Commit | Line | Data |
a0d0e21e |
1 | =head1 NAME |
2 | |
3 | perlop - Perl operators and precedence |
4 | |
5 | =head1 SYNOPSIS |
6 | |
7 | Perl operators have the following associativity and precedence, |
19799a22 |
8 | listed from highest precedence to lowest. Operators borrowed from |
9 | C keep the same precedence relationship with each other, even where |
10 | C's precedence is slightly screwy. (This makes learning Perl easier |
11 | for C folks.) With very few exceptions, these all operate on scalar |
12 | values only, not array values. |
a0d0e21e |
13 | |
14 | left terms and list operators (leftward) |
15 | left -> |
16 | nonassoc ++ -- |
17 | right ** |
18 | right ! ~ \ and unary + and - |
54310121 |
19 | left =~ !~ |
a0d0e21e |
20 | left * / % x |
21 | left + - . |
22 | left << >> |
23 | nonassoc named unary operators |
24 | nonassoc < > <= >= lt gt le ge |
25 | nonassoc == != <=> eq ne cmp |
26 | left & |
27 | left | ^ |
28 | left && |
29 | left || |
137443ea |
30 | nonassoc .. ... |
a0d0e21e |
31 | right ?: |
32 | right = += -= *= etc. |
33 | left , => |
34 | nonassoc list operators (rightward) |
a5f75d66 |
35 | right not |
a0d0e21e |
36 | left and |
37 | left or xor |
38 | |
39 | In the following sections, these operators are covered in precedence order. |
40 | |
5a964f20 |
41 | Many operators can be overloaded for objects. See L<overload>. |
42 | |
cb1a09d0 |
43 | =head1 DESCRIPTION |
a0d0e21e |
44 | |
45 | =head2 Terms and List Operators (Leftward) |
46 | |
62c18ce2 |
47 | A TERM has the highest precedence in Perl. They include variables, |
5f05dabc |
48 | quote and quote-like operators, any expression in parentheses, |
a0d0e21e |
49 | and any function whose arguments are parenthesized. Actually, there |
50 | aren't really functions in this sense, just list operators and unary |
51 | operators behaving as functions because you put parentheses around |
52 | the arguments. These are all documented in L<perlfunc>. |
53 | |
54 | If any list operator (print(), etc.) or any unary operator (chdir(), etc.) |
55 | is followed by a left parenthesis as the next token, the operator and |
56 | arguments within parentheses are taken to be of highest precedence, |
57 | just like a normal function call. |
58 | |
59 | In the absence of parentheses, the precedence of list operators such as |
60 | C<print>, C<sort>, or C<chmod> is either very high or very low depending on |
54310121 |
61 | whether you are looking at the left side or the right side of the operator. |
a0d0e21e |
62 | For example, in |
63 | |
64 | @ary = (1, 3, sort 4, 2); |
65 | print @ary; # prints 1324 |
66 | |
19799a22 |
67 | the commas on the right of the sort are evaluated before the sort, |
68 | but the commas on the left are evaluated after. In other words, |
69 | list operators tend to gobble up all arguments that follow, and |
a0d0e21e |
70 | then act like a simple TERM with regard to the preceding expression. |
19799a22 |
71 | Be careful with parentheses: |
a0d0e21e |
72 | |
73 | # These evaluate exit before doing the print: |
74 | print($foo, exit); # Obviously not what you want. |
75 | print $foo, exit; # Nor is this. |
76 | |
77 | # These do the print before evaluating exit: |
78 | (print $foo), exit; # This is what you want. |
79 | print($foo), exit; # Or this. |
80 | print ($foo), exit; # Or even this. |
81 | |
82 | Also note that |
83 | |
84 | print ($foo & 255) + 1, "\n"; |
85 | |
54310121 |
86 | probably doesn't do what you expect at first glance. See |
a0d0e21e |
87 | L<Named Unary Operators> for more discussion of this. |
88 | |
89 | Also parsed as terms are the C<do {}> and C<eval {}> constructs, as |
54310121 |
90 | well as subroutine and method calls, and the anonymous |
a0d0e21e |
91 | constructors C<[]> and C<{}>. |
92 | |
2ae324a7 |
93 | See also L<Quote and Quote-like Operators> toward the end of this section, |
c07a80fd |
94 | as well as L<"I/O Operators">. |
a0d0e21e |
95 | |
96 | =head2 The Arrow Operator |
97 | |
19799a22 |
98 | "C<-E<gt>>" is an infix dereference operator, just as it is in C |
99 | and C++. If the right side is either a C<[...]>, C<{...}>, or a |
100 | C<(...)> subscript, then the left side must be either a hard or |
101 | symbolic reference to an array, a hash, or a subroutine respectively. |
102 | (Or technically speaking, a location capable of holding a hard |
103 | reference, if it's an array or hash reference being used for |
104 | assignment.) See L<perlreftut> and L<perlref>. |
a0d0e21e |
105 | |
19799a22 |
106 | Otherwise, the right side is a method name or a simple scalar |
107 | variable containing either the method name or a subroutine reference, |
108 | and the left side must be either an object (a blessed reference) |
109 | or a class name (that is, a package name). See L<perlobj>. |
a0d0e21e |
110 | |
5f05dabc |
111 | =head2 Auto-increment and Auto-decrement |
a0d0e21e |
112 | |
113 | "++" and "--" work as in C. That is, if placed before a variable, they |
114 | increment or decrement the variable before returning the value, and if |
115 | placed after, increment or decrement the variable after returning the value. |
116 | |
54310121 |
117 | The auto-increment operator has a little extra builtin magic to it. If |
a0d0e21e |
118 | you increment a variable that is numeric, or that has ever been used in |
119 | a numeric context, you get a normal increment. If, however, the |
5f05dabc |
120 | variable has been used in only string contexts since it was set, and |
5a964f20 |
121 | has a value that is not the empty string and matches the pattern |
a0d0e21e |
122 | C</^[a-zA-Z]*[0-9]*$/>, the increment is done as a string, preserving each |
123 | character within its range, with carry: |
124 | |
125 | print ++($foo = '99'); # prints '100' |
126 | print ++($foo = 'a0'); # prints 'a1' |
127 | print ++($foo = 'Az'); # prints 'Ba' |
128 | print ++($foo = 'zz'); # prints 'aaa' |
129 | |
5f05dabc |
130 | The auto-decrement operator is not magical. |
a0d0e21e |
131 | |
132 | =head2 Exponentiation |
133 | |
19799a22 |
134 | Binary "**" is the exponentiation operator. It binds even more |
cb1a09d0 |
135 | tightly than unary minus, so -2**4 is -(2**4), not (-2)**4. (This is |
136 | implemented using C's pow(3) function, which actually works on doubles |
137 | internally.) |
a0d0e21e |
138 | |
139 | =head2 Symbolic Unary Operators |
140 | |
5f05dabc |
141 | Unary "!" performs logical negation, i.e., "not". See also C<not> for a lower |
a0d0e21e |
142 | precedence version of this. |
143 | |
144 | Unary "-" performs arithmetic negation if the operand is numeric. If |
145 | the operand is an identifier, a string consisting of a minus sign |
146 | concatenated with the identifier is returned. Otherwise, if the string |
147 | starts with a plus or minus, a string starting with the opposite sign |
148 | is returned. One effect of these rules is that C<-bareword> is equivalent |
149 | to C<"-bareword">. |
150 | |
5a964f20 |
151 | Unary "~" performs bitwise negation, i.e., 1's complement. For example, |
152 | C<0666 &~ 027> is 0640. (See also L<Integer Arithmetic> and L<Bitwise |
153 | String Operators>.) |
a0d0e21e |
154 | |
155 | Unary "+" has no effect whatsoever, even on strings. It is useful |
156 | syntactically for separating a function name from a parenthesized expression |
157 | that would otherwise be interpreted as the complete list of function |
5ba421f6 |
158 | arguments. (See examples above under L<Terms and List Operators (Leftward)>.) |
a0d0e21e |
159 | |
19799a22 |
160 | Unary "\" creates a reference to whatever follows it. See L<perlreftut> |
161 | and L<perlref>. Do not confuse this behavior with the behavior of |
162 | backslash within a string, although both forms do convey the notion |
163 | of protecting the next thing from interpolation. |
a0d0e21e |
164 | |
165 | =head2 Binding Operators |
166 | |
c07a80fd |
167 | Binary "=~" binds a scalar expression to a pattern match. Certain operations |
cb1a09d0 |
168 | search or modify the string $_ by default. This operator makes that kind |
169 | of operation work on some other string. The right argument is a search |
2c268ad5 |
170 | pattern, substitution, or transliteration. The left argument is what is |
171 | supposed to be searched, substituted, or transliterated instead of the default |
cb1a09d0 |
172 | $_. The return value indicates the success of the operation. (If the |
173 | right argument is an expression rather than a search pattern, |
2c268ad5 |
174 | substitution, or transliteration, it is interpreted as a search pattern at run |
aa689395 |
175 | time. This can be is less efficient than an explicit search, because the |
176 | pattern must be compiled every time the expression is evaluated. |
a0d0e21e |
177 | |
178 | Binary "!~" is just like "=~" except the return value is negated in |
179 | the logical sense. |
180 | |
181 | =head2 Multiplicative Operators |
182 | |
183 | Binary "*" multiplies two numbers. |
184 | |
185 | Binary "/" divides two numbers. |
186 | |
54310121 |
187 | Binary "%" computes the modulus of two numbers. Given integer |
188 | operands C<$a> and C<$b>: If C<$b> is positive, then C<$a % $b> is |
189 | C<$a> minus the largest multiple of C<$b> that is not greater than |
190 | C<$a>. If C<$b> is negative, then C<$a % $b> is C<$a> minus the |
191 | smallest multiple of C<$b> that is not less than C<$a> (i.e. the |
6bb4e6d4 |
192 | result will be less than or equal to zero). |
5a964f20 |
193 | Note than when C<use integer> is in scope, "%" give you direct access |
55d729e4 |
194 | to the modulus operator as implemented by your C compiler. This |
195 | operator is not as well defined for negative operands, but it will |
196 | execute faster. |
197 | |
5a964f20 |
198 | Binary "x" is the repetition operator. In scalar context, it |
a0d0e21e |
199 | returns a string consisting of the left operand repeated the number of |
5a964f20 |
200 | times specified by the right operand. In list context, if the left |
5f05dabc |
201 | operand is a list in parentheses, it repeats the list. |
a0d0e21e |
202 | |
203 | print '-' x 80; # print row of dashes |
204 | |
205 | print "\t" x ($tab/8), ' ' x ($tab%8); # tab over |
206 | |
207 | @ones = (1) x 80; # a list of 80 1's |
208 | @ones = (5) x @ones; # set all elements to 5 |
209 | |
210 | |
211 | =head2 Additive Operators |
212 | |
213 | Binary "+" returns the sum of two numbers. |
214 | |
215 | Binary "-" returns the difference of two numbers. |
216 | |
217 | Binary "." concatenates two strings. |
218 | |
219 | =head2 Shift Operators |
220 | |
55497cff |
221 | Binary "<<" returns the value of its left argument shifted left by the |
222 | number of bits specified by the right argument. Arguments should be |
223 | integers. (See also L<Integer Arithmetic>.) |
a0d0e21e |
224 | |
55497cff |
225 | Binary ">>" returns the value of its left argument shifted right by |
226 | the number of bits specified by the right argument. Arguments should |
227 | be integers. (See also L<Integer Arithmetic>.) |
a0d0e21e |
228 | |
229 | =head2 Named Unary Operators |
230 | |
231 | The various named unary operators are treated as functions with one |
232 | argument, with optional parentheses. These include the filetest |
233 | operators, like C<-f>, C<-M>, etc. See L<perlfunc>. |
234 | |
235 | If any list operator (print(), etc.) or any unary operator (chdir(), etc.) |
