't think the C<...> is the yada yada
866because it's expecting an expression instead of a statement:
867
868 my @transformed = map { ... } @input; # syntax error
869
870You can use a C<;> inside your block to denote that the C<{ ... }> is
871a block and not a hash reference constructor. Now the yada yada works:
872
873 my @transformed = map {; ... } @input; # ; disambiguates
874
875 my @transformed = map { ...; } @input; # ; disambiguates
be25f609 876
a0d0e21e 877=head2 List Operators (Rightward)
d74e8afc 878X<operator, list, rightward> X<list operator>
a0d0e21e
LW
879
880On the right side of a list operator, it has very low precedence,
881such that it controls all comma-separated expressions found there.
882The only operators with lower precedence are the logical operators
883"and", "or", and "not", which may be used to evaluate calls to list
884operators without the need for extra parentheses:
885
886 open HANDLE, "filename"
887 or die "Can't open: $!\n";
888
5ba421f6 889See also discussion of list operators in L<Terms and List Operators (Leftward)>.
a0d0e21e
LW
890
891=head2 Logical Not
d74e8afc 892X<operator, logical, not> X<not>
a0d0e21e
LW
893
894Unary "not" returns the logical negation of the expression to its right.
895It's the equivalent of "!" except for the very low precedence.
896
897=head2 Logical And
d74e8afc 898X<operator, logical, and> X<and>
a0d0e21e
LW
899
900Binary "and" returns the logical conjunction of the two surrounding
901expressions. It's equivalent to && except for the very low
5f05dabc 902precedence. This means that it short-circuits: i.e., the right
a0d0e21e
LW
903expression is evaluated only if the left expression is true.
904
c963b151 905=head2 Logical or, Defined or, and Exclusive Or
f23102e2 906X<operator, logical, or> X<operator, logical, xor>
d74e8afc 907X<operator, logical, defined or> X<operator, logical, exclusive or>
f23102e2 908X<or> X<xor>
a0d0e21e
LW
909
910Binary "or" returns the logical disjunction of the two surrounding
5a964f20
TC
911expressions. It's equivalent to || except for the very low precedence.
912This makes it useful for control flow
913
914 print FH $data or die "Can't write to FH: $!";
915
916This means that it short-circuits: i.e., the right expression is evaluated
917only if the left expression is false. Due to its precedence, you should
918probably avoid using this for assignment, only for control flow.
919
920 $a = $b or $c; # bug: this is wrong
921 ($a = $b) or $c; # really means this
922 $a = $b || $c; # better written this way
923
19799a22 924However, when it's a list-context assignment and you're trying to use
5a964f20
TC
925"||" for control flow, you probably need "or" so that the assignment
926takes higher precedence.
927
928 @info = stat($file) || die; # oops, scalar sense of stat!
929 @info = stat($file) or die; # better, now @info gets its due
930
c963b151
BD
931Then again, you could always use parentheses.
932
a0d0e21e
LW
933Binary "xor" returns the exclusive-OR of the two surrounding expressions.
934It cannot short circuit, of course.
935
936=head2 C Operators Missing From Perl
d74e8afc
ITB
937X<operator, missing from perl> X<&> X<*>
938X<typecasting> X<(TYPE)>
a0d0e21e
LW
939
940Here is what C has that Perl doesn't:
941
942=over 8
943
944=item unary &
945
946Address-of operator. (But see the "\" operator for taking a reference.)
947
948=item unary *
949
54310121 950Dereference-address operator. (Perl's prefix dereferencing
a0d0e21e
LW
951operators are typed: $, @, %, and &.)
952
953=item (TYPE)
954
19799a22 955Type-casting operator.
a0d0e21e
LW
956
957=back
958
5f05dabc 959=head2 Quote and Quote-like Operators
89d205f2 960X<operator, quote> X<operator, quote-like> X<q> X<qq> X<qx> X<qw> X<m>
d74e8afc
ITB
961X<qr> X<s> X<tr> X<'> X<''> X<"> X<""> X<//> X<`> X<``> X<<< << >>>
962X<escape sequence> X<escape>
963
a0d0e21e
LW
964
965While we usually think of quotes as literal values, in Perl they
966function as operators, providing various kinds of interpolating and
967pattern matching capabilities. Perl provides customary quote characters
968for these behaviors, but also provides a way for you to choose your
969quote character for any of them. In the following table, a C<{}> represents
9f10b797 970any pair of delimiters you choose.
a0d0e21e 971
2c268ad5
TP
972 Customary Generic Meaning Interpolates
973 '' q{} Literal no
974 "" qq{} Literal yes
af9219ee 975 `` qx{} Command yes*
2c268ad5 976 qw{} Word list no
af9219ee
MG
977 // m{} Pattern match yes*
978 qr{} Pattern yes*
979 s{}{} Substitution yes*
2c268ad5 980 tr{}{} Transliteration no (but see below)
7e3b091d 981 <<EOF here-doc yes*
a0d0e21e 982
af9219ee
MG
983 * unless the delimiter is ''.
984
87275199
GS
985Non-bracketing delimiters use the same character fore and aft, but the four
986sorts of brackets (round, angle, square, curly) will all nest, which means
9f10b797 987that
87275199 988
9f10b797 989 q{foo{bar}baz}
35f2feb0 990
9f10b797 991is the same as
87275199
GS
992
993 'foo{bar}baz'
994
995Note, however, that this does not always work for quoting Perl code:
996
997 $s = q{ if($a eq "}") ... }; # WRONG
998
83df6a1d
JH
999is a syntax error. The C<Text::Balanced> module (from CPAN, and
1000starting from Perl 5.8 part of the standard distribution) is able
1001to do this properly.
87275199 1002
19799a22 1003There can be whitespace between the operator and the quoting
fb73857a 1004characters, except when C<#> is being used as the quoting character.
19799a22
GS
1005C<q#foo#> is parsed as the string C<foo>, while C<q #foo#> is the
1006operator C<q> followed by a comment. Its argument will be taken
1007from the next line. This allows you to write:
fb73857a
PP
1008
1009 s {foo} # Replace foo
1010 {bar} # with bar.
1011
904501ec
MG
1012The following escape sequences are available in constructs that interpolate
1013and in transliterations.
d74e8afc 1014X<\t> X<\n> X<\r> X<\f> X<\b> X<\a> X<\e> X<\x> X<\0> X<\c> X<\N>
a0d0e21e 1015
6ee5d4e7 1016 \t tab (HT, TAB)
5a964f20 1017 \n newline (NL)
6ee5d4e7
PP
1018 \r return (CR)
1019 \f form feed (FF)
1020 \b backspace (BS)
1021 \a alarm (bell) (BEL)
1022 \e escape (ESC)
ee9f418e
WL
1023 \033 octal char (example: ESC)
1024 \x1b hex char (example: ESC)
1025 \x{263a} wide hex char (example: SMILEY)
1026 \c[ control char (example: ESC)
95cc3e0c 1027 \N{name} named Unicode character
e526e8bb 1028 \N{U+263D} Unicode character (example: FIRST QUARTER MOON)
2c268ad5 1029
ee9f418e
WL
1030The character following C<\c> is mapped to some other character by
1031converting letters to upper case and then (on ASCII systems) by inverting
1032the 7th bit (0x40). The most interesting range is from '@' to '_'
1033(0x40 through 0x5F), resulting in a control character from 0x00
1034through 0x1F. A '?' maps to the DEL character. On EBCDIC systems only
1035'@', the letters, '[', '\', ']', '^', '_' and '?' will work, resulting
1036in 0x00 through 0x1F and 0x7F.
1037
e526e8bb
KW
1038C<\N{U+I<wide hex char>}> means the Unicode character whose Unicode ordinal
1039number is I<wide hex char>.
1040For documentation of C<\N{name}>, see L<charnames>.
4c77eaa2 1041
e526e8bb
KW
1042B<NOTE>: Unlike C and other languages, Perl has no C<\v> escape sequence for
1043the vertical tab (VT - ASCII 11), but you may use C<\ck> or C<\x0b>. (C<\v>
1044does have meaning in regular expression patterns in Perl, see L<perlre>.)
1045
1046The following escape sequences are available in constructs that interpolate,
904501ec 1047but not in transliterations.
d74e8afc 1048X<\l> X<\u> X<\L> X<\U> X<\E> X<\Q>
904501ec 1049
a0d0e21e
LW
1050 \l lowercase next char
1051 \u uppercase next char
1052 \L lowercase till \E
1053 \U uppercase till \E
1054 \E end case modification
1d2dff63 1055 \Q quote non-word characters till \E
a0d0e21e 1056
95cc3e0c
JH
1057If C<use locale> is in effect, the case map used by C<\l>, C<\L>,
1058C<\u> and C<\U> is taken from the current locale. See L<perllocale>.
1059If Unicode (for example, C<\N{}> or wide hex characters of 0x100 or
1060beyond) is being used, the case map used by C<\l>, C<\L>, C<\u> and
e526e8bb 1061C<\U> is as defined by Unicode.
a034a98d 1062
5a964f20
TC
1063All systems use the virtual C<"\n"> to represent a line terminator,
1064called a "newline". There is no such thing as an unvarying, physical
19799a22 1065newline character. It is only an illusion that the operating system,
5a964f20
TC
1066device drivers, C libraries, and Perl all conspire to preserve. Not all
1067systems read C<"\r"> as ASCII CR and C<"\n"> as ASCII LF. For example,
1068on a Mac, these are reversed, and on systems without line terminator,
1069printing C<"\n"> may emit no actual data. In general, use C<"\n"> when
1070you mean a "newline" for your system, but use the literal ASCII when you
1071need an exact character. For example, most networking protocols expect
2a380090 1072and prefer a CR+LF (C<"\015\012"> or C<"\cM\cJ">) for line terminators,
5a964f20
TC
1073and although they often accept just C<"\012">, they seldom tolerate just
1074C<"\015">. If you get in the habit of using C<"\n"> for networking,
1075you may be burned some day.
d74e8afc
ITB
1076X<newline> X<line terminator> X<eol> X<end of line>
1077X<\n> X<\r> X<\r\n>
5a964f20 1078
904501ec
MG
1079For constructs that do interpolate, variables beginning with "C<$>"
1080or "C<@>" are interpolated. Subscripted variables such as C<$a[3]> or
ad0f383a
A
1081C<< $href->{key}[0] >> are also interpolated, as are array and hash slices.
