4 perlop - Perl operators and precedence
8 =head2 Operator Precedence and Associativity
9 X<operator, precedence> X<precedence> X<associativity>
11 Operator precedence and associativity work in Perl more or less like
12 they do in mathematics.
14 I<Operator precedence> means some operators are evaluated before
15 others. For example, in C<2 + 4 * 5>, the multiplication has higher
16 precedence so C<4 * 5> is evaluated first yielding C<2 + 20 ==
17 22> and not C<6 * 5 == 30>.
19 I<Operator associativity> defines what happens if a sequence of the
20 same operators is used one after another: whether the evaluator will
21 evaluate the left operations first or the right. For example, in C<8
22 - 4 - 2>, subtraction is left associative so Perl evaluates the
23 expression left to right. C<8 - 4> is evaluated first making the
24 expression C<4 - 2 == 2> and not C<8 - 2 == 6>.
26 Perl operators have the following associativity and precedence,
27 listed from highest precedence to lowest. Operators borrowed from
28 C keep the same precedence relationship with each other, even where
29 C's precedence is slightly screwy. (This makes learning Perl easier
30 for C folks.) With very few exceptions, these all operate on scalar
31 values only, not array values.
33 left terms and list operators (leftward)
37 right ! ~ \ and unary + and -
42 nonassoc named unary operators
43 nonassoc < > <= >= lt gt le ge
44 nonassoc == != <=> eq ne cmp ~~
53 nonassoc list operators (rightward)
58 In the following sections, these operators are covered in precedence order.
60 Many operators can be overloaded for objects. See L<overload>.
62 =head2 Terms and List Operators (Leftward)
63 X<list operator> X<operator, list> X<term>
65 A TERM has the highest precedence in Perl. They include variables,
66 quote and quote-like operators, any expression in parentheses,
67 and any function whose arguments are parenthesized. Actually, there
68 aren't really functions in this sense, just list operators and unary
69 operators behaving as functions because you put parentheses around
70 the arguments. These are all documented in L<perlfunc>.
72 If any list operator (print(), etc.) or any unary operator (chdir(), etc.)
73 is followed by a left parenthesis as the next token, the operator and
74 arguments within parentheses are taken to be of highest precedence,
75 just like a normal function call.
77 In the absence of parentheses, the precedence of list operators such as
78 C<print>, C<sort>, or C<chmod> is either very high or very low depending on
79 whether you are looking at the left side or the right side of the operator.
82 @ary = (1, 3, sort 4, 2);
83 print @ary; # prints 1324
85 the commas on the right of the sort are evaluated before the sort,
86 but the commas on the left are evaluated after. In other words,
87 list operators tend to gobble up all arguments that follow, and
88 then act like a simple TERM with regard to the preceding expression.
89 Be careful with parentheses:
91 # These evaluate exit before doing the print:
92 print($foo, exit); # Obviously not what you want.
93 print $foo, exit; # Nor is this.
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 ($foo), exit; # Or even this.
102 print ($foo & 255) + 1, "\n";
104 probably doesn't do what you expect at first glance. The parentheses
105 enclose the argument list for C<print> which is evaluated (printing
106 the result of C<$foo & 255>). Then one is added to the return value
107 of C<print> (usually 1). The result is something like this:
109 1 + 1, "\n"; # Obviously not what you meant.
111 To do what you meant properly, you must write:
113 print(($foo & 255) + 1, "\n");
115 See L<Named Unary Operators> for more discussion of this.
117 Also parsed as terms are the C<do {}> and C<eval {}> constructs, as
118 well as subroutine and method calls, and the anonymous
119 constructors C<[]> and C<{}>.
121 See also L<Quote and Quote-like Operators> toward the end of this section,
122 as well as L</"I/O Operators">.
124 =head2 The Arrow Operator
125 X<arrow> X<dereference> X<< -> >>
127 "C<< -> >>" is an infix dereference operator, just as it is in C
128 and C++. If the right side is either a C<[...]>, C<{...}>, or a
129 C<(...)> subscript, then the left side must be either a hard or
130 symbolic reference to an array, a hash, or a subroutine respectively.
131 (Or technically speaking, a location capable of holding a hard
132 reference, if it's an array or hash reference being used for
133 assignment.) See L<perlreftut> and L<perlref>.
135 Otherwise, the right side is a method name or a simple scalar
136 variable containing either the method name or a subroutine reference,
137 and the left side must be either an object (a blessed reference)
138 or a class name (that is, a package name). See L<perlobj>.
140 =head2 Auto-increment and Auto-decrement
141 X<increment> X<auto-increment> X<++> X<decrement> X<auto-decrement> X<-->
143 "++" and "--" work as in C. That is, if placed before a variable,
144 they increment or decrement the variable by one before returning the
145 value, and if placed after, increment or decrement after returning the
149 print $i++; # prints 0
150 print ++$j; # prints 1
152 Note that just as in C, Perl doesn't define B<when> the variable is
153 incremented or decremented. You just know it will be done sometime
154 before or after the value is returned. This also means that modifying
155 a variable twice in the same statement will lead to undefined behaviour.
156 Avoid statements like:
161 Perl will not guarantee what the result of the above statements is.
163 The auto-increment operator has a little extra builtin magic to it. If
164 you increment a variable that is numeric, or that has ever been used in
165 a numeric context, you get a normal increment. If, however, the
166 variable has been used in only string contexts since it was set, and
167 has a value that is not the empty string and matches the pattern
168 C</^[a-zA-Z]*[0-9]*\z/>, the increment is done as a string, preserving each
169 character within its range, with carry:
171 print ++($foo = '99'); # prints '100'
172 print ++($foo = 'a0'); # prints 'a1'
173 print ++($foo = 'Az'); # prints 'Ba'
174 print ++($foo = 'zz'); # prints 'aaa'
176 C<undef> is always treated as numeric, and in particular is changed
177 to C<0> before incrementing (so that a post-increment of an undef value
178 will return C<0> rather than C<undef>).
180 The auto-decrement operator is not magical.
182 =head2 Exponentiation
183 X<**> X<exponentiation> X<power>
185 Binary "**" is the exponentiation operator. It binds even more
186 tightly than unary minus, so -2**4 is -(2**4), not (-2)**4. (This is
187 implemented using C's pow(3) function, which actually works on doubles
190 =head2 Symbolic Unary Operators
191 X<unary operator> X<operator, unary>
193 Unary "!" performs logical negation, i.e., "not". See also C<not> for a lower
194 precedence version of this.
197 Unary "-" performs arithmetic negation if the operand is numeric. If
198 the operand is an identifier, a string consisting of a minus sign
199 concatenated with the identifier is returned. Otherwise, if the string
200 starts with a plus or minus, a string starting with the opposite sign
201 is returned. One effect of these rules is that -bareword is equivalent
202 to the string "-bareword". If, however, the string begins with a
203 non-alphabetic character (excluding "+" or "-"), Perl will attempt to convert
204 the string to a numeric and the arithmetic negation is performed. If the
205 string cannot be cleanly converted to a numeric, Perl will give the warning
206 B<Argument "the string" isn't numeric in negation (-) at ...>.
207 X<-> X<negation, arithmetic>
209 Unary "~" performs bitwise negation, i.e., 1's complement. For
210 example, C<0666 & ~027> is 0640. (See also L<Integer Arithmetic> and
211 L<Bitwise String Operators>.) Note that the width of the result is
212 platform-dependent: ~0 is 32 bits wide on a 32-bit platform, but 64
213 bits wide on a 64-bit platform, so if you are expecting a certain bit
214 width, remember to use the & operator to mask off the excess bits.
215 X<~> X<negation, binary>
217 Unary "+" has no effect whatsoever, even on strings. It is useful
218 syntactically for separating a function name from a parenthesized expression
219 that would otherwise be interpreted as the complete list of function
220 arguments. (See examples above under L<Terms and List Operators (Leftward)>.)
223 Unary "\" creates a reference to whatever follows it. See L<perlreftut>
224 and L<perlref>. Do not confuse this behavior with the behavior of
225 backslash within a string, although both forms do convey the notion
226 of protecting the next thing from interpolation.
227 X<\> X<reference> X<backslash>
229 =head2 Binding Operators
230 X<binding> X<operator, binding> X<=~> X<!~>
232 Binary "=~" binds a scalar expression to a pattern match. Certain operations
233 search or modify the string $_ by default. This operator makes that kind
234 of operation work on some other string. The right argument is a search
235 pattern, substitution, or transliteration. The left argument is what is
236 supposed to be searched, substituted, or transliterated instead of the default
237 $_. When used in scalar context, the return value generally indicates the
238 success of the operation. Behavior in list context depends on the particular
239 operator. See L</"Regexp Quote-Like Operators"> for details and
240 L<perlretut> for examples using these operators.
242 If the right argument is an expression rather than a search pattern,
243 substitution, or transliteration, it is interpreted as a search pattern at run
244 time. Note that this means that its contents will be interpolated twice, so
248 is not ok, as the regex engine will end up trying to compile the
249 pattern C<\>, which it will consider a syntax error.
251 Binary "!~" is just like "=~" except the return value is negated in
254 =head2 Multiplicative Operators
255 X<operator, multiplicative>
257 Binary "*" multiplies two numbers.
260 Binary "/" divides two numbers.
263 Binary "%" is the modulo operator, which computes the division
264 remainder of its first argument with respect to its second argument.
266 operands C<$a> and C<$b>: If C<$b> is positive, then C<$a % $b> is
267 C<$a> minus the largest multiple of C<$b> less than or equal to
268 C<$a>. If C<$b> is negative, then C<$a % $b> is C<$a> minus the
269 smallest multiple of C<$b> that is not less than C<$a> (i.e. the
270 result will be less than or equal to zero). If the operands
271 C<$a> and C<$b> are floating point values and the absolute value of
272 C<$b> (that is C<abs($b)>) is less than C<(UV_MAX + 1)>, only
273 the 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).
275 If the absolute value of the right operand (C<abs($b)>) is greater than
276 or equal to C<(UV_MAX + 1)>, "%" computes the floating-point remainder
277 C<$r> in the equation C<($r = $a - $i*$b)> where C<$i> is a certain
278 integer that makes C<$r> have the same sign as the right operand
279 C<$b> (B<not> as the left operand C<$a> like C function C<fmod()>)
280 and the absolute value less than that of C<$b>.
281 Note that when C<use integer> is in scope, "%" gives you direct access
282 to the modulo operator as implemented by your C compiler. This
283 operator is not as well defined for negative operands, but it will
285 X<%> X<remainder> X<modulo> X<mod>
287 Binary "x" is the repetition operator. In scalar context or if the left
288 operand is not enclosed in parentheses, it returns a string consisting
289 of the left operand repeated the number of times specified by the right
290 operand. In list context, if the left operand is enclosed in
291 parentheses or is a list formed by C<qw/STRING/>, it repeats the list.
292 If the right operand is zero or negative, it returns an empty string
293 or an empty list, depending on the context.
296 print '-' x 80; # print row of dashes
298 print "\t" x ($tab/8), ' ' x ($tab%8); # tab over
300 @ones = (1) x 80; # a list of 80 1's
301 @ones = (5) x @ones; # set all elements to 5
304 =head2 Additive Operators
305 X<operator, additive>
307 Binary "+" returns the sum of two numbers.
310 Binary "-" returns the difference of two numbers.
313 Binary "." concatenates two strings.
314 X<string, concatenation> X<concatenation>
315 X<cat> X<concat> X<concatenate> X<.>
317 =head2 Shift Operators
318 X<shift operator> X<operator, shift> X<<< << >>>
319 X<<< >> >>> X<right shift> X<left shift> X<bitwise shift>
320 X<shl> X<shr> X<shift, right> X<shift, left>
322 Binary "<<" returns the value of its left argument shifted left by the
323 number of bits specified by the right argument. Arguments should be
324 integers. (See also L<Integer Arithmetic>.)
