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, |
8 | listed from highest precedence to lowest. Note that all operators |
9 | borrowed from C keep the same precedence relationship with each other, |
10 | even where C's precedence is slightly screwy. (This makes learning |
54310121 |
11 | Perl easier for C folks.) With very few exceptions, these all |
c07a80fd |
12 | operate on scalar 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 | |
54310121 |
47 | A TERM has the highest precedence in Perl. They includes 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 | |
67 | the commas on the right of the sort are evaluated before the sort, but |
68 | the commas on the left are evaluated after. In other words, list |
69 | operators tend to gobble up all the arguments that follow them, and |
70 | then act like a simple TERM with regard to the preceding expression. |
5f05dabc |
71 | Note that you have to 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 | |
98 | Just as in C and C++, "C<-E<gt>>" is an infix dereference operator. If the |
99 | right side is either a C<[...]> or C<{...}> subscript, then the left side |
100 | must be either a hard or symbolic reference to an array or hash (or |
101 | a location capable of holding a hard reference, if it's an lvalue (assignable)). |
102 | See L<perlref>. |
103 | |
104 | Otherwise, the right side is a method name or a simple scalar variable |
105 | containing the method name, and the left side must either be an object |
106 | (a blessed reference) or a class name (that is, a package name). |
107 | See L<perlobj>. |
108 | |
5f05dabc |
109 | =head2 Auto-increment and Auto-decrement |
a0d0e21e |
110 | |
111 | "++" and "--" work as in C. That is, if placed before a variable, they |
112 | increment or decrement the variable before returning the value, and if |
113 | placed after, increment or decrement the variable after returning the value. |
114 | |
54310121 |
115 | The auto-increment operator has a little extra builtin magic to it. If |
a0d0e21e |
116 | you increment a variable that is numeric, or that has ever been used in |
117 | a numeric context, you get a normal increment. If, however, the |
5f05dabc |
118 | variable has been used in only string contexts since it was set, and |
5a964f20 |
119 | has a value that is not the empty string and matches the pattern |
a0d0e21e |
120 | C</^[a-zA-Z]*[0-9]*$/>, the increment is done as a string, preserving each |
121 | character within its range, with carry: |
122 | |
123 | print ++($foo = '99'); # prints '100' |
124 | print ++($foo = 'a0'); # prints 'a1' |
125 | print ++($foo = 'Az'); # prints 'Ba' |
126 | print ++($foo = 'zz'); # prints 'aaa' |
127 | |
5f05dabc |
128 | The auto-decrement operator is not magical. |
a0d0e21e |
129 | |
130 | =head2 Exponentiation |
131 | |
132 | Binary "**" is the exponentiation operator. Note that it binds even more |
cb1a09d0 |
133 | tightly than unary minus, so -2**4 is -(2**4), not (-2)**4. (This is |
134 | implemented using C's pow(3) function, which actually works on doubles |
135 | internally.) |
a0d0e21e |
136 | |
137 | =head2 Symbolic Unary Operators |
138 | |
5f05dabc |
139 | Unary "!" performs logical negation, i.e., "not". See also C<not> for a lower |
a0d0e21e |
140 | precedence version of this. |
141 | |
142 | Unary "-" performs arithmetic negation if the operand is numeric. If |
143 | the operand is an identifier, a string consisting of a minus sign |
144 | concatenated with the identifier is returned. Otherwise, if the string |
145 | starts with a plus or minus, a string starting with the opposite sign |
146 | is returned. One effect of these rules is that C<-bareword> is equivalent |
147 | to C<"-bareword">. |
148 | |
5a964f20 |
149 | Unary "~" performs bitwise negation, i.e., 1's complement. For example, |
150 | C<0666 &~ 027> is 0640. (See also L<Integer Arithmetic> and L<Bitwise |
151 | String Operators>.) |
a0d0e21e |
152 | |
153 | Unary "+" has no effect whatsoever, even on strings. It is useful |
154 | syntactically for separating a function name from a parenthesized expression |
155 | that would otherwise be interpreted as the complete list of function |
5ba421f6 |
156 | arguments. (See examples above under L<Terms and List Operators (Leftward)>.) |
a0d0e21e |
157 | |
158 | Unary "\" creates a reference to whatever follows it. See L<perlref>. |
159 | Do not confuse this behavior with the behavior of backslash within a |
160 | string, although both forms do convey the notion of protecting the next |
161 | thing from interpretation. |
162 | |
163 | =head2 Binding Operators |
164 | |
c07a80fd |
165 | Binary "=~" binds a scalar expression to a pattern match. Certain operations |
cb1a09d0 |
166 | search or modify the string $_ by default. This operator makes that kind |
167 | of operation work on some other string. The right argument is a search |
2c268ad5 |
168 | pattern, substitution, or transliteration. The left argument is what is |
169 | supposed to be searched, substituted, or transliterated instead of the default |
cb1a09d0 |
170 | $_. The return value indicates the success of the operation. (If the |
171 | right argument is an expression rather than a search pattern, |
2c268ad5 |
172 | substitution, or transliteration, it is interpreted as a search pattern at run |
aa689395 |
173 | time. This can be is less efficient than an explicit search, because the |
174 | pattern must be compiled every time the expression is evaluated. |
a0d0e21e |
175 | |
176 | Binary "!~" is just like "=~" except the return value is negated in |
177 | the logical sense. |
178 | |
179 | =head2 Multiplicative Operators |
180 | |
181 | Binary "*" multiplies two numbers. |
182 | |
183 | Binary "/" divides two numbers. |
184 | |
54310121 |
185 | Binary "%" computes the modulus of two numbers. Given integer |
186 | operands C<$a> and C<$b>: If C<$b> is positive, then C<$a % $b> is |
187 | C<$a> minus the largest multiple of C<$b> that is not greater than |
188 | C<$a>. If C<$b> is negative, then C<$a % $b> is C<$a> minus the |
189 | smallest multiple of C<$b> that is not less than C<$a> (i.e. the |
6bb4e6d4 |
190 | result will be less than or equal to zero). |
5a964f20 |
191 | Note than when C<use integer> is in scope, "%" give you direct access |
55d729e4 |
192 | to the modulus operator as implemented by your C compiler. This |
193 | operator is not as well defined for negative operands, but it will |
194 | execute faster. |
195 | |
5a964f20 |
196 | Binary "x" is the repetition operator. In scalar context, it |
a0d0e21e |
197 | returns a string consisting of the left operand repeated the number of |
5a964f20 |
198 | times specified by the right operand. In list context, if the left |
5f05dabc |
199 | operand is a list in parentheses, it repeats the list. |
a0d0e21e |
200 | |
201 | print '-' x 80; # print row of dashes |
202 | |
203 | print "\t" x ($tab/8), ' ' x ($tab%8); # tab over |
204 | |
205 | @ones = (1) x 80; # a list of 80 1's |
206 | @ones = (5) x @ones; # set all elements to 5 |
207 | |
208 | |
209 | =head2 Additive Operators |
210 | |
211 | Binary "+" returns the sum of two numbers. |
212 | |
213 | Binary "-" returns the difference of two numbers. |
214 | |
215 | Binary "." concatenates two strings. |
216 | |
217 | =head2 Shift Operators |
218 | |
55497cff |
219 | Binary "<<" returns the value of its left argument shifted left by the |
220 | number of bits specified by the right argument. Arguments should be |
221 | integers. (See also L<Integer Arithmetic>.) |
a0d0e21e |
222 | |
55497cff |
223 | Binary ">>" returns the value of its left argument shifted right by |
224 | the number of bits specified by the right argument. Arguments should |
225 | be integers. (See also L<Integer Arithmetic>.) |
a0d0e21e |
226 | |
227 | =head2 Named Unary Operators |
228 | |
229 | The various named unary operators are treated as functions with one |
230 | argument, with optional parentheses. These include the filetest |
231 | operators, like C<-f>, C<-M>, etc. See L<perlfunc>. |
232 | |
233 | If any list operator (print(), etc.) or any unary operator (chdir(), etc.) |
234 | is followed by a left parenthesis as the next token, the operator and |
235 | arguments within parentheses are taken to be of highest precedence, |
236 | just like a normal function call. Examples: |
237 | |
238 | chdir $foo || die; # (chdir $foo) || die |
239 | chdir($foo) || die; # (chdir $foo) || die |
240 | chdir ($foo) || die; # (chdir $foo) || die |
241 | chdir +($foo) || die; # (chdir $foo) || die |
242 | |
243 | but, because * is higher precedence than ||: |
244 | |
245 | chdir $foo * 20; # chdir ($foo * 20) |
246 | chdir($foo) * 20; # (chdir $foo) * 20 |
247 | chdir ($foo) * 20; # (chdir $foo) * 20 |
248 | chdir +($foo) * 20; # chdir ($foo * 20) |
249 | |
250 | rand 10 * 20; # rand (10 * 20) |
251 | rand(10) * 20; # (rand 10) * 20 |
252 | rand (10) * 20; # (rand 10) * 20 |
253 | rand +(10) * 20; # rand (10 * 20) |
254 | |
5ba421f6 |
255 | See also L<"Terms and List Operators (Leftward)">. |
a0d0e21e |
256 | |
257 | =head2 Relational Operators |
258 | |
6ee5d4e7 |
259 | Binary "E<lt>" returns true if the left argument is numerically less than |
a0d0e21e |
260 | the right argument. |
261 | |
6ee5d4e7 |
262 | Binary "E<gt>" returns true if the left argument is numerically greater |
a0d0e21e |
263 | than the right argument. |
264 | |
6ee5d4e7 |
265 | Binary "E<lt>=" returns true if the left argument is numerically less than |