1082But method calls such as C<< $obj->meth >> are not.
af9219ee
MG
1083
1084Interpolating an array or slice interpolates the elements in order,
1085separated by the value of C<$">, so is equivalent to interpolating
6deea57f
ST
1086C<join $", @array>. "Punctuation" arrays such as C<@*> are only
1087interpolated if the name is enclosed in braces C<@{*}>, but special
1088arrays C<@_>, C<@+>, and C<@-> are interpolated, even without braces.
af9219ee 1089
89d205f2
YO
1090You cannot include a literal C<$> or C<@> within a C<\Q> sequence.
1091An unescaped C<$> or C<@> interpolates the corresponding variable,
1d2dff63 1092while escaping will cause the literal string C<\$> to be inserted.
89d205f2 1093You'll need to write something like C<m/\Quser\E\@\Qhost/>.
1d2dff63 1094
a0d0e21e
LW
1095Patterns are subject to an additional level of interpretation as a
1096regular expression. This is done as a second pass, after variables are
1097interpolated, so that regular expressions may be incorporated into the
1098pattern from the variables. If this is not what you want, use C<\Q> to
1099interpolate a variable literally.
1100
19799a22
GS
1101Apart from the behavior described above, Perl does not expand
1102multiple levels of interpolation. In particular, contrary to the
1103expectations of shell programmers, back-quotes do I<NOT> interpolate
1104within double quotes, nor do single quotes impede evaluation of
1105variables when used within double quotes.
a0d0e21e 1106
5f05dabc 1107=head2 Regexp Quote-Like Operators
d74e8afc 1108X<operator, regexp>
cb1a09d0 1109
5f05dabc 1110Here are the quote-like operators that apply to pattern
cb1a09d0
AD
1111matching and related activities.
1112
a0d0e21e
LW
1113=over 8
1114
87e95b7f 1115=item qr/STRING/msixpo
01c6f5f4 1116X<qr> X</i> X</m> X</o> X</s> X</x> X</p>
a0d0e21e 1117
87e95b7f
YO
1118This operator quotes (and possibly compiles) its I<STRING> as a regular
1119expression. I<STRING> is interpolated the same way as I<PATTERN>
1120in C<m/PATTERN/>. If "'" is used as the delimiter, no interpolation
1121is done. Returns a Perl value which may be used instead of the
64c5a566 1122corresponding C</STRING/msixpo> expression. The returned value is a
85dd5c8b 1123normalized version of the original pattern. It magically differs from
64c5a566 1124a string containing the same characters: C<ref(qr/x/)> returns "Regexp",
85dd5c8b 1125even though dereferencing the result returns undef.
a0d0e21e 1126
87e95b7f
YO
1127For example,
1128
1129 $rex = qr/my.STRING/is;
85dd5c8b 1130 print $rex; # prints (?si-xm:my.STRING)
87e95b7f
YO
1131 s/$rex/foo/;
1132
1133is equivalent to
1134
1135 s/my.STRING/foo/is;
1136
1137The result may be used as a subpattern in a match:
1138
1139 $re = qr/$pattern/;
1140 $string =~ /foo${re}bar/; # can be interpolated in other patterns
1141 $string =~ $re; # or used standalone
1142 $string =~ /$re/; # or this way
1143
1144Since Perl may compile the pattern at the moment of execution of qr()
1145operator, using qr() may have speed advantages in some situations,
1146notably if the result of qr() is used standalone:
1147
1148 sub match {
1149 my $patterns = shift;
1150 my @compiled = map qr/$_/i, @$patterns;
1151 grep {
1152 my $success = 0;
1153 foreach my $pat (@compiled) {
1154 $success = 1, last if /$pat/;
1155 }
1156 $success;
1157 } @_;
5a964f20
TC
1158 }
1159
87e95b7f
YO
1160Precompilation of the pattern into an internal representation at
1161the moment of qr() avoids a need to recompile the pattern every
1162time a match C</$pat/> is attempted. (Perl has many other internal
1163optimizations, but none would be triggered in the above example if
1164we did not use qr() operator.)
1165
1166Options are:
1167
1168 m Treat string as multiple lines.
1169 s Treat string as single line. (Make . match a newline)
1170 i Do case-insensitive pattern matching.
1171 x Use extended regular expressions.
1172 p When matching preserve a copy of the matched string so
1173 that ${^PREMATCH}, ${^MATCH}, ${^POSTMATCH} will be defined.
1174 o Compile pattern only once.
1175
1176If a precompiled pattern is embedded in a larger pattern then the effect
1177of 'msixp' will be propagated appropriately. The effect of the 'o'
1178modifier has is not propagated, being restricted to those patterns
1179explicitly using it.
1180
1181See L<perlre> for additional information on valid syntax for STRING, and
1182for a detailed look at the semantics of regular expressions.
a0d0e21e 1183
87e95b7f 1184=item m/PATTERN/msixpogc
89d205f2
YO
1185X<m> X<operator, match>
1186X<regexp, options> X<regexp> X<regex, options> X<regex>
01c6f5f4 1187X</m> X</s> X</i> X</x> X</p> X</o> X</g> X</c>
a0d0e21e 1188
87e95b7f 1189=item /PATTERN/msixpogc
a0d0e21e 1190
5a964f20 1191Searches a string for a pattern match, and in scalar context returns
19799a22
GS
1192true if it succeeds, false if it fails. If no string is specified
1193via the C<=~> or C<!~> operator, the $_ string is searched. (The
1194string specified with C<=~> need not be an lvalue--it may be the
1195result of an expression evaluation, but remember the C<=~> binds
1196rather tightly.) See also L<perlre>. See L<perllocale> for
1197discussion of additional considerations that apply when C<use locale>
1198is in effect.
a0d0e21e 1199
01c6f5f4
RGS
1200Options are as described in C<qr//>; in addition, the following match
1201process modifiers are available:
a0d0e21e 1202
cde0cee5
YO
1203 g Match globally, i.e., find all occurrences.
1204 c Do not reset search position on a failed match when /g is in effect.
a0d0e21e
LW
1205
1206If "/" is the delimiter then the initial C<m> is optional. With the C<m>
ed02a3bf 1207you can use any pair of non-whitespace characters
19799a22
GS
1208as delimiters. This is particularly useful for matching path names
1209that contain "/", to avoid LTS (leaning toothpick syndrome). If "?" is
7bac28a0 1210the delimiter, then the match-only-once rule of C<?PATTERN?> applies.
19799a22 1211If "'" is the delimiter, no interpolation is performed on the PATTERN.
ed02a3bf
DN
1212When using a character valid in an identifier, whitespace is required
1213after the C<m>.
a0d0e21e
LW
1214
1215PATTERN may contain variables, which will be interpolated (and the
f70b4f9c 1216pattern recompiled) every time the pattern search is evaluated, except
1f247705
GS
1217for when the delimiter is a single quote. (Note that C<$(>, C<$)>, and
1218C<$|> are not interpolated because they look like end-of-string tests.)
f70b4f9c
AB
1219If you want such a pattern to be compiled only once, add a C</o> after
1220the trailing delimiter. This avoids expensive run-time recompilations,
1221and is useful when the value you are interpolating won't change over
1222the life of the script. However, mentioning C</o> constitutes a promise
1223that you won't change the variables in the pattern. If you change them,
01c6f5f4 1224Perl won't even notice. See also L<"qr/STRING/msixpo">.
a0d0e21e 1225
e9d89077
DN
1226=item The empty pattern //
1227
5a964f20 1228If the PATTERN evaluates to the empty string, the last
d65afb4b
HS
1229I<successfully> matched regular expression is used instead. In this
1230case, only the C<g> and C<c> flags on the empty pattern is honoured -
1231the other flags are taken from the original pattern. If no match has
1232previously succeeded, this will (silently) act instead as a genuine
1233empty pattern (which will always match).
a0d0e21e 1234
89d205f2
YO
1235Note that it's possible to confuse Perl into thinking C<//> (the empty
1236regex) is really C<//> (the defined-or operator). Perl is usually pretty
1237good about this, but some pathological cases might trigger this, such as
1238C<$a///> (is that C<($a) / (//)> or C<$a // />?) and C<print $fh //>
1239(C<print $fh(//> or C<print($fh //>?). In all of these examples, Perl
1240will assume you meant defined-or. If you meant the empty regex, just
1241use parentheses or spaces to disambiguate, or even prefix the empty
c963b151
BD
1242regex with an C<m> (so C<//> becomes C<m//>).
1243
e9d89077
DN
1244=item Matching in list context
1245
19799a22 1246If the C</g> option is not used, C<m//> in list context returns a
a0d0e21e 1247list consisting of the subexpressions matched by the parentheses in the
f7e33566
GS
1248pattern, i.e., (C<$1>, C<$2>, C<$3>...). (Note that here C<$1> etc. are
1249also set, and that this differs from Perl 4's behavior.) When there are
1250no parentheses in the pattern, the return value is the list C<(1)> for
1251success. With or without parentheses, an empty list is returned upon
1252failure.
a0d0e21e
LW
1253
1254Examples:
1255
1256 open(TTY, '/dev/tty');
1257 <TTY> =~ /^y/i && foo(); # do foo if desired
1258
1259 if (/Version: *([0-9.]*)/) { $version = $1; }
1260
1261 next if m#^/usr/spool/uucp#;
1262
1263 # poor man's grep
1264 $arg = shift;
1265 while (<>) {
1266 print if /$arg/o; # compile only once
1267 }
1268
1269 if (($F1, $F2, $Etc) = ($foo =~ /^(\S+)\s+(\S+)\s*(.*)/))
1270
1271This last example splits $foo into the first two words and the
5f05dabc
PP
1272remainder of the line, and assigns those three fields to $F1, $F2, and
1273$Etc. The conditional is true if any variables were assigned, i.e., if
a0d0e21e
LW
1274the pattern matched.
1275
19799a22
GS
1276The C</g> modifier specifies global pattern matching--that is,
1277matching as many times as possible within the string. How it behaves
1278depends on the context. In list context, it returns a list of the
1279substrings matched by any capturing parentheses in the regular
1280expression. If there are no parentheses, it returns a list of all
1281the matched strings, as if there were parentheses around the whole
1282pattern.
a0d0e21e 1283
7e86de3e 1284In scalar context, each execution of C<m//g> finds the next match,
19799a22 1285returning true if it matches, and false if there is no further match.