236 | is followed by a left parenthesis as the next token, the operator and |
237 | arguments within parentheses are taken to be of highest precedence, |
238 | just like a normal function call. Examples: |
239 | |
240 | chdir $foo || die; # (chdir $foo) || die |
241 | chdir($foo) || die; # (chdir $foo) || die |
242 | chdir ($foo) || die; # (chdir $foo) || die |
243 | chdir +($foo) || die; # (chdir $foo) || die |
244 | |
245 | but, because * is higher precedence than ||: |
246 | |
247 | chdir $foo * 20; # chdir ($foo * 20) |
248 | chdir($foo) * 20; # (chdir $foo) * 20 |
249 | chdir ($foo) * 20; # (chdir $foo) * 20 |
250 | chdir +($foo) * 20; # chdir ($foo * 20) |
251 | |
252 | rand 10 * 20; # rand (10 * 20) |
253 | rand(10) * 20; # (rand 10) * 20 |
254 | rand (10) * 20; # (rand 10) * 20 |
255 | rand +(10) * 20; # rand (10 * 20) |
256 | |
5ba421f6 |
257 | See also L<"Terms and List Operators (Leftward)">. |
a0d0e21e |
258 | |
259 | =head2 Relational Operators |
260 | |
6ee5d4e7 |
261 | Binary "E<lt>" returns true if the left argument is numerically less than |
a0d0e21e |
262 | the right argument. |
263 | |
6ee5d4e7 |
264 | Binary "E<gt>" returns true if the left argument is numerically greater |
a0d0e21e |
265 | than the right argument. |
266 | |
6ee5d4e7 |
267 | Binary "E<lt>=" returns true if the left argument is numerically less than |
a0d0e21e |
268 | or equal to the right argument. |
269 | |
6ee5d4e7 |
270 | Binary "E<gt>=" returns true if the left argument is numerically greater |
a0d0e21e |
271 | than or equal to the right argument. |
272 | |
273 | Binary "lt" returns true if the left argument is stringwise less than |
274 | the right argument. |
275 | |
276 | Binary "gt" returns true if the left argument is stringwise greater |
277 | than the right argument. |
278 | |
279 | Binary "le" returns true if the left argument is stringwise less than |
280 | or equal to the right argument. |
281 | |
282 | Binary "ge" returns true if the left argument is stringwise greater |
283 | than or equal to the right argument. |
284 | |
285 | =head2 Equality Operators |
286 | |
287 | Binary "==" returns true if the left argument is numerically equal to |
288 | the right argument. |
289 | |
290 | Binary "!=" returns true if the left argument is numerically not equal |
291 | to the right argument. |
292 | |
6ee5d4e7 |
293 | Binary "E<lt>=E<gt>" returns -1, 0, or 1 depending on whether the left |
294 | argument is numerically less than, equal to, or greater than the right |
295 | argument. |
a0d0e21e |
296 | |
297 | Binary "eq" returns true if the left argument is stringwise equal to |
298 | the right argument. |
299 | |
300 | Binary "ne" returns true if the left argument is stringwise not equal |
301 | to the right argument. |
302 | |
303 | Binary "cmp" returns -1, 0, or 1 depending on whether the left argument is stringwise |
304 | less than, equal to, or greater than the right argument. |
305 | |
a034a98d |
306 | "lt", "le", "ge", "gt" and "cmp" use the collation (sort) order specified |
307 | by the current locale if C<use locale> is in effect. See L<perllocale>. |
308 | |
a0d0e21e |
309 | =head2 Bitwise And |
310 | |
311 | Binary "&" returns its operators ANDed together bit by bit. |
2c268ad5 |
312 | (See also L<Integer Arithmetic> and L<Bitwise String Operators>.) |
a0d0e21e |
313 | |
314 | =head2 Bitwise Or and Exclusive Or |
315 | |
316 | Binary "|" returns its operators ORed together bit by bit. |
2c268ad5 |
317 | (See also L<Integer Arithmetic> and L<Bitwise String Operators>.) |
a0d0e21e |
318 | |
319 | Binary "^" returns its operators XORed together bit by bit. |
2c268ad5 |
320 | (See also L<Integer Arithmetic> and L<Bitwise String Operators>.) |
a0d0e21e |
321 | |
322 | =head2 C-style Logical And |
323 | |
324 | Binary "&&" performs a short-circuit logical AND operation. That is, |
325 | if the left operand is false, the right operand is not even evaluated. |
326 | Scalar or list context propagates down to the right operand if it |
327 | is evaluated. |
328 | |
329 | =head2 C-style Logical Or |
330 | |
331 | Binary "||" performs a short-circuit logical OR operation. That is, |
332 | if the left operand is true, the right operand is not even evaluated. |
333 | Scalar or list context propagates down to the right operand if it |
334 | is evaluated. |
335 | |
336 | The C<||> and C<&&> operators differ from C's in that, rather than returning |
337 | 0 or 1, they return the last value evaluated. Thus, a reasonably portable |
338 | way to find out the home directory (assuming it's not "0") might be: |
339 | |
340 | $home = $ENV{'HOME'} || $ENV{'LOGDIR'} || |
341 | (getpwuid($<))[7] || die "You're homeless!\n"; |
342 | |
5a964f20 |
343 | In particular, this means that you shouldn't use this |
344 | for selecting between two aggregates for assignment: |
345 | |
346 | @a = @b || @c; # this is wrong |
347 | @a = scalar(@b) || @c; # really meant this |
348 | @a = @b ? @b : @c; # this works fine, though |
349 | |
350 | As more readable alternatives to C<&&> and C<||> when used for |
351 | control flow, Perl provides C<and> and C<or> operators (see below). |
352 | The short-circuit behavior is identical. The precedence of "and" and |
353 | "or" is much lower, however, so that you can safely use them after a |
354 | list operator without the need for parentheses: |
a0d0e21e |
355 | |
356 | unlink "alpha", "beta", "gamma" |
357 | or gripe(), next LINE; |
358 | |
359 | With the C-style operators that would have been written like this: |
360 | |
361 | unlink("alpha", "beta", "gamma") |
362 | || (gripe(), next LINE); |
363 | |
5a964f20 |
364 | Use "or" for assignment is unlikely to do what you want; see below. |
365 | |
366 | =head2 Range Operators |
a0d0e21e |
367 | |
368 | Binary ".." is the range operator, which is really two different |
5a964f20 |
369 | operators depending on the context. In list context, it returns an |
2cdbc966 |
370 | array of values counting (up by ones) from the left value to the right |
371 | value. If the left value is greater than the right value then it |
372 | returns the empty array. The range operator is useful for writing |
373 | C<foreach (1..10)> loops and for doing slice operations on arrays. In |
374 | the current implementation, no temporary array is created when the |
375 | range operator is used as the expression in C<foreach> loops, but older |
376 | versions of Perl might burn a lot of memory when you write something |
377 | like this: |
a0d0e21e |
378 | |
379 | for (1 .. 1_000_000) { |
380 | # code |
54310121 |
381 | } |
a0d0e21e |
382 | |
5a964f20 |
383 | In scalar context, ".." returns a boolean value. The operator is |
a0d0e21e |
384 | bistable, like a flip-flop, and emulates the line-range (comma) operator |
385 | of B<sed>, B<awk>, and various editors. Each ".." operator maintains its |
386 | own boolean state. It is false as long as its left operand is false. |
387 | Once the left operand is true, the range operator stays true until the |
388 | right operand is true, I<AFTER> which the range operator becomes false |
19799a22 |
389 | again. It doesn't become false till the next time the range operator is |
a0d0e21e |
390 | evaluated. It can test the right operand and become false on the same |
391 | evaluation it became true (as in B<awk>), but it still returns true once. |
19799a22 |
392 | If you don't want it to test the right operand till the next |
393 | evaluation, as in B<sed>, just use three dots ("...") instead of |
394 | two. In all other regards, "..." behaves just like ".." does. |
395 | |
396 | The right operand is not evaluated while the operator is in the |
397 | "false" state, and the left operand is not evaluated while the |
398 | operator is in the "true" state. The precedence is a little lower |
399 | than || and &&. The value returned is either the empty string for |
400 | false, or a sequence number (beginning with 1) for true. The |
401 | sequence number is reset for each range encountered. The final |
402 | sequence number in a range has the string "E0" appended to it, which |
403 | doesn't affect its numeric value, but gives you something to search |
404 | for if you want to exclude the endpoint. You can exclude the |
405 | beginning point by waiting for the sequence number to be greater |
406 | than 1. If either operand of scalar ".." is a constant expression, |
407 | that operand is implicitly compared to the C<$.> variable, the |
408 | current line number. Examples: |
a0d0e21e |
409 | |
410 | As a scalar operator: |
411 | |
412 | if (101 .. 200) { print; } # print 2nd hundred lines |
413 | next line if (1 .. /^$/); # skip header lines |
414 | s/^/> / if (/^$/ .. eof()); # quote body |
415 | |
5a964f20 |
416 | # parse mail messages |
417 | while (<>) { |
418 | $in_header = 1 .. /^$/; |
419 | $in_body = /^$/ .. eof(); |
420 | # do something based on those |
421 | } continue { |
422 | close ARGV if eof; # reset $. each file |
423 | } |
424 | |
a0d0e21e |
425 | As a list operator: |
426 | |
427 | for (101 .. 200) { print; } # print $_ 100 times |
3e3baf6d |
428 | @foo = @foo[0 .. $#foo]; # an expensive no-op |
a0d0e21e |
429 | @foo = @foo[$#foo-4 .. $#foo]; # slice last 5 items |
430 | |
5a964f20 |
431 | The range operator (in list context) makes use of the magical |
5f05dabc |
432 | auto-increment algorithm if the operands are strings. You |
a0d0e21e |
433 | can say |
434 | |
435 | @alphabet = ('A' .. 'Z'); |
436 | |
19799a22 |
437 | to get all normal letters of the alphabet, or |
a0d0e21e |
438 | |
439 | $hexdigit = (0 .. 9, 'a' .. 'f')[$num & 15]; |
440 | |
441 | to get a hexadecimal digit, or |
442 | |
443 | @z2 = ('01' .. '31'); print $z2[$mday]; |
444 | |
445 | to get dates with leading zeros. If the final value specified is not |
446 | in the sequence that the magical increment would produce, the sequence |
447 | goes until the next value would be longer than the final value |
448 | specified. |
449 | |
450 | =head2 Conditional Operator |
451 | |
452 | Ternary "?:" is the conditional operator, just as in C. It works much |
453 | like an if-then-else. If the argument before the ? is true, the |
454 | argument before the : is returned, otherwise the argument after the : |
cb1a09d0 |
455 | is returned. For example: |
456 | |
54310121 |
457 | printf "I have %d dog%s.\n", $n, |
cb1a09d0 |
458 | ($n == 1) ? '' : "s"; |
459 | |
460 | Scalar or list context propagates downward into the 2nd |
54310121 |
461 | or 3rd argument, whichever is selected. |
cb1a09d0 |
462 | |
463 | $a = $ok ? $b : $c; # get a scalar |
464 | @a = $ok ? @b : @c; # get an array |
465 | $a = $ok ? @b : @c; # oops, that's just a count! |
466 | |
467 | The operator may be assigned to if both the 2nd and 3rd arguments are |
468 | legal lvalues (meaning that you can assign to them): |
a0d0e21e |
469 | |
470 | ($a_or_b ? $a : $b) = $c; |
471 | |
5a964f20 |
472 | Because this operator produces an assignable result, using assignments |
473 | without parentheses will get you in trouble. For example, this: |
474 | |
475 | $a % 2 ? $a += 10 : $a += 2 |
476 | |
477 | Really means this: |
478 | |
479 | (($a % 2) ? ($a += 10) : $a) += 2 |
480 | |
481 | Rather than this: |
482 | |
483 | ($a % 2) ? ($a += 10) : ($a += 2) |
484 | |
19799a22 |
485 | That should probably be written more simply as: |
486 | |
487 | $a += ($a % 2) ? 10 : 2; |
488 | |