a0d0e21e |
266 | or equal to the right argument. |
267 | |
6ee5d4e7 |
268 | Binary "E<gt>=" returns true if the left argument is numerically greater |
a0d0e21e |
269 | than or equal to the right argument. |
270 | |
271 | Binary "lt" returns true if the left argument is stringwise less than |
272 | the right argument. |
273 | |
274 | Binary "gt" returns true if the left argument is stringwise greater |
275 | than the right argument. |
276 | |
277 | Binary "le" returns true if the left argument is stringwise less than |
278 | or equal to the right argument. |
279 | |
280 | Binary "ge" returns true if the left argument is stringwise greater |
281 | than or equal to the right argument. |
282 | |
283 | =head2 Equality Operators |
284 | |
285 | Binary "==" returns true if the left argument is numerically equal to |
286 | the right argument. |
287 | |
288 | Binary "!=" returns true if the left argument is numerically not equal |
289 | to the right argument. |
290 | |
6ee5d4e7 |
291 | Binary "E<lt>=E<gt>" returns -1, 0, or 1 depending on whether the left |
292 | argument is numerically less than, equal to, or greater than the right |
293 | argument. |
a0d0e21e |
294 | |
295 | Binary "eq" returns true if the left argument is stringwise equal to |
296 | the right argument. |
297 | |
298 | Binary "ne" returns true if the left argument is stringwise not equal |
299 | to the right argument. |
300 | |
301 | Binary "cmp" returns -1, 0, or 1 depending on whether the left argument is stringwise |
302 | less than, equal to, or greater than the right argument. |
303 | |
a034a98d |
304 | "lt", "le", "ge", "gt" and "cmp" use the collation (sort) order specified |
305 | by the current locale if C<use locale> is in effect. See L<perllocale>. |
306 | |
a0d0e21e |
307 | =head2 Bitwise And |
308 | |
309 | Binary "&" returns its operators ANDed together bit by bit. |
2c268ad5 |
310 | (See also L<Integer Arithmetic> and L<Bitwise String Operators>.) |
a0d0e21e |
311 | |
312 | =head2 Bitwise Or and Exclusive Or |
313 | |
314 | Binary "|" returns its operators ORed together bit by bit. |
2c268ad5 |
315 | (See also L<Integer Arithmetic> and L<Bitwise String Operators>.) |
a0d0e21e |
316 | |
317 | Binary "^" returns its operators XORed together bit by bit. |
2c268ad5 |
318 | (See also L<Integer Arithmetic> and L<Bitwise String Operators>.) |
a0d0e21e |
319 | |
320 | =head2 C-style Logical And |
321 | |
322 | Binary "&&" performs a short-circuit logical AND operation. That is, |
323 | if the left operand is false, the right operand is not even evaluated. |
324 | Scalar or list context propagates down to the right operand if it |
325 | is evaluated. |
326 | |
327 | =head2 C-style Logical Or |
328 | |
329 | Binary "||" performs a short-circuit logical OR operation. That is, |
330 | if the left operand is true, the right operand is not even evaluated. |
331 | Scalar or list context propagates down to the right operand if it |
332 | is evaluated. |
333 | |
334 | The C<||> and C<&&> operators differ from C's in that, rather than returning |
335 | 0 or 1, they return the last value evaluated. Thus, a reasonably portable |
336 | way to find out the home directory (assuming it's not "0") might be: |
337 | |
338 | $home = $ENV{'HOME'} || $ENV{'LOGDIR'} || |
339 | (getpwuid($<))[7] || die "You're homeless!\n"; |
340 | |
5a964f20 |
341 | In particular, this means that you shouldn't use this |
342 | for selecting between two aggregates for assignment: |
343 | |
344 | @a = @b || @c; # this is wrong |
345 | @a = scalar(@b) || @c; # really meant this |
346 | @a = @b ? @b : @c; # this works fine, though |
347 | |
348 | As more readable alternatives to C<&&> and C<||> when used for |
349 | control flow, Perl provides C<and> and C<or> operators (see below). |
350 | The short-circuit behavior is identical. The precedence of "and" and |
351 | "or" is much lower, however, so that you can safely use them after a |
352 | list operator without the need for parentheses: |
a0d0e21e |
353 | |
354 | unlink "alpha", "beta", "gamma" |
355 | or gripe(), next LINE; |
356 | |
357 | With the C-style operators that would have been written like this: |
358 | |
359 | unlink("alpha", "beta", "gamma") |
360 | || (gripe(), next LINE); |
361 | |
5a964f20 |
362 | Use "or" for assignment is unlikely to do what you want; see below. |
363 | |
364 | =head2 Range Operators |
a0d0e21e |
365 | |
366 | Binary ".." is the range operator, which is really two different |
5a964f20 |
367 | operators depending on the context. In list context, it returns an |
a0d0e21e |
368 | array of values counting (by ones) from the left value to the right |
89ea2908 |
369 | value. This is useful for writing C<foreach (1..10)> loops and for |
370 | doing slice operations on arrays. In the current implementation, no |
371 | temporary array is created when the range operator is used as the |
372 | expression in C<foreach> loops, but older versions of Perl might burn |
373 | a lot of memory when you write something like this: |
a0d0e21e |
374 | |
375 | for (1 .. 1_000_000) { |
376 | # code |
54310121 |
377 | } |
a0d0e21e |
378 | |
5a964f20 |
379 | In scalar context, ".." returns a boolean value. The operator is |
a0d0e21e |
380 | bistable, like a flip-flop, and emulates the line-range (comma) operator |
381 | of B<sed>, B<awk>, and various editors. Each ".." operator maintains its |
382 | own boolean state. It is false as long as its left operand is false. |
383 | Once the left operand is true, the range operator stays true until the |
384 | right operand is true, I<AFTER> which the range operator becomes false |
385 | again. (It doesn't become false till the next time the range operator is |
386 | evaluated. It can test the right operand and become false on the same |
387 | evaluation it became true (as in B<awk>), but it still returns true once. |
388 | If you don't want it to test the right operand till the next evaluation |
389 | (as in B<sed>), use three dots ("...") instead of two.) The right |
390 | operand is not evaluated while the operator is in the "false" state, and |
391 | the left operand is not evaluated while the operator is in the "true" |
392 | state. The precedence is a little lower than || and &&. The value |
5a964f20 |
393 | returned is either the empty string for false, or a sequence number |
a0d0e21e |
394 | (beginning with 1) for true. The sequence number is reset for each range |
395 | encountered. The final sequence number in a range has the string "E0" |
396 | appended to it, which doesn't affect its numeric value, but gives you |
397 | something to search for if you want to exclude the endpoint. You can |
398 | exclude the beginning point by waiting for the sequence number to be |
0a528a35 |
399 | greater than 1. If either operand of scalar ".." is a constant expression, |
a0d0e21e |
400 | that operand is implicitly compared to the C<$.> variable, the current |
401 | line number. Examples: |
402 | |
403 | As a scalar operator: |
404 | |
405 | if (101 .. 200) { print; } # print 2nd hundred lines |
406 | next line if (1 .. /^$/); # skip header lines |
407 | s/^/> / if (/^$/ .. eof()); # quote body |
408 | |
5a964f20 |
409 | # parse mail messages |
410 | while (<>) { |
411 | $in_header = 1 .. /^$/; |
412 | $in_body = /^$/ .. eof(); |
413 | # do something based on those |
414 | } continue { |
415 | close ARGV if eof; # reset $. each file |
416 | } |
417 | |
a0d0e21e |
418 | As a list operator: |
419 | |
420 | for (101 .. 200) { print; } # print $_ 100 times |
3e3baf6d |
421 | @foo = @foo[0 .. $#foo]; # an expensive no-op |
a0d0e21e |
422 | @foo = @foo[$#foo-4 .. $#foo]; # slice last 5 items |
423 | |
5a964f20 |
424 | The range operator (in list context) makes use of the magical |
5f05dabc |
425 | auto-increment algorithm if the operands are strings. You |
a0d0e21e |
426 | can say |
427 | |
428 | @alphabet = ('A' .. 'Z'); |
429 | |
430 | to get all the letters of the alphabet, or |
431 | |
432 | $hexdigit = (0 .. 9, 'a' .. 'f')[$num & 15]; |
433 | |
434 | to get a hexadecimal digit, or |
435 | |
436 | @z2 = ('01' .. '31'); print $z2[$mday]; |
437 | |
438 | to get dates with leading zeros. If the final value specified is not |
439 | in the sequence that the magical increment would produce, the sequence |
440 | goes until the next value would be longer than the final value |
441 | specified. |
442 | |
443 | =head2 Conditional Operator |
444 | |
445 | Ternary "?:" is the conditional operator, just as in C. It works much |
446 | like an if-then-else. If the argument before the ? is true, the |
447 | argument before the : is returned, otherwise the argument after the : |
cb1a09d0 |
448 | is returned. For example: |
449 | |
54310121 |
450 | printf "I have %d dog%s.\n", $n, |
cb1a09d0 |
451 | ($n == 1) ? '' : "s"; |
452 | |
453 | Scalar or list context propagates downward into the 2nd |
54310121 |
454 | or 3rd argument, whichever is selected. |
cb1a09d0 |
455 | |
456 | $a = $ok ? $b : $c; # get a scalar |
457 | @a = $ok ? @b : @c; # get an array |
458 | $a = $ok ? @b : @c; # oops, that's just a count! |
459 | |
460 | The operator may be assigned to if both the 2nd and 3rd arguments are |
461 | legal lvalues (meaning that you can assign to them): |
a0d0e21e |
462 | |
463 | ($a_or_b ? $a : $b) = $c; |
464 | |
cb1a09d0 |
465 | This is not necessarily guaranteed to contribute to the readability of your program. |
a0d0e21e |
466 | |
5a964f20 |