7e86de3e
G
1286The position after the last match can be read or set using the pos()
1287function; see L<perlfunc/pos>. A failed match normally resets the
1288search position to the beginning of the string, but you can avoid that
1289by adding the C</c> modifier (e.g. C<m//gc>). Modifying the target
1290string also resets the search position.
c90c0ff4 1291
e9d89077
DN
1292=item \G assertion
1293
c90c0ff4
PP
1294You can intermix C<m//g> matches with C<m/\G.../g>, where C<\G> is a
1295zero-width assertion that matches the exact position where the previous
5d43e42d
DC
1296C<m//g>, if any, left off. Without the C</g> modifier, the C<\G> assertion
1297still anchors at pos(), but the match is of course only attempted once.
1298Using C<\G> without C</g> on a target string that has not previously had a
1299C</g> match applied to it is the same as using the C<\A> assertion to match
fe4b3f22
RGS
1300the beginning of the string. Note also that, currently, C<\G> is only
1301properly supported when anchored at the very beginning of the pattern.
c90c0ff4
PP
1302
1303Examples:
a0d0e21e
LW
1304
1305 # list context
1306 ($one,$five,$fifteen) = (`uptime` =~ /(\d+\.\d+)/g);
1307
1308 # scalar context
5d43e42d 1309 $/ = "";
19799a22
GS
1310 while (defined($paragraph = <>)) {
1311 while ($paragraph =~ /[a-z]['")]*[.!?]+['")]*\s/g) {
1312 $sentences++;
a0d0e21e
LW
1313 }
1314 }
1315 print "$sentences\n";
1316
c90c0ff4 1317 # using m//gc with \G
137443ea 1318 $_ = "ppooqppqq";
44a8e56a
PP
1319 while ($i++ < 2) {
1320 print "1: '";
c90c0ff4 1321 print $1 while /(o)/gc; print "', pos=", pos, "\n";
44a8e56a 1322 print "2: '";
c90c0ff4 1323 print $1 if /\G(q)/gc; print "', pos=", pos, "\n";
44a8e56a 1324 print "3: '";
c90c0ff4 1325 print $1 while /(p)/gc; print "', pos=", pos, "\n";
44a8e56a 1326 }
5d43e42d 1327 print "Final: '$1', pos=",pos,"\n" if /\G(.)/;
44a8e56a
PP
1328
1329The last example should print:
1330
1331 1: 'oo', pos=4
137443ea 1332 2: 'q', pos=5
44a8e56a
PP
1333 3: 'pp', pos=7
1334 1: '', pos=7
137443ea
PP
1335 2: 'q', pos=8
1336 3: '', pos=8
5d43e42d
DC
1337 Final: 'q', pos=8
1338
1339Notice that the final match matched C<q> instead of C<p>, which a match
1340without the C<\G> anchor would have done. Also note that the final match
ac036724 1341did not update C<pos>. C<pos> is only updated on a C</g> match. If the
5d43e42d
DC
1342final match did indeed match C<p>, it's a good bet that you're running an
1343older (pre-5.6.0) Perl.
44a8e56a 1344
c90c0ff4 1345A useful idiom for C<lex>-like scanners is C</\G.../gc>. You can
e7ea3e70 1346combine several regexps like this to process a string part-by-part,
c90c0ff4
PP
1347doing different actions depending on which regexp matched. Each
1348regexp tries to match where the previous one leaves off.
e7ea3e70 1349
3fe9a6f1 1350 $_ = <<'EOL';
46c3340e 1351 $url = URI::URL->new( "http://example.com/" ); die if $url eq "xXx";
3fe9a6f1
PP
1352 EOL
1353 LOOP:
e7ea3e70 1354 {
c90c0ff4
PP
1355 print(" digits"), redo LOOP if /\G\d+\b[,.;]?\s*/gc;
1356 print(" lowercase"), redo LOOP if /\G[a-z]+\b[,.;]?\s*/gc;
1357 print(" UPPERCASE"), redo LOOP if /\G[A-Z]+\b[,.;]?\s*/gc;
1358 print(" Capitalized"), redo LOOP if /\G[A-Z][a-z]+\b[,.;]?\s*/gc;
1359 print(" MiXeD"), redo LOOP if /\G[A-Za-z]+\b[,.;]?\s*/gc;
1360 print(" alphanumeric"), redo LOOP if /\G[A-Za-z0-9]+\b[,.;]?\s*/gc;
1361 print(" line-noise"), redo LOOP if /\G[^A-Za-z0-9]+/gc;
e7ea3e70
IZ
1362 print ". That's all!\n";
1363 }
1364
1365Here is the output (split into several lines):
1366
1367 line-noise lowercase line-noise lowercase UPPERCASE line-noise
1368 UPPERCASE line-noise lowercase line-noise lowercase line-noise
1369 lowercase lowercase line-noise lowercase lowercase line-noise
1370 MiXeD line-noise. That's all!
44a8e56a 1371
87e95b7f
YO
1372=item ?PATTERN?
1373X<?>
1374
1375This is just like the C</pattern/> search, except that it matches only
1376once between calls to the reset() operator. This is a useful
1377optimization when you want to see only the first occurrence of
1378something in each file of a set of files, for instance. Only C<??>
1379patterns local to the current package are reset.
1380
1381 while (<>) {
1382 if (?^$?) {
1383 # blank line between header and body
1384 }
1385 } continue {
1386 reset if eof; # clear ?? status for next file
1387 }
1388
1389This usage is vaguely deprecated, which means it just might possibly
1390be removed in some distant future version of Perl, perhaps somewhere
1391around the year 2168.
1392
1393=item s/PATTERN/REPLACEMENT/msixpogce
1394X<substitute> X<substitution> X<replace> X<regexp, replace>
01c6f5f4 1395X<regexp, substitute> X</m> X</s> X</i> X</x> X</p> X</o> X</g> X</c> X</e>
87e95b7f
YO
1396
1397Searches a string for a pattern, and if found, replaces that pattern
1398with the replacement text and returns the number of substitutions
1399made. Otherwise it returns false (specifically, the empty string).
1400
1401If no string is specified via the C<=~> or C<!~> operator, the C<$_>
1402variable is searched and modified. (The string specified with C<=~> must
1403be scalar variable, an array element, a hash element, or an assignment
1404to one of those, i.e., an lvalue.)
1405
1406If the delimiter chosen is a single quote, no interpolation is
1407done on either the PATTERN or the REPLACEMENT. Otherwise, if the
1408PATTERN contains a $ that looks like a variable rather than an
1409end-of-string test, the variable will be interpolated into the pattern
1410at run-time. If you want the pattern compiled only once the first time
1411the variable is interpolated, use the C</o> option. If the pattern
1412evaluates to the empty string, the last successfully executed regular
1413expression is used instead. See L<perlre> for further explanation on these.
1414See L<perllocale> for discussion of additional considerations that apply
1415when C<use locale> is in effect.
1416
1417Options are as with m// with the addition of the following replacement
1418specific options:
1419
1420 e Evaluate the right side as an expression.
1421 ee Evaluate the right side as a string then eval the result
1422
ed02a3bf
DN
1423Any non-whitespace delimiter may replace the slashes. Add space after
1424the C<s> when using a character allowed in identifiers. If single quotes
1425are used, no interpretation is done on the replacement string (the C</e>
1426modifier overrides this, however). Unlike Perl 4, Perl 5 treats backticks
1427as normal delimiters; the replacement text is not evaluated as a command.
1428If the PATTERN is delimited by bracketing quotes, the REPLACEMENT has
1429its own pair of quotes, which may or may not be bracketing quotes, e.g.,
87e95b7f
YO
1430C<s(foo)(bar)> or C<< s<foo>/bar/ >>. A C</e> will cause the
1431replacement portion to be treated as a full-fledged Perl expression
1432and evaluated right then and there. It is, however, syntax checked at
1433compile-time. A second C<e> modifier will cause the replacement portion
1434to be C<eval>ed before being run as a Perl expression.
1435
1436Examples:
1437
1438 s/\bgreen\b/mauve/g; # don't change wintergreen
1439
1440 $path =~ s|/usr/bin|/usr/local/bin|;
1441
1442 s/Login: $foo/Login: $bar/; # run-time pattern
1443
1444 ($foo = $bar) =~ s/this/that/; # copy first, then change
1445
1446 $count = ($paragraph =~ s/Mister\b/Mr./g); # get change-count
1447
1448 $_ = 'abc123xyz';
1449 s/\d+/$&*2/e; # yields 'abc246xyz'
1450 s/\d+/sprintf("%5d",$&)/e; # yields 'abc 246xyz'
1451 s/\w/$& x 2/eg; # yields 'aabbcc 224466xxyyzz'
1452
1453 s/%(.)/$percent{$1}/g; # change percent escapes; no /e
1454 s/%(.)/$percent{$1} || $&/ge; # expr now, so /e
1455 s/^=(\w+)/pod($1)/ge; # use function call
1456
1457 # expand variables in $_, but dynamics only, using
1458 # symbolic dereferencing
1459 s/\$(\w+)/${$1}/g;
1460
1461 # Add one to the value of any numbers in the string
1462 s/(\d+)/1 + $1/eg;
1463
1464 # This will expand any embedded scalar variable
1465 # (including lexicals) in $_ : First $1 is interpolated
1466 # to the variable name, and then evaluated
1467 s/(\$\w+)/$1/eeg;
1468
1469 # Delete (most) C comments.
1470 $program =~ s {
1471 /\* # Match the opening delimiter.
1472 .*? # Match a minimal number of characters.
1473 \*/ # Match the closing delimiter.
1474 } []gsx;
1475
1476 s/^\s*(.*?)\s*$/$1/; # trim whitespace in $_, expensively
1477
1478 for ($variable) { # trim whitespace in $variable, cheap
1479 s/^\s+//;
1480 s/\s+$//;
1481 }
1482
1483 s/([^ ]*) *([^ ]*)/$2 $1/; # reverse 1st two fields
1484
1485Note the use of $ instead of \ in the last example. Unlike
1486B<sed>, we use the \<I<digit>> form in only the left hand side.
1487Anywhere else it's $<I<digit>>.