4633a7c4 |
489 | =head2 Assignment Operators |
a0d0e21e |
490 | |
491 | "=" is the ordinary assignment operator. |
492 | |
493 | Assignment operators work as in C. That is, |
494 | |
495 | $a += 2; |
496 | |
497 | is equivalent to |
498 | |
499 | $a = $a + 2; |
500 | |
501 | although without duplicating any side effects that dereferencing the lvalue |
54310121 |
502 | might trigger, such as from tie(). Other assignment operators work similarly. |
503 | The following are recognized: |
a0d0e21e |
504 | |
505 | **= += *= &= <<= &&= |
506 | -= /= |= >>= ||= |
507 | .= %= ^= |
508 | x= |
509 | |
19799a22 |
510 | Although these are grouped by family, they all have the precedence |
a0d0e21e |
511 | of assignment. |
512 | |
513 | Unlike in C, the assignment operator produces a valid lvalue. Modifying |
514 | an assignment is equivalent to doing the assignment and then modifying |
515 | the variable that was assigned to. This is useful for modifying |
516 | a copy of something, like this: |
517 | |
518 | ($tmp = $global) =~ tr [A-Z] [a-z]; |
519 | |
520 | Likewise, |
521 | |
522 | ($a += 2) *= 3; |
523 | |
524 | is equivalent to |
525 | |
526 | $a += 2; |
527 | $a *= 3; |
528 | |
748a9306 |
529 | =head2 Comma Operator |
a0d0e21e |
530 | |
5a964f20 |
531 | Binary "," is the comma operator. In scalar context it evaluates |
a0d0e21e |
532 | its left argument, throws that value away, then evaluates its right |
533 | argument and returns that value. This is just like C's comma operator. |
534 | |
5a964f20 |
535 | In list context, it's just the list argument separator, and inserts |
a0d0e21e |
536 | both its arguments into the list. |
537 | |
6ee5d4e7 |
538 | The =E<gt> digraph is mostly just a synonym for the comma operator. It's useful for |
cb1a09d0 |
539 | documenting arguments that come in pairs. As of release 5.001, it also forces |
4633a7c4 |
540 | any word to the left of it to be interpreted as a string. |
748a9306 |
541 | |
a0d0e21e |
542 | =head2 List Operators (Rightward) |
543 | |
544 | On the right side of a list operator, it has very low precedence, |
545 | such that it controls all comma-separated expressions found there. |
546 | The only operators with lower precedence are the logical operators |
547 | "and", "or", and "not", which may be used to evaluate calls to list |
548 | operators without the need for extra parentheses: |
549 | |
550 | open HANDLE, "filename" |
551 | or die "Can't open: $!\n"; |
552 | |
5ba421f6 |
553 | See also discussion of list operators in L<Terms and List Operators (Leftward)>. |
a0d0e21e |
554 | |
555 | =head2 Logical Not |
556 | |
557 | Unary "not" returns the logical negation of the expression to its right. |
558 | It's the equivalent of "!" except for the very low precedence. |
559 | |
560 | =head2 Logical And |
561 | |
562 | Binary "and" returns the logical conjunction of the two surrounding |
563 | expressions. It's equivalent to && except for the very low |
5f05dabc |
564 | precedence. This means that it short-circuits: i.e., the right |
a0d0e21e |
565 | expression is evaluated only if the left expression is true. |
566 | |
567 | =head2 Logical or and Exclusive Or |
568 | |
569 | Binary "or" returns the logical disjunction of the two surrounding |
5a964f20 |
570 | expressions. It's equivalent to || except for the very low precedence. |
571 | This makes it useful for control flow |
572 | |
573 | print FH $data or die "Can't write to FH: $!"; |
574 | |
575 | This means that it short-circuits: i.e., the right expression is evaluated |
576 | only if the left expression is false. Due to its precedence, you should |
577 | probably avoid using this for assignment, only for control flow. |
578 | |
579 | $a = $b or $c; # bug: this is wrong |
580 | ($a = $b) or $c; # really means this |
581 | $a = $b || $c; # better written this way |
582 | |
19799a22 |
583 | However, when it's a list-context assignment and you're trying to use |
5a964f20 |
584 | "||" for control flow, you probably need "or" so that the assignment |
585 | takes higher precedence. |
586 | |
587 | @info = stat($file) || die; # oops, scalar sense of stat! |
588 | @info = stat($file) or die; # better, now @info gets its due |
589 | |
19799a22 |
590 | Then again, you could always use parentheses. |
a0d0e21e |
591 | |
592 | Binary "xor" returns the exclusive-OR of the two surrounding expressions. |
593 | It cannot short circuit, of course. |
594 | |
595 | =head2 C Operators Missing From Perl |
596 | |
597 | Here is what C has that Perl doesn't: |
598 | |
599 | =over 8 |
600 | |
601 | =item unary & |
602 | |
603 | Address-of operator. (But see the "\" operator for taking a reference.) |
604 | |
605 | =item unary * |
606 | |
54310121 |
607 | Dereference-address operator. (Perl's prefix dereferencing |
a0d0e21e |
608 | operators are typed: $, @, %, and &.) |
609 | |
610 | =item (TYPE) |
611 | |
19799a22 |
612 | Type-casting operator. |
a0d0e21e |
613 | |
614 | =back |
615 | |
5f05dabc |
616 | =head2 Quote and Quote-like Operators |
a0d0e21e |
617 | |
618 | While we usually think of quotes as literal values, in Perl they |
619 | function as operators, providing various kinds of interpolating and |
620 | pattern matching capabilities. Perl provides customary quote characters |
621 | for these behaviors, but also provides a way for you to choose your |
622 | quote character for any of them. In the following table, a C<{}> represents |
87275199 |
623 | any pair of delimiters you choose. |
a0d0e21e |
624 | |
2c268ad5 |
625 | Customary Generic Meaning Interpolates |
626 | '' q{} Literal no |
627 | "" qq{} Literal yes |
01ae956f |
628 | `` qx{} Command yes (unless '' is delimiter) |
2c268ad5 |
629 | qw{} Word list no |
f70b4f9c |
630 | // m{} Pattern match yes (unless '' is delimiter) |
631 | qr{} Pattern yes (unless '' is delimiter) |
632 | s{}{} Substitution yes (unless '' is delimiter) |
2c268ad5 |
633 | tr{}{} Transliteration no (but see below) |
a0d0e21e |
634 | |
87275199 |
635 | Non-bracketing delimiters use the same character fore and aft, but the four |
636 | sorts of brackets (round, angle, square, curly) will all nest, which means |
637 | that |
638 | |
639 | q{foo{bar}baz} |
640 | |
641 | is the same as |
642 | |
643 | 'foo{bar}baz' |
644 | |
645 | Note, however, that this does not always work for quoting Perl code: |
646 | |
647 | $s = q{ if($a eq "}") ... }; # WRONG |
648 | |
649 | is a syntax error. The C<Text::Balanced> module on CPAN is able to do this |
650 | properly. |
651 | |
19799a22 |
652 | There can be whitespace between the operator and the quoting |
fb73857a |
653 | characters, except when C<#> is being used as the quoting character. |
19799a22 |
654 | C<q#foo#> is parsed as the string C<foo>, while C<q #foo#> is the |
655 | operator C<q> followed by a comment. Its argument will be taken |
656 | from the next line. This allows you to write: |
fb73857a |
657 | |
658 | s {foo} # Replace foo |
659 | {bar} # with bar. |
660 | |
19799a22 |
661 | For constructs that do interpolate, variables beginning with "C<$>" |
662 | or "C<@>" are interpolated, as are the following escape sequences. Within |
a0ed51b3 |
663 | a transliteration, the first eleven of these sequences may be used. |
a0d0e21e |
664 | |
6ee5d4e7 |
665 | \t tab (HT, TAB) |
5a964f20 |
666 | \n newline (NL) |
6ee5d4e7 |
667 | \r return (CR) |
668 | \f form feed (FF) |
669 | \b backspace (BS) |
670 | \a alarm (bell) (BEL) |
671 | \e escape (ESC) |
a0ed51b3 |
672 | \033 octal char (ESC) |
673 | \x1b hex char (ESC) |
674 | \x{263a} wide hex char (SMILEY) |
19799a22 |
675 | \c[ control char (ESC) |
423cee85 |
676 | \C{name} named char |
2c268ad5 |
677 | |
a0d0e21e |
678 | \l lowercase next char |
679 | \u uppercase next char |
680 | \L lowercase till \E |
681 | \U uppercase till \E |
682 | \E end case modification |
1d2dff63 |
683 | \Q quote non-word characters till \E |
a0d0e21e |
684 | |
a034a98d |
685 | If C<use locale> is in effect, the case map used by C<\l>, C<\L>, C<\u> |
423cee85 |
686 | and C<\U> is taken from the current locale. See L<perllocale>. For |
687 | documentation of C<\C{name}>, see L<charnames>. |
a034a98d |
688 | |
5a964f20 |
689 | All systems use the virtual C<"\n"> to represent a line terminator, |
690 | called a "newline". There is no such thing as an unvarying, physical |
19799a22 |
691 | newline character. It is only an illusion that the operating system, |
5a964f20 |
692 | device drivers, C libraries, and Perl all conspire to preserve. Not all |
693 | systems read C<"\r"> as ASCII CR and C<"\n"> as ASCII LF. For example, |
694 | on a Mac, these are reversed, and on systems without line terminator, |
695 | printing C<"\n"> may emit no actual data. In general, use C<"\n"> when |
696 | you mean a "newline" for your system, but use the literal ASCII when you |
697 | need an exact character. For example, most networking protocols expect |
698 | and prefer a CR+LF (C<"\012\015"> or C<"\cJ\cM">) for line terminators, |
699 | and although they often accept just C<"\012">, they seldom tolerate just |
700 | C<"\015">. If you get in the habit of using C<"\n"> for networking, |
701 | you may be burned some day. |
702 | |
1d2dff63 |
703 | You cannot include a literal C<$> or C<@> within a C<\Q> sequence. |
704 | An unescaped C<$> or C<@> interpolates the corresponding variable, |
705 | while escaping will cause the literal string C<\$> to be inserted. |
706 | You'll need to write something like C<m/\Quser\E\@\Qhost/>. |
707 | |
a0d0e21e |
708 | Patterns are subject to an additional level of interpretation as a |
709 | regular expression. This is done as a second pass, after variables are |
710 | interpolated, so that regular expressions may be incorporated into the |
711 | pattern from the variables. If this is not what you want, use C<\Q> to |
712 | interpolate a variable literally. |
713 | |
19799a22 |
714 | Apart from the behavior described above, Perl does not expand |
715 | multiple levels of interpolation. In particular, contrary to the |
716 | expectations of shell programmers, back-quotes do I<NOT> interpolate |
717 | within double quotes, nor do single quotes impede evaluation of |
718 | variables when used within double quotes. |
a0d0e21e |
719 | |
5f05dabc |
720 | =head2 Regexp Quote-Like Operators |
cb1a09d0 |
721 | |
5f05dabc |
722 | Here are the quote-like operators that apply to pattern |
cb1a09d0 |
723 | matching and related activities. |
724 | |
a0d0e21e |
725 | =over 8 |
726 | |
727 | =item ?PATTERN? |
728 | |
729 | This is just like the C</pattern/> search, except that it matches only |
730 | once between calls to the reset() operator. This is a useful |
5f05dabc |
731 | optimization when you want to see only the first occurrence of |
a0d0e21e |
732 | something in each file of a set of files, for instance. Only C<??> |
733 | patterns local to the current package are reset. |
734 | |
5a964f20 |
735 | while (<>) { |
736 | if (?^$?) { |
737 | # blank line between header and body |
738 | } |
739 | } continue { |
740 | reset if eof; # clear ?? status for next file |
741 | } |
742 | |
19799a22 |
743 | This usage is vaguely depreciated, which means it just might possibly |
744 | be removed in some distant future version of Perl, perhaps somewhere |
745 | around the year 2168. |
a0d0e21e |
746 | |
fb73857a |
747 | =item m/PATTERN/cgimosx |
a0d0e21e |
748 | |
fb73857a |
749 | =item /PATTERN/cgimosx |