467 | Because this operator produces an assignable result, using assignments |
468 | without parentheses will get you in trouble. For example, this: |
469 | |
470 | $a % 2 ? $a += 10 : $a += 2 |
471 | |
472 | Really means this: |
473 | |
474 | (($a % 2) ? ($a += 10) : $a) += 2 |
475 | |
476 | Rather than this: |
477 | |
478 | ($a % 2) ? ($a += 10) : ($a += 2) |
479 | |
4633a7c4 |
480 | =head2 Assignment Operators |
a0d0e21e |
481 | |
482 | "=" is the ordinary assignment operator. |
483 | |
484 | Assignment operators work as in C. That is, |
485 | |
486 | $a += 2; |
487 | |
488 | is equivalent to |
489 | |
490 | $a = $a + 2; |
491 | |
492 | although without duplicating any side effects that dereferencing the lvalue |
54310121 |
493 | might trigger, such as from tie(). Other assignment operators work similarly. |
494 | The following are recognized: |
a0d0e21e |
495 | |
496 | **= += *= &= <<= &&= |
497 | -= /= |= >>= ||= |
498 | .= %= ^= |
499 | x= |
500 | |
501 | Note that while these are grouped by family, they all have the precedence |
502 | of assignment. |
503 | |
504 | Unlike in C, the assignment operator produces a valid lvalue. Modifying |
505 | an assignment is equivalent to doing the assignment and then modifying |
506 | the variable that was assigned to. This is useful for modifying |
507 | a copy of something, like this: |
508 | |
509 | ($tmp = $global) =~ tr [A-Z] [a-z]; |
510 | |
511 | Likewise, |
512 | |
513 | ($a += 2) *= 3; |
514 | |
515 | is equivalent to |
516 | |
517 | $a += 2; |
518 | $a *= 3; |
519 | |
748a9306 |
520 | =head2 Comma Operator |
a0d0e21e |
521 | |
5a964f20 |
522 | Binary "," is the comma operator. In scalar context it evaluates |
a0d0e21e |
523 | its left argument, throws that value away, then evaluates its right |
524 | argument and returns that value. This is just like C's comma operator. |
525 | |
5a964f20 |
526 | In list context, it's just the list argument separator, and inserts |
a0d0e21e |
527 | both its arguments into the list. |
528 | |
6ee5d4e7 |
529 | The =E<gt> digraph is mostly just a synonym for the comma operator. It's useful for |
cb1a09d0 |
530 | documenting arguments that come in pairs. As of release 5.001, it also forces |
4633a7c4 |
531 | any word to the left of it to be interpreted as a string. |
748a9306 |
532 | |
a0d0e21e |
533 | =head2 List Operators (Rightward) |
534 | |
535 | On the right side of a list operator, it has very low precedence, |
536 | such that it controls all comma-separated expressions found there. |
537 | The only operators with lower precedence are the logical operators |
538 | "and", "or", and "not", which may be used to evaluate calls to list |
539 | operators without the need for extra parentheses: |
540 | |
541 | open HANDLE, "filename" |
542 | or die "Can't open: $!\n"; |
543 | |
5ba421f6 |
544 | See also discussion of list operators in L<Terms and List Operators (Leftward)>. |
a0d0e21e |
545 | |
546 | =head2 Logical Not |
547 | |
548 | Unary "not" returns the logical negation of the expression to its right. |
549 | It's the equivalent of "!" except for the very low precedence. |
550 | |
551 | =head2 Logical And |
552 | |
553 | Binary "and" returns the logical conjunction of the two surrounding |
554 | expressions. It's equivalent to && except for the very low |
5f05dabc |
555 | precedence. This means that it short-circuits: i.e., the right |
a0d0e21e |
556 | expression is evaluated only if the left expression is true. |
557 | |
558 | =head2 Logical or and Exclusive Or |
559 | |
560 | Binary "or" returns the logical disjunction of the two surrounding |
5a964f20 |
561 | expressions. It's equivalent to || except for the very low precedence. |
562 | This makes it useful for control flow |
563 | |
564 | print FH $data or die "Can't write to FH: $!"; |
565 | |
566 | This means that it short-circuits: i.e., the right expression is evaluated |
567 | only if the left expression is false. Due to its precedence, you should |
568 | probably avoid using this for assignment, only for control flow. |
569 | |
570 | $a = $b or $c; # bug: this is wrong |
571 | ($a = $b) or $c; # really means this |
572 | $a = $b || $c; # better written this way |
573 | |
574 | However, when it's a list context assignment and you're trying to use |
575 | "||" for control flow, you probably need "or" so that the assignment |
576 | takes higher precedence. |
577 | |
578 | @info = stat($file) || die; # oops, scalar sense of stat! |
579 | @info = stat($file) or die; # better, now @info gets its due |
580 | |
581 | Then again, you could always use parentheses. |
a0d0e21e |
582 | |
583 | Binary "xor" returns the exclusive-OR of the two surrounding expressions. |
584 | It cannot short circuit, of course. |
585 | |
586 | =head2 C Operators Missing From Perl |
587 | |
588 | Here is what C has that Perl doesn't: |
589 | |
590 | =over 8 |
591 | |
592 | =item unary & |
593 | |
594 | Address-of operator. (But see the "\" operator for taking a reference.) |
595 | |
596 | =item unary * |
597 | |
54310121 |
598 | Dereference-address operator. (Perl's prefix dereferencing |
a0d0e21e |
599 | operators are typed: $, @, %, and &.) |
600 | |
601 | =item (TYPE) |
602 | |
54310121 |
603 | Type casting operator. |
a0d0e21e |
604 | |
605 | =back |
606 | |
5f05dabc |
607 | =head2 Quote and Quote-like Operators |
a0d0e21e |
608 | |
609 | While we usually think of quotes as literal values, in Perl they |
610 | function as operators, providing various kinds of interpolating and |
611 | pattern matching capabilities. Perl provides customary quote characters |
612 | for these behaviors, but also provides a way for you to choose your |
613 | quote character for any of them. In the following table, a C<{}> represents |
614 | any pair of delimiters you choose. Non-bracketing delimiters use |
54310121 |
615 | the same character fore and aft, but the 4 sorts of brackets |
a0d0e21e |
616 | (round, angle, square, curly) will all nest. |
617 | |
2c268ad5 |
618 | Customary Generic Meaning Interpolates |
619 | '' q{} Literal no |
620 | "" qq{} Literal yes |
01ae956f |
621 | `` qx{} Command yes (unless '' is delimiter) |
2c268ad5 |
622 | qw{} Word list no |
623 | // m{} Pattern match yes |
eec2d3df |
624 | qr{} Pattern yes |
2c268ad5 |
625 | s{}{} Substitution yes |
626 | tr{}{} Transliteration no (but see below) |
a0d0e21e |
627 | |
fb73857a |
628 | Note that there can be whitespace between the operator and the quoting |
629 | characters, except when C<#> is being used as the quoting character. |
a3cb178b |
630 | C<q#foo#> is parsed as being the string C<foo>, while C<q #foo#> is the |
fb73857a |
631 | operator C<q> followed by a comment. Its argument will be taken from the |
632 | next line. This allows you to write: |
633 | |
634 | s {foo} # Replace foo |
635 | {bar} # with bar. |
636 | |
2c268ad5 |
637 | For constructs that do interpolation, variables beginning with "C<$>" |
638 | or "C<@>" are interpolated, as are the following sequences. Within |
639 | a transliteration, the first ten of these sequences may be used. |
a0d0e21e |
640 | |
6ee5d4e7 |
641 | \t tab (HT, TAB) |
5a964f20 |
642 | \n newline (NL) |
6ee5d4e7 |
643 | \r return (CR) |
644 | \f form feed (FF) |
645 | \b backspace (BS) |
646 | \a alarm (bell) (BEL) |
647 | \e escape (ESC) |
a0d0e21e |
648 | \033 octal char |
649 | \x1b hex char |
650 | \c[ control char |
2c268ad5 |
651 | |
a0d0e21e |
652 | \l lowercase next char |
653 | \u uppercase next char |
654 | \L lowercase till \E |
655 | \U uppercase till \E |
656 | \E end case modification |
1d2dff63 |
657 | \Q quote non-word characters till \E |
a0d0e21e |
658 | |
a034a98d |
659 | If C<use locale> is in effect, the case map used by C<\l>, C<\L>, C<\u> |
7b8d334a |
660 | and C<\U> is taken from the current locale. See L<perllocale>. |
a034a98d |
661 | |
5a964f20 |
662 | All systems use the virtual C<"\n"> to represent a line terminator, |
663 | called a "newline". There is no such thing as an unvarying, physical |
664 | newline character. It is an illusion that the operating system, |
665 | device drivers, C libraries, and Perl all conspire to preserve. Not all |
666 | systems read C<"\r"> as ASCII CR and C<"\n"> as ASCII LF. For example, |
667 | on a Mac, these are reversed, and on systems without line terminator, |
668 | printing C<"\n"> may emit no actual data. In general, use C<"\n"> when |
669 | you mean a "newline" for your system, but use the literal ASCII when you |
670 | need an exact character. For example, most networking protocols expect |
671 | and prefer a CR+LF (C<"\012\015"> or C<"\cJ\cM">) for line terminators, |
672 | and although they often accept just C<"\012">, they seldom tolerate just |
673 | C<"\015">. If you get in the habit of using C<"\n"> for networking, |
674 | you may be burned some day. |
675 | |
1d2dff63 |
676 | You cannot include a literal C<$> or C<@> within a C<\Q> sequence. |
677 | An unescaped C<$> or C<@> interpolates the corresponding variable, |
678 | while escaping will cause the literal string C<\$> to be inserted. |
679 | You'll need to write something like C<m/\Quser\E\@\Qhost/>. |
680 | |
a0d0e21e |
681 | Patterns are subject to an additional level of interpretation as a |
682 | regular expression. This is done as a second pass, after variables are |
683 | interpolated, so that regular expressions may be incorporated into the |
684 | pattern from the variables. If this is not what you want, use C<\Q> to |
685 | interpolate a variable literally. |
686 | |