1488
1489Occasionally, you can't use just a C</g> to get all the changes
1490to occur that you might want. Here are two common cases:
1491
1492 # put commas in the right places in an integer
1493 1 while s/(\d)(\d\d\d)(?!\d)/$1,$2/g;
1494
1495 # expand tabs to 8-column spacing
1496 1 while s/\t+/' ' x (length($&)*8 - length($`)%8)/e;
1497
1498=back
1499
1500=head2 Quote-Like Operators
1501X<operator, quote-like>
1502
01c6f5f4
RGS
1503=over 4
1504
a0d0e21e 1505=item q/STRING/
5d44bfff 1506X<q> X<quote, single> X<'> X<''>
a0d0e21e 1507
5d44bfff 1508=item 'STRING'
a0d0e21e 1509
19799a22 1510A single-quoted, literal string. A backslash represents a backslash
68dc0745
PP
1511unless followed by the delimiter or another backslash, in which case
1512the delimiter or backslash is interpolated.
a0d0e21e
LW
1513
1514 $foo = q!I said, "You said, 'She said it.'"!;
1515 $bar = q('This is it.');
68dc0745 1516 $baz = '\n'; # a two-character string
a0d0e21e
LW
1517
1518=item qq/STRING/
d74e8afc 1519X<qq> X<quote, double> X<"> X<"">
a0d0e21e
LW
1520
1521=item "STRING"
1522
1523A double-quoted, interpolated string.
1524
1525 $_ .= qq
1526 (*** The previous line contains the naughty word "$1".\n)
19799a22 1527 if /\b(tcl|java|python)\b/i; # :-)
68dc0745 1528 $baz = "\n"; # a one-character string
a0d0e21e
LW
1529
1530=item qx/STRING/
d74e8afc 1531X<qx> X<`> X<``> X<backtick>
a0d0e21e
LW
1532
1533=item `STRING`
1534
43dd4d21
JH
1535A string which is (possibly) interpolated and then executed as a
1536system command with C</bin/sh> or its equivalent. Shell wildcards,
1537pipes, and redirections will be honored. The collected standard
1538output of the command is returned; standard error is unaffected. In
1539scalar context, it comes back as a single (potentially multi-line)
1540string, or undef if the command failed. In list context, returns a
1541list of lines (however you've defined lines with $/ or
1542$INPUT_RECORD_SEPARATOR), or an empty list if the command failed.
5a964f20
TC
1543
1544Because backticks do not affect standard error, use shell file descriptor
1545syntax (assuming the shell supports this) if you care to address this.
1546To capture a command's STDERR and STDOUT together:
a0d0e21e 1547
5a964f20
TC
1548 $output = `cmd 2>&1`;
1549
1550To capture a command's STDOUT but discard its STDERR:
1551
1552 $output = `cmd 2>/dev/null`;
1553
1554To capture a command's STDERR but discard its STDOUT (ordering is
1555important here):
1556
1557 $output = `cmd 2>&1 1>/dev/null`;
1558
1559To exchange a command's STDOUT and STDERR in order to capture the STDERR
1560but leave its STDOUT to come out the old STDERR:
1561
1562 $output = `cmd 3>&1 1>&2 2>&3 3>&-`;
1563
1564To read both a command's STDOUT and its STDERR separately, it's easiest
2359510d
SD
1565to redirect them separately to files, and then read from those files
1566when the program is done:
5a964f20 1567
2359510d 1568 system("program args 1>program.stdout 2>program.stderr");
5a964f20 1569
30398227
SP
1570The STDIN filehandle used by the command is inherited from Perl's STDIN.
1571For example:
1572
1573 open BLAM, "blam" || die "Can't open: $!";
1574 open STDIN, "<&BLAM";
1575 print `sort`;
1576
1577will print the sorted contents of the file "blam".
1578
5a964f20
TC
1579Using single-quote as a delimiter protects the command from Perl's
1580double-quote interpolation, passing it on to the shell instead:
1581
1582 $perl_info = qx(ps $$); # that's Perl's $$
1583 $shell_info = qx'ps $$'; # that's the new shell's $$
1584
19799a22 1585How that string gets evaluated is entirely subject to the command
5a964f20
TC
1586interpreter on your system. On most platforms, you will have to protect
1587shell metacharacters if you want them treated literally. This is in
1588practice difficult to do, as it's unclear how to escape which characters.
1589See L<perlsec> for a clean and safe example of a manual fork() and exec()
1590to emulate backticks safely.
a0d0e21e 1591
bb32b41a
GS
1592On some platforms (notably DOS-like ones), the shell may not be
1593capable of dealing with multiline commands, so putting newlines in
1594the string may not get you what you want. You may be able to evaluate
1595multiple commands in a single line by separating them with the command
1596separator character, if your shell supports that (e.g. C<;> on many Unix
1597shells; C<&> on the Windows NT C<cmd> shell).
1598
0f897271
GS
1599Beginning with v5.6.0, Perl will attempt to flush all files opened for
1600output before starting the child process, but this may not be supported
1601on some platforms (see L<perlport>). To be safe, you may need to set
1602C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method of
1603C<IO::Handle> on any open handles.
1604
bb32b41a
GS
1605Beware that some command shells may place restrictions on the length
1606of the command line. You must ensure your strings don't exceed this
1607limit after any necessary interpolations. See the platform-specific
1608release notes for more details about your particular environment.
1609
5a964f20
TC
1610Using this operator can lead to programs that are difficult to port,
1611because the shell commands called vary between systems, and may in
1612fact not be present at all. As one example, the C<type> command under
1613the POSIX shell is very different from the C<type> command under DOS.
1614That doesn't mean you should go out of your way to avoid backticks
1615when they're the right way to get something done. Perl was made to be
1616a glue language, and one of the things it glues together is commands.
1617Just understand what you're getting yourself into.
bb32b41a 1618
da87341d 1619See L</"I/O Operators"> for more discussion.
a0d0e21e 1620
945c54fd 1621=item qw/STRING/
d74e8afc 1622X<qw> X<quote, list> X<quote, words>
945c54fd
JH
1623
1624Evaluates to a list of the words extracted out of STRING, using embedded
1625whitespace as the word delimiters. It can be understood as being roughly
1626equivalent to:
1627
1628 split(' ', q/STRING/);
1629
efb1e162
CW
1630the differences being that it generates a real list at compile time, and
1631in scalar context it returns the last element in the list. So
945c54fd
JH
1632this expression:
1633
1634 qw(foo bar baz)
1635
1636is semantically equivalent to the list:
1637
1638 'foo', 'bar', 'baz'
1639
1640Some frequently seen examples:
1641
1642 use POSIX qw( setlocale localeconv )
1643 @EXPORT = qw( foo bar baz );
1644
1645A common mistake is to try to separate the words with comma or to
1646put comments into a multi-line C<qw>-string. For this reason, the
89d205f2 1647C<use warnings> pragma and the B<-w> switch (that is, the C<$^W> variable)
945c54fd
JH
1648produces warnings if the STRING contains the "," or the "#" character.
1649
a0d0e21e 1650
6940069f 1651=item tr/SEARCHLIST/REPLACEMENTLIST/cds
d74e8afc 1652X<tr> X<y> X<transliterate> X</c> X</d> X</s>
a0d0e21e 1653
6940069f 1654=item y/SEARCHLIST/REPLACEMENTLIST/cds
a0d0e21e 1655
2c268ad5 1656Transliterates all occurrences of the characters found in the search list
a0d0e21e
LW
1657with the corresponding character in the replacement list. It returns
1658the number of characters replaced or deleted. If no string is
2c268ad5 1659specified via the =~ or !~ operator, the $_ string is transliterated. (The
54310121
PP
1660string specified with =~ must be a scalar variable, an array element, a
1661hash element, or an assignment to one of those, i.e., an lvalue.)
8ada0baa 1662
89d205f2 1663A character range may be specified with a hyphen, so C<tr/A-J/0-9/>
2c268ad5 1664does the same replacement as C<tr/ACEGIBDFHJ/0246813579/>.
54310121
PP
1665For B<sed> devotees, C<y> is provided as a synonym for C<tr>. If the
1666SEARCHLIST is delimited by bracketing quotes, the REPLACEMENTLIST has
1667its own pair of quotes, which may or may not be bracketing quotes,
2c268ad5 1668e.g., C<tr[A-Z][a-z]> or C<tr(+\-*/)/ABCD/>.
a0d0e21e 1669
cc255d5f 1670Note that C<tr> does B<not> do regular expression character classes
e0c83546 1671such as C<\d> or C<[:lower:]>. The C<tr> operator is not equivalent to
cc255d5f
JH
1672the tr(1) utility. If you want to map strings between lower/upper
1673cases, see L<perlfunc/lc> and L<perlfunc/uc>, and in general consider
1674using the C<s> operator if you need regular expressions.
1675
8ada0baa
JH
1676Note also that the whole range idea is rather unportable between
1677character sets--and even within character sets they may cause results
1678you probably didn't expect. A sound principle is to use only ranges
1679that begin from and end at either alphabets of equal case (a-e, A-E),
1680or digits (0-4). Anything else is unsafe. If in doubt, spell out the
1681character sets in full.
1682
a0d0e21e
LW
1683Options:
1684
1685 c Complement the SEARCHLIST.
1686 d Delete found but unreplaced characters.
1687 s Squash duplicate replaced characters.
1688
19799a22
GS
1689If the C</c> modifier is specified, the SEARCHLIST character set
1690is complemented. If the C</d> modifier is specified, any characters
1691specified by SEARCHLIST not found in REPLACEMENTLIST are deleted.
1692(Note that this is slightly more flexible than the behavior of some
1693B<tr> programs, which delete anything they find in the SEARCHLIST,
1694period.) If the C</s> modifier is specified, sequences of characters
1695that were transliterated to the same character are squashed down
1696to a single instance of the character.
a0d0e21e
LW
1697
1698If the C</d> modifier is used, the REPLACEMENTLIST is always interpreted
1699exactly as specified. Otherwise, if the REPLACEMENTLIST is shorter
1700than the SEARCHLIST, the final character is replicated till it is long
5a964f20 1701enough. If the REPLACEMENTLIST is empty, the SEARCHLIST is replicated.
a0d0e21e
LW
1702This latter is useful for counting characters in a class or for
1703squashing character sequences in a class.
1704
1705Examples:
1706
1707 $ARGV[1] =~ tr/A-Z/a-z/; # canonicalize to lower case
1708
1709 $cnt = tr/*/*/; # count the stars in $_
1710
1711 $cnt = $sky =~ tr/*/*/; # count the stars in $sky
1712
1713 $cnt = tr/0-9//; # count the digits in $_
1714
1715 tr/a-zA-Z//s; # bookkeeper -> bokeper
1716
1717 ($HOST = $host) =~ tr/a-z/A-Z/;
1718
1719 tr/a-zA-Z/ /cs; # change non-alphas to single space
1720
1721 tr [\200-\377]
1722 [\000-\177]; # delete 8th bit
1723
19799a22
GS
1724If multiple transliterations are given for a character, only the
1725first one is used:
748a9306
LW
1726
1727 tr/AAA/XYZ/
1728
2c268ad5 1729will transliterate any A to X.