326 Binary ">>" returns the value of its left argument shifted right by
327 the number of bits specified by the right argument. Arguments should
328 be integers. (See also L<Integer Arithmetic>.)
330 Note that both "<<" and ">>" in Perl are implemented directly using
331 "<<" and ">>" in C. If C<use integer> (see L<Integer Arithmetic>) is
332 in force then signed C integers are used, else unsigned C integers are
333 used. Either way, the implementation isn't going to generate results
334 larger than the size of the integer type Perl was built with (32 bits
337 The result of overflowing the range of the integers is undefined
338 because it is undefined also in C. In other words, using 32-bit
339 integers, C<< 1 << 32 >> is undefined. Shifting by a negative number
340 of bits is also undefined.
342 =head2 Named Unary Operators
343 X<operator, named unary>
345 The various named unary operators are treated as functions with one
346 argument, with optional parentheses.
348 If any list operator (print(), etc.) or any unary operator (chdir(), etc.)
349 is followed by a left parenthesis as the next token, the operator and
350 arguments within parentheses are taken to be of highest precedence,
351 just like a normal function call. For example,
352 because named unary operators are higher precedence than ||:
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
359 but, because * is higher precedence than named operators:
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 ($foo * 20)
366 rand 10 * 20; # rand (10 * 20)
367 rand(10) * 20; # (rand 10) * 20
368 rand (10) * 20; # (rand 10) * 20
369 rand +(10) * 20; # rand (10 * 20)
371 Regarding precedence, the filetest operators, like C<-f>, C<-M>, etc. are
372 treated like named unary operators, but they don't follow this functional
373 parenthesis rule. That means, for example, that C<-f($file).".bak"> is
374 equivalent to C<-f "$file.bak">.
375 X<-X> X<filetest> X<operator, filetest>
377 See also L<"Terms and List Operators (Leftward)">.
379 =head2 Relational Operators
380 X<relational operator> X<operator, relational>
382 Binary "<" returns true if the left argument is numerically less than
386 Binary ">" returns true if the left argument is numerically greater
387 than the right argument.
390 Binary "<=" returns true if the left argument is numerically less than
391 or equal to the right argument.
394 Binary ">=" returns true if the left argument is numerically greater
395 than or equal to the right argument.
398 Binary "lt" returns true if the left argument is stringwise less than
402 Binary "gt" returns true if the left argument is stringwise greater
403 than the right argument.
406 Binary "le" returns true if the left argument is stringwise less than
407 or equal to the right argument.
410 Binary "ge" returns true if the left argument is stringwise greater
411 than or equal to the right argument.
414 =head2 Equality Operators
415 X<equality> X<equal> X<equals> X<operator, equality>
417 Binary "==" returns true if the left argument is numerically equal to
421 Binary "!=" returns true if the left argument is numerically not equal
422 to the right argument.
425 Binary "<=>" returns -1, 0, or 1 depending on whether the left
426 argument is numerically less than, equal to, or greater than the right
427 argument. If your platform supports NaNs (not-a-numbers) as numeric
428 values, using them with "<=>" returns undef. NaN is not "<", "==", ">",
429 "<=" or ">=" anything (even NaN), so those 5 return false. NaN != NaN
430 returns true, as does NaN != anything else. If your platform doesn't
431 support NaNs then NaN is just a string with numeric value 0.
432 X<< <=> >> X<spaceship>
434 perl -le '$a = "NaN"; print "No NaN support here" if $a == $a'
435 perl -le '$a = "NaN"; print "NaN support here" if $a != $a'
437 Binary "eq" returns true if the left argument is stringwise equal to
441 Binary "ne" returns true if the left argument is stringwise not equal
442 to the right argument.
445 Binary "cmp" returns -1, 0, or 1 depending on whether the left
446 argument is stringwise less than, equal to, or greater than the right
450 Binary "~~" does a smart match between its arguments. Smart matching
451 is described in L<perlsyn/"Smart matching in detail">.
454 "lt", "le", "ge", "gt" and "cmp" use the collation (sort) order specified
455 by the current locale if C<use locale> is in effect. See L<perllocale>.
458 X<operator, bitwise, and> X<bitwise and> X<&>
460 Binary "&" returns its operands ANDed together bit by bit.
461 (See also L<Integer Arithmetic> and L<Bitwise String Operators>.)
463 Note that "&" has lower priority than relational operators, so for example
464 the brackets are essential in a test like
466 print "Even\n" if ($x & 1) == 0;
468 =head2 Bitwise Or and Exclusive Or
469 X<operator, bitwise, or> X<bitwise or> X<|> X<operator, bitwise, xor>
472 Binary "|" returns its operands ORed together bit by bit.
473 (See also L<Integer Arithmetic> and L<Bitwise String Operators>.)
475 Binary "^" returns its operands XORed together bit by bit.
476 (See also L<Integer Arithmetic> and L<Bitwise String Operators>.)
478 Note that "|" and "^" have lower priority than relational operators, so
479 for example the brackets are essential in a test like
481 print "false\n" if (8 | 2) != 10;
483 =head2 C-style Logical And
484 X<&&> X<logical and> X<operator, logical, and>
486 Binary "&&" performs a short-circuit logical AND operation. That is,
487 if the left operand is false, the right operand is not even evaluated.
488 Scalar or list context propagates down to the right operand if it
491 =head2 C-style Logical Or
492 X<||> X<operator, logical, or>
494 Binary "||" performs a short-circuit logical OR operation. That is,
495 if the left operand is true, the right operand is not even evaluated.
496 Scalar or list context propagates down to the right operand if it
499 =head2 C-style Logical Defined-Or
500 X<//> X<operator, logical, defined-or>
502 Although it has no direct equivalent in C, Perl's C<//> operator is related
503 to its C-style or. In fact, it's exactly the same as C<||>, except that it
504 tests the left hand side's definedness instead of its truth. Thus, C<$a // $b>
505 is similar to C<defined($a) || $b> (except that it returns the value of C<$a>
506 rather than the value of C<defined($a)>) and is exactly equivalent to
507 C<defined($a) ? $a : $b>. This is very useful for providing default values
508 for variables. If you actually want to test if at least one of C<$a> and
509 C<$b> is defined, use C<defined($a // $b)>.
511 The 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
513 portable way to find out the home directory might be:
515 $home = $ENV{'HOME'} // $ENV{'LOGDIR'} //
516 (getpwuid($<))[7] // die "You're homeless!\n";
518 In particular, this means that you shouldn't use this
519 for selecting between two aggregates for assignment:
521 @a = @b || @c; # this is wrong
522 @a = scalar(@b) || @c; # really meant this
523 @a = @b ? @b : @c; # this works fine, though
525 As more readable alternatives to C<&&> and C<||> when used for
526 control flow, Perl provides the C<and> and C<or> operators (see below).
527 The short-circuit behavior is identical. The precedence of "and"
528 and "or" is much lower, however, so that you can safely use them after a
529 list operator without the need for parentheses:
531 unlink "alpha", "beta", "gamma"
532 or gripe(), next LINE;
534 With the C-style operators that would have been written like this:
536 unlink("alpha", "beta", "gamma")
537 || (gripe(), next LINE);
539 Using "or" for assignment is unlikely to do what you want; see below.
541 =head2 Range Operators
542 X<operator, range> X<range> X<..> X<...>
544 Binary ".." is the range operator, which is really two different
545 operators depending on the context. In list context, it returns a
546 list of values counting (up by ones) from the left value to the right
547 value. If the left value is greater than the right value then it
548 returns the empty list. The range operator is useful for writing
549 C<foreach (1..10)> loops and for doing slice operations on arrays. In
550 the current implementation, no temporary array is created when the
551 range operator is used as the expression in C<foreach> loops, but older
552 versions of Perl might burn a lot of memory when you write something
555 for (1 .. 1_000_000) {
559 The range operator also works on strings, using the magical
560 auto-increment, see below.
562 In scalar context, ".." returns a boolean value. The operator is
563 bistable, like a flip-flop, and emulates the line-range (comma)
564 operator of B<sed>, B<awk>, and various editors. Each ".." operator
565 maintains its own boolean state, even across calls to a subroutine
566 that contains it. It is false as long as its left operand is false.
567 Once the left operand is true, the range operator stays true until the
568 right operand is true, I<AFTER> which the range operator becomes false
569 again. It doesn't become false till the next time the range operator
570 is evaluated. It can test the right operand and become false on the
571 same evaluation it became true (as in B<awk>), but it still returns
572 true once. If you don't want it to test the right operand until the
573 next evaluation, as in B<sed>, just use three dots ("...") instead of
574 two. In all other regards, "..." behaves just like ".." does.
576 The right operand is not evaluated while the operator is in the
577 "false" state, and the left operand is not evaluated while the
578 operator is in the "true" state. The precedence is a little lower
579 than || and &&. The value returned is either the empty string for
580 false, or a sequence number (beginning with 1) for true. The sequence
581 number is reset for each range encountered. The final sequence number
582 in a range has the string "E0" appended to it, which doesn't affect
583 its numeric value, but gives you something to search for if you want
584 to exclude the endpoint. You can exclude the beginning point by
585 waiting for the sequence number to be greater than 1.
587 If either operand of scalar ".." is a constant expression,
588 that operand is considered true if it is equal (C<==>) to the current
589 input line number (the C<$.> variable).
591 To be pedantic, the comparison is actually C<int(EXPR) == int(EXPR)>,
592 but that is only an issue if you use a floating point expression; when
593 implicitly using C<$.> as described in the previous paragraph, the
594 comparison is C<int(EXPR) == int($.)> which is only an issue when C<$.>
595 is set to a floating point value and you are not reading from a file.
596 Furthermore, C<"span" .. "spat"> or C<2.18 .. 3.14> will not do what
597 you want in scalar context because each of the operands are evaluated
598 using their integer representation.
602 As a scalar operator:
604 if (101 .. 200) { print; } # print 2nd hundred lines, short for
605 # if ($. == 101 .. $. == 200) { print; }
607 next LINE if (1 .. /^$/); # skip header lines, short for
608 # next LINE if ($. == 1 .. /^$/);
609 # (typically in a loop labeled LINE)
611 s/^/> / if (/^$/ .. eof()); # quote body
613 # parse mail messages
615 $in_header = 1 .. /^$/;
616 $in_body = /^$/ .. eof;
623 close ARGV if eof; # reset $. each file
626 Here's a simple example to illustrate the difference between
627 the two range operators:
640 This program will print only the line containing "Bar". If
641 the range operator is changed to C<...>, it will also print the
644 And now some examples as a list operator:
646 for (101 .. 200) { print; } # print $_ 100 times
647 @foo = @foo[0 .. $#foo]; # an expensive no-op
648 @foo = @foo[$#foo-4 .. $#foo]; # slice last 5 items
650 The range operator (in list context) makes use of the magical
651 auto-increment algorithm if the operands are strings. You
654 @alphabet = ('A' .. 'Z');
656 to get all normal letters of the English alphabet, or
658 $hexdigit = (0 .. 9, 'a' .. 'f')[$num & 15];
660 to get a hexadecimal digit, or
662 @z2 = ('01' .. '31'); print $z2[$mday];
664 to get dates with leading zeros.
666 If the final value specified is not in the sequence that the magical
667 increment would produce, the sequence goes until the next value would
668 be longer than the final value specified.
670 If the initial value specified isn't part of a magical increment
671 sequence (that is, a non-empty string matching "/^[a-zA-Z]*[0-9]*\z/"),
672 only the initial value will be returned. So the following will only
675 use charnames 'greek';
676 my @greek_small = ("\N{alpha}" .. "\N{omega}");
678 To get lower-case greek letters, use this instead:
680 my @greek_small = map { chr } ( ord("\N{alpha}") ..