687 | Apart from the above, there are no multiple levels of interpolation. In |
5f05dabc |
688 | particular, contrary to the expectations of shell programmers, back-quotes |
a0d0e21e |
689 | do I<NOT> interpolate within double quotes, nor do single quotes impede |
690 | evaluation of variables when used within double quotes. |
691 | |
5f05dabc |
692 | =head2 Regexp Quote-Like Operators |
cb1a09d0 |
693 | |
5f05dabc |
694 | Here are the quote-like operators that apply to pattern |
cb1a09d0 |
695 | matching and related activities. |
696 | |
75e14d17 |
697 | Most of this section is related to use of regular expressions from Perl. |
698 | Such a use may be considered from two points of view: Perl handles a |
699 | a string and a "pattern" to RE (regular expression) engine to match, |
700 | RE engine finds (or does not find) the match, and Perl uses the findings |
701 | of RE engine for its operation, possibly asking the engine for other matches. |
702 | |
703 | RE engine has no idea what Perl is going to do with what it finds, |
704 | similarly, the rest of Perl has no idea what a particular regular expression |
705 | means to RE engine. This creates a clean separation, and in this section |
706 | we discuss matching from Perl point of view only. The other point of |
707 | view may be found in L<perlre>. |
708 | |
a0d0e21e |
709 | =over 8 |
710 | |
711 | =item ?PATTERN? |
712 | |
713 | This is just like the C</pattern/> search, except that it matches only |
714 | once between calls to the reset() operator. This is a useful |
5f05dabc |
715 | optimization when you want to see only the first occurrence of |
a0d0e21e |
716 | something in each file of a set of files, for instance. Only C<??> |
717 | patterns local to the current package are reset. |
718 | |
5a964f20 |
719 | while (<>) { |
720 | if (?^$?) { |
721 | # blank line between header and body |
722 | } |
723 | } continue { |
724 | reset if eof; # clear ?? status for next file |
725 | } |
726 | |
a0d0e21e |
727 | This usage is vaguely deprecated, and may be removed in some future |
728 | version of Perl. |
729 | |
fb73857a |
730 | =item m/PATTERN/cgimosx |
a0d0e21e |
731 | |
fb73857a |
732 | =item /PATTERN/cgimosx |
a0d0e21e |
733 | |
5a964f20 |
734 | Searches a string for a pattern match, and in scalar context returns |
a0d0e21e |
735 | true (1) or false (''). If no string is specified via the C<=~> or |
736 | C<!~> operator, the $_ string is searched. (The string specified with |
737 | C<=~> need not be an lvalue--it may be the result of an expression |
738 | evaluation, but remember the C<=~> binds rather tightly.) See also |
739 | L<perlre>. |
5a964f20 |
740 | See L<perllocale> for discussion of additional considerations that apply |
a034a98d |
741 | when C<use locale> is in effect. |
a0d0e21e |
742 | |
743 | Options are: |
744 | |
fb73857a |
745 | c Do not reset search position on a failed match when /g is in effect. |
5f05dabc |
746 | g Match globally, i.e., find all occurrences. |
a0d0e21e |
747 | i Do case-insensitive pattern matching. |
748 | m Treat string as multiple lines. |
5f05dabc |
749 | o Compile pattern only once. |
a0d0e21e |
750 | s Treat string as single line. |
751 | x Use extended regular expressions. |
752 | |
753 | If "/" is the delimiter then the initial C<m> is optional. With the C<m> |
01ae956f |
754 | you can use any pair of non-alphanumeric, non-whitespace characters |
755 | as delimiters (if single quotes are used, no interpretation is done |
756 | on the replacement string. Unlike Perl 4, Perl 5 treats backticks as normal |
757 | delimiters; the replacement text is not evaluated as a command). |
758 | This is particularly useful for matching Unix path names |
7bac28a0 |
759 | that contain "/", to avoid LTS (leaning toothpick syndrome). If "?" is |
760 | the delimiter, then the match-only-once rule of C<?PATTERN?> applies. |
a0d0e21e |
761 | |
762 | PATTERN may contain variables, which will be interpolated (and the |
763 | pattern recompiled) every time the pattern search is evaluated. (Note |
764 | that C<$)> and C<$|> might not be interpolated because they look like |
765 | end-of-string tests.) If you want such a pattern to be compiled only |
766 | once, add a C</o> after the trailing delimiter. This avoids expensive |
767 | run-time recompilations, and is useful when the value you are |
768 | interpolating won't change over the life of the script. However, mentioning |
769 | C</o> constitutes a promise that you won't change the variables in the pattern. |
770 | If you change them, Perl won't even notice. |
771 | |
5a964f20 |
772 | If the PATTERN evaluates to the empty string, the last |
773 | I<successfully> matched regular expression is used instead. |
a0d0e21e |
774 | |
a2008d6d |
775 | If the C</g> option is not used, C<m//> in a list context returns a |
a0d0e21e |
776 | list consisting of the subexpressions matched by the parentheses in the |
f7e33566 |
777 | pattern, i.e., (C<$1>, C<$2>, C<$3>...). (Note that here C<$1> etc. are |
778 | also set, and that this differs from Perl 4's behavior.) When there are |
779 | no parentheses in the pattern, the return value is the list C<(1)> for |
780 | success. With or without parentheses, an empty list is returned upon |
781 | failure. |
a0d0e21e |
782 | |
783 | Examples: |
784 | |
785 | open(TTY, '/dev/tty'); |
786 | <TTY> =~ /^y/i && foo(); # do foo if desired |
787 | |
788 | if (/Version: *([0-9.]*)/) { $version = $1; } |
789 | |
790 | next if m#^/usr/spool/uucp#; |
791 | |
792 | # poor man's grep |
793 | $arg = shift; |
794 | while (<>) { |
795 | print if /$arg/o; # compile only once |
796 | } |
797 | |
798 | if (($F1, $F2, $Etc) = ($foo =~ /^(\S+)\s+(\S+)\s*(.*)/)) |
799 | |
800 | This last example splits $foo into the first two words and the |
5f05dabc |
801 | remainder of the line, and assigns those three fields to $F1, $F2, and |
802 | $Etc. The conditional is true if any variables were assigned, i.e., if |
a0d0e21e |
803 | the pattern matched. |
804 | |
805 | The C</g> modifier specifies global pattern matching--that is, matching |
806 | as many times as possible within the string. How it behaves depends on |
5a964f20 |
807 | the context. In list context, it returns a list of all the |
a0d0e21e |
808 | substrings matched by all the parentheses in the regular expression. |
809 | If there are no parentheses, it returns a list of all the matched |
810 | strings, as if there were parentheses around the whole pattern. |
811 | |
7e86de3e |
812 | In scalar context, each execution of C<m//g> finds the next match, |
813 | returning TRUE if it matches, and FALSE if there is no further match. |
814 | The position after the last match can be read or set using the pos() |
815 | function; see L<perlfunc/pos>. A failed match normally resets the |
816 | search position to the beginning of the string, but you can avoid that |
817 | by adding the C</c> modifier (e.g. C<m//gc>). Modifying the target |
818 | string also resets the search position. |
c90c0ff4 |
819 | |
820 | You can intermix C<m//g> matches with C<m/\G.../g>, where C<\G> is a |
821 | zero-width assertion that matches the exact position where the previous |
822 | C<m//g>, if any, left off. The C<\G> assertion is not supported without |
823 | the C</g> modifier; currently, without C</g>, C<\G> behaves just like |
824 | C<\A>, but that's accidental and may change in the future. |
825 | |
826 | Examples: |
a0d0e21e |
827 | |
828 | # list context |
829 | ($one,$five,$fifteen) = (`uptime` =~ /(\d+\.\d+)/g); |
830 | |
831 | # scalar context |
5f05dabc |
832 | $/ = ""; $* = 1; # $* deprecated in modern perls |
54310121 |
833 | while (defined($paragraph = <>)) { |
a0d0e21e |
834 | while ($paragraph =~ /[a-z]['")]*[.!?]+['")]*\s/g) { |
835 | $sentences++; |
836 | } |
837 | } |
838 | print "$sentences\n"; |
839 | |
c90c0ff4 |
840 | # using m//gc with \G |
137443ea |
841 | $_ = "ppooqppqq"; |
44a8e56a |
842 | while ($i++ < 2) { |
843 | print "1: '"; |
c90c0ff4 |
844 | print $1 while /(o)/gc; print "', pos=", pos, "\n"; |
44a8e56a |
845 | print "2: '"; |
c90c0ff4 |
846 | print $1 if /\G(q)/gc; print "', pos=", pos, "\n"; |
44a8e56a |
847 | print "3: '"; |
c90c0ff4 |
848 | print $1 while /(p)/gc; print "', pos=", pos, "\n"; |
44a8e56a |
849 | } |
850 | |
851 | The last example should print: |
852 | |
853 | 1: 'oo', pos=4 |
137443ea |
854 | 2: 'q', pos=5 |
44a8e56a |
855 | 3: 'pp', pos=7 |
856 | 1: '', pos=7 |
137443ea |
857 | 2: 'q', pos=8 |
858 | 3: '', pos=8 |
44a8e56a |
859 | |
c90c0ff4 |
860 | A useful idiom for C<lex>-like scanners is C</\G.../gc>. You can |
e7ea3e70 |
861 | combine several regexps like this to process a string part-by-part, |
c90c0ff4 |
862 | doing different actions depending on which regexp matched. Each |
863 | regexp tries to match where the previous one leaves off. |
e7ea3e70 |
864 | |
3fe9a6f1 |
865 | $_ = <<'EOL'; |
e7ea3e70 |
866 | $url = new URI::URL "http://www/"; die if $url eq "xXx"; |
3fe9a6f1 |
867 | EOL |
868 | LOOP: |
e7ea3e70 |
869 | { |
c90c0ff4 |
870 | print(" digits"), redo LOOP if /\G\d+\b[,.;]?\s*/gc; |
871 | print(" lowercase"), redo LOOP if /\G[a-z]+\b[,.;]?\s*/gc; |
872 | print(" UPPERCASE"), redo LOOP if /\G[A-Z]+\b[,.;]?\s*/gc; |
873 | print(" Capitalized"), redo LOOP if /\G[A-Z][a-z]+\b[,.;]?\s*/gc; |
874 | print(" MiXeD"), redo LOOP if /\G[A-Za-z]+\b[,.;]?\s*/gc; |
875 | print(" alphanumeric"), redo LOOP if /\G[A-Za-z0-9]+\b[,.;]?\s*/gc; |
876 | print(" line-noise"), redo LOOP if /\G[^A-Za-z0-9]+/gc; |