748a9306 1730
19799a22 1731Because the transliteration table is built at compile time, neither
a0d0e21e 1732the SEARCHLIST nor the REPLACEMENTLIST are subjected to double quote
19799a22
GS
1733interpolation. That means that if you want to use variables, you
1734must use an eval():
a0d0e21e
LW
1735
1736 eval "tr/$oldlist/$newlist/";
1737 die $@ if $@;
1738
1739 eval "tr/$oldlist/$newlist/, 1" or die $@;
1740
7e3b091d 1741=item <<EOF
d74e8afc 1742X<here-doc> X<heredoc> X<here-document> X<<< << >>>
7e3b091d
DA
1743
1744A line-oriented form of quoting is based on the shell "here-document"
1745syntax. Following a C<< << >> you specify a string to terminate
1746the quoted material, and all lines following the current line down to
89d205f2
YO
1747the terminating string are the value of the item.
1748
1749The terminating string may be either an identifier (a word), or some
1750quoted text. An unquoted identifier works like double quotes.
1751There may not be a space between the C<< << >> and the identifier,
1752unless the identifier is explicitly quoted. (If you put a space it
1753will be treated as a null identifier, which is valid, and matches the
1754first empty line.) The terminating string must appear by itself
1755(unquoted and with no surrounding whitespace) on the terminating line.
1756
1757If the terminating string is quoted, the type of quotes used determine
1758the treatment of the text.
1759
1760=over 4
1761
1762=item Double Quotes
1763
1764Double quotes indicate that the text will be interpolated using exactly
1765the same rules as normal double quoted strings.
7e3b091d
DA
1766
1767 print <<EOF;
1768 The price is $Price.
1769 EOF
1770
1771 print << "EOF"; # same as above
1772 The price is $Price.
1773 EOF
1774
89d205f2
YO
1775
1776=item Single Quotes
1777
1778Single quotes indicate the text is to be treated literally with no
1779interpolation of its content. This is similar to single quoted
1780strings except that backslashes have no special meaning, with C<\\>
1781being treated as two backslashes and not one as they would in every
1782other quoting construct.
1783
1784This is the only form of quoting in perl where there is no need
1785to worry about escaping content, something that code generators
1786can and do make good use of.
1787
1788=item Backticks
1789
1790The content of the here doc is treated just as it would be if the
1791string were embedded in backticks. Thus the content is interpolated
1792as though it were double quoted and then executed via the shell, with
1793the results of the execution returned.
1794
1795 print << `EOC`; # execute command and get results
7e3b091d 1796 echo hi there
7e3b091d
DA
1797 EOC
1798
89d205f2
YO
1799=back
1800
1801It is possible to stack multiple here-docs in a row:
1802
7e3b091d
DA
1803 print <<"foo", <<"bar"; # you can stack them
1804 I said foo.
1805 foo
1806 I said bar.
1807 bar
1808
1809 myfunc(<< "THIS", 23, <<'THAT');
1810 Here's a line
1811 or two.
1812 THIS
1813 and here's another.
1814 THAT
1815
1816Just don't forget that you have to put a semicolon on the end
1817to finish the statement, as Perl doesn't know you're not going to
1818try to do this:
1819
1820 print <<ABC
1821 179231
1822 ABC
1823 + 20;
1824
872d7e53
ST
1825If you want to remove the line terminator from your here-docs,
1826use C<chomp()>.
1827
1828 chomp($string = <<'END');
1829 This is a string.
1830 END
1831
1832If you want your here-docs to be indented with the rest of the code,
1833you'll need to remove leading whitespace from each line manually:
7e3b091d
DA
1834
1835 ($quote = <<'FINIS') =~ s/^\s+//gm;
89d205f2 1836 The Road goes ever on and on,
7e3b091d
DA
1837 down from the door where it began.
1838 FINIS
1839
1840If you use a here-doc within a delimited construct, such as in C<s///eg>,
1841the quoted material must come on the lines following the final delimiter.
1842So instead of
1843
1844 s/this/<<E . 'that'
1845 the other
1846 E
1847 . 'more '/eg;
1848
1849you have to write
1850
89d205f2
YO
1851 s/this/<<E . 'that'
1852 . 'more '/eg;
1853 the other
1854 E
7e3b091d
DA
1855
1856If the terminating identifier is on the last line of the program, you
1857must be sure there is a newline after it; otherwise, Perl will give the
1858warning B<Can't find string terminator "END" anywhere before EOF...>.
1859
89d205f2 1860Additionally, the quoting rules for the end of string identifier are not
ac036724 1861related to Perl's quoting rules. C<q()>, C<qq()>, and the like are not
89d205f2
YO
1862supported in place of C<''> and C<"">, and the only interpolation is for
1863backslashing the quoting character:
7e3b091d
DA
1864
1865 print << "abc\"def";
1866 testing...
1867 abc"def
1868
1869Finally, quoted strings cannot span multiple lines. The general rule is
1870that the identifier must be a string literal. Stick with that, and you
1871should be safe.
1872
a0d0e21e
LW
1873=back
1874
75e14d17 1875=head2 Gory details of parsing quoted constructs
d74e8afc 1876X<quote, gory details>
75e14d17 1877
19799a22
GS
1878When presented with something that might have several different
1879interpretations, Perl uses the B<DWIM> (that's "Do What I Mean")
1880principle to pick the most probable interpretation. This strategy
1881is so successful that Perl programmers often do not suspect the
1882ambivalence of what they write. But from time to time, Perl's
1883notions differ substantially from what the author honestly meant.
1884
1885This section hopes to clarify how Perl handles quoted constructs.
1886Although the most common reason to learn this is to unravel labyrinthine
1887regular expressions, because the initial steps of parsing are the
1888same for all quoting operators, they are all discussed together.
1889
1890The most important Perl parsing rule is the first one discussed
1891below: when processing a quoted construct, Perl first finds the end
1892of that construct, then interprets its contents. If you understand
1893this rule, you may skip the rest of this section on the first
1894reading. The other rules are likely to contradict the user's
1895expectations much less frequently than this first one.
1896
1897Some passes discussed below are performed concurrently, but because
1898their results are the same, we consider them individually. For different
1899quoting constructs, Perl performs different numbers of passes, from
6deea57f 1900one to four, but these passes are always performed in the same order.
75e14d17 1901
13a2d996 1902=over 4
75e14d17
IZ
1903
1904=item Finding the end
1905
6deea57f
ST
1906The first pass is finding the end of the quoted construct, where
1907the information about the delimiters is used in parsing.
1908During this search, text between the starting and ending delimiters
1909is copied to a safe location. The text copied gets delimiter-independent.
1910
1911If the construct is a here-doc, the ending delimiter is a line
1912that has a terminating string as the content. Therefore C<<<EOF> is
1913terminated by C<EOF> immediately followed by C<"\n"> and starting
1914from the first column of the terminating line.
1915When searching for the terminating line of a here-doc, nothing
1916is skipped. In other words, lines after the here-doc syntax
1917are compared with the terminating string line by line.
1918
1919For the constructs except here-docs, single characters are used as starting
1920and ending delimiters. If the starting delimiter is an opening punctuation
1921(that is C<(>, C<[>, C<{>, or C<< < >>), the ending delimiter is the
1922corresponding closing punctuation (that is C<)>, C<]>, C<}>, or C<< > >>).
1923If the starting delimiter is an unpaired character like C</> or a closing
1924punctuation, the ending delimiter is same as the starting delimiter.
1925Therefore a C</> terminates a C<qq//> construct, while a C<]> terminates
1926C<qq[]> and C<qq]]> constructs.
1927
1928When searching for single-character delimiters, escaped delimiters
1929and C<\\> are skipped. For example, while searching for terminating C</>,
1930combinations of C<\\> and C<\/> are skipped. If the delimiters are
1931bracketing, nested pairs are also skipped. For example, while searching
1932for closing C<]> paired with the opening C<[>, combinations of C<\\>, C<\]>,
1933and C<\[> are all skipped, and nested C<[> and C<]> are skipped as well.
1934However, when backslashes are used as the delimiters (like C<qq\\> and
1935C<tr\\\>), nothing is skipped.
1936During the search for the end, backslashes that escape delimiters
1937are removed (exactly speaking, they are not copied to the safe location).
75e14d17 1938
19799a22
GS
1939For constructs with three-part delimiters (C<s///>, C<y///>, and
1940C<tr///>), the search is repeated once more.
6deea57f
ST
1941If the first delimiter is not an opening punctuation, three delimiters must
1942be same such as C<s!!!> and C<tr)))>, in which case the second delimiter
1943terminates the left part and starts the right part at once.
1944If the left part is delimited by bracketing punctuations (that is C<()>,
1945C<[]>, C<{}>, or C<< <> >>), the right part needs another pair of
1946delimiters such as C<s(){}> and C<tr[]//>. In these cases, whitespaces
1947and comments are allowed between both parts, though the comment must follow
1948at least one whitespace; otherwise a character expected as the start of
1949the comment may be regarded as the starting delimiter of the right part.
75e14d17 1950
19799a22
GS
1951During this search no attention is paid to the semantics of the construct.
1952Thus:
75e14d17
IZ
1953
1954 "$hash{"$foo/$bar"}"
1955
2a94b7ce 1956or:
75e14d17 1957
89d205f2 1958 m/
2a94b7ce 1959 bar # NOT a comment, this slash / terminated m//!
75e14d17
IZ
1960 /x
1961
19799a22
GS
1962do not form legal quoted expressions. The quoted part ends on the
1963first C<"> and C</>, and the rest happens to be a syntax error.
1964Because the slash that terminated C<m//> was followed by a C<SPACE>,
1965the example above is not C<m//x>, but rather C<m//> with no C</x>
1966modifier. So the embedded C<#> is interpreted as a literal C<#>.
75e14d17 1967
89d205f2
YO
1968Also no attention is paid to C<\c\> (multichar control char syntax) during
1969this search. Thus the second C<\> in C<qq/\c\/> is interpreted as a part
1970of C<\/>, and the following C</> is not recognized as a delimiter.