683 Because each operand is evaluated in integer form, C<2.18 .. 3.14> will
684 return two elements in list context.
686 @list = (2.18 .. 3.14); # same as @list = (2 .. 3);
688 =head2 Conditional Operator
689 X<operator, conditional> X<operator, ternary> X<ternary> X<?:>
691 Ternary "?:" is the conditional operator, just as in C. It works much
692 like an if-then-else. If the argument before the ? is true, the
693 argument before the : is returned, otherwise the argument after the :
694 is returned. For example:
696 printf "I have %d dog%s.\n", $n,
697 ($n == 1) ? '' : "s";
699 Scalar or list context propagates downward into the 2nd
700 or 3rd argument, whichever is selected.
702 $a = $ok ? $b : $c; # get a scalar
703 @a = $ok ? @b : @c; # get an array
704 $a = $ok ? @b : @c; # oops, that's just a count!
706 The operator may be assigned to if both the 2nd and 3rd arguments are
707 legal lvalues (meaning that you can assign to them):
709 ($a_or_b ? $a : $b) = $c;
711 Because this operator produces an assignable result, using assignments
712 without parentheses will get you in trouble. For example, this:
714 $a % 2 ? $a += 10 : $a += 2
718 (($a % 2) ? ($a += 10) : $a) += 2
722 ($a % 2) ? ($a += 10) : ($a += 2)
724 That should probably be written more simply as:
726 $a += ($a % 2) ? 10 : 2;
728 =head2 Assignment Operators
729 X<assignment> X<operator, assignment> X<=> X<**=> X<+=> X<*=> X<&=>
730 X<<< <<= >>> X<&&=> X<-=> X</=> X<|=> X<<< >>= >>> X<||=> X<//=> X<.=>
733 "=" is the ordinary assignment operator.
735 Assignment operators work as in C. That is,
743 although without duplicating any side effects that dereferencing the lvalue
744 might trigger, such as from tie(). Other assignment operators work similarly.
745 The following are recognized:
752 Although these are grouped by family, they all have the precedence
755 Unlike in C, the scalar assignment operator produces a valid lvalue.
756 Modifying an assignment is equivalent to doing the assignment and
757 then modifying the variable that was assigned to. This is useful
758 for modifying a copy of something, like this:
760 ($tmp = $global) =~ tr [A-Z] [a-z];
771 Similarly, a list assignment in list context produces the list of
772 lvalues assigned to, and a list assignment in scalar context returns
773 the number of elements produced by the expression on the right hand
774 side of the assignment.
776 =head2 Comma Operator
777 X<comma> X<operator, comma> X<,>
779 Binary "," is the comma operator. In scalar context it evaluates
780 its left argument, throws that value away, then evaluates its right
781 argument and returns that value. This is just like C's comma operator.
783 In list context, it's just the list argument separator, and inserts
784 both its arguments into the list. These arguments are also evaluated
787 The C<< => >> operator is a synonym for the comma except that it causes
788 its left operand to be interpreted as a string if it begins with a letter
789 or underscore and is composed only of letters, digits and underscores.
790 This includes operands that might otherwise be interpreted as operators,
791 constants, single number v-strings or function calls. If in doubt about
792 this behaviour, the left operand can be quoted explicitly.
794 Otherwise, the C<< => >> operator behaves exactly as the comma operator
795 or list argument separator, according to context.
799 use constant FOO => "something";
801 my %h = ( FOO => 23 );
809 my %h = ("something", 23);
811 The C<< => >> operator is helpful in documenting the correspondence
812 between keys and values in hashes, and other paired elements in lists.
814 %hash = ( $key => $value );
815 login( $username => $password );
817 =head2 Yada Yada Operator
818 X<...> X<... operator> X<yada yada operator>
820 The yada yada operator (noted C<...>) is a placeholder for code. Perl
821 parses it without error, but when you try to execute a yada yada, it
822 throws an exception with the text C<Unimplemented>:
824 sub unimplemented { ... }
826 eval { unimplemented() };
827 if( $@ eq 'Unimplemented' ) {
828 print "I found the yada yada!\n";
831 You can only use the yada yada to stand in for a complete statement.
832 These examples of the yada yada work:
848 do { my $n; ...; print 'Hurrah!' };
850 The yada yada cannot stand in for an expression that is part of a
851 larger statement since the C<...> is also the three-dot version of the
852 range operator (see L<Range Operators>). These examples of the yada
853 yada are still syntax errors:
857 open my($fh), '>', '/dev/passwd' or ...;
859 if( $condition && ... ) { print "Hello\n" };
861 There are some cases where Perl can't immediately tell the difference
862 between an expression and a statement. For instance, the syntax for a
863 block and an anonymous hash reference constructor look the same unless
864 there's something in the braces that give Perl a hint. The yada yada
865 is a syntax error if Perl doesn't guess that the C<{ ... }> is a
866 block. In that case, it doesn't think the C<...> is the yada yada
867 because it's expecting an expression instead of a statement:
869 my @transformed = map { ... } @input; # syntax error
871 You can use a C<;> inside your block to denote that the C<{ ... }> is
872 a block and not a hash reference constructor. Now the yada yada works:
874 my @transformed = map {; ... } @input; # ; disambiguates
876 my @transformed = map { ...; } @input; # ; disambiguates
878 =head2 List Operators (Rightward)
879 X<operator, list, rightward> X<list operator>
881 On the right side of a list operator, it has very low precedence,
882 such that it controls all comma-separated expressions found there.
883 The only operators with lower precedence are the logical operators
884 "and", "or", and "not", which may be used to evaluate calls to list
885 operators without the need for extra parentheses:
887 open HANDLE, "filename"
888 or die "Can't open: $!\n";
890 See also discussion of list operators in L<Terms and List Operators (Leftward)>.
893 X<operator, logical, not> X<not>
895 Unary "not" returns the logical negation of the expression to its right.
896 It's the equivalent of "!" except for the very low precedence.
899 X<operator, logical, and> X<and>
901 Binary "and" returns the logical conjunction of the two surrounding
902 expressions. It's equivalent to && except for the very low
903 precedence. This means that it short-circuits: i.e., the right
904 expression is evaluated only if the left expression is true.
906 =head2 Logical or, Defined or, and Exclusive Or
907 X<operator, logical, or> X<operator, logical, xor>
908 X<operator, logical, defined or> X<operator, logical, exclusive or>
911 Binary "or" returns the logical disjunction of the two surrounding
912 expressions. It's equivalent to || except for the very low precedence.
913 This makes it useful for control flow
915 print FH $data or die "Can't write to FH: $!";
917 This means that it short-circuits: i.e., the right expression is evaluated
918 only if the left expression is false. Due to its precedence, you should
919 probably avoid using this for assignment, only for control flow.
921 $a = $b or $c; # bug: this is wrong
922 ($a = $b) or $c; # really means this
923 $a = $b || $c; # better written this way
925 However, when it's a list-context assignment and you're trying to use
926 "||" for control flow, you probably need "or" so that the assignment
927 takes higher precedence.
929 @info = stat($file) || die; # oops, scalar sense of stat!
930 @info = stat($file) or die; # better, now @info gets its due
932 Then again, you could always use parentheses.
934 Binary "xor" returns the exclusive-OR of the two surrounding expressions.
935 It cannot short circuit, of course.
937 =head2 C Operators Missing From Perl
938 X<operator, missing from perl> X<&> X<*>
939 X<typecasting> X<(TYPE)>
941 Here is what C has that Perl doesn't:
947 Address-of operator. (But see the "\" operator for taking a reference.)
951 Dereference-address operator. (Perl's prefix dereferencing
952 operators are typed: $, @, %, and &.)
956 Type-casting operator.
960 =head2 Quote and Quote-like Operators
961 X<operator, quote> X<operator, quote-like> X<q> X<qq> X<qx> X<qw> X<m>
962 X<qr> X<s> X<tr> X<'> X<''> X<"> X<""> X<//> X<`> X<``> X<<< << >>>
963 X<escape sequence> X<escape>
966 While we usually think of quotes as literal values, in Perl they
967 function as operators, providing various kinds of interpolating and
968 pattern matching capabilities. Perl provides customary quote characters
969 for these behaviors, but also provides a way for you to choose your
970 quote character for any of them. In the following table, a C<{}> represents
971 any pair of delimiters you choose.
973 Customary Generic Meaning Interpolates
978 // m{} Pattern match yes*
980 s{}{} Substitution yes*
981 tr{}{} Transliteration no (but see below)
984 * unless the delimiter is ''.
986 Non-bracketing delimiters use the same character fore and aft, but the four
987 sorts of brackets (round, angle, square, curly) will all nest, which means
996 Note, however, that this does not always work for quoting Perl code:
998 $s = q{ if($a eq "}") ... }; # WRONG
1000 is a syntax error. The C<Text::Balanced> module (from CPAN, and
1001 starting from Perl 5.8 part of the standard distribution) is able
1002 to do this properly.
1004 There can be whitespace between the operator and the quoting
1005 characters, except when C<#> is being used as the quoting character.
1006 C<q#foo#> is parsed as the string C<foo>, while C<q #foo#> is the
1007 operator C<q> followed by a comment. Its argument will be taken
1008 from the next line. This allows you to write:
1010 s {foo} # Replace foo
1013 The following escape sequences are available in constructs that interpolate
1014 and in transliterations.
1015 X<\t> X<\n> X<\r> X<\f> X<\b> X<\a> X<\e> X<\x> X<\0> X<\c> X<\N> X<\N{}>
1017 Sequence Note Description
1023 \a alarm (bell) (BEL)
1025 \033 octal char (example: ESC)
1026 \x1b hex char (example: ESC)
1027 \x{263a} wide hex char (example: SMILEY)
1028 \c[ [1] control char (example: chr(27))
1029 \N{name} [2] named Unicode character
1030 \N{U+263D} [3] Unicode character (example: FIRST QUARTER MOON)
1036 The character following C<\c> is mapped to some other character as shown in the
1053 Also, C<\c\I<X>> yields C< chr(28) . "I<X>"> for any I<X>, but cannot come at the
1054 end of a string, because the backslash would be parsed as escaping the end
1057 On ASCII platforms, the resulting characters from the list above are the
1058 complete set of ASCII controls. This isn't the case on EBCDIC platforms; see
1059 L<perlebcdic/OPERATOR DIFFERENCES> for the complete list of what these
1060 sequences mean on both ASCII and EBCDIC platforms.
1062 Use of any other character following the "c" besides those listed above is
1063 prohibited on EBCDIC platforms, and discouraged (and may become deprecated or
1064 forbidden) on ASCII ones. What happens for those other characters currently
1065 though, is that the value is derived by inverting the 7th bit (0x40).
1067 To get platform independent controls, you can use C<\N{...}>.
1071 For documentation of C<\N{name}>, see L<charnames>.
1075 C<\N{U+I<wide hex char>}> means the Unicode character whose Unicode ordinal
1076 number is I<wide hex char>.
1080 B<NOTE>: Unlike C and other languages, Perl has no C<\v> escape sequence for
1081 the vertical tab (VT - ASCII 11), but you may use C<\ck> or C<\x0b>. (C<\v>
1082 does have meaning in regular expression patterns in Perl, see L<perlre>.)
1084 The following escape sequences are available in constructs that interpolate,
1085 but not in transliterations.