a0d0e21e |
750 | |
5a964f20 |
751 | Searches a string for a pattern match, and in scalar context returns |
19799a22 |
752 | true if it succeeds, false if it fails. If no string is specified |
753 | via the C<=~> or C<!~> operator, the $_ string is searched. (The |
754 | string specified with C<=~> need not be an lvalue--it may be the |
755 | result of an expression evaluation, but remember the C<=~> binds |
756 | rather tightly.) See also L<perlre>. See L<perllocale> for |
757 | discussion of additional considerations that apply when C<use locale> |
758 | is in effect. |
a0d0e21e |
759 | |
760 | Options are: |
761 | |
fb73857a |
762 | c Do not reset search position on a failed match when /g is in effect. |
5f05dabc |
763 | g Match globally, i.e., find all occurrences. |
a0d0e21e |
764 | i Do case-insensitive pattern matching. |
765 | m Treat string as multiple lines. |
5f05dabc |
766 | o Compile pattern only once. |
a0d0e21e |
767 | s Treat string as single line. |
768 | x Use extended regular expressions. |
769 | |
770 | If "/" is the delimiter then the initial C<m> is optional. With the C<m> |
01ae956f |
771 | you can use any pair of non-alphanumeric, non-whitespace characters |
19799a22 |
772 | as delimiters. This is particularly useful for matching path names |
773 | that contain "/", to avoid LTS (leaning toothpick syndrome). If "?" is |
7bac28a0 |
774 | the delimiter, then the match-only-once rule of C<?PATTERN?> applies. |
19799a22 |
775 | If "'" is the delimiter, no interpolation is performed on the PATTERN. |
a0d0e21e |
776 | |
777 | PATTERN may contain variables, which will be interpolated (and the |
f70b4f9c |
778 | pattern recompiled) every time the pattern search is evaluated, except |
779 | for when the delimiter is a single quote. (Note that C<$)> and C<$|> |
780 | might not be interpolated because they look like end-of-string tests.) |
781 | If you want such a pattern to be compiled only once, add a C</o> after |
782 | the trailing delimiter. This avoids expensive run-time recompilations, |
783 | and is useful when the value you are interpolating won't change over |
784 | the life of the script. However, mentioning C</o> constitutes a promise |
785 | that you won't change the variables in the pattern. If you change them, |
19799a22 |
786 | Perl won't even notice. See also L<qr//>. |
a0d0e21e |
787 | |
5a964f20 |
788 | If the PATTERN evaluates to the empty string, the last |
789 | I<successfully> matched regular expression is used instead. |
a0d0e21e |
790 | |
19799a22 |
791 | If the C</g> option is not used, C<m//> in list context returns a |
a0d0e21e |
792 | list consisting of the subexpressions matched by the parentheses in the |
f7e33566 |
793 | pattern, i.e., (C<$1>, C<$2>, C<$3>...). (Note that here C<$1> etc. are |
794 | also set, and that this differs from Perl 4's behavior.) When there are |
795 | no parentheses in the pattern, the return value is the list C<(1)> for |
796 | success. With or without parentheses, an empty list is returned upon |
797 | failure. |
a0d0e21e |
798 | |
799 | Examples: |
800 | |
801 | open(TTY, '/dev/tty'); |
802 | <TTY> =~ /^y/i && foo(); # do foo if desired |
803 | |
804 | if (/Version: *([0-9.]*)/) { $version = $1; } |
805 | |
806 | next if m#^/usr/spool/uucp#; |
807 | |
808 | # poor man's grep |
809 | $arg = shift; |
810 | while (<>) { |
811 | print if /$arg/o; # compile only once |
812 | } |
813 | |
814 | if (($F1, $F2, $Etc) = ($foo =~ /^(\S+)\s+(\S+)\s*(.*)/)) |
815 | |
816 | This last example splits $foo into the first two words and the |
5f05dabc |
817 | remainder of the line, and assigns those three fields to $F1, $F2, and |
818 | $Etc. The conditional is true if any variables were assigned, i.e., if |
a0d0e21e |
819 | the pattern matched. |
820 | |
19799a22 |
821 | The C</g> modifier specifies global pattern matching--that is, |
822 | matching as many times as possible within the string. How it behaves |
823 | depends on the context. In list context, it returns a list of the |
824 | substrings matched by any capturing parentheses in the regular |
825 | expression. If there are no parentheses, it returns a list of all |
826 | the matched strings, as if there were parentheses around the whole |
827 | pattern. |
a0d0e21e |
828 | |
7e86de3e |
829 | In scalar context, each execution of C<m//g> finds the next match, |
19799a22 |
830 | returning true if it matches, and false if there is no further match. |
7e86de3e |
831 | The position after the last match can be read or set using the pos() |
832 | function; see L<perlfunc/pos>. A failed match normally resets the |
833 | search position to the beginning of the string, but you can avoid that |
834 | by adding the C</c> modifier (e.g. C<m//gc>). Modifying the target |
835 | string also resets the search position. |
c90c0ff4 |
836 | |
837 | You can intermix C<m//g> matches with C<m/\G.../g>, where C<\G> is a |
838 | zero-width assertion that matches the exact position where the previous |
839 | C<m//g>, if any, left off. The C<\G> assertion is not supported without |
19799a22 |
840 | the C</g> modifier. (Currently, without C</g>, C<\G> behaves just like |
841 | C<\A>, but that's accidental and may change in the future.) |
c90c0ff4 |
842 | |
843 | Examples: |
a0d0e21e |
844 | |
845 | # list context |
846 | ($one,$five,$fifteen) = (`uptime` =~ /(\d+\.\d+)/g); |
847 | |
848 | # scalar context |
19799a22 |
849 | $/ = ""; $* = 1; # $* deprecated in modern perls |
850 | while (defined($paragraph = <>)) { |
851 | while ($paragraph =~ /[a-z]['")]*[.!?]+['")]*\s/g) { |
852 | $sentences++; |
a0d0e21e |
853 | } |
854 | } |
855 | print "$sentences\n"; |
856 | |
c90c0ff4 |
857 | # using m//gc with \G |
137443ea |
858 | $_ = "ppooqppqq"; |
44a8e56a |
859 | while ($i++ < 2) { |
860 | print "1: '"; |
c90c0ff4 |
861 | print $1 while /(o)/gc; print "', pos=", pos, "\n"; |
44a8e56a |
862 | print "2: '"; |
c90c0ff4 |
863 | print $1 if /\G(q)/gc; print "', pos=", pos, "\n"; |
44a8e56a |
864 | print "3: '"; |
c90c0ff4 |
865 | print $1 while /(p)/gc; print "', pos=", pos, "\n"; |
44a8e56a |
866 | } |
867 | |
868 | The last example should print: |
869 | |
870 | 1: 'oo', pos=4 |
137443ea |
871 | 2: 'q', pos=5 |
44a8e56a |
872 | 3: 'pp', pos=7 |
873 | 1: '', pos=7 |
137443ea |
874 | 2: 'q', pos=8 |
875 | 3: '', pos=8 |
44a8e56a |
876 | |
c90c0ff4 |
877 | A useful idiom for C<lex>-like scanners is C</\G.../gc>. You can |
e7ea3e70 |
878 | combine several regexps like this to process a string part-by-part, |
c90c0ff4 |
879 | doing different actions depending on which regexp matched. Each |
880 | regexp tries to match where the previous one leaves off. |
e7ea3e70 |
881 | |
3fe9a6f1 |
882 | $_ = <<'EOL'; |
e7ea3e70 |
883 | $url = new URI::URL "http://www/"; die if $url eq "xXx"; |
3fe9a6f1 |
884 | EOL |
885 | LOOP: |
e7ea3e70 |
886 | { |
c90c0ff4 |
887 | print(" digits"), redo LOOP if /\G\d+\b[,.;]?\s*/gc; |
888 | print(" lowercase"), redo LOOP if /\G[a-z]+\b[,.;]?\s*/gc; |
889 | print(" UPPERCASE"), redo LOOP if /\G[A-Z]+\b[,.;]?\s*/gc; |
890 | print(" Capitalized"), redo LOOP if /\G[A-Z][a-z]+\b[,.;]?\s*/gc; |
891 | print(" MiXeD"), redo LOOP if /\G[A-Za-z]+\b[,.;]?\s*/gc; |
892 | print(" alphanumeric"), redo LOOP if /\G[A-Za-z0-9]+\b[,.;]?\s*/gc; |
893 | print(" line-noise"), redo LOOP if /\G[^A-Za-z0-9]+/gc; |
e7ea3e70 |
894 | print ". That's all!\n"; |
895 | } |
896 | |
897 | Here is the output (split into several lines): |
898 | |
899 | line-noise lowercase line-noise lowercase UPPERCASE line-noise |
900 | UPPERCASE line-noise lowercase line-noise lowercase line-noise |
901 | lowercase lowercase line-noise lowercase lowercase line-noise |
902 | MiXeD line-noise. That's all! |
44a8e56a |
903 | |
a0d0e21e |
904 | =item q/STRING/ |
905 | |
906 | =item C<'STRING'> |
907 | |
19799a22 |
908 | A single-quoted, literal string. A backslash represents a backslash |
68dc0745 |
909 | unless followed by the delimiter or another backslash, in which case |
910 | the delimiter or backslash is interpolated. |
a0d0e21e |
911 | |
912 | $foo = q!I said, "You said, 'She said it.'"!; |
913 | $bar = q('This is it.'); |
68dc0745 |
914 | $baz = '\n'; # a two-character string |
a0d0e21e |
915 | |
916 | =item qq/STRING/ |
917 | |
918 | =item "STRING" |
919 | |
920 | A double-quoted, interpolated string. |
921 | |
922 | $_ .= qq |
923 | (*** The previous line contains the naughty word "$1".\n) |
19799a22 |
924 | if /\b(tcl|java|python)\b/i; # :-) |
68dc0745 |
925 | $baz = "\n"; # a one-character string |
a0d0e21e |
926 | |
eec2d3df |
927 | =item qr/STRING/imosx |
928 | |
19799a22 |
929 | This operators quotes--and compiles--its I<STRING> as a regular |
930 | expression. I<STRING> is interpolated the same way as I<PATTERN> |
931 | in C<m/PATTERN/>. If "'" is used as the delimiter, no interpolation |
932 | is done. Returns a Perl value which may be used instead of the |
933 | corresponding C</STRING/imosx> expression. |
4b6a7270 |
934 | |
935 | For example, |
936 | |
937 | $rex = qr/my.STRING/is; |
938 | s/$rex/foo/; |
939 | |
940 | is equivalent to |
941 | |
942 | s/my.STRING/foo/is; |
943 | |
944 | The result may be used as a subpattern in a match: |
eec2d3df |
945 | |
946 | $re = qr/$pattern/; |
0a92e3a8 |
947 | $string =~ /foo${re}bar/; # can be interpolated in other patterns |
948 | $string =~ $re; # or used standalone |
4b6a7270 |
949 | $string =~ /$re/; # or this way |
950 | |
951 | Since Perl may compile the pattern at the moment of execution of qr() |
19799a22 |
952 | operator, using qr() may have speed advantages in some situations, |
4b6a7270 |
953 | notably if the result of qr() is used standalone: |
954 | |
955 | sub match { |
956 | my $patterns = shift; |
957 | my @compiled = map qr/$_/i, @$patterns; |
958 | grep { |
959 | my $success = 0; |
960 | foreach my $pat @compiled { |
961 | $success = 1, last if /$pat/; |
962 | } |
963 | $success; |
964 | } @_; |
965 | } |
966 | |
19799a22 |
967 | Precompilation of the pattern into an internal representation at |
968 | the moment of qr() avoids a need to recompile the pattern every |
969 | time a match C</$pat/> is attempted. (Perl has many other internal |
970 | optimizations, but none would be triggered in the above example if |
971 | we did not use qr() operator.) |
eec2d3df |
972 | |
973 | Options are: |
974 | |
975 | i Do case-insensitive pattern matching. |
976 | m Treat string as multiple lines. |
977 | o Compile pattern only once. |
978 | s Treat string as single line. |
979 | x Use extended regular expressions. |
980 | |
0a92e3a8 |
981 | See L<perlre> for additional information on valid syntax for STRING, and |
982 | for a detailed look at the semantics of regular expressions. |
983 | |
a0d0e21e |
984 | =item qx/STRING/ |
985 | |
986 | =item `STRING` |
987 | |
5a964f20 |
988 | A string which is (possibly) interpolated and then executed as a system |
989 | command with C</bin/sh> or its equivalent. Shell wildcards, pipes, |
990 | and redirections will be honored. The collected standard output of the |
991 | command is returned; standard error is unaffected. In scalar context, |
992 | it comes back as a single (potentially multi-line) string. In list |
993 | context, returns a list of lines (however you've defined lines with $/ |
994 | or $INPUT_RECORD_SEPARATOR). |
995 | |
996 | Because backticks do not affect standard error, use shell file descriptor |