0d594e51
ST
1971Instead, use C<\034> or C<\x1c> at the end of quoted constructs.
1972
75e14d17 1973=item Interpolation
d74e8afc 1974X<interpolation>
75e14d17 1975
19799a22 1976The next step is interpolation in the text obtained, which is now
89d205f2 1977delimiter-independent. There are multiple cases.
75e14d17 1978
13a2d996 1979=over 4
75e14d17 1980
89d205f2 1981=item C<<<'EOF'>
75e14d17
IZ
1982
1983No interpolation is performed.
6deea57f
ST
1984Note that the combination C<\\> is left intact, since escaped delimiters
1985are not available for here-docs.
75e14d17 1986
6deea57f 1987=item C<m''>, the pattern of C<s'''>
89d205f2 1988
6deea57f
ST
1989No interpolation is performed at this stage.
1990Any backslashed sequences including C<\\> are treated at the stage
1991to L</"parsing regular expressions">.
89d205f2 1992
6deea57f 1993=item C<''>, C<q//>, C<tr'''>, C<y'''>, the replacement of C<s'''>
75e14d17 1994
89d205f2 1995The only interpolation is removal of C<\> from pairs of C<\\>.
6deea57f
ST
1996Therefore C<-> in C<tr'''> and C<y'''> is treated literally
1997as a hyphen and no character range is available.
1998C<\1> in the replacement of C<s'''> does not work as C<$1>.
89d205f2
YO
1999
2000=item C<tr///>, C<y///>
2001
6deea57f
ST
2002No variable interpolation occurs. String modifying combinations for
2003case and quoting such as C<\Q>, C<\U>, and C<\E> are not recognized.
2004The other escape sequences such as C<\200> and C<\t> and backslashed
2005characters such as C<\\> and C<\-> are converted to appropriate literals.
89d205f2
YO
2006The character C<-> is treated specially and therefore C<\-> is treated
2007as a literal C<->.
75e14d17 2008
89d205f2 2009=item C<"">, C<``>, C<qq//>, C<qx//>, C<< <file*glob> >>, C<<<"EOF">
75e14d17 2010
19799a22
GS
2011C<\Q>, C<\U>, C<\u>, C<\L>, C<\l> (possibly paired with C<\E>) are
2012converted to corresponding Perl constructs. Thus, C<"$foo\Qbaz$bar">
2013is converted to C<$foo . (quotemeta("baz" . $bar))> internally.
6deea57f
ST
2014The other escape sequences such as C<\200> and C<\t> and backslashed
2015characters such as C<\\> and C<\-> are replaced with appropriate
2016expansions.
2a94b7ce 2017
19799a22
GS
2018Let it be stressed that I<whatever falls between C<\Q> and C<\E>>
2019is interpolated in the usual way. Something like C<"\Q\\E"> has
2020no C<\E> inside. instead, it has C<\Q>, C<\\>, and C<E>, so the
2021result is the same as for C<"\\\\E">. As a general rule, backslashes
2022between C<\Q> and C<\E> may lead to counterintuitive results. So,
2023C<"\Q\t\E"> is converted to C<quotemeta("\t")>, which is the same
2024as C<"\\\t"> (since TAB is not alphanumeric). Note also that:
2a94b7ce
IZ
2025
2026 $str = '\t';
2027 return "\Q$str";
2028
2029may be closer to the conjectural I<intention> of the writer of C<"\Q\t\E">.
2030
19799a22 2031Interpolated scalars and arrays are converted internally to the C<join> and
92d29cee 2032C<.> catenation operations. Thus, C<"$foo XXX '@arr'"> becomes:
75e14d17 2033
19799a22 2034 $foo . " XXX '" . (join $", @arr) . "'";
75e14d17 2035
19799a22 2036All operations above are performed simultaneously, left to right.
75e14d17 2037
19799a22
GS
2038Because the result of C<"\Q STRING \E"> has all metacharacters
2039quoted, there is no way to insert a literal C<$> or C<@> inside a
2040C<\Q\E> pair. If protected by C<\>, C<$> will be quoted to became
2041C<"\\\$">; if not, it is interpreted as the start of an interpolated
2042scalar.
75e14d17 2043
19799a22 2044Note also that the interpolation code needs to make a decision on
89d205f2 2045where the interpolated scalar ends. For instance, whether
35f2feb0 2046C<< "a $b -> {c}" >> really means:
75e14d17
IZ
2047
2048 "a " . $b . " -> {c}";
2049
2a94b7ce 2050or:
75e14d17
IZ
2051
2052 "a " . $b -> {c};
2053
19799a22
GS
2054Most of the time, the longest possible text that does not include
2055spaces between components and which contains matching braces or
2056brackets. because the outcome may be determined by voting based
2057on heuristic estimators, the result is not strictly predictable.
2058Fortunately, it's usually correct for ambiguous cases.
75e14d17 2059
6deea57f 2060=item the replacement of C<s///>
75e14d17 2061
19799a22 2062Processing of C<\Q>, C<\U>, C<\u>, C<\L>, C<\l>, and interpolation
6deea57f
ST
2063happens as with C<qq//> constructs.
2064
2065It is at this step that C<\1> is begrudgingly converted to C<$1> in
2066the replacement text of C<s///>, in order to correct the incorrigible
2067I<sed> hackers who haven't picked up the saner idiom yet. A warning
2068is emitted if the C<use warnings> pragma or the B<-w> command-line flag
2069(that is, the C<$^W> variable) was set.
2070
2071=item C<RE> in C<?RE?>, C</RE/>, C<m/RE/>, C<s/RE/foo/>,
2072
cc74c5bd
ST
2073Processing of C<\Q>, C<\U>, C<\u>, C<\L>, C<\l>, C<\E>,
2074and interpolation happens (almost) as with C<qq//> constructs.
2075
5d03b57c
KW
2076Processing of C<\N{...}> is also done here, and compiled into an intermediate
2077form for the regex compiler. (This is because, as mentioned below, the regex
2078compilation may be done at execution time, and C<\N{...}> is a compile-time
2079construct.)
2080
cc74c5bd
ST
2081However any other combinations of C<\> followed by a character
2082are not substituted but only skipped, in order to parse them
2083as regular expressions at the following step.
6deea57f 2084As C<\c> is skipped at this step, C<@> of C<\c@> in RE is possibly
1749ea0d 2085treated as an array symbol (for example C<@foo>),
6deea57f 2086even though the same text in C<qq//> gives interpolation of C<\c@>.
6deea57f
ST
2087
2088Moreover, inside C<(?{BLOCK})>, C<(?# comment )>, and
19799a22
GS
2089a C<#>-comment in a C<//x>-regular expression, no processing is
2090performed whatsoever. This is the first step at which the presence
2091of the C<//x> modifier is relevant.
2092
1749ea0d
ST
2093Interpolation in patterns has several quirks: C<$|>, C<$(>, C<$)>, C<@+>
2094and C<@-> are not interpolated, and constructs C<$var[SOMETHING]> are
2095voted (by several different estimators) to be either an array element
2096or C<$var> followed by an RE alternative. This is where the notation
19799a22
GS
2097C<${arr[$bar]}> comes handy: C</${arr[0-9]}/> is interpreted as
2098array element C<-9>, not as a regular expression from the variable
2099C<$arr> followed by a digit, which would be the interpretation of
2100C</$arr[0-9]/>. Since voting among different estimators may occur,
2101the result is not predictable.
2102
19799a22
GS
2103The lack of processing of C<\\> creates specific restrictions on
2104the post-processed text. If the delimiter is C</>, one cannot get
2105the combination C<\/> into the result of this step. C</> will
2106finish the regular expression, C<\/> will be stripped to C</> on
2107the previous step, and C<\\/> will be left as is. Because C</> is
2108equivalent to C<\/> inside a regular expression, this does not
2109matter unless the delimiter happens to be character special to the
2110RE engine, such as in C<s*foo*bar*>, C<m[foo]>, or C<?foo?>; or an
2111alphanumeric char, as in:
2a94b7ce
IZ
2112
2113 m m ^ a \s* b mmx;
2114
19799a22 2115In the RE above, which is intentionally obfuscated for illustration, the
6deea57f 2116delimiter is C<m>, the modifier is C<mx>, and after delimiter-removal the
89d205f2 2117RE is the same as for C<m/ ^ a \s* b /mx>. There's more than one
19799a22
GS
2118reason you're encouraged to restrict your delimiters to non-alphanumeric,
2119non-whitespace choices.
75e14d17
IZ
2120
2121=back
2122
19799a22 2123This step is the last one for all constructs except regular expressions,
75e14d17
IZ
2124which are processed further.
2125
6deea57f
ST
2126=item parsing regular expressions
2127X<regexp, parse>
75e14d17 2128
19799a22 2129Previous steps were performed during the compilation of Perl code,
ac036724 2130but this one happens at run time, although it may be optimized to
19799a22 2131be calculated at compile time if appropriate. After preprocessing
6deea57f 2132described above, and possibly after evaluation if concatenation,
19799a22
GS
2133joining, casing translation, or metaquoting are involved, the
2134resulting I<string> is passed to the RE engine for compilation.
2135
2136Whatever happens in the RE engine might be better discussed in L<perlre>,
2137but for the sake of continuity, we shall do so here.
2138
2139This is another step where the presence of the C<//x> modifier is
2140relevant. The RE engine scans the string from left to right and
2141converts it to a finite automaton.
2142
2143Backslashed characters are either replaced with corresponding
2144literal strings (as with C<\{>), or else they generate special nodes
2145in the finite automaton (as with C<\b>). Characters special to the
2146RE engine (such as C<|>) generate corresponding nodes or groups of
2147nodes. C<(?#...)> comments are ignored. All the rest is either
2148converted to literal strings to match, or else is ignored (as is
2149whitespace and C<#>-style comments if C<//x> is present).
2150
2151Parsing of the bracketed character class construct, C<[...]>, is
2152rather different than the rule used for the rest of the pattern.
2153The terminator of this construct is found using the same rules as
2154for finding the terminator of a C<{}>-delimited construct, the only
2155exception being that C<]> immediately following C<[> is treated as
2156though preceded by a backslash. Similarly, the terminator of
2157C<(?{...})> is found using the same rules as for finding the
2158terminator of a C<{}>-delimited construct.
2159
2160It is possible to inspect both the string given to RE engine and the
2161resulting finite automaton. See the arguments C<debug>/C<debugcolor>
2162in the C<use L<re>> pragma, as well as Perl's B<-Dr> command-line
4a4eefd0 2163switch documented in L<perlrun/"Command Switches">.