1086 X<\l> X<\u> X<\L> X<\U> X<\E> X<\Q>
1088 \l lowercase next char
1089 \u uppercase next char
1090 \L lowercase till \E
1091 \U uppercase till \E
1092 \E end case modification
1093 \Q quote non-word characters till \E
1095 If C<use locale> is in effect, the case map used by C<\l>, C<\L>,
1096 C<\u> and C<\U> is taken from the current locale. See L<perllocale>.
1097 If Unicode (for example, C<\N{}> or wide hex characters of 0x100 or
1098 beyond) is being used, the case map used by C<\l>, C<\L>, C<\u> and
1099 C<\U> is as defined by Unicode.
1101 All systems use the virtual C<"\n"> to represent a line terminator,
1102 called a "newline". There is no such thing as an unvarying, physical
1103 newline character. It is only an illusion that the operating system,
1104 device drivers, C libraries, and Perl all conspire to preserve. Not all
1105 systems read C<"\r"> as ASCII CR and C<"\n"> as ASCII LF. For example,
1106 on a Mac, these are reversed, and on systems without line terminator,
1107 printing C<"\n"> may emit no actual data. In general, use C<"\n"> when
1108 you mean a "newline" for your system, but use the literal ASCII when you
1109 need an exact character. For example, most networking protocols expect
1110 and prefer a CR+LF (C<"\015\012"> or C<"\cM\cJ">) for line terminators,
1111 and although they often accept just C<"\012">, they seldom tolerate just
1112 C<"\015">. If you get in the habit of using C<"\n"> for networking,
1113 you may be burned some day.
1114 X<newline> X<line terminator> X<eol> X<end of line>
1117 For constructs that do interpolate, variables beginning with "C<$>"
1118 or "C<@>" are interpolated. Subscripted variables such as C<$a[3]> or
1119 C<< $href->{key}[0] >> are also interpolated, as are array and hash slices.
1120 But method calls such as C<< $obj->meth >> are not.
1122 Interpolating an array or slice interpolates the elements in order,
1123 separated by the value of C<$">, so is equivalent to interpolating
1124 C<join $", @array>. "Punctuation" arrays such as C<@*> are only
1125 interpolated if the name is enclosed in braces C<@{*}>, but special
1126 arrays C<@_>, C<@+>, and C<@-> are interpolated, even without braces.
1128 You cannot include a literal C<$> or C<@> within a C<\Q> sequence.
1129 An unescaped C<$> or C<@> interpolates the corresponding variable,
1130 while escaping will cause the literal string C<\$> to be inserted.
1131 You'll need to write something like C<m/\Quser\E\@\Qhost/>.
1133 Patterns are subject to an additional level of interpretation as a
1134 regular expression. This is done as a second pass, after variables are
1135 interpolated, so that regular expressions may be incorporated into the
1136 pattern from the variables. If this is not what you want, use C<\Q> to
1137 interpolate a variable literally.
1139 Apart from the behavior described above, Perl does not expand
1140 multiple levels of interpolation. In particular, contrary to the
1141 expectations of shell programmers, back-quotes do I<NOT> interpolate
1142 within double quotes, nor do single quotes impede evaluation of
1143 variables when used within double quotes.
1145 =head2 Regexp Quote-Like Operators
1148 Here are the quote-like operators that apply to pattern
1149 matching and related activities.
1153 =item qr/STRING/msixpo
1154 X<qr> X</i> X</m> X</o> X</s> X</x> X</p>
1156 This operator quotes (and possibly compiles) its I<STRING> as a regular
1157 expression. I<STRING> is interpolated the same way as I<PATTERN>
1158 in C<m/PATTERN/>. If "'" is used as the delimiter, no interpolation
1159 is done. Returns a Perl value which may be used instead of the
1160 corresponding C</STRING/msixpo> expression. The returned value is a
1161 normalized version of the original pattern. It magically differs from
1162 a string containing the same characters: C<ref(qr/x/)> returns "Regexp",
1163 even though dereferencing the result returns undef.
1167 $rex = qr/my.STRING/is;
1168 print $rex; # prints (?si-xm:my.STRING)
1175 The result may be used as a subpattern in a match:
1178 $string =~ /foo${re}bar/; # can be interpolated in other patterns
1179 $string =~ $re; # or used standalone
1180 $string =~ /$re/; # or this way
1182 Since Perl may compile the pattern at the moment of execution of qr()
1183 operator, using qr() may have speed advantages in some situations,
1184 notably if the result of qr() is used standalone:
1187 my $patterns = shift;
1188 my @compiled = map qr/$_/i, @$patterns;
1191 foreach my $pat (@compiled) {
1192 $success = 1, last if /$pat/;
1198 Precompilation of the pattern into an internal representation at
1199 the moment of qr() avoids a need to recompile the pattern every
1200 time a match C</$pat/> is attempted. (Perl has many other internal
1201 optimizations, but none would be triggered in the above example if
1202 we did not use qr() operator.)
1206 m Treat string as multiple lines.
1207 s Treat string as single line. (Make . match a newline)
1208 i Do case-insensitive pattern matching.
1209 x Use extended regular expressions.
1210 p When matching preserve a copy of the matched string so
1211 that ${^PREMATCH}, ${^MATCH}, ${^POSTMATCH} will be defined.
1212 o Compile pattern only once.
1214 If a precompiled pattern is embedded in a larger pattern then the effect
1215 of 'msixp' will be propagated appropriately. The effect of the 'o'
1216 modifier has is not propagated, being restricted to those patterns
1217 explicitly using it.
1219 See L<perlre> for additional information on valid syntax for STRING, and
1220 for a detailed look at the semantics of regular expressions.
1222 =item m/PATTERN/msixpogc
1223 X<m> X<operator, match>
1224 X<regexp, options> X<regexp> X<regex, options> X<regex>
1225 X</m> X</s> X</i> X</x> X</p> X</o> X</g> X</c>
1227 =item /PATTERN/msixpogc
1229 Searches a string for a pattern match, and in scalar context returns
1230 true if it succeeds, false if it fails. If no string is specified
1231 via the C<=~> or C<!~> operator, the $_ string is searched. (The
1232 string specified with C<=~> need not be an lvalue--it may be the
1233 result of an expression evaluation, but remember the C<=~> binds
1234 rather tightly.) See also L<perlre>. See L<perllocale> for
1235 discussion of additional considerations that apply when C<use locale>
1238 Options are as described in C<qr//>; in addition, the following match
1239 process modifiers are available:
1241 g Match globally, i.e., find all occurrences.
1242 c Do not reset search position on a failed match when /g is in effect.
1244 If "/" is the delimiter then the initial C<m> is optional. With the C<m>
1245 you can use any pair of non-whitespace characters
1246 as delimiters. This is particularly useful for matching path names
1247 that contain "/", to avoid LTS (leaning toothpick syndrome). If "?" is
1248 the delimiter, then the match-only-once rule of C<?PATTERN?> applies.
1249 If "'" is the delimiter, no interpolation is performed on the PATTERN.
1250 When using a character valid in an identifier, whitespace is required
1253 PATTERN may contain variables, which will be interpolated (and the
1254 pattern recompiled) every time the pattern search is evaluated, except
1255 for when the delimiter is a single quote. (Note that C<$(>, C<$)>, and
1256 C<$|> are not interpolated because they look like end-of-string tests.)
1257 If you want such a pattern to be compiled only once, add a C</o> after
1258 the trailing delimiter. This avoids expensive run-time recompilations,
1259 and is useful when the value you are interpolating won't change over
1260 the life of the script. However, mentioning C</o> constitutes a promise
1261 that you won't change the variables in the pattern. If you change them,
1262 Perl won't even notice. See also L<"qr/STRING/msixpo">.
1264 =item The empty pattern //
1266 If the PATTERN evaluates to the empty string, the last
1267 I<successfully> matched regular expression is used instead. In this
1268 case, only the C<g> and C<c> flags on the empty pattern is honoured -
1269 the other flags are taken from the original pattern. If no match has
1270 previously succeeded, this will (silently) act instead as a genuine
1271 empty pattern (which will always match).
1273 Note that it's possible to confuse Perl into thinking C<//> (the empty
1274 regex) is really C<//> (the defined-or operator). Perl is usually pretty
1275 good about this, but some pathological cases might trigger this, such as
1276 C<$a///> (is that C<($a) / (//)> or C<$a // />?) and C<print $fh //>
1277 (C<print $fh(//> or C<print($fh //>?). In all of these examples, Perl
1278 will assume you meant defined-or. If you meant the empty regex, just
1279 use parentheses or spaces to disambiguate, or even prefix the empty
1280 regex with an C<m> (so C<//> becomes C<m//>).
1282 =item Matching in list context
1284 If the C</g> option is not used, C<m//> in list context returns a
1285 list consisting of the subexpressions matched by the parentheses in the
1286 pattern, i.e., (C<$1>, C<$2>, C<$3>...). (Note that here C<$1> etc. are
1287 also set, and that this differs from Perl 4's behavior.) When there are
1288 no parentheses in the pattern, the return value is the list C<(1)> for
1289 success. With or without parentheses, an empty list is returned upon
1294 open(TTY, '/dev/tty');
1295 <TTY> =~ /^y/i && foo(); # do foo if desired
1297 if (/Version: *([0-9.]*)/) { $version = $1; }
1299 next if m#^/usr/spool/uucp#;
1304 print if /$arg/o; # compile only once
1307 if (($F1, $F2, $Etc) = ($foo =~ /^(\S+)\s+(\S+)\s*(.*)/))
1309 This last example splits $foo into the first two words and the
1310 remainder of the line, and assigns those three fields to $F1, $F2, and
1311 $Etc. The conditional is true if any variables were assigned, i.e., if
1312 the pattern matched.
1314 The C</g> modifier specifies global pattern matching--that is,
1315 matching as many times as possible within the string. How it behaves
1316 depends on the context. In list context, it returns a list of the
1317 substrings matched by any capturing parentheses in the regular
1318 expression. If there are no parentheses, it returns a list of all
1319 the matched strings, as if there were parentheses around the whole
1322 In scalar context, each execution of C<m//g> finds the next match,
1323 returning true if it matches, and false if there is no further match.
1324 The position after the last match can be read or set using the pos()
1325 function; see L<perlfunc/pos>. A failed match normally resets the
1326 search position to the beginning of the string, but you can avoid that
1327 by adding the C</c> modifier (e.g. C<m//gc>). Modifying the target
1328 string also resets the search position.
1332 You can intermix C<m//g> matches with C<m/\G.../g>, where C<\G> is a
1333 zero-width assertion that matches the exact position where the previous
1334 C<m//g>, if any, left off. Without the C</g> modifier, the C<\G> assertion
1335 still anchors at pos(), but the match is of course only attempted once.
1336 Using C<\G> without C</g> on a target string that has not previously had a
1337 C</g> match applied to it is the same as using the C<\A> assertion to match
1338 the beginning of the string. Note also that, currently, C<\G> is only
1339 properly supported when anchored at the very beginning of the pattern.
1344 ($one,$five,$fifteen) = (`uptime` =~ /(\d+\.\d+)/g);
1348 while (defined($paragraph = <>)) {
1349 while ($paragraph =~ /[a-z]['")]*[.!?]+['")]*\s/g) {
1353 print "$sentences\n";
1355 # using m//gc with \G
1359 print $1 while /(o)/gc; print "', pos=", pos, "\n";
1361 print $1 if /\G(q)/gc; print "', pos=", pos, "\n";
1363 print $1 while /(p)/gc; print "', pos=", pos, "\n";
1365 print "Final: '$1', pos=",pos,"\n" if /\G(.)/;
1367 The last example should print:
1377 Notice that the final match matched C<q> instead of C<p>, which a match
1378 without the C<\G> anchor would have done. Also note that the final match
1379 did not update C<pos>. C<pos> is only updated on a C</g> match. If the
1380 final match did indeed match C<p>, it's a good bet that you're running an
1381 older (pre-5.6.0) Perl.
1383 A useful idiom for C<lex>-like scanners is C</\G.../gc>. You can
1384 combine several regexps like this to process a string part-by-part,
1385 doing different actions depending on which regexp matched. Each
1386 regexp tries to match where the previous one leaves off.