997 | syntax (assuming the shell supports this) if you care to address this. |
998 | To capture a command's STDERR and STDOUT together: |
a0d0e21e |
999 | |
5a964f20 |
1000 | $output = `cmd 2>&1`; |
1001 | |
1002 | To capture a command's STDOUT but discard its STDERR: |
1003 | |
1004 | $output = `cmd 2>/dev/null`; |
1005 | |
1006 | To capture a command's STDERR but discard its STDOUT (ordering is |
1007 | important here): |
1008 | |
1009 | $output = `cmd 2>&1 1>/dev/null`; |
1010 | |
1011 | To exchange a command's STDOUT and STDERR in order to capture the STDERR |
1012 | but leave its STDOUT to come out the old STDERR: |
1013 | |
1014 | $output = `cmd 3>&1 1>&2 2>&3 3>&-`; |
1015 | |
1016 | To read both a command's STDOUT and its STDERR separately, it's easiest |
1017 | and safest to redirect them separately to files, and then read from those |
1018 | files when the program is done: |
1019 | |
1020 | system("program args 1>/tmp/program.stdout 2>/tmp/program.stderr"); |
1021 | |
1022 | Using single-quote as a delimiter protects the command from Perl's |
1023 | double-quote interpolation, passing it on to the shell instead: |
1024 | |
1025 | $perl_info = qx(ps $$); # that's Perl's $$ |
1026 | $shell_info = qx'ps $$'; # that's the new shell's $$ |
1027 | |
19799a22 |
1028 | How that string gets evaluated is entirely subject to the command |
5a964f20 |
1029 | interpreter on your system. On most platforms, you will have to protect |
1030 | shell metacharacters if you want them treated literally. This is in |
1031 | practice difficult to do, as it's unclear how to escape which characters. |
1032 | See L<perlsec> for a clean and safe example of a manual fork() and exec() |
1033 | to emulate backticks safely. |
a0d0e21e |
1034 | |
bb32b41a |
1035 | On some platforms (notably DOS-like ones), the shell may not be |
1036 | capable of dealing with multiline commands, so putting newlines in |
1037 | the string may not get you what you want. You may be able to evaluate |
1038 | multiple commands in a single line by separating them with the command |
1039 | separator character, if your shell supports that (e.g. C<;> on many Unix |
1040 | shells; C<&> on the Windows NT C<cmd> shell). |
1041 | |
1042 | Beware that some command shells may place restrictions on the length |
1043 | of the command line. You must ensure your strings don't exceed this |
1044 | limit after any necessary interpolations. See the platform-specific |
1045 | release notes for more details about your particular environment. |
1046 | |
5a964f20 |
1047 | Using this operator can lead to programs that are difficult to port, |
1048 | because the shell commands called vary between systems, and may in |
1049 | fact not be present at all. As one example, the C<type> command under |
1050 | the POSIX shell is very different from the C<type> command under DOS. |
1051 | That doesn't mean you should go out of your way to avoid backticks |
1052 | when they're the right way to get something done. Perl was made to be |
1053 | a glue language, and one of the things it glues together is commands. |
1054 | Just understand what you're getting yourself into. |
bb32b41a |
1055 | |
dc848c6f |
1056 | See L<"I/O Operators"> for more discussion. |
a0d0e21e |
1057 | |
1058 | =item qw/STRING/ |
1059 | |
8127e0e3 |
1060 | Evaluates to a list of the words extracted out of STRING, using embedded |
1061 | whitespace as the word delimiters. It can be understood as being roughly |
1062 | equivalent to: |
a0d0e21e |
1063 | |
1064 | split(' ', q/STRING/); |
1065 | |
26ef7447 |
1066 | the difference being that it generates a real list at compile time. So |
1067 | this expression: |
1068 | |
1069 | qw(foo bar baz) |
1070 | |
1071 | is exactly equivalent to the list: |
1072 | |
1073 | ('foo', 'bar', 'baz') |
5a964f20 |
1074 | |
a0d0e21e |
1075 | Some frequently seen examples: |
1076 | |
1077 | use POSIX qw( setlocale localeconv ) |
1078 | @EXPORT = qw( foo bar baz ); |
1079 | |
19799a22 |
1080 | A common mistake is to try to separate the words with comma or to |
1081 | put comments into a multi-line C<qw>-string. For this reason, the |
1082 | B<-w> switch (that is, the C<$^W> variable) produces warnings if |
1083 | the STRING contains the "," or the "#" character. |
7bac28a0 |
1084 | |
a0d0e21e |
1085 | =item s/PATTERN/REPLACEMENT/egimosx |
1086 | |
1087 | Searches a string for a pattern, and if found, replaces that pattern |
1088 | with the replacement text and returns the number of substitutions |
e37d713d |
1089 | made. Otherwise it returns false (specifically, the empty string). |
a0d0e21e |
1090 | |
1091 | If no string is specified via the C<=~> or C<!~> operator, the C<$_> |
1092 | variable is searched and modified. (The string specified with C<=~> must |
5a964f20 |
1093 | be scalar variable, an array element, a hash element, or an assignment |
5f05dabc |
1094 | to one of those, i.e., an lvalue.) |
a0d0e21e |
1095 | |
19799a22 |
1096 | If the delimiter chosen is a single quote, no interpolation is |
a0d0e21e |
1097 | done on either the PATTERN or the REPLACEMENT. Otherwise, if the |
1098 | PATTERN contains a $ that looks like a variable rather than an |
1099 | end-of-string test, the variable will be interpolated into the pattern |
5f05dabc |
1100 | at run-time. If you want the pattern compiled only once the first time |
a0d0e21e |
1101 | the variable is interpolated, use the C</o> option. If the pattern |
5a964f20 |
1102 | evaluates to the empty string, the last successfully executed regular |
a0d0e21e |
1103 | expression is used instead. See L<perlre> for further explanation on these. |
5a964f20 |
1104 | See L<perllocale> for discussion of additional considerations that apply |
a034a98d |
1105 | when C<use locale> is in effect. |
a0d0e21e |
1106 | |
1107 | Options are: |
1108 | |
1109 | e Evaluate the right side as an expression. |
5f05dabc |
1110 | g Replace globally, i.e., all occurrences. |
a0d0e21e |
1111 | i Do case-insensitive pattern matching. |
1112 | m Treat string as multiple lines. |
5f05dabc |
1113 | o Compile pattern only once. |
a0d0e21e |
1114 | s Treat string as single line. |
1115 | x Use extended regular expressions. |
1116 | |
1117 | Any non-alphanumeric, non-whitespace delimiter may replace the |
1118 | slashes. If single quotes are used, no interpretation is done on the |
e37d713d |
1119 | replacement string (the C</e> modifier overrides this, however). Unlike |
54310121 |
1120 | Perl 4, Perl 5 treats backticks as normal delimiters; the replacement |
e37d713d |
1121 | text is not evaluated as a command. If the |
a0d0e21e |
1122 | PATTERN is delimited by bracketing quotes, the REPLACEMENT has its own |
5f05dabc |
1123 | pair of quotes, which may or may not be bracketing quotes, e.g., |
a0d0e21e |
1124 | C<s(foo)(bar)> or C<sE<lt>fooE<gt>/bar/>. A C</e> will cause the |
7b8d334a |
1125 | replacement portion to be interpreted as a full-fledged Perl expression |
a0d0e21e |
1126 | and eval()ed right then and there. It is, however, syntax checked at |
1127 | compile-time. |
1128 | |
1129 | Examples: |
1130 | |
1131 | s/\bgreen\b/mauve/g; # don't change wintergreen |
1132 | |
1133 | $path =~ s|/usr/bin|/usr/local/bin|; |
1134 | |
1135 | s/Login: $foo/Login: $bar/; # run-time pattern |
1136 | |
5a964f20 |
1137 | ($foo = $bar) =~ s/this/that/; # copy first, then change |
a0d0e21e |
1138 | |
5a964f20 |
1139 | $count = ($paragraph =~ s/Mister\b/Mr./g); # get change-count |
a0d0e21e |
1140 | |
1141 | $_ = 'abc123xyz'; |
1142 | s/\d+/$&*2/e; # yields 'abc246xyz' |
1143 | s/\d+/sprintf("%5d",$&)/e; # yields 'abc 246xyz' |
1144 | s/\w/$& x 2/eg; # yields 'aabbcc 224466xxyyzz' |
1145 | |
1146 | s/%(.)/$percent{$1}/g; # change percent escapes; no /e |
1147 | s/%(.)/$percent{$1} || $&/ge; # expr now, so /e |
1148 | s/^=(\w+)/&pod($1)/ge; # use function call |
1149 | |
5a964f20 |
1150 | # expand variables in $_, but dynamics only, using |
1151 | # symbolic dereferencing |
1152 | s/\$(\w+)/${$1}/g; |
1153 | |
a0d0e21e |
1154 | # /e's can even nest; this will expand |
5a964f20 |
1155 | # any embedded scalar variable (including lexicals) in $_ |
a0d0e21e |
1156 | s/(\$\w+)/$1/eeg; |
1157 | |
5a964f20 |
1158 | # Delete (most) C comments. |
a0d0e21e |
1159 | $program =~ s { |
4633a7c4 |
1160 | /\* # Match the opening delimiter. |
1161 | .*? # Match a minimal number of characters. |
1162 | \*/ # Match the closing delimiter. |
a0d0e21e |
1163 | } []gsx; |
1164 | |
5a964f20 |
1165 | s/^\s*(.*?)\s*$/$1/; # trim white space in $_, expensively |
1166 | |
1167 | for ($variable) { # trim white space in $variable, cheap |
1168 | s/^\s+//; |
1169 | s/\s+$//; |
1170 | } |
a0d0e21e |
1171 | |
1172 | s/([^ ]*) *([^ ]*)/$2 $1/; # reverse 1st two fields |
1173 | |
54310121 |
1174 | Note the use of $ instead of \ in the last example. Unlike |
5f05dabc |
1175 | B<sed>, we use the \E<lt>I<digit>E<gt> form in only the left hand side. |
6ee5d4e7 |
1176 | Anywhere else it's $E<lt>I<digit>E<gt>. |
a0d0e21e |
1177 | |
5f05dabc |
1178 | Occasionally, you can't use just a C</g> to get all the changes |
19799a22 |
1179 | to occur that you might want. Here are two common cases: |
a0d0e21e |
1180 | |
1181 | # put commas in the right places in an integer |
19799a22 |
1182 | 1 while s/(\d)(\d\d\d)(?!\d)/$1,$2/g; |
a0d0e21e |
1183 | |
1184 | # expand tabs to 8-column spacing |
1185 | 1 while s/\t+/' ' x (length($&)*8 - length($`)%8)/e; |
1186 | |
a0ed51b3 |
1187 | =item tr/SEARCHLIST/REPLACEMENTLIST/cdsUC |
a0d0e21e |
1188 | |
a0ed51b3 |
1189 | =item y/SEARCHLIST/REPLACEMENTLIST/cdsUC |
a0d0e21e |
1190 | |
2c268ad5 |
1191 | Transliterates all occurrences of the characters found in the search list |
a0d0e21e |
1192 | with the corresponding character in the replacement list. It returns |
1193 | the number of characters replaced or deleted. If no string is |
2c268ad5 |
1194 | specified via the =~ or !~ operator, the $_ string is transliterated. (The |
54310121 |
1195 | string specified with =~ must be a scalar variable, an array element, a |
1196 | hash element, or an assignment to one of those, i.e., an lvalue.) |
8ada0baa |
1197 | |
2c268ad5 |
1198 | A character range may be specified with a hyphen, so C<tr/A-J/0-9/> |
1199 | does the same replacement as C<tr/ACEGIBDFHJ/0246813579/>. |
54310121 |
1200 | For B<sed> devotees, C<y> is provided as a synonym for C<tr>. If the |
1201 | SEARCHLIST is delimited by bracketing quotes, the REPLACEMENTLIST has |
1202 | its own pair of quotes, which may or may not be bracketing quotes, |
2c268ad5 |
1203 | e.g., C<tr[A-Z][a-z]> or C<tr(+\-*/)/ABCD/>. |
a0d0e21e |
1204 | |
8ada0baa |
1205 | Note also that the whole range idea is rather unportable between |
1206 | character sets--and even within character sets they may cause results |
1207 | you probably didn't expect. A sound principle is to use only ranges |
1208 | that begin from and end at either alphabets of equal case (a-e, A-E), |
1209 | or digits (0-4). Anything else is unsafe. If in doubt, spell out the |
1210 | character sets in full. |
1211 | |
a0d0e21e |
1212 | Options: |
1213 | |
1214 | c Complement the SEARCHLIST. |
1215 | d Delete found but unreplaced characters. |
1216 | s Squash duplicate replaced characters. |
a0ed51b3 |
1217 | U Translate to/from UTF-8. |
1218 | C Translate to/from 8-bit char (octet). |
a0d0e21e |
1219 | |
19799a22 |