e7ea3e70 |
877 | print ". That's all!\n"; |
878 | } |
879 | |
880 | Here is the output (split into several lines): |
881 | |
882 | line-noise lowercase line-noise lowercase UPPERCASE line-noise |
883 | UPPERCASE line-noise lowercase line-noise lowercase line-noise |
884 | lowercase lowercase line-noise lowercase lowercase line-noise |
885 | MiXeD line-noise. That's all! |
44a8e56a |
886 | |
a0d0e21e |
887 | =item q/STRING/ |
888 | |
889 | =item C<'STRING'> |
890 | |
68dc0745 |
891 | A single-quoted, literal string. A backslash represents a backslash |
892 | unless followed by the delimiter or another backslash, in which case |
893 | the delimiter or backslash is interpolated. |
a0d0e21e |
894 | |
895 | $foo = q!I said, "You said, 'She said it.'"!; |
896 | $bar = q('This is it.'); |
68dc0745 |
897 | $baz = '\n'; # a two-character string |
a0d0e21e |
898 | |
899 | =item qq/STRING/ |
900 | |
901 | =item "STRING" |
902 | |
903 | A double-quoted, interpolated string. |
904 | |
905 | $_ .= qq |
906 | (*** The previous line contains the naughty word "$1".\n) |
907 | if /(tcl|rexx|python)/; # :-) |
68dc0745 |
908 | $baz = "\n"; # a one-character string |
a0d0e21e |
909 | |
eec2d3df |
910 | =item qr/STRING/imosx |
911 | |
912 | A string which is (possibly) interpolated and then compiled as a |
913 | regular expression. The result may be used as a pattern in a match |
914 | |
915 | $re = qr/$pattern/; |
0a92e3a8 |
916 | $string =~ /foo${re}bar/; # can be interpolated in other patterns |
917 | $string =~ $re; # or used standalone |
eec2d3df |
918 | |
919 | Options are: |
920 | |
921 | i Do case-insensitive pattern matching. |
922 | m Treat string as multiple lines. |
923 | o Compile pattern only once. |
924 | s Treat string as single line. |
925 | x Use extended regular expressions. |
926 | |
0a92e3a8 |
927 | The benefit from this is that the pattern is precompiled into an internal |
928 | representation, and does not need to be recompiled every time a match |
929 | is attempted. This makes it very efficient to do something like: |
eec2d3df |
930 | |
931 | foreach $pattern (@pattern_list) { |
932 | my $re = qr/$pattern/; |
933 | foreach $line (@lines) { |
934 | if($line =~ /$re/) { |
935 | do_something($line); |
936 | } |
937 | } |
938 | } |
939 | |
0a92e3a8 |
940 | See L<perlre> for additional information on valid syntax for STRING, and |
941 | for a detailed look at the semantics of regular expressions. |
942 | |
a0d0e21e |
943 | =item qx/STRING/ |
944 | |
945 | =item `STRING` |
946 | |
5a964f20 |
947 | A string which is (possibly) interpolated and then executed as a system |
948 | command with C</bin/sh> or its equivalent. Shell wildcards, pipes, |
949 | and redirections will be honored. The collected standard output of the |
950 | command is returned; standard error is unaffected. In scalar context, |
951 | it comes back as a single (potentially multi-line) string. In list |
952 | context, returns a list of lines (however you've defined lines with $/ |
953 | or $INPUT_RECORD_SEPARATOR). |
954 | |
955 | Because backticks do not affect standard error, use shell file descriptor |
956 | syntax (assuming the shell supports this) if you care to address this. |
957 | To capture a command's STDERR and STDOUT together: |
a0d0e21e |
958 | |
5a964f20 |
959 | $output = `cmd 2>&1`; |
960 | |
961 | To capture a command's STDOUT but discard its STDERR: |
962 | |
963 | $output = `cmd 2>/dev/null`; |
964 | |
965 | To capture a command's STDERR but discard its STDOUT (ordering is |
966 | important here): |
967 | |
968 | $output = `cmd 2>&1 1>/dev/null`; |
969 | |
970 | To exchange a command's STDOUT and STDERR in order to capture the STDERR |
971 | but leave its STDOUT to come out the old STDERR: |
972 | |
973 | $output = `cmd 3>&1 1>&2 2>&3 3>&-`; |
974 | |
975 | To read both a command's STDOUT and its STDERR separately, it's easiest |
976 | and safest to redirect them separately to files, and then read from those |
977 | files when the program is done: |
978 | |
979 | system("program args 1>/tmp/program.stdout 2>/tmp/program.stderr"); |
980 | |
981 | Using single-quote as a delimiter protects the command from Perl's |
982 | double-quote interpolation, passing it on to the shell instead: |
983 | |
984 | $perl_info = qx(ps $$); # that's Perl's $$ |
985 | $shell_info = qx'ps $$'; # that's the new shell's $$ |
986 | |
987 | Note that how the string gets evaluated is entirely subject to the command |
988 | interpreter on your system. On most platforms, you will have to protect |
989 | shell metacharacters if you want them treated literally. This is in |
990 | practice difficult to do, as it's unclear how to escape which characters. |
991 | See L<perlsec> for a clean and safe example of a manual fork() and exec() |
992 | to emulate backticks safely. |
a0d0e21e |
993 | |
bb32b41a |
994 | On some platforms (notably DOS-like ones), the shell may not be |
995 | capable of dealing with multiline commands, so putting newlines in |
996 | the string may not get you what you want. You may be able to evaluate |
997 | multiple commands in a single line by separating them with the command |
998 | separator character, if your shell supports that (e.g. C<;> on many Unix |
999 | shells; C<&> on the Windows NT C<cmd> shell). |
1000 | |
1001 | Beware that some command shells may place restrictions on the length |
1002 | of the command line. You must ensure your strings don't exceed this |
1003 | limit after any necessary interpolations. See the platform-specific |
1004 | release notes for more details about your particular environment. |
1005 | |
5a964f20 |
1006 | Using this operator can lead to programs that are difficult to port, |
1007 | because the shell commands called vary between systems, and may in |
1008 | fact not be present at all. As one example, the C<type> command under |
1009 | the POSIX shell is very different from the C<type> command under DOS. |
1010 | That doesn't mean you should go out of your way to avoid backticks |
1011 | when they're the right way to get something done. Perl was made to be |
1012 | a glue language, and one of the things it glues together is commands. |
1013 | Just understand what you're getting yourself into. |
bb32b41a |
1014 | |
dc848c6f |
1015 | See L<"I/O Operators"> for more discussion. |
a0d0e21e |
1016 | |
1017 | =item qw/STRING/ |
1018 | |
1019 | Returns a list of the words extracted out of STRING, using embedded |
1020 | whitespace as the word delimiters. It is exactly equivalent to |
1021 | |
1022 | split(' ', q/STRING/); |
1023 | |
5a964f20 |
1024 | This equivalency means that if used in scalar context, you'll get split's |
1025 | (unfortunate) scalar context behavior, complete with mysterious warnings. |
1026 | |
a0d0e21e |
1027 | Some frequently seen examples: |
1028 | |
1029 | use POSIX qw( setlocale localeconv ) |
1030 | @EXPORT = qw( foo bar baz ); |
1031 | |
7bac28a0 |
1032 | A common mistake is to try to separate the words with comma or to put |
5a964f20 |
1033 | comments into a multi-line C<qw>-string. For this reason the C<-w> |
7bac28a0 |
1034 | switch produce warnings if the STRING contains the "," or the "#" |
1035 | character. |
1036 | |
a0d0e21e |
1037 | =item s/PATTERN/REPLACEMENT/egimosx |
1038 | |
1039 | Searches a string for a pattern, and if found, replaces that pattern |
1040 | with the replacement text and returns the number of substitutions |
e37d713d |
1041 | made. Otherwise it returns false (specifically, the empty string). |
a0d0e21e |
1042 | |
1043 | If no string is specified via the C<=~> or C<!~> operator, the C<$_> |
1044 | variable is searched and modified. (The string specified with C<=~> must |
5a964f20 |
1045 | be scalar variable, an array element, a hash element, or an assignment |
5f05dabc |
1046 | to one of those, i.e., an lvalue.) |
a0d0e21e |
1047 | |
1048 | If the delimiter chosen is single quote, no variable interpolation is |
1049 | done on either the PATTERN or the REPLACEMENT. Otherwise, if the |
1050 | PATTERN contains a $ that looks like a variable rather than an |
1051 | end-of-string test, the variable will be interpolated into the pattern |
5f05dabc |
1052 | at run-time. If you want the pattern compiled only once the first time |
a0d0e21e |
1053 | the variable is interpolated, use the C</o> option. If the pattern |
5a964f20 |
1054 | evaluates to the empty string, the last successfully executed regular |
a0d0e21e |
1055 | expression is used instead. See L<perlre> for further explanation on these. |
5a964f20 |
1056 | See L<perllocale> for discussion of additional considerations that apply |
a034a98d |
1057 | when C<use locale> is in effect. |
a0d0e21e |
1058 | |
1059 | Options are: |
1060 | |
1061 | e Evaluate the right side as an expression. |
5f05dabc |
1062 | g Replace globally, i.e., all occurrences. |
a0d0e21e |
1063 | i Do case-insensitive pattern matching. |
1064 | m Treat string as multiple lines. |
5f05dabc |
1065 | o Compile pattern only once. |
a0d0e21e |
1066 | s Treat string as single line. |
1067 | x Use extended regular expressions. |
1068 | |
1069 | Any non-alphanumeric, non-whitespace delimiter may replace the |
1070 | slashes. If single quotes are used, no interpretation is done on the |
e37d713d |
1071 | replacement string (the C</e> modifier overrides this, however). Unlike |
54310121 |
1072 | Perl 4, Perl 5 treats backticks as normal delimiters; the replacement |
e37d713d |