1389 $url = URI::URL->new( "http://example.com/" ); die if $url eq "xXx";
1393 print(" digits"), redo LOOP if /\G\d+\b[,.;]?\s*/gc;
1394 print(" lowercase"), redo LOOP if /\G[a-z]+\b[,.;]?\s*/gc;
1395 print(" UPPERCASE"), redo LOOP if /\G[A-Z]+\b[,.;]?\s*/gc;
1396 print(" Capitalized"), redo LOOP if /\G[A-Z][a-z]+\b[,.;]?\s*/gc;
1397 print(" MiXeD"), redo LOOP if /\G[A-Za-z]+\b[,.;]?\s*/gc;
1398 print(" alphanumeric"), redo LOOP if /\G[A-Za-z0-9]+\b[,.;]?\s*/gc;
1399 print(" line-noise"), redo LOOP if /\G[^A-Za-z0-9]+/gc;
1400 print ". That's all!\n";
1403 Here is the output (split into several lines):
1405 line-noise lowercase line-noise lowercase UPPERCASE line-noise
1406 UPPERCASE line-noise lowercase line-noise lowercase line-noise
1407 lowercase lowercase line-noise lowercase lowercase line-noise
1408 MiXeD line-noise. That's all!
1413 This is just like the C</pattern/> search, except that it matches only
1414 once between calls to the reset() operator. This is a useful
1415 optimization when you want to see only the first occurrence of
1416 something in each file of a set of files, for instance. Only C<??>
1417 patterns local to the current package are reset.
1421 # blank line between header and body
1424 reset if eof; # clear ?? status for next file
1427 This usage is vaguely deprecated, which means it just might possibly
1428 be removed in some distant future version of Perl, perhaps somewhere
1429 around the year 2168.
1431 =item s/PATTERN/REPLACEMENT/msixpogce
1432 X<substitute> X<substitution> X<replace> X<regexp, replace>
1433 X<regexp, substitute> X</m> X</s> X</i> X</x> X</p> X</o> X</g> X</c> X</e>
1435 Searches a string for a pattern, and if found, replaces that pattern
1436 with the replacement text and returns the number of substitutions
1437 made. Otherwise it returns false (specifically, the empty string).
1439 If no string is specified via the C<=~> or C<!~> operator, the C<$_>
1440 variable is searched and modified. (The string specified with C<=~> must
1441 be scalar variable, an array element, a hash element, or an assignment
1442 to one of those, i.e., an lvalue.)
1444 If the delimiter chosen is a single quote, no interpolation is
1445 done on either the PATTERN or the REPLACEMENT. Otherwise, if the
1446 PATTERN contains a $ that looks like a variable rather than an
1447 end-of-string test, the variable will be interpolated into the pattern
1448 at run-time. If you want the pattern compiled only once the first time
1449 the variable is interpolated, use the C</o> option. If the pattern
1450 evaluates to the empty string, the last successfully executed regular
1451 expression is used instead. See L<perlre> for further explanation on these.
1452 See L<perllocale> for discussion of additional considerations that apply
1453 when C<use locale> is in effect.
1455 Options are as with m// with the addition of the following replacement
1458 e Evaluate the right side as an expression.
1459 ee Evaluate the right side as a string then eval the result
1461 Any non-whitespace delimiter may replace the slashes. Add space after
1462 the C<s> when using a character allowed in identifiers. If single quotes
1463 are used, no interpretation is done on the replacement string (the C</e>
1464 modifier overrides this, however). Unlike Perl 4, Perl 5 treats backticks
1465 as normal delimiters; the replacement text is not evaluated as a command.
1466 If the PATTERN is delimited by bracketing quotes, the REPLACEMENT has
1467 its own pair of quotes, which may or may not be bracketing quotes, e.g.,
1468 C<s(foo)(bar)> or C<< s<foo>/bar/ >>. A C</e> will cause the
1469 replacement portion to be treated as a full-fledged Perl expression
1470 and evaluated right then and there. It is, however, syntax checked at
1471 compile-time. A second C<e> modifier will cause the replacement portion
1472 to be C<eval>ed before being run as a Perl expression.
1476 s/\bgreen\b/mauve/g; # don't change wintergreen
1478 $path =~ s|/usr/bin|/usr/local/bin|;
1480 s/Login: $foo/Login: $bar/; # run-time pattern
1482 ($foo = $bar) =~ s/this/that/; # copy first, then change
1484 $count = ($paragraph =~ s/Mister\b/Mr./g); # get change-count
1487 s/\d+/$&*2/e; # yields 'abc246xyz'
1488 s/\d+/sprintf("%5d",$&)/e; # yields 'abc 246xyz'
1489 s/\w/$& x 2/eg; # yields 'aabbcc 224466xxyyzz'
1491 s/%(.)/$percent{$1}/g; # change percent escapes; no /e
1492 s/%(.)/$percent{$1} || $&/ge; # expr now, so /e
1493 s/^=(\w+)/pod($1)/ge; # use function call
1495 # expand variables in $_, but dynamics only, using
1496 # symbolic dereferencing
1499 # Add one to the value of any numbers in the string
1502 # This will expand any embedded scalar variable
1503 # (including lexicals) in $_ : First $1 is interpolated
1504 # to the variable name, and then evaluated
1507 # Delete (most) C comments.
1509 /\* # Match the opening delimiter.
1510 .*? # Match a minimal number of characters.
1511 \*/ # Match the closing delimiter.
1514 s/^\s*(.*?)\s*$/$1/; # trim whitespace in $_, expensively
1516 for ($variable) { # trim whitespace in $variable, cheap
1521 s/([^ ]*) *([^ ]*)/$2 $1/; # reverse 1st two fields
1523 Note the use of $ instead of \ in the last example. Unlike
1524 B<sed>, we use the \<I<digit>> form in only the left hand side.
1525 Anywhere else it's $<I<digit>>.
1527 Occasionally, you can't use just a C</g> to get all the changes
1528 to occur that you might want. Here are two common cases:
1530 # put commas in the right places in an integer
1531 1 while s/(\d)(\d\d\d)(?!\d)/$1,$2/g;
1533 # expand tabs to 8-column spacing
1534 1 while s/\t+/' ' x (length($&)*8 - length($`)%8)/e;
1538 =head2 Quote-Like Operators
1539 X<operator, quote-like>
1544 X<q> X<quote, single> X<'> X<''>
1548 A single-quoted, literal string. A backslash represents a backslash
1549 unless followed by the delimiter or another backslash, in which case
1550 the delimiter or backslash is interpolated.
1552 $foo = q!I said, "You said, 'She said it.'"!;
1553 $bar = q('This is it.');
1554 $baz = '\n'; # a two-character string
1557 X<qq> X<quote, double> X<"> X<"">
1561 A double-quoted, interpolated string.
1564 (*** The previous line contains the naughty word "$1".\n)
1565 if /\b(tcl|java|python)\b/i; # :-)
1566 $baz = "\n"; # a one-character string
1569 X<qx> X<`> X<``> X<backtick>
1573 A string which is (possibly) interpolated and then executed as a
1574 system command with C</bin/sh> or its equivalent. Shell wildcards,
1575 pipes, and redirections will be honored. The collected standard
1576 output of the command is returned; standard error is unaffected. In
1577 scalar context, it comes back as a single (potentially multi-line)
1578 string, or undef if the command failed. In list context, returns a
1579 list of lines (however you've defined lines with $/ or
1580 $INPUT_RECORD_SEPARATOR), or an empty list if the command failed.
1582 Because backticks do not affect standard error, use shell file descriptor
1583 syntax (assuming the shell supports this) if you care to address this.
1584 To capture a command's STDERR and STDOUT together:
1586 $output = `cmd 2>&1`;
1588 To capture a command's STDOUT but discard its STDERR:
1590 $output = `cmd 2>/dev/null`;
1592 To capture a command's STDERR but discard its STDOUT (ordering is
1595 $output = `cmd 2>&1 1>/dev/null`;
1597 To exchange a command's STDOUT and STDERR in order to capture the STDERR
1598 but leave its STDOUT to come out the old STDERR:
1600 $output = `cmd 3>&1 1>&2 2>&3 3>&-`;
1602 To read both a command's STDOUT and its STDERR separately, it's easiest
1603 to redirect them separately to files, and then read from those files
1604 when the program is done:
1606 system("program args 1>program.stdout 2>program.stderr");
1608 The STDIN filehandle used by the command is inherited from Perl's STDIN.
1611 open BLAM, "blam" || die "Can't open: $!";
1612 open STDIN, "<&BLAM";
1615 will print the sorted contents of the file "blam".
1617 Using single-quote as a delimiter protects the command from Perl's
1618 double-quote interpolation, passing it on to the shell instead:
1620 $perl_info = qx(ps $$); # that's Perl's $$
1621 $shell_info = qx'ps $$'; # that's the new shell's $$
1623 How that string gets evaluated is entirely subject to the command
1624 interpreter on your system. On most platforms, you will have to protect
1625 shell metacharacters if you want them treated literally. This is in
1626 practice difficult to do, as it's unclear how to escape which characters.
1627 See L<perlsec> for a clean and safe example of a manual fork() and exec()
1628 to emulate backticks safely.
1630 On some platforms (notably DOS-like ones), the shell may not be
1631 capable of dealing with multiline commands, so putting newlines in
1632 the string may not get you what you want. You may be able to evaluate
1633 multiple commands in a single line by separating them with the command
1634 separator character, if your shell supports that (e.g. C<;> on many Unix
1635 shells; C<&> on the Windows NT C<cmd> shell).
1637 Beginning with v5.6.0, Perl will attempt to flush all files opened for
1638 output before starting the child process, but this may not be supported
1639 on some platforms (see L<perlport>). To be safe, you may need to set
1640 C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method of
1641 C<IO::Handle> on any open handles.
1643 Beware that some command shells may place restrictions on the length
1644 of the command line. You must ensure your strings don't exceed this
1645 limit after any necessary interpolations. See the platform-specific
1646 release notes for more details about your particular environment.
1648 Using this operator can lead to programs that are difficult to port,
1649 because the shell commands called vary between systems, and may in
1650 fact not be present at all. As one example, the C<type> command under
1651 the POSIX shell is very different from the C<type> command under DOS.
1652 That doesn't mean you should go out of your way to avoid backticks
1653 when they're the right way to get something done. Perl was made to be
1654 a glue language, and one of the things it glues together is commands.
1655 Just understand what you're getting yourself into.
1657 See L</"I/O Operators"> for more discussion.
1660 X<qw> X<quote, list> X<quote, words>
1662 Evaluates to a list of the words extracted out of STRING, using embedded
1663 whitespace as the word delimiters. It can be understood as being roughly
1666 split(' ', q/STRING/);
1668 the differences being that it generates a real list at compile time, and
1669 in scalar context it returns the last element in the list. So
1674 is semantically equivalent to the list:
1678 Some frequently seen examples:
1680 use POSIX qw( setlocale localeconv )
1681 @EXPORT = qw( foo bar baz );
1683 A common mistake is to try to separate the words with comma or to
1684 put comments into a multi-line C<qw>-string. For this reason, the
1685 C<use warnings> pragma and the B<-w> switch (that is, the C<$^W> variable)
1686 produces warnings if the STRING contains the "," or the "#" character.
1689 =item tr/SEARCHLIST/REPLACEMENTLIST/cds
1690 X<tr> X<y> X<transliterate> X</c> X</d> X</s>
1692 =item y/SEARCHLIST/REPLACEMENTLIST/cds
1694 Transliterates all occurrences of the characters found in the search list
1695 with the corresponding character in the replacement list. It returns
1696 the number of characters replaced or deleted. If no string is
1697 specified via the =~ or !~ operator, the $_ string is transliterated. (The
1698 string specified with =~ must be a scalar variable, an array element, a
1699 hash element, or an assignment to one of those, i.e., an lvalue.)