75e14d17
IZ
2164
2165=item Optimization of regular expressions
d74e8afc 2166X<regexp, optimization>
75e14d17 2167
7522fed5 2168This step is listed for completeness only. Since it does not change
75e14d17 2169semantics, details of this step are not documented and are subject
19799a22
GS
2170to change without notice. This step is performed over the finite
2171automaton that was generated during the previous pass.
2a94b7ce 2172
19799a22
GS
2173It is at this stage that C<split()> silently optimizes C</^/> to
2174mean C</^/m>.
75e14d17
IZ
2175
2176=back
2177
a0d0e21e 2178=head2 I/O Operators
d74e8afc
ITB
2179X<operator, i/o> X<operator, io> X<io> X<while> X<filehandle>
2180X<< <> >> X<@ARGV>
a0d0e21e 2181
54310121 2182There are several I/O operators you should know about.
fbad3eb5 2183
7b8d334a 2184A string enclosed by backticks (grave accents) first undergoes
19799a22
GS
2185double-quote interpolation. It is then interpreted as an external
2186command, and the output of that command is the value of the
e9c56f9b
JH
2187backtick string, like in a shell. In scalar context, a single string
2188consisting of all output is returned. In list context, a list of
2189values is returned, one per line of output. (You can set C<$/> to use
2190a different line terminator.) The command is executed each time the
2191pseudo-literal is evaluated. The status value of the command is
2192returned in C<$?> (see L<perlvar> for the interpretation of C<$?>).
2193Unlike in B<csh>, no translation is done on the return data--newlines
2194remain newlines. Unlike in any of the shells, single quotes do not
2195hide variable names in the command from interpretation. To pass a
2196literal dollar-sign through to the shell you need to hide it with a
2197backslash. The generalized form of backticks is C<qx//>. (Because
2198backticks always undergo shell expansion as well, see L<perlsec> for
2199security concerns.)
d74e8afc 2200X<qx> X<`> X<``> X<backtick> X<glob>
19799a22
GS
2201
2202In scalar context, evaluating a filehandle in angle brackets yields
2203the next line from that file (the newline, if any, included), or
2204C<undef> at end-of-file or on error. When C<$/> is set to C<undef>
2205(sometimes known as file-slurp mode) and the file is empty, it
2206returns C<''> the first time, followed by C<undef> subsequently.
2207
2208Ordinarily you must assign the returned value to a variable, but
2209there is one situation where an automatic assignment happens. If
2210and only if the input symbol is the only thing inside the conditional
2211of a C<while> statement (even if disguised as a C<for(;;)> loop),
2212the value is automatically assigned to the global variable $_,
2213destroying whatever was there previously. (This may seem like an
2214odd thing to you, but you'll use the construct in almost every Perl
17b829fa 2215script you write.) The $_ variable is not implicitly localized.
19799a22
GS
2216You'll have to put a C<local $_;> before the loop if you want that
2217to happen.
2218
2219The following lines are equivalent:
a0d0e21e 2220
748a9306 2221 while (defined($_ = <STDIN>)) { print; }
7b8d334a 2222 while ($_ = <STDIN>) { print; }
a0d0e21e
LW
2223 while (<STDIN>) { print; }
2224 for (;<STDIN>;) { print; }
748a9306 2225 print while defined($_ = <STDIN>);
7b8d334a 2226 print while ($_ = <STDIN>);
a0d0e21e
LW
2227 print while <STDIN>;
2228
19799a22 2229This also behaves similarly, but avoids $_ :
7b8d334a 2230
89d205f2 2231 while (my $line = <STDIN>) { print $line }
7b8d334a 2232
19799a22
GS
2233In these loop constructs, the assigned value (whether assignment
2234is automatic or explicit) is then tested to see whether it is
2235defined. The defined test avoids problems where line has a string
2236value that would be treated as false by Perl, for example a "" or
2237a "0" with no trailing newline. If you really mean for such values
2238to terminate the loop, they should be tested for explicitly:
7b8d334a
GS
2239
2240 while (($_ = <STDIN>) ne '0') { ... }
2241 while (<STDIN>) { last unless $_; ... }
2242
5ef4d93e 2243In other boolean contexts, C<< <filehandle> >> without an
2244explicit C<defined> test or comparison elicits a warning if the
9f1b1f2d 2245C<use warnings> pragma or the B<-w>
19799a22 2246command-line switch (the C<$^W> variable) is in effect.
7b8d334a 2247
5f05dabc 2248The filehandles STDIN, STDOUT, and STDERR are predefined. (The
19799a22
GS
2249filehandles C<stdin>, C<stdout>, and C<stderr> will also work except
2250in packages, where they would be interpreted as local identifiers
2251rather than global.) Additional filehandles may be created with
2252the open() function, amongst others. See L<perlopentut> and
2253L<perlfunc/open> for details on this.
d74e8afc 2254X<stdin> X<stdout> X<sterr>
a0d0e21e 2255
35f2feb0 2256If a <FILEHANDLE> is used in a context that is looking for
19799a22
GS
2257a list, a list comprising all input lines is returned, one line per
2258list element. It's easy to grow to a rather large data space this
2259way, so use with care.
a0d0e21e 2260
35f2feb0 2261<FILEHANDLE> may also be spelled C<readline(*FILEHANDLE)>.
19799a22 2262See L<perlfunc/readline>.
fbad3eb5 2263
35f2feb0
GS
2264The null filehandle <> is special: it can be used to emulate the
2265behavior of B<sed> and B<awk>. Input from <> comes either from
a0d0e21e 2266standard input, or from each file listed on the command line. Here's
35f2feb0 2267how it works: the first time <> is evaluated, the @ARGV array is
5a964f20 2268checked, and if it is empty, C<$ARGV[0]> is set to "-", which when opened
a0d0e21e
LW
2269gives you standard input. The @ARGV array is then processed as a list
2270of filenames. The loop
2271
2272 while (<>) {
2273 ... # code for each line
2274 }
2275
2276is equivalent to the following Perl-like pseudo code:
2277
3e3baf6d 2278 unshift(@ARGV, '-') unless @ARGV;
a0d0e21e
LW
2279 while ($ARGV = shift) {
2280 open(ARGV, $ARGV);
2281 while (<ARGV>) {
2282 ... # code for each line
2283 }
2284 }
2285
19799a22
GS
2286except that it isn't so cumbersome to say, and will actually work.
2287It really does shift the @ARGV array and put the current filename
2288into the $ARGV variable. It also uses filehandle I<ARGV>
ac036724 2289internally. <> is just a synonym for <ARGV>, which
19799a22 2290is magical. (The pseudo code above doesn't work because it treats
35f2feb0 2291<ARGV> as non-magical.)
a0d0e21e 2292
48ab5743
ML
2293Since the null filehandle uses the two argument form of L<perlfunc/open>
2294it interprets special characters, so if you have a script like this:
2295
2296 while (<>) {
2297 print;
2298 }
2299
2300and call it with C<perl dangerous.pl 'rm -rfv *|'>, it actually opens a
2301pipe, executes the C<rm> command and reads C<rm>'s output from that pipe.
2302If you want all items in C<@ARGV> to be interpreted as file names, you
2303can use the module C<ARGV::readonly> from CPAN.
2304
35f2feb0 2305You can modify @ARGV before the first <> as long as the array ends up
a0d0e21e 2306containing the list of filenames you really want. Line numbers (C<$.>)
19799a22
GS
2307continue as though the input were one big happy file. See the example
2308in L<perlfunc/eof> for how to reset line numbers on each file.
5a964f20 2309
89d205f2 2310If you want to set @ARGV to your own list of files, go right ahead.
5a964f20
TC
2311This sets @ARGV to all plain text files if no @ARGV was given:
2312
2313 @ARGV = grep { -f && -T } glob('*') unless @ARGV;
a0d0e21e 2314
5a964f20
TC
2315You can even set them to pipe commands. For example, this automatically
2316filters compressed arguments through B<gzip>:
2317
2318 @ARGV = map { /\.(gz|Z)$/ ? "gzip -dc < $_ |" : $_ } @ARGV;
2319
2320If you want to pass switches into your script, you can use one of the
a0d0e21e
LW
2321Getopts modules or put a loop on the front like this:
2322
2323 while ($_ = $ARGV[0], /^-/) {
2324 shift;
2325 last if /^--$/;
2326 if (/^-D(.*)/) { $debug = $1 }
2327 if (/^-v/) { $verbose++ }
5a964f20 2328 # ... # other switches
a0d0e21e 2329 }
5a964f20 2330
a0d0e21e 2331 while (<>) {
5a964f20 2332 # ... # code for each line
a0d0e21e
LW
2333 }
2334
89d205f2
YO
2335The <> symbol will return C<undef> for end-of-file only once.
2336If you call it again after this, it will assume you are processing another
19799a22 2337@ARGV list, and if you haven't set @ARGV, will read input from STDIN.
a0d0e21e 2338
b159ebd3 2339If what the angle brackets contain is a simple scalar variable (e.g.,
35f2feb0 2340<$foo>), then that variable contains the name of the
19799a22
GS
2341filehandle to input from, or its typeglob, or a reference to the
2342same. For example:
cb1a09d0
AD
2343
2344 $fh = \*STDIN;
2345 $line = <$fh>;
a0d0e21e 2346
5a964f20
TC
2347If what's within the angle brackets is neither a filehandle nor a simple
2348scalar variable containing a filehandle name, typeglob, or typeglob
2349reference, it is interpreted as a filename pattern to be globbed, and
2350either a list of filenames or the next filename in the list is returned,
19799a22 2351depending on context. This distinction is determined on syntactic
35f2feb0
GS
2352grounds alone. That means C<< <$x> >> is always a readline() from
2353an indirect handle, but C<< <$hash{key}> >> is always a glob().
5a964f20 2354That's because $x is a simple scalar variable, but C<$hash{key}> is
ef191992
YST
2355not--it's a hash element. Even C<< <$x > >> (note the extra space)
2356is treated as C<glob("$x ")>, not C<readline($x)>.