1073 | text is not evaluated as a command. If the |
a0d0e21e |
1074 | PATTERN is delimited by bracketing quotes, the REPLACEMENT has its own |
5f05dabc |
1075 | pair of quotes, which may or may not be bracketing quotes, e.g., |
a0d0e21e |
1076 | C<s(foo)(bar)> or C<sE<lt>fooE<gt>/bar/>. A C</e> will cause the |
7b8d334a |
1077 | replacement portion to be interpreted as a full-fledged Perl expression |
a0d0e21e |
1078 | and eval()ed right then and there. It is, however, syntax checked at |
1079 | compile-time. |
1080 | |
1081 | Examples: |
1082 | |
1083 | s/\bgreen\b/mauve/g; # don't change wintergreen |
1084 | |
1085 | $path =~ s|/usr/bin|/usr/local/bin|; |
1086 | |
1087 | s/Login: $foo/Login: $bar/; # run-time pattern |
1088 | |
5a964f20 |
1089 | ($foo = $bar) =~ s/this/that/; # copy first, then change |
a0d0e21e |
1090 | |
5a964f20 |
1091 | $count = ($paragraph =~ s/Mister\b/Mr./g); # get change-count |
a0d0e21e |
1092 | |
1093 | $_ = 'abc123xyz'; |
1094 | s/\d+/$&*2/e; # yields 'abc246xyz' |
1095 | s/\d+/sprintf("%5d",$&)/e; # yields 'abc 246xyz' |
1096 | s/\w/$& x 2/eg; # yields 'aabbcc 224466xxyyzz' |
1097 | |
1098 | s/%(.)/$percent{$1}/g; # change percent escapes; no /e |
1099 | s/%(.)/$percent{$1} || $&/ge; # expr now, so /e |
1100 | s/^=(\w+)/&pod($1)/ge; # use function call |
1101 | |
5a964f20 |
1102 | # expand variables in $_, but dynamics only, using |
1103 | # symbolic dereferencing |
1104 | s/\$(\w+)/${$1}/g; |
1105 | |
a0d0e21e |
1106 | # /e's can even nest; this will expand |
5a964f20 |
1107 | # any embedded scalar variable (including lexicals) in $_ |
a0d0e21e |
1108 | s/(\$\w+)/$1/eeg; |
1109 | |
5a964f20 |
1110 | # Delete (most) C comments. |
a0d0e21e |
1111 | $program =~ s { |
4633a7c4 |
1112 | /\* # Match the opening delimiter. |
1113 | .*? # Match a minimal number of characters. |
1114 | \*/ # Match the closing delimiter. |
a0d0e21e |
1115 | } []gsx; |
1116 | |
5a964f20 |
1117 | s/^\s*(.*?)\s*$/$1/; # trim white space in $_, expensively |
1118 | |
1119 | for ($variable) { # trim white space in $variable, cheap |
1120 | s/^\s+//; |
1121 | s/\s+$//; |
1122 | } |
a0d0e21e |
1123 | |
1124 | s/([^ ]*) *([^ ]*)/$2 $1/; # reverse 1st two fields |
1125 | |
54310121 |
1126 | Note the use of $ instead of \ in the last example. Unlike |
5f05dabc |
1127 | B<sed>, we use the \E<lt>I<digit>E<gt> form in only the left hand side. |
6ee5d4e7 |
1128 | Anywhere else it's $E<lt>I<digit>E<gt>. |
a0d0e21e |
1129 | |
5f05dabc |
1130 | Occasionally, you can't use just a C</g> to get all the changes |
a0d0e21e |
1131 | to occur. Here are two common cases: |
1132 | |
1133 | # put commas in the right places in an integer |
1134 | 1 while s/(.*\d)(\d\d\d)/$1,$2/g; # perl4 |
1135 | 1 while s/(\d)(\d\d\d)(?!\d)/$1,$2/g; # perl5 |
1136 | |
1137 | # expand tabs to 8-column spacing |
1138 | 1 while s/\t+/' ' x (length($&)*8 - length($`)%8)/e; |
1139 | |
1140 | |
1141 | =item tr/SEARCHLIST/REPLACEMENTLIST/cds |
1142 | |
1143 | =item y/SEARCHLIST/REPLACEMENTLIST/cds |
1144 | |
2c268ad5 |
1145 | Transliterates all occurrences of the characters found in the search list |
a0d0e21e |
1146 | with the corresponding character in the replacement list. It returns |
1147 | the number of characters replaced or deleted. If no string is |
2c268ad5 |
1148 | specified via the =~ or !~ operator, the $_ string is transliterated. (The |
54310121 |
1149 | string specified with =~ must be a scalar variable, an array element, a |
1150 | hash element, or an assignment to one of those, i.e., an lvalue.) |
2c268ad5 |
1151 | A character range may be specified with a hyphen, so C<tr/A-J/0-9/> |
1152 | does the same replacement as C<tr/ACEGIBDFHJ/0246813579/>. |
54310121 |
1153 | For B<sed> devotees, C<y> is provided as a synonym for C<tr>. If the |
1154 | SEARCHLIST is delimited by bracketing quotes, the REPLACEMENTLIST has |
1155 | its own pair of quotes, which may or may not be bracketing quotes, |
2c268ad5 |
1156 | e.g., C<tr[A-Z][a-z]> or C<tr(+\-*/)/ABCD/>. |
a0d0e21e |
1157 | |
1158 | Options: |
1159 | |
1160 | c Complement the SEARCHLIST. |
1161 | d Delete found but unreplaced characters. |
1162 | s Squash duplicate replaced characters. |
1163 | |
1164 | If the C</c> modifier is specified, the SEARCHLIST character set is |
1165 | complemented. If the C</d> modifier is specified, any characters specified |
1166 | by SEARCHLIST not found in REPLACEMENTLIST are deleted. (Note |
1167 | that this is slightly more flexible than the behavior of some B<tr> |
1168 | programs, which delete anything they find in the SEARCHLIST, period.) |
1169 | If the C</s> modifier is specified, sequences of characters that were |
2c268ad5 |
1170 | transliterated to the same character are squashed down to a single instance of the |
a0d0e21e |
1171 | character. |
1172 | |
1173 | If the C</d> modifier is used, the REPLACEMENTLIST is always interpreted |
1174 | exactly as specified. Otherwise, if the REPLACEMENTLIST is shorter |
1175 | than the SEARCHLIST, the final character is replicated till it is long |
5a964f20 |
1176 | enough. If the REPLACEMENTLIST is empty, the SEARCHLIST is replicated. |
a0d0e21e |
1177 | This latter is useful for counting characters in a class or for |
1178 | squashing character sequences in a class. |
1179 | |
1180 | Examples: |
1181 | |
1182 | $ARGV[1] =~ tr/A-Z/a-z/; # canonicalize to lower case |
1183 | |
1184 | $cnt = tr/*/*/; # count the stars in $_ |
1185 | |
1186 | $cnt = $sky =~ tr/*/*/; # count the stars in $sky |
1187 | |
1188 | $cnt = tr/0-9//; # count the digits in $_ |
1189 | |
1190 | tr/a-zA-Z//s; # bookkeeper -> bokeper |
1191 | |
1192 | ($HOST = $host) =~ tr/a-z/A-Z/; |
1193 | |
1194 | tr/a-zA-Z/ /cs; # change non-alphas to single space |
1195 | |
1196 | tr [\200-\377] |
1197 | [\000-\177]; # delete 8th bit |
1198 | |
2c268ad5 |
1199 | If multiple transliterations are given for a character, only the first one is used: |
748a9306 |
1200 | |
1201 | tr/AAA/XYZ/ |
1202 | |
2c268ad5 |
1203 | will transliterate any A to X. |
748a9306 |
1204 | |
2c268ad5 |
1205 | Note that because the transliteration table is built at compile time, neither |
a0d0e21e |
1206 | the SEARCHLIST nor the REPLACEMENTLIST are subjected to double quote |
1207 | interpolation. That means that if you want to use variables, you must use |
1208 | an eval(): |
1209 | |
1210 | eval "tr/$oldlist/$newlist/"; |
1211 | die $@ if $@; |
1212 | |
1213 | eval "tr/$oldlist/$newlist/, 1" or die $@; |
1214 | |
1215 | =back |
1216 | |
75e14d17 |
1217 | =head2 Gory details of parsing quoted constructs |
1218 | |
1219 | When presented with something which may have several different |
1220 | interpretations, Perl uses the principle B<DWIM> (expanded to Do What I Mean |
1221 | - not what I wrote) to pick up the most probable interpretation of the |
1222 | source. This strategy is so successful that Perl users usually do not |
1223 | suspect ambivalence of what they write. However, time to time Perl's ideas |
1224 | differ from what the author meant. |
1225 | |
1226 | The target of this section is to clarify the Perl's way of interpreting |
1227 | quoted constructs. The most frequent reason one may have to want to know the |
1228 | details discussed in this section is hairy regular expressions. However, the |
1229 | first steps of parsing are the same for all Perl quoting operators, so here |
1230 | they are discussed together. |
1231 | |
1232 | Some of the passes discussed below are performed concurrently, but as |
1233 | far as results are the same, we consider them one-by-one. For different |
1234 | quoting constructs Perl performs different number of passes, from |
1235 | one to five, but they are always performed in the same order. |
1236 | |
1237 | =over |
1238 | |
1239 | =item Finding the end |
1240 | |
1241 | First pass is finding the end of the quoted construct, be it multichar ender |
1242 | C<"\nEOF\n"> of C<<<EOF> construct, C</> which terminates C<qq/> construct, |
7522fed5 |
1243 | C<]> which terminates C<qq[> construct, or C<E<gt>> which terminates a |
75e14d17 |
1244 | fileglob started with C<<>. |
1245 | |
1246 | When searching for multichar construct no skipping is performed. When |
1247 | searching for one-char non-matching delimiter, such as C</>, combinations |
1248 | C<\\> and C<\/> are skipped. When searching for one-char matching delimiter, |
1249 | such as C<]>, combinations C<\\>, C<\]> and C<\[> are skipped, and |
1250 | nested C<[>, C<]> are skipped as well. |
1251 | |
7522fed5 |
1252 | For 3-parts constructs, C<s///> etc. the search is repeated once more. |
75e14d17 |
1253 | |
7522fed5 |
1254 | During this search no attention is paid to the semantic of the construct, thus |
75e14d17 |
1255 | |
1256 | "$hash{"$foo/$bar"}" |
1257 | |
1258 | or |
1259 | |
1260 | m/ |
1261 | bar # This is not a comment, this slash / terminated m//! |
1262 | /x |
1263 | |
1264 | do not form legal quoted expressions. Note that since the slash which |
1265 | terminated C<m//> was followed by a C<SPACE>, this is not C<m//x>, |
1266 | thus C<#> was interpreted as a literal C<#>. |
1267 | |
1268 | =item Removal of backslashes before delimiters |
1269 | |
1270 | During the second pass the text between the starting delimiter and |
1271 | the ending delimiter is copied to a safe location, and the C<\> is |
1272 | removed from combinations consisting of C<\> and delimiter(s) (both starting |
1273 | and ending delimiter if they differ). |
1274 | |
1275 | The removal does not happen for multi-char delimiters. |