1701 A character range may be specified with a hyphen, so C<tr/A-J/0-9/>
1702 does the same replacement as C<tr/ACEGIBDFHJ/0246813579/>.
1703 For B<sed> devotees, C<y> is provided as a synonym for C<tr>. If the
1704 SEARCHLIST is delimited by bracketing quotes, the REPLACEMENTLIST has
1705 its own pair of quotes, which may or may not be bracketing quotes,
1706 e.g., C<tr[A-Z][a-z]> or C<tr(+\-*/)/ABCD/>.
1708 Note that C<tr> does B<not> do regular expression character classes
1709 such as C<\d> or C<[:lower:]>. The C<tr> operator is not equivalent to
1710 the tr(1) utility. If you want to map strings between lower/upper
1711 cases, see L<perlfunc/lc> and L<perlfunc/uc>, and in general consider
1712 using the C<s> operator if you need regular expressions.
1714 Note also that the whole range idea is rather unportable between
1715 character sets--and even within character sets they may cause results
1716 you probably didn't expect. A sound principle is to use only ranges
1717 that begin from and end at either alphabets of equal case (a-e, A-E),
1718 or digits (0-4). Anything else is unsafe. If in doubt, spell out the
1719 character sets in full.
1723 c Complement the SEARCHLIST.
1724 d Delete found but unreplaced characters.
1725 s Squash duplicate replaced characters.
1727 If the C</c> modifier is specified, the SEARCHLIST character set
1728 is complemented. If the C</d> modifier is specified, any characters
1729 specified by SEARCHLIST not found in REPLACEMENTLIST are deleted.
1730 (Note that this is slightly more flexible than the behavior of some
1731 B<tr> programs, which delete anything they find in the SEARCHLIST,
1732 period.) If the C</s> modifier is specified, sequences of characters
1733 that were transliterated to the same character are squashed down
1734 to a single instance of the character.
1736 If the C</d> modifier is used, the REPLACEMENTLIST is always interpreted
1737 exactly as specified. Otherwise, if the REPLACEMENTLIST is shorter
1738 than the SEARCHLIST, the final character is replicated till it is long
1739 enough. If the REPLACEMENTLIST is empty, the SEARCHLIST is replicated.
1740 This latter is useful for counting characters in a class or for
1741 squashing character sequences in a class.
1745 $ARGV[1] =~ tr/A-Z/a-z/; # canonicalize to lower case
1747 $cnt = tr/*/*/; # count the stars in $_
1749 $cnt = $sky =~ tr/*/*/; # count the stars in $sky
1751 $cnt = tr/0-9//; # count the digits in $_
1753 tr/a-zA-Z//s; # bookkeeper -> bokeper
1755 ($HOST = $host) =~ tr/a-z/A-Z/;
1757 tr/a-zA-Z/ /cs; # change non-alphas to single space
1760 [\000-\177]; # delete 8th bit
1762 If multiple transliterations are given for a character, only the
1767 will transliterate any A to X.
1769 Because the transliteration table is built at compile time, neither
1770 the SEARCHLIST nor the REPLACEMENTLIST are subjected to double quote
1771 interpolation. That means that if you want to use variables, you
1774 eval "tr/$oldlist/$newlist/";
1777 eval "tr/$oldlist/$newlist/, 1" or die $@;
1780 X<here-doc> X<heredoc> X<here-document> X<<< << >>>
1782 A line-oriented form of quoting is based on the shell "here-document"
1783 syntax. Following a C<< << >> you specify a string to terminate
1784 the quoted material, and all lines following the current line down to
1785 the terminating string are the value of the item.
1787 The terminating string may be either an identifier (a word), or some
1788 quoted text. An unquoted identifier works like double quotes.
1789 There may not be a space between the C<< << >> and the identifier,
1790 unless the identifier is explicitly quoted. (If you put a space it
1791 will be treated as a null identifier, which is valid, and matches the
1792 first empty line.) The terminating string must appear by itself
1793 (unquoted and with no surrounding whitespace) on the terminating line.
1795 If the terminating string is quoted, the type of quotes used determine
1796 the treatment of the text.
1802 Double quotes indicate that the text will be interpolated using exactly
1803 the same rules as normal double quoted strings.
1806 The price is $Price.
1809 print << "EOF"; # same as above
1810 The price is $Price.
1816 Single quotes indicate the text is to be treated literally with no
1817 interpolation of its content. This is similar to single quoted
1818 strings except that backslashes have no special meaning, with C<\\>
1819 being treated as two backslashes and not one as they would in every
1820 other quoting construct.
1822 This is the only form of quoting in perl where there is no need
1823 to worry about escaping content, something that code generators
1824 can and do make good use of.
1828 The content of the here doc is treated just as it would be if the
1829 string were embedded in backticks. Thus the content is interpolated
1830 as though it were double quoted and then executed via the shell, with
1831 the results of the execution returned.
1833 print << `EOC`; # execute command and get results
1839 It is possible to stack multiple here-docs in a row:
1841 print <<"foo", <<"bar"; # you can stack them
1847 myfunc(<< "THIS", 23, <<'THAT');
1854 Just don't forget that you have to put a semicolon on the end
1855 to finish the statement, as Perl doesn't know you're not going to
1863 If you want to remove the line terminator from your here-docs,
1866 chomp($string = <<'END');
1870 If you want your here-docs to be indented with the rest of the code,
1871 you'll need to remove leading whitespace from each line manually:
1873 ($quote = <<'FINIS') =~ s/^\s+//gm;
1874 The Road goes ever on and on,
1875 down from the door where it began.
1878 If you use a here-doc within a delimited construct, such as in C<s///eg>,
1879 the quoted material must come on the lines following the final delimiter.
1894 If the terminating identifier is on the last line of the program, you
1895 must be sure there is a newline after it; otherwise, Perl will give the
1896 warning B<Can't find string terminator "END" anywhere before EOF...>.
1898 Additionally, the quoting rules for the end of string identifier are not
1899 related to Perl's quoting rules. C<q()>, C<qq()>, and the like are not
1900 supported in place of C<''> and C<"">, and the only interpolation is for
1901 backslashing the quoting character:
1903 print << "abc\"def";
1907 Finally, quoted strings cannot span multiple lines. The general rule is
1908 that the identifier must be a string literal. Stick with that, and you
1913 =head2 Gory details of parsing quoted constructs
1914 X<quote, gory details>
1916 When presented with something that might have several different
1917 interpretations, Perl uses the B<DWIM> (that's "Do What I Mean")
1918 principle to pick the most probable interpretation. This strategy
1919 is so successful that Perl programmers often do not suspect the
1920 ambivalence of what they write. But from time to time, Perl's
1921 notions differ substantially from what the author honestly meant.
1923 This section hopes to clarify how Perl handles quoted constructs.
1924 Although the most common reason to learn this is to unravel labyrinthine
1925 regular expressions, because the initial steps of parsing are the
1926 same for all quoting operators, they are all discussed together.
1928 The most important Perl parsing rule is the first one discussed
1929 below: when processing a quoted construct, Perl first finds the end
1930 of that construct, then interprets its contents. If you understand
1931 this rule, you may skip the rest of this section on the first
1932 reading. The other rules are likely to contradict the user's
1933 expectations much less frequently than this first one.
1935 Some passes discussed below are performed concurrently, but because
1936 their results are the same, we consider them individually. For different
1937 quoting constructs, Perl performs different numbers of passes, from
1938 one to four, but these passes are always performed in the same order.
1942 =item Finding the end
1944 The first pass is finding the end of the quoted construct, where
1945 the information about the delimiters is used in parsing.
1946 During this search, text between the starting and ending delimiters
1947 is copied to a safe location. The text copied gets delimiter-independent.
1949 If the construct is a here-doc, the ending delimiter is a line
1950 that has a terminating string as the content. Therefore C<<<EOF> is
1951 terminated by C<EOF> immediately followed by C<"\n"> and starting
1952 from the first column of the terminating line.
1953 When searching for the terminating line of a here-doc, nothing
1954 is skipped. In other words, lines after the here-doc syntax
1955 are compared with the terminating string line by line.
1957 For the constructs except here-docs, single characters are used as starting
1958 and ending delimiters. If the starting delimiter is an opening punctuation
1959 (that is C<(>, C<[>, C<{>, or C<< < >>), the ending delimiter is the
1960 corresponding closing punctuation (that is C<)>, C<]>, C<}>, or C<< > >>).
1961 If the starting delimiter is an unpaired character like C</> or a closing
1962 punctuation, the ending delimiter is same as the starting delimiter.
1963 Therefore a C</> terminates a C<qq//> construct, while a C<]> terminates
1964 C<qq[]> and C<qq]]> constructs.
1966 When searching for single-character delimiters, escaped delimiters
1967 and C<\\> are skipped. For example, while searching for terminating C</>,
1968 combinations of C<\\> and C<\/> are skipped. If the delimiters are
1969 bracketing, nested pairs are also skipped. For example, while searching
1970 for closing C<]> paired with the opening C<[>, combinations of C<\\>, C<\]>,
1971 and C<\[> are all skipped, and nested C<[> and C<]> are skipped as well.
1972 However, when backslashes are used as the delimiters (like C<qq\\> and
1973 C<tr\\\>), nothing is skipped.
1974 During the search for the end, backslashes that escape delimiters
1975 are removed (exactly speaking, they are not copied to the safe location).
1977 For constructs with three-part delimiters (C<s///>, C<y///>, and
1978 C<tr///>), the search is repeated once more.
1979 If the first delimiter is not an opening punctuation, three delimiters must
1980 be same such as C<s!!!> and C<tr)))>, in which case the second delimiter
1981 terminates the left part and starts the right part at once.
1982 If the left part is delimited by bracketing punctuations (that is C<()>,
1983 C<[]>, C<{}>, or C<< <> >>), the right part needs another pair of
1984 delimiters such as C<s(){}> and C<tr[]//>. In these cases, whitespaces
1985 and comments are allowed between both parts, though the comment must follow
1986 at least one whitespace; otherwise a character expected as the start of
1987 the comment may be regarded as the starting delimiter of the right part.
1989 During this search no attention is paid to the semantics of the construct.
1992 "$hash{"$foo/$bar"}"
1997 bar # NOT a comment, this slash / terminated m//!
2000 do not form legal quoted expressions. The quoted part ends on the
2001 first C<"> and C</>, and the rest happens to be a syntax error.
2002 Because the slash that terminated C<m//> was followed by a C<SPACE>,
2003 the example above is not C<m//x>, but rather C<m//> with no C</x>
2004 modifier. So the embedded C<#> is interpreted as a literal C<#>.
2006 Also no attention is paid to C<\c\> (multichar control char syntax) during
2007 this search. Thus the second C<\> in C<qq/\c\/> is interpreted as a part
2008 of C<\/>, and the following C</> is not recognized as a delimiter.
2009 Instead, use C<\034> or C<\x1c> at the end of quoted constructs.
2014 The next step is interpolation in the text obtained, which is now
2015 delimiter-independent. There are multiple cases.
2021 No interpolation is performed.
2022 Note that the combination C<\\> is left intact, since escaped delimiters
2023 are not available for here-docs.
2025 =item C<m''>, the pattern of C<s'''>
2027 No interpolation is performed at this stage.
2028 Any backslashed sequences including C<\\> are treated at the stage
2029 to L</"parsing regular expressions">.
2031 =item C<''>, C<q//>, C<tr'''>, C<y'''>, the replacement of C<s'''>
2033 The only interpolation is removal of C<\> from pairs of C<\\>.
2034 Therefore C<-> in C<tr'''> and C<y'''> is treated literally
2035 as a hyphen and no character range is available.