1220 | If the C</c> modifier is specified, the SEARCHLIST character set |
1221 | is complemented. If the C</d> modifier is specified, any characters |
1222 | specified by SEARCHLIST not found in REPLACEMENTLIST are deleted. |
1223 | (Note that this is slightly more flexible than the behavior of some |
1224 | B<tr> programs, which delete anything they find in the SEARCHLIST, |
1225 | period.) If the C</s> modifier is specified, sequences of characters |
1226 | that were transliterated to the same character are squashed down |
1227 | to a single instance of the character. |
a0d0e21e |
1228 | |
1229 | If the C</d> modifier is used, the REPLACEMENTLIST is always interpreted |
1230 | exactly as specified. Otherwise, if the REPLACEMENTLIST is shorter |
1231 | than the SEARCHLIST, the final character is replicated till it is long |
5a964f20 |
1232 | enough. If the REPLACEMENTLIST is empty, the SEARCHLIST is replicated. |
a0d0e21e |
1233 | This latter is useful for counting characters in a class or for |
1234 | squashing character sequences in a class. |
1235 | |
a0ed51b3 |
1236 | The first C</U> or C</C> modifier applies to the left side of the translation. |
1237 | The second one applies to the right side. If present, these modifiers override |
1238 | the current utf8 state. |
1239 | |
a0d0e21e |
1240 | Examples: |
1241 | |
1242 | $ARGV[1] =~ tr/A-Z/a-z/; # canonicalize to lower case |
1243 | |
1244 | $cnt = tr/*/*/; # count the stars in $_ |
1245 | |
1246 | $cnt = $sky =~ tr/*/*/; # count the stars in $sky |
1247 | |
1248 | $cnt = tr/0-9//; # count the digits in $_ |
1249 | |
1250 | tr/a-zA-Z//s; # bookkeeper -> bokeper |
1251 | |
1252 | ($HOST = $host) =~ tr/a-z/A-Z/; |
1253 | |
1254 | tr/a-zA-Z/ /cs; # change non-alphas to single space |
1255 | |
1256 | tr [\200-\377] |
1257 | [\000-\177]; # delete 8th bit |
1258 | |
19799a22 |
1259 | tr/\0-\xFF//CU; # change Latin-1 to Unicode |
1260 | tr/\0-\x{FF}//UC; # change Unicode to Latin-1 |
a0ed51b3 |
1261 | |
19799a22 |
1262 | If multiple transliterations are given for a character, only the |
1263 | first one is used: |
748a9306 |
1264 | |
1265 | tr/AAA/XYZ/ |
1266 | |
2c268ad5 |
1267 | will transliterate any A to X. |
748a9306 |
1268 | |
19799a22 |
1269 | Because the transliteration table is built at compile time, neither |
a0d0e21e |
1270 | the SEARCHLIST nor the REPLACEMENTLIST are subjected to double quote |
19799a22 |
1271 | interpolation. That means that if you want to use variables, you |
1272 | must use an eval(): |
a0d0e21e |
1273 | |
1274 | eval "tr/$oldlist/$newlist/"; |
1275 | die $@ if $@; |
1276 | |
1277 | eval "tr/$oldlist/$newlist/, 1" or die $@; |
1278 | |
1279 | =back |
1280 | |
75e14d17 |
1281 | =head2 Gory details of parsing quoted constructs |
1282 | |
19799a22 |
1283 | When presented with something that might have several different |
1284 | interpretations, Perl uses the B<DWIM> (that's "Do What I Mean") |
1285 | principle to pick the most probable interpretation. This strategy |
1286 | is so successful that Perl programmers often do not suspect the |
1287 | ambivalence of what they write. But from time to time, Perl's |
1288 | notions differ substantially from what the author honestly meant. |
1289 | |
1290 | This section hopes to clarify how Perl handles quoted constructs. |
1291 | Although the most common reason to learn this is to unravel labyrinthine |
1292 | regular expressions, because the initial steps of parsing are the |
1293 | same for all quoting operators, they are all discussed together. |
1294 | |
1295 | The most important Perl parsing rule is the first one discussed |
1296 | below: when processing a quoted construct, Perl first finds the end |
1297 | of that construct, then interprets its contents. If you understand |
1298 | this rule, you may skip the rest of this section on the first |
1299 | reading. The other rules are likely to contradict the user's |
1300 | expectations much less frequently than this first one. |
1301 | |
1302 | Some passes discussed below are performed concurrently, but because |
1303 | their results are the same, we consider them individually. For different |
1304 | quoting constructs, Perl performs different numbers of passes, from |
1305 | one to five, but these passes are always performed in the same order. |
75e14d17 |
1306 | |
1307 | =over |
1308 | |
1309 | =item Finding the end |
1310 | |
19799a22 |
1311 | The first pass is finding the end of the quoted construct, whether |
1312 | it be a multicharacter delimiter C<"\nEOF\n"> in the C<<<EOF> |
1313 | construct, a C</> that terminates a C<qq//> construct, a C<]> which |
1314 | terminates C<qq[]> construct, or a C<E<gt>> which terminates a |
1315 | fileglob started with C<E<lt>>. |
75e14d17 |
1316 | |
19799a22 |
1317 | When searching for single-character non-pairing delimiters, such |
1318 | as C</>, combinations of C<\\> and C<\/> are skipped. However, |
1319 | when searching for single-character pairing delimiter like C<[>, |
1320 | combinations of C<\\>, C<\]>, and C<\[> are all skipped, and nested |
1321 | C<[>, C<]> are skipped as well. When searching for multicharacter |
1322 | delimiters, nothing is skipped. |
75e14d17 |
1323 | |
19799a22 |
1324 | For constructs with three-part delimiters (C<s///>, C<y///>, and |
1325 | C<tr///>), the search is repeated once more. |
75e14d17 |
1326 | |
19799a22 |
1327 | During this search no attention is paid to the semantics of the construct. |
1328 | Thus: |
75e14d17 |
1329 | |
1330 | "$hash{"$foo/$bar"}" |
1331 | |
2a94b7ce |
1332 | or: |
75e14d17 |
1333 | |
1334 | m/ |
2a94b7ce |
1335 | bar # NOT a comment, this slash / terminated m//! |
75e14d17 |
1336 | /x |
1337 | |
19799a22 |
1338 | do not form legal quoted expressions. The quoted part ends on the |
1339 | first C<"> and C</>, and the rest happens to be a syntax error. |
1340 | Because the slash that terminated C<m//> was followed by a C<SPACE>, |
1341 | the example above is not C<m//x>, but rather C<m//> with no C</x> |
1342 | modifier. So the embedded C<#> is interpreted as a literal C<#>. |
75e14d17 |
1343 | |
1344 | =item Removal of backslashes before delimiters |
1345 | |
19799a22 |
1346 | During the second pass, text between the starting and ending |
1347 | delimiters is copied to a safe location, and the C<\> is removed |
1348 | from combinations consisting of C<\> and delimiter--or delimiters, |
1349 | meaning both starting and ending delimiters will should these differ. |
1350 | This removal does not happen for multi-character delimiters. |
1351 | Note that the combination C<\\> is left intact, just as it was. |
75e14d17 |
1352 | |
19799a22 |
1353 | Starting from this step no information about the delimiters is |
1354 | used in parsing. |
75e14d17 |
1355 | |
1356 | =item Interpolation |
1357 | |
19799a22 |
1358 | The next step is interpolation in the text obtained, which is now |
1359 | delimiter-independent. There are four different cases. |
75e14d17 |
1360 | |
1361 | =over |
1362 | |
1363 | =item C<<<'EOF'>, C<m''>, C<s'''>, C<tr///>, C<y///> |
1364 | |
1365 | No interpolation is performed. |
1366 | |
1367 | =item C<''>, C<q//> |
1368 | |
1369 | The only interpolation is removal of C<\> from pairs C<\\>. |
1370 | |
1371 | =item C<"">, C<``>, C<qq//>, C<qx//>, C<<file*globE<gt>> |
1372 | |
19799a22 |
1373 | C<\Q>, C<\U>, C<\u>, C<\L>, C<\l> (possibly paired with C<\E>) are |
1374 | converted to corresponding Perl constructs. Thus, C<"$foo\Qbaz$bar"> |
1375 | is converted to C<$foo . (quotemeta("baz" . $bar))> internally. |
1376 | The other combinations are replaced with appropriate expansions. |
2a94b7ce |
1377 | |
19799a22 |
1378 | Let it be stressed that I<whatever falls between C<\Q> and C<\E>> |
1379 | is interpolated in the usual way. Something like C<"\Q\\E"> has |
1380 | no C<\E> inside. instead, it has C<\Q>, C<\\>, and C<E>, so the |
1381 | result is the same as for C<"\\\\E">. As a general rule, backslashes |
1382 | between C<\Q> and C<\E> may lead to counterintuitive results. So, |
1383 | C<"\Q\t\E"> is converted to C<quotemeta("\t")>, which is the same |
1384 | as C<"\\\t"> (since TAB is not alphanumeric). Note also that: |
2a94b7ce |
1385 | |
1386 | $str = '\t'; |
1387 | return "\Q$str"; |
1388 | |
1389 | may be closer to the conjectural I<intention> of the writer of C<"\Q\t\E">. |
1390 | |
19799a22 |
1391 | Interpolated scalars and arrays are converted internally to the C<join> and |
1392 | C<.> catentation operations. Thus, C<"$foo XXX '@arr'"> becomes: |
75e14d17 |
1393 | |
19799a22 |
1394 | $foo . " XXX '" . (join $", @arr) . "'"; |
75e14d17 |
1395 | |
19799a22 |
1396 | All operations above are performed simultaneously, left to right. |
75e14d17 |
1397 | |
19799a22 |
1398 | Because the result of C<"\Q STRING \E"> has all metacharacters |
1399 | quoted, there is no way to insert a literal C<$> or C<@> inside a |
1400 | C<\Q\E> pair. If protected by C<\>, C<$> will be quoted to became |
1401 | C<"\\\$">; if not, it is interpreted as the start of an interpolated |
1402 | scalar. |
75e14d17 |
1403 | |
19799a22 |
1404 | Note also that the interpolation code needs to make a decision on |
1405 | where the interpolated scalar ends. For instance, whether |
1406 | C<"a $b -E<gt> {c}"> really means: |
75e14d17 |
1407 | |
1408 | "a " . $b . " -> {c}"; |
1409 | |
2a94b7ce |
1410 | or: |
75e14d17 |
1411 | |
1412 | "a " . $b -> {c}; |
1413 | |
19799a22 |
1414 | Most of the time, the longest possible text that does not include |
1415 | spaces between components and which contains matching braces or |
1416 | brackets. because the outcome may be determined by voting based |
1417 | on heuristic estimators, the result is not strictly predictable. |
1418 | Fortunately, it's usually correct for ambiguous cases. |
75e14d17 |
1419 | |
1420 | =item C<?RE?>, C</RE/>, C<m/RE/>, C<s/RE/foo/>, |
1421 | |
19799a22 |
1422 | Processing of C<\Q>, C<\U>, C<\u>, C<\L>, C<\l>, and interpolation |
1423 | happens (almost) as with C<qq//> constructs, but the substitution |
1424 | of C<\> followed by RE-special chars (including C<\>) is not |
1425 | performed. Moreover, inside C<(?{BLOCK})>, C<(?# comment )>, and |
1426 | a C<#>-comment in a C<//x>-regular expression, no processing is |
1427 | performed whatsoever. This is the first step at which the presence |
1428 | of the C<//x> modifier is relevant. |
1429 | |
1430 | Interpolation has several quirks: C<$|>, C<$(>, and C<$)> are not |
1431 | interpolated, and constructs C<$var[SOMETHING]> are voted (by several |
1432 | different estimators) to be either an array element or C<$var> |
1433 | followed by an RE alternative. This is where the notation |
1434 | C<${arr[$bar]}> comes handy: C</${arr[0-9]}/> is interpreted as |
1435 | array element C<-9>, not as a regular expression from the variable |
1436 | C<$arr> followed by a digit, which would be the interpretation of |
1437 | C</$arr[0-9]/>. Since voting among different estimators may occur, |
1438 | the result is not predictable. |
1439 | |
1440 | It is at this step that C<\1> is begrudgingly converted to C<$1> in |
1441 | the replacement text of C<s///> to correct the incorrigible |
1442 | I<sed> hackers who haven't picked up the saner idiom yet. A warning |
1443 | is emitted if the B<-w> command-line flag (that is, the C<$^W> variable) |
1444 | was set. |
1445 | |
1446 | The lack of processing of C<\\> creates specific restrictions on |