1276 | |
1277 | Note that the combination C<\\> is left as it was! |
1278 | |
1279 | Starting from this step no information about the delimiter(s) is used in the |
1280 | parsing. |
1281 | |
1282 | =item Interpolation |
1283 | |
1284 | Next step is interpolation in the obtained delimiter-independent text. |
7522fed5 |
1285 | There are four different cases. |
75e14d17 |
1286 | |
1287 | =over |
1288 | |
1289 | =item C<<<'EOF'>, C<m''>, C<s'''>, C<tr///>, C<y///> |
1290 | |
1291 | No interpolation is performed. |
1292 | |
1293 | =item C<''>, C<q//> |
1294 | |
1295 | The only interpolation is removal of C<\> from pairs C<\\>. |
1296 | |
1297 | =item C<"">, C<``>, C<qq//>, C<qx//>, C<<file*globE<gt>> |
1298 | |
1299 | C<\Q>, C<\U>, C<\u>, C<\L>, C<\l> (possibly paired with C<\E>) are converted |
1300 | to corresponding Perl constructs, thus C<"$foo\Qbaz$bar"> is converted to |
1301 | |
1302 | $foo . (quotemeta("baz" . $bar)); |
1303 | |
1304 | Other combinations of C<\> with following chars are substituted with |
1305 | appropriate expansions. |
1306 | |
1307 | Interpolated scalars and arrays are converted to C<join> and C<.> Perl |
1308 | constructs, thus C<"'@arr'"> becomes |
1309 | |
1310 | "'" . (join $", @arr) . "'"; |
1311 | |
1312 | Since all three above steps are performed simultaneously left-to-right, |
1313 | the is no way to insert a literal C<$> or C<@> inside C<\Q\E> pair: it |
1314 | cannot be protected by C<\>, since any C<\> (except in C<\E>) is |
7522fed5 |
1315 | interpreted as a literal inside C<\Q\E>, and any C<$> is |
75e14d17 |
1316 | interpreted as starting an interpolated scalar. |
1317 | |
1318 | Note also that the interpolating code needs to make decision where the |
7522fed5 |
1319 | interpolated scalar ends, say, whether C<"a $b -E<gt> {c}"> means |
75e14d17 |
1320 | |
1321 | "a " . $b . " -> {c}"; |
1322 | |
1323 | or |
1324 | |
1325 | "a " . $b -> {c}; |
1326 | |
1327 | Most the time the decision is to take the longest possible text which does |
1328 | not include spaces between components and contains matching braces/brackets. |
1329 | |
1330 | =item C<?RE?>, C</RE/>, C<m/RE/>, C<s/RE/foo/>, |
1331 | |
1332 | Processing of C<\Q>, C<\U>, C<\u>, C<\L>, C<\l> and interpolation happens |
7522fed5 |
1333 | (almost) as with C<qq//> constructs, but I<the substitution of C<\> followed by |
75e14d17 |
1334 | other chars is not performed>! Moreover, inside C<(?{BLOCK})> no processing |
1335 | is performed at all. |
1336 | |
7522fed5 |
1337 | Interpolation has several quirks: C<$|>, C<$(> and C<$)> are not interpolated, and |
75e14d17 |
1338 | constructs C<$var[SOMETHING]> are I<voted> (by several different estimators) |
7522fed5 |
1339 | to be an array element or C<$var> followed by a RE alternative. This is |
75e14d17 |
1340 | the place where the notation C<${arr[$bar]}> comes handy: C</${arr[0-9]}/> |
7522fed5 |
1341 | is interpreted as an array element C<-9>, not as a regular expression from |
1342 | variable C<$arr> followed by a digit, which is the interpretation of |
75e14d17 |
1343 | C</$arr[0-9]/>. |
1344 | |
7522fed5 |
1345 | Note that absence of processing of C<\\> creates specific restrictions on the |
1346 | post-processed text: if the delimiter is C</>, one cannot get the combination |
75e14d17 |
1347 | C<\/> into the result of this step: C</> will finish the regular expression, |
1348 | C<\/> will be stripped to C</> on the previous step, and C<\\/> will be left |
1349 | as is. Since C</> is equivalent to C<\/> inside a regular expression, this |
7522fed5 |
1350 | does not matter unless the delimiter is special character for the RE engine, as |
75e14d17 |
1351 | in C<s*foo*bar*>, C<m[foo]>, or C<?foo?>. |
1352 | |
1353 | =back |
1354 | |
1355 | This step is the last one for all the constructs except regular expressions, |
1356 | which are processed further. |
1357 | |
1358 | =item Interpolation of regular expressions |
1359 | |
1360 | All the previous steps were performed during the compilation of Perl code, |
1361 | this one happens in run time (though it may be optimized to be calculated |
1362 | at compile time if appropriate). After all the preprocessing performed |
1363 | above (and possibly after evaluation if catenation, joining, up/down-casing |
7522fed5 |
1364 | and C<quotemeta()>ing are involved) the resulting I<string> is passed to RE |
75e14d17 |
1365 | engine for compilation. |
1366 | |
1367 | Whatever happens in the RE engine is better be discussed in L<perlre>, |
1368 | but for the sake of continuity let us do it here. |
1369 | |
7522fed5 |
1370 | This is the first step where presence of the C<//x> switch is relevant. |
1371 | The RE engine scans the string left-to-right, and converts it to a finite |
75e14d17 |
1372 | automaton. |
1373 | |
1374 | Backslashed chars are either substituted by corresponding literal |
1375 | strings, or generate special nodes of the finite automaton. Characters |
7522fed5 |
1376 | which are special to the RE engine generate corresponding nodes. C<(?#...)> |
75e14d17 |
1377 | comments are ignored. All the rest is either converted to literal strings |
1378 | to match, or is ignored (as is whitespace and C<#>-style comments if |
1379 | C<//x> is present). |
1380 | |
1381 | Note that the parsing of the construct C<[...]> is performed using |
1382 | absolutely different rules than the rest of the regular expression. |
1383 | Similarly, the C<(?{...})> is only checked for matching braces. |
1384 | |
1385 | =item Optimization of regular expressions |
1386 | |
7522fed5 |
1387 | This step is listed for completeness only. Since it does not change |
75e14d17 |
1388 | semantics, details of this step are not documented and are subject |
1389 | to change. |
1390 | |
1391 | =back |
1392 | |
a0d0e21e |
1393 | =head2 I/O Operators |
1394 | |
54310121 |
1395 | There are several I/O operators you should know about. |
7b8d334a |
1396 | A string enclosed by backticks (grave accents) first undergoes |
a0d0e21e |
1397 | variable substitution just like a double quoted string. It is then |
1398 | interpreted as a command, and the output of that command is the value |
5a964f20 |
1399 | of the pseudo-literal, like in a shell. In scalar context, a single |
1400 | string consisting of all the output is returned. In list context, |
a0d0e21e |
1401 | a list of values is returned, one for each line of output. (You can |
1402 | set C<$/> to use a different line terminator.) The command is executed |
1403 | each time the pseudo-literal is evaluated. The status value of the |
1404 | command is returned in C<$?> (see L<perlvar> for the interpretation |
1405 | of C<$?>). Unlike in B<csh>, no translation is done on the return |
1406 | data--newlines remain newlines. Unlike in any of the shells, single |
1407 | quotes do not hide variable names in the command from interpretation. |
1408 | To pass a $ through to the shell you need to hide it with a backslash. |
54310121 |
1409 | The generalized form of backticks is C<qx//>. (Because backticks |
1410 | always undergo shell expansion as well, see L<perlsec> for |
cb1a09d0 |
1411 | security concerns.) |
a0d0e21e |
1412 | |
1413 | Evaluating a filehandle in angle brackets yields the next line from |
aa689395 |
1414 | that file (newline, if any, included), or C<undef> at end of file. |
1415 | Ordinarily you must assign that value to a variable, but there is one |
1416 | situation where an automatic assignment happens. I<If and ONLY if> the |
1417 | input symbol is the only thing inside the conditional of a C<while> or |
1418 | C<for(;;)> loop, the value is automatically assigned to the variable |
7b8d334a |
1419 | C<$_>. In these loop constructs, the assigned value (whether assignment |
5a964f20 |
1420 | is automatic or explicit) is then tested to see if it is defined. |
7b8d334a |
1421 | The defined test avoids problems where line has a string value |
1422 | that would be treated as false by perl e.g. "" or "0" with no trailing |
1423 | newline. (This may seem like an odd thing to you, but you'll use the |
1424 | construct in almost every Perl script you write.) Anyway, the following |
1425 | lines are equivalent to each other: |
a0d0e21e |
1426 | |
748a9306 |
1427 | while (defined($_ = <STDIN>)) { print; } |
7b8d334a |
1428 | while ($_ = <STDIN>) { print; } |
a0d0e21e |
1429 | while (<STDIN>) { print; } |
1430 | for (;<STDIN>;) { print; } |
748a9306 |
1431 | print while defined($_ = <STDIN>); |
7b8d334a |
1432 | print while ($_ = <STDIN>); |
a0d0e21e |
1433 | print while <STDIN>; |
1434 | |
7b8d334a |
1435 | and this also behaves similarly, but avoids the use of $_ : |
1436 | |
1437 | while (my $line = <STDIN>) { print $line } |
1438 | |
1439 | If you really mean such values to terminate the loop they should be |
5a964f20 |
1440 | tested for explicitly: |
7b8d334a |
1441 | |
1442 | while (($_ = <STDIN>) ne '0') { ... } |
1443 | while (<STDIN>) { last unless $_; ... } |
1444 | |
5a964f20 |
1445 | In other boolean contexts, C<E<lt>I<filehandle>E<gt>> without explicit C<defined> |
7b8d334a |
1446 | test or comparison will solicit a warning if C<-w> is in effect. |
1447 | |
5f05dabc |
1448 | The filehandles STDIN, STDOUT, and STDERR are predefined. (The |
1449 | filehandles C<stdin>, C<stdout>, and C<stderr> will also work except in |
a0d0e21e |
1450 | packages, where they would be interpreted as local identifiers rather |
1451 | than global.) Additional filehandles may be created with the open() |