2036 C<\1> in the replacement of C<s'''> does not work as C<$1>.
2038 =item C<tr///>, C<y///>
2040 No variable interpolation occurs. String modifying combinations for
2041 case and quoting such as C<\Q>, C<\U>, and C<\E> are not recognized.
2042 The other escape sequences such as C<\200> and C<\t> and backslashed
2043 characters such as C<\\> and C<\-> are converted to appropriate literals.
2044 The character C<-> is treated specially and therefore C<\-> is treated
2047 =item C<"">, C<``>, C<qq//>, C<qx//>, C<< <file*glob> >>, C<<<"EOF">
2049 C<\Q>, C<\U>, C<\u>, C<\L>, C<\l> (possibly paired with C<\E>) are
2050 converted to corresponding Perl constructs. Thus, C<"$foo\Qbaz$bar">
2051 is converted to C<$foo . (quotemeta("baz" . $bar))> internally.
2052 The other escape sequences such as C<\200> and C<\t> and backslashed
2053 characters such as C<\\> and C<\-> are replaced with appropriate
2056 Let it be stressed that I<whatever falls between C<\Q> and C<\E>>
2057 is interpolated in the usual way. Something like C<"\Q\\E"> has
2058 no C<\E> inside. instead, it has C<\Q>, C<\\>, and C<E>, so the
2059 result is the same as for C<"\\\\E">. As a general rule, backslashes
2060 between C<\Q> and C<\E> may lead to counterintuitive results. So,
2061 C<"\Q\t\E"> is converted to C<quotemeta("\t")>, which is the same
2062 as C<"\\\t"> (since TAB is not alphanumeric). Note also that:
2067 may be closer to the conjectural I<intention> of the writer of C<"\Q\t\E">.
2069 Interpolated scalars and arrays are converted internally to the C<join> and
2070 C<.> catenation operations. Thus, C<"$foo XXX '@arr'"> becomes:
2072 $foo . " XXX '" . (join $", @arr) . "'";
2074 All operations above are performed simultaneously, left to right.
2076 Because the result of C<"\Q STRING \E"> has all metacharacters
2077 quoted, there is no way to insert a literal C<$> or C<@> inside a
2078 C<\Q\E> pair. If protected by C<\>, C<$> will be quoted to became
2079 C<"\\\$">; if not, it is interpreted as the start of an interpolated
2082 Note also that the interpolation code needs to make a decision on
2083 where the interpolated scalar ends. For instance, whether
2084 C<< "a $b -> {c}" >> really means:
2086 "a " . $b . " -> {c}";
2092 Most of the time, the longest possible text that does not include
2093 spaces between components and which contains matching braces or
2094 brackets. because the outcome may be determined by voting based
2095 on heuristic estimators, the result is not strictly predictable.
2096 Fortunately, it's usually correct for ambiguous cases.
2098 =item the replacement of C<s///>
2100 Processing of C<\Q>, C<\U>, C<\u>, C<\L>, C<\l>, and interpolation
2101 happens as with C<qq//> constructs.
2103 It is at this step that C<\1> is begrudgingly converted to C<$1> in
2104 the replacement text of C<s///>, in order to correct the incorrigible
2105 I<sed> hackers who haven't picked up the saner idiom yet. A warning
2106 is emitted if the C<use warnings> pragma or the B<-w> command-line flag
2107 (that is, the C<$^W> variable) was set.
2109 =item C<RE> in C<?RE?>, C</RE/>, C<m/RE/>, C<s/RE/foo/>,
2111 Processing of C<\Q>, C<\U>, C<\u>, C<\L>, C<\l>, C<\E>,
2112 and interpolation happens (almost) as with C<qq//> constructs.
2114 Processing of C<\N{...}> is also done here, and compiled into an intermediate
2115 form for the regex compiler. (This is because, as mentioned below, the regex
2116 compilation may be done at execution time, and C<\N{...}> is a compile-time
2119 However any other combinations of C<\> followed by a character
2120 are not substituted but only skipped, in order to parse them
2121 as regular expressions at the following step.
2122 As C<\c> is skipped at this step, C<@> of C<\c@> in RE is possibly
2123 treated as an array symbol (for example C<@foo>),
2124 even though the same text in C<qq//> gives interpolation of C<\c@>.
2126 Moreover, inside C<(?{BLOCK})>, C<(?# comment )>, and
2127 a C<#>-comment in a C<//x>-regular expression, no processing is
2128 performed whatsoever. This is the first step at which the presence
2129 of the C<//x> modifier is relevant.
2131 Interpolation in patterns has several quirks: C<$|>, C<$(>, C<$)>, C<@+>
2132 and C<@-> are not interpolated, and constructs C<$var[SOMETHING]> are
2133 voted (by several different estimators) to be either an array element
2134 or C<$var> followed by an RE alternative. This is where the notation
2135 C<${arr[$bar]}> comes handy: C</${arr[0-9]}/> is interpreted as
2136 array element C<-9>, not as a regular expression from the variable
2137 C<$arr> followed by a digit, which would be the interpretation of
2138 C</$arr[0-9]/>. Since voting among different estimators may occur,
2139 the result is not predictable.
2141 The lack of processing of C<\\> creates specific restrictions on
2142 the post-processed text. If the delimiter is C</>, one cannot get
2143 the combination C<\/> into the result of this step. C</> will
2144 finish the regular expression, C<\/> will be stripped to C</> on
2145 the previous step, and C<\\/> will be left as is. Because C</> is
2146 equivalent to C<\/> inside a regular expression, this does not
2147 matter unless the delimiter happens to be character special to the
2148 RE engine, such as in C<s*foo*bar*>, C<m[foo]>, or C<?foo?>; or an
2149 alphanumeric char, as in:
2153 In the RE above, which is intentionally obfuscated for illustration, the
2154 delimiter is C<m>, the modifier is C<mx>, and after delimiter-removal the
2155 RE is the same as for C<m/ ^ a \s* b /mx>. There's more than one
2156 reason you're encouraged to restrict your delimiters to non-alphanumeric,
2157 non-whitespace choices.
2161 This step is the last one for all constructs except regular expressions,
2162 which are processed further.
2164 =item parsing regular expressions
2167 Previous steps were performed during the compilation of Perl code,
2168 but this one happens at run time, although it may be optimized to
2169 be calculated at compile time if appropriate. After preprocessing
2170 described above, and possibly after evaluation if concatenation,
2171 joining, casing translation, or metaquoting are involved, the
2172 resulting I<string> is passed to the RE engine for compilation.
2174 Whatever happens in the RE engine might be better discussed in L<perlre>,
2175 but for the sake of continuity, we shall do so here.
2177 This is another step where the presence of the C<//x> modifier is
2178 relevant. The RE engine scans the string from left to right and
2179 converts it to a finite automaton.
2181 Backslashed characters are either replaced with corresponding
2182 literal strings (as with C<\{>), or else they generate special nodes
2183 in the finite automaton (as with C<\b>). Characters special to the
2184 RE engine (such as C<|>) generate corresponding nodes or groups of
2185 nodes. C<(?#...)> comments are ignored. All the rest is either
2186 converted to literal strings to match, or else is ignored (as is
2187 whitespace and C<#>-style comments if C<//x> is present).
2189 Parsing of the bracketed character class construct, C<[...]>, is
2190 rather different than the rule used for the rest of the pattern.
2191 The terminator of this construct is found using the same rules as
2192 for finding the terminator of a C<{}>-delimited construct, the only
2193 exception being that C<]> immediately following C<[> is treated as
2194 though preceded by a backslash. Similarly, the terminator of
2195 C<(?{...})> is found using the same rules as for finding the
2196 terminator of a C<{}>-delimited construct.
2198 It is possible to inspect both the string given to RE engine and the
2199 resulting finite automaton. See the arguments C<debug>/C<debugcolor>
2200 in the C<use L<re>> pragma, as well as Perl's B<-Dr> command-line
2201 switch documented in L<perlrun/"Command Switches">.
2203 =item Optimization of regular expressions
2204 X<regexp, optimization>
2206 This step is listed for completeness only. Since it does not change
2207 semantics, details of this step are not documented and are subject
2208 to change without notice. This step is performed over the finite
2209 automaton that was generated during the previous pass.
2211 It is at this stage that C<split()> silently optimizes C</^/> to
2216 =head2 I/O Operators
2217 X<operator, i/o> X<operator, io> X<io> X<while> X<filehandle>
2220 There are several I/O operators you should know about.
2222 A string enclosed by backticks (grave accents) first undergoes
2223 double-quote interpolation. It is then interpreted as an external
2224 command, and the output of that command is the value of the
2225 backtick string, like in a shell. In scalar context, a single string
2226 consisting of all output is returned. In list context, a list of
2227 values is returned, one per line of output. (You can set C<$/> to use
2228 a different line terminator.) The command is executed each time the
2229 pseudo-literal is evaluated. The status value of the command is
2230 returned in C<$?> (see L<perlvar> for the interpretation of C<$?>).
2231 Unlike in B<csh>, no translation is done on the return data--newlines
2232 remain newlines. Unlike in any of the shells, single quotes do not
2233 hide variable names in the command from interpretation. To pass a
2234 literal dollar-sign through to the shell you need to hide it with a
2235 backslash. The generalized form of backticks is C<qx//>. (Because
2236 backticks always undergo shell expansion as well, see L<perlsec> for
2238 X<qx> X<`> X<``> X<backtick> X<glob>
2240 In scalar context, evaluating a filehandle in angle brackets yields
2241 the next line from that file (the newline, if any, included), or
2242 C<undef> at end-of-file or on error. When C<$/> is set to C<undef>
2243 (sometimes known as file-slurp mode) and the file is empty, it
2244 returns C<''> the first time, followed by C<undef> subsequently.
2246 Ordinarily you must assign the returned value to a variable, but
2247 there is one situation where an automatic assignment happens. If
2248 and only if the input symbol is the only thing inside the conditional
2249 of a C<while> statement (even if disguised as a C<for(;;)> loop),
2250 the value is automatically assigned to the global variable $_,
2251 destroying whatever was there previously. (This may seem like an
2252 odd thing to you, but you'll use the construct in almost every Perl
2253 script you write.) The $_ variable is not implicitly localized.
2254 You'll have to put a C<local $_;> before the loop if you want that
2257 The following lines are equivalent:
2259 while (defined($_ = <STDIN>)) { print; }
2260 while ($_ = <STDIN>) { print; }
2261 while (<STDIN>) { print; }
2262 for (;<STDIN>;) { print; }
2263 print while defined($_ = <STDIN>);
2264 print while ($_ = <STDIN>);
2265 print while <STDIN>;
2267 This also behaves similarly, but avoids $_ :
2269 while (my $line = <STDIN>) { print $line }
2271 In these loop constructs, the assigned value (whether assignment
2272 is automatic or explicit) is then tested to see whether it is
2273 defined. The defined test avoids problems where line has a string
2274 value that would be treated as false by Perl, for example a "" or
2275 a "0" with no trailing newline. If you really mean for such values
2276 to terminate the loop, they should be tested for explicitly:
2278 while (($_ = <STDIN>) ne '0') { ... }
2279 while (<STDIN>) { last unless $_; ... }
2281 In other boolean contexts, C<< <filehandle> >> without an
2282 explicit C<defined> test or comparison elicits a warning if the
2283 C<use warnings> pragma or the B<-w>
2284 command-line switch (the C<$^W> variable) is in effect.
2286 The filehandles STDIN, STDOUT, and STDERR are predefined. (The
2287 filehandles C<stdin>, C<stdout>, and C<stderr> will also work except
2288 in packages, where they would be interpreted as local identifiers
2289 rather than global.) Additional filehandles may be created with
2290 the open() function, amongst others. See L<perlopentut> and
2291 L<perlfunc/open> for details on this.
2292 X<stdin> X<stdout> X<sterr>
2294 If a <FILEHANDLE> is used in a context that is looking for
2295 a list, a list comprising all input lines is returned, one line per
2296 list element. It's easy to grow to a rather large data space this
2297 way, so use with care.