1447 | the post-processed text. If the delimiter is C</>, one cannot get |
1448 | the combination C<\/> into the result of this step. C</> will |
1449 | finish the regular expression, C<\/> will be stripped to C</> on |
1450 | the previous step, and C<\\/> will be left as is. Because C</> is |
1451 | equivalent to C<\/> inside a regular expression, this does not |
1452 | matter unless the delimiter happens to be character special to the |
1453 | RE engine, such as in C<s*foo*bar*>, C<m[foo]>, or C<?foo?>; or an |
1454 | alphanumeric char, as in: |
2a94b7ce |
1455 | |
1456 | m m ^ a \s* b mmx; |
1457 | |
19799a22 |
1458 | In the RE above, which is intentionally obfuscated for illustration, the |
2a94b7ce |
1459 | delimiter is C<m>, the modifier is C<mx>, and after backslash-removal the |
19799a22 |
1460 | RE is the same as for C<m/ ^ a s* b /mx>). There's more than one |
1461 | reason you're encouraged to restrict your delimiters to non-alphanumeric, |
1462 | non-whitespace choices. |
75e14d17 |
1463 | |
1464 | =back |
1465 | |
19799a22 |
1466 | This step is the last one for all constructs except regular expressions, |
75e14d17 |
1467 | which are processed further. |
1468 | |
1469 | =item Interpolation of regular expressions |
1470 | |
19799a22 |
1471 | Previous steps were performed during the compilation of Perl code, |
1472 | but this one happens at run time--although it may be optimized to |
1473 | be calculated at compile time if appropriate. After preprocessing |
1474 | described above, and possibly after evaluation if catenation, |
1475 | joining, casing translation, or metaquoting are involved, the |
1476 | resulting I<string> is passed to the RE engine for compilation. |
1477 | |
1478 | Whatever happens in the RE engine might be better discussed in L<perlre>, |
1479 | but for the sake of continuity, we shall do so here. |
1480 | |
1481 | This is another step where the presence of the C<//x> modifier is |
1482 | relevant. The RE engine scans the string from left to right and |
1483 | converts it to a finite automaton. |
1484 | |
1485 | Backslashed characters are either replaced with corresponding |
1486 | literal strings (as with C<\{>), or else they generate special nodes |
1487 | in the finite automaton (as with C<\b>). Characters special to the |
1488 | RE engine (such as C<|>) generate corresponding nodes or groups of |
1489 | nodes. C<(?#...)> comments are ignored. All the rest is either |
1490 | converted to literal strings to match, or else is ignored (as is |
1491 | whitespace and C<#>-style comments if C<//x> is present). |
1492 | |
1493 | Parsing of the bracketed character class construct, C<[...]>, is |
1494 | rather different than the rule used for the rest of the pattern. |
1495 | The terminator of this construct is found using the same rules as |
1496 | for finding the terminator of a C<{}>-delimited construct, the only |
1497 | exception being that C<]> immediately following C<[> is treated as |
1498 | though preceded by a backslash. Similarly, the terminator of |
1499 | C<(?{...})> is found using the same rules as for finding the |
1500 | terminator of a C<{}>-delimited construct. |
1501 | |
1502 | It is possible to inspect both the string given to RE engine and the |
1503 | resulting finite automaton. See the arguments C<debug>/C<debugcolor> |
1504 | in the C<use L<re>> pragma, as well as Perl's B<-Dr> command-line |
1505 | switch documented in L<perlrun/Switches>. |
75e14d17 |
1506 | |
1507 | =item Optimization of regular expressions |
1508 | |
7522fed5 |
1509 | This step is listed for completeness only. Since it does not change |
75e14d17 |
1510 | semantics, details of this step are not documented and are subject |
19799a22 |
1511 | to change without notice. This step is performed over the finite |
1512 | automaton that was generated during the previous pass. |
2a94b7ce |
1513 | |
19799a22 |
1514 | It is at this stage that C<split()> silently optimizes C</^/> to |
1515 | mean C</^/m>. |
75e14d17 |
1516 | |
1517 | =back |
1518 | |
a0d0e21e |
1519 | =head2 I/O Operators |
1520 | |
54310121 |
1521 | There are several I/O operators you should know about. |
fbad3eb5 |
1522 | |
7b8d334a |
1523 | A string enclosed by backticks (grave accents) first undergoes |
19799a22 |
1524 | double-quote interpolation. It is then interpreted as an external |
1525 | command, and the output of that command is the value of the |
1526 | pseudo-literal, j |
1527 | string consisting of all output is returned. In list context, a |
1528 | list of values is returned, one per line of output. (You can set |
1529 | C<$/> to use a different line terminator.) The command is executed |
a0d0e21e |
1530 | each time the pseudo-literal is evaluated. The status value of the |
1531 | command is returned in C<$?> (see L<perlvar> for the interpretation |
1532 | of C<$?>). Unlike in B<csh>, no translation is done on the return |
1533 | data--newlines remain newlines. Unlike in any of the shells, single |
1534 | quotes do not hide variable names in the command from interpretation. |
19799a22 |
1535 | To pass a literal dollar-sign through to the shell you need to hide |
1536 | it with a backslash. The generalized form of backticks is C<qx//>. |
1537 | (Because backticks always undergo shell expansion as well, see |
1538 | L<perlsec> for security concerns.) |
1539 | |
1540 | In scalar context, evaluating a filehandle in angle brackets yields |
1541 | the next line from that file (the newline, if any, included), or |
1542 | C<undef> at end-of-file or on error. When C<$/> is set to C<undef> |
1543 | (sometimes known as file-slurp mode) and the file is empty, it |
1544 | returns C<''> the first time, followed by C<undef> subsequently. |
1545 | |
1546 | Ordinarily you must assign the returned value to a variable, but |
1547 | there is one situation where an automatic assignment happens. If |
1548 | and only if the input symbol is the only thing inside the conditional |
1549 | of a C<while> statement (even if disguised as a C<for(;;)> loop), |
1550 | the value is automatically assigned to the global variable $_, |
1551 | destroying whatever was there previously. (This may seem like an |
1552 | odd thing to you, but you'll use the construct in almost every Perl |
1553 | script you write.) The $_ variables is not implicitly localized. |
1554 | You'll have to put a C<local $_;> before the loop if you want that |
1555 | to happen. |
1556 | |
1557 | The following lines are equivalent: |
a0d0e21e |
1558 | |
748a9306 |
1559 | while (defined($_ = <STDIN>)) { print; } |
7b8d334a |
1560 | while ($_ = <STDIN>) { print; } |
a0d0e21e |
1561 | while (<STDIN>) { print; } |
1562 | for (;<STDIN>;) { print; } |
748a9306 |
1563 | print while defined($_ = <STDIN>); |
7b8d334a |
1564 | print while ($_ = <STDIN>); |
a0d0e21e |
1565 | print while <STDIN>; |
1566 | |
19799a22 |
1567 | This also behaves similarly, but avoids $_ : |
7b8d334a |
1568 | |
1569 | while (my $line = <STDIN>) { print $line } |
1570 | |
19799a22 |
1571 | In these loop constructs, the assigned value (whether assignment |
1572 | is automatic or explicit) is then tested to see whether it is |
1573 | defined. The defined test avoids problems where line has a string |
1574 | value that would be treated as false by Perl, for example a "" or |
1575 | a "0" with no trailing newline. If you really mean for such values |
1576 | to terminate the loop, they should be tested for explicitly: |
7b8d334a |
1577 | |
1578 | while (($_ = <STDIN>) ne '0') { ... } |
1579 | while (<STDIN>) { last unless $_; ... } |
1580 | |
19799a22 |
1581 | In other boolean contexts, C<E<lt>I<filehandle>E<gt>> without an |
1582 | explicit C<defined> test or comparison elicit a warning if the B<-w> |
1583 | command-line switch (the C<$^W> variable) is in effect. |
7b8d334a |
1584 | |
5f05dabc |
1585 | The filehandles STDIN, STDOUT, and STDERR are predefined. (The |
19799a22 |
1586 | filehandles C<stdin>, C<stdout>, and C<stderr> will also work except |
1587 | in packages, where they would be interpreted as local identifiers |
1588 | rather than global.) Additional filehandles may be created with |
1589 | the open() function, amongst others. See L<perlopentut> and |
1590 | L<perlfunc/open> for details on this. |
a0d0e21e |
1591 | |
19799a22 |
1592 | If a E<lt>FILEHANDLEE<gt> is used in a context that is looking for |
1593 | a list, a list comprising all input lines is returned, one line per |
1594 | list element. It's easy to grow to a rather large data space this |
1595 | way, so use with care. |
a0d0e21e |
1596 | |
19799a22 |
1597 | E<lt>FILEHANDLEE<gt> may also be spelled C<readline(*FILEHANDLE)>. |
1598 | See L<perlfunc/readline>. |
fbad3eb5 |
1599 | |
19799a22 |
1600 | The null filehandle E<lt>E<gt> is special: it can be used to emulate the |
d28ebecd |
1601 | behavior of B<sed> and B<awk>. Input from E<lt>E<gt> comes either from |
a0d0e21e |
1602 | standard input, or from each file listed on the command line. Here's |
d28ebecd |
1603 | how it works: the first time E<lt>E<gt> is evaluated, the @ARGV array is |
5a964f20 |
1604 | checked, and if it is empty, C<$ARGV[0]> is set to "-", which when opened |
a0d0e21e |
1605 | gives you standard input. The @ARGV array is then processed as a list |
1606 | of filenames. The loop |
1607 | |
1608 | while (<>) { |
1609 | ... # code for each line |
1610 | } |
1611 | |
1612 | is equivalent to the following Perl-like pseudo code: |
1613 | |
3e3baf6d |
1614 | unshift(@ARGV, '-') unless @ARGV; |
a0d0e21e |
1615 | while ($ARGV = shift) { |
1616 | open(ARGV, $ARGV); |
1617 | while (<ARGV>) { |
1618 | ... # code for each line |
1619 | } |
1620 | } |
1621 | |
19799a22 |
1622 | except that it isn't so cumbersome to say, and will actually work. |
1623 | It really does shift the @ARGV array and put the current filename |
1624 | into the $ARGV variable. It also uses filehandle I<ARGV> |
1625 | internally--E<lt>E<gt> is just a synonym for E<lt>ARGVE<gt>, which |
1626 | is magical. (The pseudo code above doesn't work because it treats |
1627 | E<lt>ARGVE<gt> as non-magical.) |
a0d0e21e |
1628 | |
d28ebecd |
1629 | You can modify @ARGV before the first E<lt>E<gt> as long as the array ends up |
a0d0e21e |
1630 | containing the list of filenames you really want. Line numbers (C<$.>) |
19799a22 |
1631 | continue as though the input were one big happy file. See the example |
1632 | in L<perlfunc/eof> for how to reset line numbers on each file. |
5a964f20 |
1633 | |
1634 | If you want to set @ARGV to your own list of files, go right ahead. |
1635 | This sets @ARGV to all plain text files if no @ARGV was given: |
1636 | |
1637 | @ARGV = grep { -f && -T } glob('*') unless @ARGV; |
a0d0e21e |
1638 | |
5a964f20 |
1639 | You can even set them to pipe commands. For example, this automatically |
1640 | filters compressed arguments through B<gzip>: |
1641 | |
1642 | @ARGV = map { /\.(gz|Z)$/ ? "gzip -dc < $_ |" : $_ } @ARGV; |
1643 | |
1644 | If you want to pass switches into your script, you can use one of the |
a0d0e21e |
1645 | Getopts modules or put a loop on the front like this: |
1646 | |
1647 | while ($_ = $ARGV[0], /^-/) { |
1648 | shift; |
1649 | last if /^--$/; |
1650 | if (/^-D(.*)/) { $debug = $1 } |
1651 | if (/^-v/) { $verbose++ } |
5a964f20 |
1652 | # ... # other switches |
a0d0e21e |
1653 | } |
5a964f20 |
1654 | |
a0d0e21e |
1655 | while (<>) { |