5a964f20
TC
2357
2358One level of double-quote interpretation is done first, but you can't
35f2feb0 2359say C<< <$foo> >> because that's an indirect filehandle as explained
5a964f20
TC
2360in the previous paragraph. (In older versions of Perl, programmers
2361would insert curly brackets to force interpretation as a filename glob:
35f2feb0 2362C<< <${foo}> >>. These days, it's considered cleaner to call the
5a964f20 2363internal function directly as C<glob($foo)>, which is probably the right
19799a22 2364way to have done it in the first place.) For example:
a0d0e21e
LW
2365
2366 while (<*.c>) {
2367 chmod 0644, $_;
2368 }
2369
3a4b19e4 2370is roughly equivalent to:
a0d0e21e
LW
2371
2372 open(FOO, "echo *.c | tr -s ' \t\r\f' '\\012\\012\\012\\012'|");
2373 while (<FOO>) {
5b3eff12 2374 chomp;
a0d0e21e
LW
2375 chmod 0644, $_;
2376 }
2377
3a4b19e4
GS
2378except that the globbing is actually done internally using the standard
2379C<File::Glob> extension. Of course, the shortest way to do the above is:
a0d0e21e
LW
2380
2381 chmod 0644, <*.c>;
2382
19799a22
GS
2383A (file)glob evaluates its (embedded) argument only when it is
2384starting a new list. All values must be read before it will start
2385over. In list context, this isn't important because you automatically
2386get them all anyway. However, in scalar context the operator returns
069e01df 2387the next value each time it's called, or C<undef> when the list has
19799a22
GS
2388run out. As with filehandle reads, an automatic C<defined> is
2389generated when the glob occurs in the test part of a C<while>,
2390because legal glob returns (e.g. a file called F<0>) would otherwise
2391terminate the loop. Again, C<undef> is returned only once. So if
2392you're expecting a single value from a glob, it is much better to
2393say
4633a7c4
LW
2394
2395 ($file) = <blurch*>;
2396
2397than
2398
2399 $file = <blurch*>;
2400
2401because the latter will alternate between returning a filename and
19799a22 2402returning false.
4633a7c4 2403
b159ebd3 2404If you're trying to do variable interpolation, it's definitely better
4633a7c4 2405to use the glob() function, because the older notation can cause people
e37d713d 2406to become confused with the indirect filehandle notation.
4633a7c4
LW
2407
2408 @files = glob("$dir/*.[ch]");
2409 @files = glob($files[$i]);
2410
a0d0e21e 2411=head2 Constant Folding
d74e8afc 2412X<constant folding> X<folding>
a0d0e21e
LW
2413
2414Like C, Perl does a certain amount of expression evaluation at
19799a22 2415compile time whenever it determines that all arguments to an
a0d0e21e
LW
2416operator are static and have no side effects. In particular, string
2417concatenation happens at compile time between literals that don't do
19799a22 2418variable substitution. Backslash interpolation also happens at
a0d0e21e
LW
2419compile time. You can say
2420
2421 'Now is the time for all' . "\n" .
2422 'good men to come to.'
2423
54310121 2424and this all reduces to one string internally. Likewise, if
a0d0e21e
LW
2425you say
2426
2427 foreach $file (@filenames) {
5a964f20 2428 if (-s $file > 5 + 100 * 2**16) { }
54310121 2429 }
a0d0e21e 2430
19799a22
GS
2431the compiler will precompute the number which that expression
2432represents so that the interpreter won't have to.
a0d0e21e 2433
fd1abbef 2434=head2 No-ops
d74e8afc 2435X<no-op> X<nop>
fd1abbef
DN
2436
2437Perl doesn't officially have a no-op operator, but the bare constants
2438C<0> and C<1> are special-cased to not produce a warning in a void
2439context, so you can for example safely do
2440
2441 1 while foo();
2442
2c268ad5 2443=head2 Bitwise String Operators
d74e8afc 2444X<operator, bitwise, string>
2c268ad5
TP
2445
2446Bitstrings of any size may be manipulated by the bitwise operators
2447(C<~ | & ^>).
2448
19799a22
GS
2449If the operands to a binary bitwise op are strings of different
2450sizes, B<|> and B<^> ops act as though the shorter operand had
2451additional zero bits on the right, while the B<&> op acts as though
2452the longer operand were truncated to the length of the shorter.
2453The granularity for such extension or truncation is one or more
2454bytes.
2c268ad5 2455
89d205f2 2456 # ASCII-based examples
2c268ad5
TP
2457 print "j p \n" ^ " a h"; # prints "JAPH\n"
2458 print "JA" | " ph\n"; # prints "japh\n"
2459 print "japh\nJunk" & '_____'; # prints "JAPH\n";
2460 print 'p N$' ^ " E<H\n"; # prints "Perl\n";
2461
19799a22 2462If you are intending to manipulate bitstrings, be certain that
2c268ad5 2463you're supplying bitstrings: If an operand is a number, that will imply
19799a22 2464a B<numeric> bitwise operation. You may explicitly show which type of
2c268ad5
TP
2465operation you intend by using C<""> or C<0+>, as in the examples below.
2466
4358a253
SS
2467 $foo = 150 | 105; # yields 255 (0x96 | 0x69 is 0xFF)
2468 $foo = '150' | 105; # yields 255
2c268ad5
TP
2469 $foo = 150 | '105'; # yields 255
2470 $foo = '150' | '105'; # yields string '155' (under ASCII)
2471
2472 $baz = 0+$foo & 0+$bar; # both ops explicitly numeric
2473 $biz = "$foo" ^ "$bar"; # both ops explicitly stringy
a0d0e21e 2474
1ae175c8
GS
2475See L<perlfunc/vec> for information on how to manipulate individual bits
2476in a bit vector.
2477
55497cff 2478=head2 Integer Arithmetic
d74e8afc 2479X<integer>
a0d0e21e 2480
19799a22 2481By default, Perl assumes that it must do most of its arithmetic in
a0d0e21e
LW
2482floating point. But by saying
2483
2484 use integer;
2485
2486you may tell the compiler that it's okay to use integer operations
19799a22
GS
2487(if it feels like it) from here to the end of the enclosing BLOCK.
2488An inner BLOCK may countermand this by saying
a0d0e21e
LW
2489
2490 no integer;
2491
19799a22
GS
2492which lasts until the end of that BLOCK. Note that this doesn't
2493mean everything is only an integer, merely that Perl may use integer
2494operations if it is so inclined. For example, even under C<use
2495integer>, if you take the C<sqrt(2)>, you'll still get C<1.4142135623731>
2496or so.
2497
2498Used on numbers, the bitwise operators ("&", "|", "^", "~", "<<",
89d205f2 2499and ">>") always produce integral results. (But see also
13a2d996 2500L<Bitwise String Operators>.) However, C<use integer> still has meaning for
19799a22
GS
2501them. By default, their results are interpreted as unsigned integers, but
2502if C<use integer> is in effect, their results are interpreted
2503as signed integers. For example, C<~0> usually evaluates to a large
0be96356 2504integral value. However, C<use integer; ~0> is C<-1> on two's-complement
19799a22 2505machines.
68dc0745
PP
2506
2507=head2 Floating-point Arithmetic
d74e8afc 2508X<floating-point> X<floating point> X<float> X<real>
68dc0745
PP
2509
2510While C<use integer> provides integer-only arithmetic, there is no
19799a22
GS
2511analogous mechanism to provide automatic rounding or truncation to a
2512certain number of decimal places. For rounding to a certain number
2513of digits, sprintf() or printf() is usually the easiest route.
2514See L<perlfaq4>.
68dc0745 2515
5a964f20
TC
2516Floating-point numbers are only approximations to what a mathematician
2517would call real numbers. There are infinitely more reals than floats,
2518so some corners must be cut. For example:
2519
2520 printf "%.20g\n", 123456789123456789;
2521 # produces 123456789123456784
2522
8548cb57
RGS
2523Testing for exact floating-point equality or inequality is not a
2524good idea. Here's a (relatively expensive) work-around to compare
5a964f20
TC
2525whether two floating-point numbers are equal to a particular number of
2526decimal places. See Knuth, volume II, for a more robust treatment of
2527this topic.
2528
2529 sub fp_equal {
2530 my ($X, $Y, $POINTS) = @_;
2531 my ($tX, $tY);
2532 $tX = sprintf("%.${POINTS}g", $X);
2533 $tY = sprintf("%.${POINTS}g", $Y);
2534 return $tX eq $tY;
2535 }
2536
68dc0745 2537The POSIX module (part of the standard perl distribution) implements
19799a22
GS
2538ceil(), floor(), and other mathematical and trigonometric functions.
2539The Math::Complex module (part of the standard perl distribution)
2540defines mathematical functions that work on both the reals and the
2541imaginary numbers. Math::Complex not as efficient as POSIX, but
68dc0745
PP
2542POSIX can't work with complex numbers.
2543
2544Rounding in financial applications can have serious implications, and
2545the rounding method used should be specified precisely. In these
2546cases, it probably pays not to trust whichever system rounding is
2547being used by Perl, but to instead implement the rounding function you
2548need yourself.
5a964f20
TC
2549
2550=head2 Bigger Numbers
d74e8afc 2551X<number, arbitrary precision>
5a964f20
TC
2552
2553The standard Math::BigInt and Math::BigFloat modules provide
19799a22 2554variable-precision arithmetic and overloaded operators, although
cd5c4fce 2555they're currently pretty slow. At the cost of some space and
19799a22
GS
2556considerable speed, they avoid the normal pitfalls associated with
2557limited-precision representations.
5a964f20
TC
2558
2559 use Math::BigInt;
2560 $x = Math::BigInt->new('123456789123456789');
2561 print $x * $x;
2562
2563 # prints +15241578780673678515622620750190521
19799a22 2564
cd5c4fce
T
2565There are several modules that let you calculate with (bound only by
2566memory and cpu-time) unlimited or fixed precision. There are also
2567some non-standard modules that provide faster implementations via
2568external C libraries.
2569
2570Here is a short, but incomplete summary:
2571
2572 Math::Fraction big, unlimited fractions like 9973 / 12967
2573 Math::String treat string sequences like numbers
2574 Math::FixedPrecision calculate with a fixed precision
2575 Math::Currency for currency calculations
2576 Bit::Vector manipulate bit vectors fast (uses C)
2577 Math::BigIntFast Bit::Vector wrapper for big numbers
2578 Math::Pari provides access to the Pari C library
2579 Math::BigInteger uses an external C library
2580 Math::Cephes uses external Cephes C library (no big numbers)
2581 Math::Cephes::Fraction fractions via the Cephes library
2582 Math::GMP another one using an external C library
2583
2584Choose wisely.
16070b82
GS
2585
2586=cut