2299 <FILEHANDLE> may also be spelled C<readline(*FILEHANDLE)>.
2300 See L<perlfunc/readline>.
2302 The null filehandle <> is special: it can be used to emulate the
2303 behavior of B<sed> and B<awk>. Input from <> comes either from
2304 standard input, or from each file listed on the command line. Here's
2305 how it works: the first time <> is evaluated, the @ARGV array is
2306 checked, and if it is empty, C<$ARGV[0]> is set to "-", which when opened
2307 gives you standard input. The @ARGV array is then processed as a list
2308 of filenames. The loop
2311 ... # code for each line
2314 is equivalent to the following Perl-like pseudo code:
2316 unshift(@ARGV, '-') unless @ARGV;
2317 while ($ARGV = shift) {
2320 ... # code for each line
2324 except that it isn't so cumbersome to say, and will actually work.
2325 It really does shift the @ARGV array and put the current filename
2326 into the $ARGV variable. It also uses filehandle I<ARGV>
2327 internally. <> is just a synonym for <ARGV>, which
2328 is magical. (The pseudo code above doesn't work because it treats
2329 <ARGV> as non-magical.)
2331 Since the null filehandle uses the two argument form of L<perlfunc/open>
2332 it interprets special characters, so if you have a script like this:
2338 and call it with C<perl dangerous.pl 'rm -rfv *|'>, it actually opens a
2339 pipe, executes the C<rm> command and reads C<rm>'s output from that pipe.
2340 If you want all items in C<@ARGV> to be interpreted as file names, you
2341 can use the module C<ARGV::readonly> from CPAN.
2343 You can modify @ARGV before the first <> as long as the array ends up
2344 containing the list of filenames you really want. Line numbers (C<$.>)
2345 continue as though the input were one big happy file. See the example
2346 in L<perlfunc/eof> for how to reset line numbers on each file.
2348 If you want to set @ARGV to your own list of files, go right ahead.
2349 This sets @ARGV to all plain text files if no @ARGV was given:
2351 @ARGV = grep { -f && -T } glob('*') unless @ARGV;
2353 You can even set them to pipe commands. For example, this automatically
2354 filters compressed arguments through B<gzip>:
2356 @ARGV = map { /\.(gz|Z)$/ ? "gzip -dc < $_ |" : $_ } @ARGV;
2358 If you want to pass switches into your script, you can use one of the
2359 Getopts modules or put a loop on the front like this:
2361 while ($_ = $ARGV[0], /^-/) {
2364 if (/^-D(.*)/) { $debug = $1 }
2365 if (/^-v/) { $verbose++ }
2366 # ... # other switches
2370 # ... # code for each line
2373 The <> symbol will return C<undef> for end-of-file only once.
2374 If you call it again after this, it will assume you are processing another
2375 @ARGV list, and if you haven't set @ARGV, will read input from STDIN.
2377 If what the angle brackets contain is a simple scalar variable (e.g.,
2378 <$foo>), then that variable contains the name of the
2379 filehandle to input from, or its typeglob, or a reference to the
2385 If what's within the angle brackets is neither a filehandle nor a simple
2386 scalar variable containing a filehandle name, typeglob, or typeglob
2387 reference, it is interpreted as a filename pattern to be globbed, and
2388 either a list of filenames or the next filename in the list is returned,
2389 depending on context. This distinction is determined on syntactic
2390 grounds alone. That means C<< <$x> >> is always a readline() from
2391 an indirect handle, but C<< <$hash{key}> >> is always a glob().
2392 That's because $x is a simple scalar variable, but C<$hash{key}> is
2393 not--it's a hash element. Even C<< <$x > >> (note the extra space)
2394 is treated as C<glob("$x ")>, not C<readline($x)>.
2396 One level of double-quote interpretation is done first, but you can't
2397 say C<< <$foo> >> because that's an indirect filehandle as explained
2398 in the previous paragraph. (In older versions of Perl, programmers
2399 would insert curly brackets to force interpretation as a filename glob:
2400 C<< <${foo}> >>. These days, it's considered cleaner to call the
2401 internal function directly as C<glob($foo)>, which is probably the right
2402 way to have done it in the first place.) For example:
2408 is roughly equivalent to:
2410 open(FOO, "echo *.c | tr -s ' \t\r\f' '\\012\\012\\012\\012'|");
2416 except that the globbing is actually done internally using the standard
2417 C<File::Glob> extension. Of course, the shortest way to do the above is:
2421 A (file)glob evaluates its (embedded) argument only when it is
2422 starting a new list. All values must be read before it will start
2423 over. In list context, this isn't important because you automatically
2424 get them all anyway. However, in scalar context the operator returns
2425 the next value each time it's called, or C<undef> when the list has
2426 run out. As with filehandle reads, an automatic C<defined> is
2427 generated when the glob occurs in the test part of a C<while>,
2428 because legal glob returns (e.g. a file called F<0>) would otherwise
2429 terminate the loop. Again, C<undef> is returned only once. So if
2430 you're expecting a single value from a glob, it is much better to
2433 ($file) = <blurch*>;
2439 because the latter will alternate between returning a filename and
2442 If you're trying to do variable interpolation, it's definitely better
2443 to use the glob() function, because the older notation can cause people
2444 to become confused with the indirect filehandle notation.
2446 @files = glob("$dir/*.[ch]");
2447 @files = glob($files[$i]);
2449 =head2 Constant Folding
2450 X<constant folding> X<folding>
2452 Like C, Perl does a certain amount of expression evaluation at
2453 compile time whenever it determines that all arguments to an
2454 operator are static and have no side effects. In particular, string
2455 concatenation happens at compile time between literals that don't do
2456 variable substitution. Backslash interpolation also happens at
2457 compile time. You can say
2459 'Now is the time for all' . "\n" .
2460 'good men to come to.'
2462 and this all reduces to one string internally. Likewise, if
2465 foreach $file (@filenames) {
2466 if (-s $file > 5 + 100 * 2**16) { }
2469 the compiler will precompute the number which that expression
2470 represents so that the interpreter won't have to.
2475 Perl doesn't officially have a no-op operator, but the bare constants
2476 C<0> and C<1> are special-cased to not produce a warning in a void
2477 context, so you can for example safely do
2481 =head2 Bitwise String Operators
2482 X<operator, bitwise, string>
2484 Bitstrings of any size may be manipulated by the bitwise operators
2487 If the operands to a binary bitwise op are strings of different
2488 sizes, B<|> and B<^> ops act as though the shorter operand had
2489 additional zero bits on the right, while the B<&> op acts as though
2490 the longer operand were truncated to the length of the shorter.
2491 The granularity for such extension or truncation is one or more
2494 # ASCII-based examples
2495 print "j p \n" ^ " a h"; # prints "JAPH\n"
2496 print "JA" | " ph\n"; # prints "japh\n"
2497 print "japh\nJunk" & '_____'; # prints "JAPH\n";
2498 print 'p N$' ^ " E<H\n"; # prints "Perl\n";
2500 If you are intending to manipulate bitstrings, be certain that
2501 you're supplying bitstrings: If an operand is a number, that will imply
2502 a B<numeric> bitwise operation. You may explicitly show which type of
2503 operation you intend by using C<""> or C<0+>, as in the examples below.
2505 $foo = 150 | 105; # yields 255 (0x96 | 0x69 is 0xFF)
2506 $foo = '150' | 105; # yields 255
2507 $foo = 150 | '105'; # yields 255
2508 $foo = '150' | '105'; # yields string '155' (under ASCII)
2510 $baz = 0+$foo & 0+$bar; # both ops explicitly numeric
2511 $biz = "$foo" ^ "$bar"; # both ops explicitly stringy
2513 See L<perlfunc/vec> for information on how to manipulate individual bits
2516 =head2 Integer Arithmetic
2519 By default, Perl assumes that it must do most of its arithmetic in
2520 floating point. But by saying
2524 you may tell the compiler that it's okay to use integer operations
2525 (if it feels like it) from here to the end of the enclosing BLOCK.
2526 An inner BLOCK may countermand this by saying
2530 which lasts until the end of that BLOCK. Note that this doesn't
2531 mean everything is only an integer, merely that Perl may use integer
2532 operations if it is so inclined. For example, even under C<use
2533 integer>, if you take the C<sqrt(2)>, you'll still get C<1.4142135623731>
2536 Used on numbers, the bitwise operators ("&", "|", "^", "~", "<<",
2537 and ">>") always produce integral results. (But see also
2538 L<Bitwise String Operators>.) However, C<use integer> still has meaning for
2539 them. By default, their results are interpreted as unsigned integers, but
2540 if C<use integer> is in effect, their results are interpreted
2541 as signed integers. For example, C<~0> usually evaluates to a large
2542 integral value. However, C<use integer; ~0> is C<-1> on two's-complement
2545 =head2 Floating-point Arithmetic
2546 X<floating-point> X<floating point> X<float> X<real>
2548 While C<use integer> provides integer-only arithmetic, there is no
2549 analogous mechanism to provide automatic rounding or truncation to a
2550 certain number of decimal places. For rounding to a certain number
2551 of digits, sprintf() or printf() is usually the easiest route.
2554 Floating-point numbers are only approximations to what a mathematician
2555 would call real numbers. There are infinitely more reals than floats,
2556 so some corners must be cut. For example:
2558 printf "%.20g\n", 123456789123456789;
2559 # produces 123456789123456784
2561 Testing for exact floating-point equality or inequality is not a
2562 good idea. Here's a (relatively expensive) work-around to compare
2563 whether two floating-point numbers are equal to a particular number of
2564 decimal places. See Knuth, volume II, for a more robust treatment of
2568 my ($X, $Y, $POINTS) = @_;
2570 $tX = sprintf("%.${POINTS}g", $X);
2571 $tY = sprintf("%.${POINTS}g", $Y);
2575 The POSIX module (part of the standard perl distribution) implements
2576 ceil(), floor(), and other mathematical and trigonometric functions.
2577 The Math::Complex module (part of the standard perl distribution)
2578 defines mathematical functions that work on both the reals and the
2579 imaginary numbers. Math::Complex not as efficient as POSIX, but
2580 POSIX can't work with complex numbers.
2582 Rounding in financial applications can have serious implications, and
2583 the rounding method used should be specified precisely. In these
2584 cases, it probably pays not to trust whichever system rounding is
2585 being used by Perl, but to instead implement the rounding function you
2588 =head2 Bigger Numbers
2589 X<number, arbitrary precision>
2591 The standard Math::BigInt and Math::BigFloat modules provide
2592 variable-precision arithmetic and overloaded operators, although
2593 they're currently pretty slow. At the cost of some space and
2594 considerable speed, they avoid the normal pitfalls associated with
2595 limited-precision representations.
2598 $x = Math::BigInt->new('123456789123456789');
2601 # prints +15241578780673678515622620750190521
2603 There are several modules that let you calculate with (bound only by
2604 memory and cpu-time) unlimited or fixed precision. There are also
2605 some non-standard modules that provide faster implementations via
2606 external C libraries.
2608 Here is a short, but incomplete summary:
2610 Math::Fraction big, unlimited fractions like 9973 / 12967
2611 Math::String treat string sequences like numbers
2612 Math::FixedPrecision calculate with a fixed precision
2613 Math::Currency for currency calculations
2614 Bit::Vector manipulate bit vectors fast (uses C)
2615 Math::BigIntFast Bit::Vector wrapper for big numbers
2616 Math::Pari provides access to the Pari C library
2617 Math::BigInteger uses an external C library
2618 Math::Cephes uses external Cephes C library (no big numbers)
2619 Math::Cephes::Fraction fractions via the Cephes library
2620 Math::GMP another one using an external C library