5a964f20 |
1656 | # ... # code for each line |
a0d0e21e |
1657 | } |
1658 | |
7b8d334a |
1659 | The E<lt>E<gt> symbol will return C<undef> for end-of-file only once. |
19799a22 |
1660 | If you call it again after this, it will assume you are processing another |
1661 | @ARGV list, and if you haven't set @ARGV, will read input from STDIN. |
a0d0e21e |
1662 | |
19799a22 |
1663 | If angle brackets contain is a simple scalar variable (e.g., |
1664 | E<lt>$fooE<gt>), then that variable contains the name of the |
1665 | filehandle to input from, or its typeglob, or a reference to the |
1666 | same. For example: |
cb1a09d0 |
1667 | |
1668 | $fh = \*STDIN; |
1669 | $line = <$fh>; |
a0d0e21e |
1670 | |
5a964f20 |
1671 | If what's within the angle brackets is neither a filehandle nor a simple |
1672 | scalar variable containing a filehandle name, typeglob, or typeglob |
1673 | reference, it is interpreted as a filename pattern to be globbed, and |
1674 | either a list of filenames or the next filename in the list is returned, |
19799a22 |
1675 | depending on context. This distinction is determined on syntactic |
1676 | grounds alone. That means C<E<lt>$xE<gt>> is always a readline() from |
1677 | an indirect handle, but C<E<lt>$hash{key}E<gt>> is always a glob(). |
5a964f20 |
1678 | That's because $x is a simple scalar variable, but C<$hash{key}> is |
1679 | not--it's a hash element. |
1680 | |
1681 | One level of double-quote interpretation is done first, but you can't |
1682 | say C<E<lt>$fooE<gt>> because that's an indirect filehandle as explained |
1683 | in the previous paragraph. (In older versions of Perl, programmers |
1684 | would insert curly brackets to force interpretation as a filename glob: |
1685 | C<E<lt>${foo}E<gt>>. These days, it's considered cleaner to call the |
1686 | internal function directly as C<glob($foo)>, which is probably the right |
19799a22 |
1687 | way to have done it in the first place.) For example: |
a0d0e21e |
1688 | |
1689 | while (<*.c>) { |
1690 | chmod 0644, $_; |
1691 | } |
1692 | |
1693 | is equivalent to |
1694 | |
1695 | open(FOO, "echo *.c | tr -s ' \t\r\f' '\\012\\012\\012\\012'|"); |
1696 | while (<FOO>) { |
1697 | chop; |
1698 | chmod 0644, $_; |
1699 | } |
1700 | |
19799a22 |
1701 | In fact, it's currently implemented that way, but this is expected |
1702 | to be made completely internal in the near future. (Which means |
1703 | it will not work on filenames with spaces in them unless you have |
1704 | csh(1) on your machine.) Of course, the shortest way to do the |
1705 | above is: |
a0d0e21e |
1706 | |
1707 | chmod 0644, <*.c>; |
1708 | |
19799a22 |
1709 | Because globbing currently invokes a shell, it's often faster to |
1710 | call readdir() yourself and do your own grep() on the filenames. |
1711 | Furthermore, due to its current implementation of using a shell, |
1712 | the glob() routine may get "Arg list too long" errors (unless you've |
1713 | installed tcsh(1L) as F</bin/csh> or hacked your F<config.sh>). |
1714 | |
1715 | A (file)glob evaluates its (embedded) argument only when it is |
1716 | starting a new list. All values must be read before it will start |
1717 | over. In list context, this isn't important because you automatically |
1718 | get them all anyway. However, in scalar context the operator returns |
1719 | the next value each time it's called, or C |
1720 | run out. As with filehandle reads, an automatic C<defined> is |
1721 | generated when the glob occurs in the test part of a C<while>, |
1722 | because legal glob returns (e.g. a file called F<0>) would otherwise |
1723 | terminate the loop. Again, C<undef> is returned only once. So if |
1724 | you're expecting a single value from a glob, it is much better to |
1725 | say |
4633a7c4 |
1726 | |
1727 | ($file) = <blurch*>; |
1728 | |
1729 | than |
1730 | |
1731 | $file = <blurch*>; |
1732 | |
1733 | because the latter will alternate between returning a filename and |
19799a22 |
1734 | returning false. |
4633a7c4 |
1735 | |
1736 | It you're trying to do variable interpolation, it's definitely better |
1737 | to use the glob() function, because the older notation can cause people |
e37d713d |
1738 | to become confused with the indirect filehandle notation. |
4633a7c4 |
1739 | |
1740 | @files = glob("$dir/*.[ch]"); |
1741 | @files = glob($files[$i]); |
1742 | |
a0d0e21e |
1743 | =head2 Constant Folding |
1744 | |
1745 | Like C, Perl does a certain amount of expression evaluation at |
19799a22 |
1746 | compile time whenever it determines that all arguments to an |
a0d0e21e |
1747 | operator are static and have no side effects. In particular, string |
1748 | concatenation happens at compile time between literals that don't do |
19799a22 |
1749 | variable substitution. Backslash interpolation also happens at |
a0d0e21e |
1750 | compile time. You can say |
1751 | |
1752 | 'Now is the time for all' . "\n" . |
1753 | 'good men to come to.' |
1754 | |
54310121 |
1755 | and this all reduces to one string internally. Likewise, if |
a0d0e21e |
1756 | you say |
1757 | |
1758 | foreach $file (@filenames) { |
5a964f20 |
1759 | if (-s $file > 5 + 100 * 2**16) { } |
54310121 |
1760 | } |
a0d0e21e |
1761 | |
19799a22 |
1762 | the compiler will precompute the number which that expression |
1763 | represents so that the interpreter won't have to. |
a0d0e21e |
1764 | |
2c268ad5 |
1765 | =head2 Bitwise String Operators |
1766 | |
1767 | Bitstrings of any size may be manipulated by the bitwise operators |
1768 | (C<~ | & ^>). |
1769 | |
19799a22 |
1770 | If the operands to a binary bitwise op are strings of different |
1771 | sizes, B<|> and B<^> ops act as though the shorter operand had |
1772 | additional zero bits on the right, while the B<&> op acts as though |
1773 | the longer operand were truncated to the length of the shorter. |
1774 | The granularity for such extension or truncation is one or more |
1775 | bytes. |
2c268ad5 |
1776 | |
1777 | # ASCII-based examples |
1778 | print "j p \n" ^ " a h"; # prints "JAPH\n" |
1779 | print "JA" | " ph\n"; # prints "japh\n" |
1780 | print "japh\nJunk" & '_____'; # prints "JAPH\n"; |
1781 | print 'p N$' ^ " E<H\n"; # prints "Perl\n"; |
1782 | |
19799a22 |
1783 | If you are intending to manipulate bitstrings, be certain that |
2c268ad5 |
1784 | you're supplying bitstrings: If an operand is a number, that will imply |
19799a22 |
1785 | a B<numeric> bitwise operation. You may explicitly show which type of |
2c268ad5 |
1786 | operation you intend by using C<""> or C<0+>, as in the examples below. |
1787 | |
1788 | $foo = 150 | 105 ; # yields 255 (0x96 | 0x69 is 0xFF) |
1789 | $foo = '150' | 105 ; # yields 255 |
1790 | $foo = 150 | '105'; # yields 255 |
1791 | $foo = '150' | '105'; # yields string '155' (under ASCII) |
1792 | |
1793 | $baz = 0+$foo & 0+$bar; # both ops explicitly numeric |
1794 | $biz = "$foo" ^ "$bar"; # both ops explicitly stringy |
a0d0e21e |
1795 | |
1ae175c8 |
1796 | See L<perlfunc/vec> for information on how to manipulate individual bits |
1797 | in a bit vector. |
1798 | |
55497cff |
1799 | =head2 Integer Arithmetic |
a0d0e21e |
1800 | |
19799a22 |
1801 | By default, Perl assumes that it must do most of its arithmetic in |
a0d0e21e |
1802 | floating point. But by saying |
1803 | |
1804 | use integer; |
1805 | |
1806 | you may tell the compiler that it's okay to use integer operations |
19799a22 |
1807 | (if it feels like it) from here to the end of the enclosing BLOCK. |
1808 | An inner BLOCK may countermand this by saying |
a0d0e21e |
1809 | |
1810 | no integer; |
1811 | |
19799a22 |
1812 | which lasts until the end of that BLOCK. Note that this doesn't |
1813 | mean everything is only an integer, merely that Perl may use integer |
1814 | operations if it is so inclined. For example, even under C<use |
1815 | integer>, if you take the C<sqrt(2)>, you'll still get C<1.4142135623731> |
1816 | or so. |
1817 | |
1818 | Used on numbers, the bitwise operators ("&", "|", "^", "~", "<<", |
1819 | and ">>") always produce integral results. (But see also L<Bitwise |
1820 | String Operators>.) However, C<use integer> still has meaning for |
1821 | them. By default, their results are interpreted as unsigned integers, but |
1822 | if C<use integer> is in effect, their results are interpreted |
1823 | as signed integers. For example, C<~0> usually evaluates to a large |
1824 | integral value. However, C<use integer; ~0> is C<-1> on twos-complement |
1825 | machines. |
68dc0745 |
1826 | |
1827 | =head2 Floating-point Arithmetic |
1828 | |
1829 | While C<use integer> provides integer-only arithmetic, there is no |
19799a22 |
1830 | analogous mechanism to provide automatic rounding or truncation to a |
1831 | certain number of decimal places. For rounding to a certain number |
1832 | of digits, sprintf() or printf() is usually the easiest route. |
1833 | See L<perlfaq4>. |
68dc0745 |
1834 | |
5a964f20 |
1835 | Floating-point numbers are only approximations to what a mathematician |
1836 | would call real numbers. There are infinitely more reals than floats, |
1837 | so some corners must be cut. For example: |
1838 | |
1839 | printf "%.20g\n", 123456789123456789; |
1840 | # produces 123456789123456784 |
1841 | |
1842 | Testing for exact equality of floating-point equality or inequality is |
1843 | not a good idea. Here's a (relatively expensive) work-around to compare |
1844 | whether two floating-point numbers are equal to a particular number of |
1845 | decimal places. See Knuth, volume II, for a more robust treatment of |
1846 | this topic. |
1847 | |
1848 | sub fp_equal { |
1849 | my ($X, $Y, $POINTS) = @_; |
1850 | my ($tX, $tY); |
1851 | $tX = sprintf("%.${POINTS}g", $X); |
1852 | $tY = sprintf("%.${POINTS}g", $Y); |
1853 | return $tX eq $tY; |
1854 | } |
1855 | |
68dc0745 |
1856 | The POSIX module (part of the standard perl distribution) implements |
19799a22 |
1857 | ceil(), floor(), and other mathematical and trigonometric functions. |
1858 | The Math::Complex module (part of the standard perl distribution) |
1859 | defines mathematical functions that work on both the reals and the |
1860 | imaginary numbers. Math::Complex not as efficient as POSIX, but |
68dc0745 |
1861 | POSIX can't work with complex numbers. |
1862 | |
1863 | Rounding in financial applications can have serious implications, and |
1864 | the rounding method used should be specified precisely. In these |
1865 | cases, it probably pays not to trust whichever system rounding is |
1866 | being used by Perl, but to instead implement the rounding function you |
1867 | need yourself. |
5a964f20 |
1868 | |
1869 | =head2 Bigger Numbers |
1870 | |
1871 | The standard Math::BigInt and Math::BigFloat modules provide |
19799a22 |
1872 | variable-precision arithmetic and overloaded operators, although |
1873 | they're currently pretty slow. At the cost of some space and |
1874 | considerable speed, they avoid the normal pitfalls associated with |
1875 | limited-precision representations. |
5a964f20 |
1876 | |
1877 | use Math::BigInt; |
1878 | $x = Math::BigInt->new('123456789123456789'); |
1879 | print $x * $x; |
1880 | |
1881 | # prints +15241578780673678515622620750190521 |
19799a22 |
1882 | |
1883 | The non-standard modules SSLeay::BN and Math::Pari provide |
1884 | equivalent functionality (and much more) with a substantial |
1885 | performance savings. |