cb1a09d0 |
1452 | function. See L<perlfunc/open()> for details on this. |
a0d0e21e |
1453 | |
6ee5d4e7 |
1454 | If a E<lt>FILEHANDLEE<gt> is used in a context that is looking for a list, a |
a0d0e21e |
1455 | list consisting of all the input lines is returned, one line per list |
1456 | element. It's easy to make a I<LARGE> data space this way, so use with |
1457 | care. |
1458 | |
d28ebecd |
1459 | The null filehandle E<lt>E<gt> is special and can be used to emulate the |
1460 | behavior of B<sed> and B<awk>. Input from E<lt>E<gt> comes either from |
a0d0e21e |
1461 | standard input, or from each file listed on the command line. Here's |
d28ebecd |
1462 | how it works: the first time E<lt>E<gt> is evaluated, the @ARGV array is |
5a964f20 |
1463 | checked, and if it is empty, C<$ARGV[0]> is set to "-", which when opened |
a0d0e21e |
1464 | gives you standard input. The @ARGV array is then processed as a list |
1465 | of filenames. The loop |
1466 | |
1467 | while (<>) { |
1468 | ... # code for each line |
1469 | } |
1470 | |
1471 | is equivalent to the following Perl-like pseudo code: |
1472 | |
3e3baf6d |
1473 | unshift(@ARGV, '-') unless @ARGV; |
a0d0e21e |
1474 | while ($ARGV = shift) { |
1475 | open(ARGV, $ARGV); |
1476 | while (<ARGV>) { |
1477 | ... # code for each line |
1478 | } |
1479 | } |
1480 | |
1481 | except that it isn't so cumbersome to say, and will actually work. It |
1482 | really does shift array @ARGV and put the current filename into variable |
5f05dabc |
1483 | $ARGV. It also uses filehandle I<ARGV> internally--E<lt>E<gt> is just a |
1484 | synonym for E<lt>ARGVE<gt>, which is magical. (The pseudo code above |
1485 | doesn't work because it treats E<lt>ARGVE<gt> as non-magical.) |
a0d0e21e |
1486 | |
d28ebecd |
1487 | You can modify @ARGV before the first E<lt>E<gt> as long as the array ends up |
a0d0e21e |
1488 | containing the list of filenames you really want. Line numbers (C<$.>) |
1489 | continue as if the input were one big happy file. (But see example |
5a964f20 |
1490 | under C<eof> for how to reset line numbers on each file.) |
1491 | |
1492 | If you want to set @ARGV to your own list of files, go right ahead. |
1493 | This sets @ARGV to all plain text files if no @ARGV was given: |
1494 | |
1495 | @ARGV = grep { -f && -T } glob('*') unless @ARGV; |
a0d0e21e |
1496 | |
5a964f20 |
1497 | You can even set them to pipe commands. For example, this automatically |
1498 | filters compressed arguments through B<gzip>: |
1499 | |
1500 | @ARGV = map { /\.(gz|Z)$/ ? "gzip -dc < $_ |" : $_ } @ARGV; |
1501 | |
1502 | If you want to pass switches into your script, you can use one of the |
a0d0e21e |
1503 | Getopts modules or put a loop on the front like this: |
1504 | |
1505 | while ($_ = $ARGV[0], /^-/) { |
1506 | shift; |
1507 | last if /^--$/; |
1508 | if (/^-D(.*)/) { $debug = $1 } |
1509 | if (/^-v/) { $verbose++ } |
5a964f20 |
1510 | # ... # other switches |
a0d0e21e |
1511 | } |
5a964f20 |
1512 | |
a0d0e21e |
1513 | while (<>) { |
5a964f20 |
1514 | # ... # code for each line |
a0d0e21e |
1515 | } |
1516 | |
7b8d334a |
1517 | The E<lt>E<gt> symbol will return C<undef> for end-of-file only once. |
1518 | If you call it again after this it will assume you are processing another |
1519 | @ARGV list, and if you haven't set @ARGV, will input from STDIN. |
a0d0e21e |
1520 | |
1521 | If the string inside the angle brackets is a reference to a scalar |
5f05dabc |
1522 | variable (e.g., E<lt>$fooE<gt>), then that variable contains the name of the |
5a964f20 |
1523 | filehandle to input from, or its typeglob, or a reference to the same. For example: |
cb1a09d0 |
1524 | |
1525 | $fh = \*STDIN; |
1526 | $line = <$fh>; |
a0d0e21e |
1527 | |
5a964f20 |
1528 | If what's within the angle brackets is neither a filehandle nor a simple |
1529 | scalar variable containing a filehandle name, typeglob, or typeglob |
1530 | reference, it is interpreted as a filename pattern to be globbed, and |
1531 | either a list of filenames or the next filename in the list is returned, |
1532 | depending on context. This distinction is determined on syntactic |
1533 | grounds alone. That means C<E<lt>$xE<gt>> is always a readline from |
1534 | an indirect handle, but C<E<lt>$hash{key}E<gt>> is always a glob. |
1535 | That's because $x is a simple scalar variable, but C<$hash{key}> is |
1536 | not--it's a hash element. |
1537 | |
1538 | One level of double-quote interpretation is done first, but you can't |
1539 | say C<E<lt>$fooE<gt>> because that's an indirect filehandle as explained |
1540 | in the previous paragraph. (In older versions of Perl, programmers |
1541 | would insert curly brackets to force interpretation as a filename glob: |
1542 | C<E<lt>${foo}E<gt>>. These days, it's considered cleaner to call the |
1543 | internal function directly as C<glob($foo)>, which is probably the right |
1544 | way to have done it in the first place.) Example: |
a0d0e21e |
1545 | |
1546 | while (<*.c>) { |
1547 | chmod 0644, $_; |
1548 | } |
1549 | |
1550 | is equivalent to |
1551 | |
1552 | open(FOO, "echo *.c | tr -s ' \t\r\f' '\\012\\012\\012\\012'|"); |
1553 | while (<FOO>) { |
1554 | chop; |
1555 | chmod 0644, $_; |
1556 | } |
1557 | |
1558 | In fact, it's currently implemented that way. (Which means it will not |
1559 | work on filenames with spaces in them unless you have csh(1) on your |
1560 | machine.) Of course, the shortest way to do the above is: |
1561 | |
1562 | chmod 0644, <*.c>; |
1563 | |
1564 | Because globbing invokes a shell, it's often faster to call readdir() yourself |
5f05dabc |
1565 | and do your own grep() on the filenames. Furthermore, due to its current |
54310121 |
1566 | implementation of using a shell, the glob() routine may get "Arg list too |
a0d0e21e |
1567 | long" errors (unless you've installed tcsh(1L) as F</bin/csh>). |
1568 | |
5f05dabc |
1569 | A glob evaluates its (embedded) argument only when it is starting a new |
4633a7c4 |
1570 | list. All values must be read before it will start over. In a list |
1571 | context this isn't important, because you automatically get them all |
5a964f20 |
1572 | anyway. In scalar context, however, the operator returns the next value |
7b8d334a |
1573 | each time it is called, or a C<undef> value if you've just run out. As |
1574 | for filehandles an automatic C<defined> is generated when the glob |
1575 | occurs in the test part of a C<while> or C<for> - because legal glob returns |
1576 | (e.g. a file called F<0>) would otherwise terminate the loop. |
1577 | Again, C<undef> is returned only once. So if you're expecting a single value |
1578 | from a glob, it is much better to say |
4633a7c4 |
1579 | |
1580 | ($file) = <blurch*>; |
1581 | |
1582 | than |
1583 | |
1584 | $file = <blurch*>; |
1585 | |
1586 | because the latter will alternate between returning a filename and |
54310121 |
1587 | returning FALSE. |
4633a7c4 |
1588 | |
1589 | It you're trying to do variable interpolation, it's definitely better |
1590 | to use the glob() function, because the older notation can cause people |
e37d713d |
1591 | to become confused with the indirect filehandle notation. |
4633a7c4 |
1592 | |
1593 | @files = glob("$dir/*.[ch]"); |
1594 | @files = glob($files[$i]); |
1595 | |
a0d0e21e |
1596 | =head2 Constant Folding |
1597 | |
1598 | Like C, Perl does a certain amount of expression evaluation at |
5a964f20 |
1599 | compile time, whenever it determines that all arguments to an |
a0d0e21e |
1600 | operator are static and have no side effects. In particular, string |
1601 | concatenation happens at compile time between literals that don't do |
1602 | variable substitution. Backslash interpretation also happens at |
1603 | compile time. You can say |
1604 | |
1605 | 'Now is the time for all' . "\n" . |
1606 | 'good men to come to.' |
1607 | |
54310121 |
1608 | and this all reduces to one string internally. Likewise, if |
a0d0e21e |
1609 | you say |
1610 | |
1611 | foreach $file (@filenames) { |
5a964f20 |
1612 | if (-s $file > 5 + 100 * 2**16) { } |
54310121 |
1613 | } |
a0d0e21e |
1614 | |
54310121 |
1615 | the compiler will precompute the number that |
a0d0e21e |
1616 | expression represents so that the interpreter |
1617 | won't have to. |
1618 | |
2c268ad5 |
1619 | =head2 Bitwise String Operators |
1620 | |
1621 | Bitstrings of any size may be manipulated by the bitwise operators |
1622 | (C<~ | & ^>). |
1623 | |
1624 | If the operands to a binary bitwise op are strings of different sizes, |
1625 | B<or> and B<xor> ops will act as if the shorter operand had additional |
1626 | zero bits on the right, while the B<and> op will act as if the longer |
1627 | operand were truncated to the length of the shorter. |
1628 | |
1629 | # ASCII-based examples |
1630 | print "j p \n" ^ " a h"; # prints "JAPH\n" |
1631 | print "JA" | " ph\n"; # prints "japh\n" |
1632 | print "japh\nJunk" & '_____'; # prints "JAPH\n"; |
1633 | print 'p N$' ^ " E<H\n"; # prints "Perl\n"; |
1634 | |
1635 | If you are intending to manipulate bitstrings, you should be certain that |
1636 | you're supplying bitstrings: If an operand is a number, that will imply |
1637 | a B<numeric> bitwise operation. You may explicitly show which type of |
1638 | operation you intend by using C<""> or C<0+>, as in the examples below. |
1639 | |
1640 | $foo = 150 | 105 ; # yields 255 (0x96 | 0x69 is 0xFF) |
1641 | $foo = '150' | 105 ; # yields 255 |
1642 | $foo = 150 | '105'; # yields 255 |
1643 | $foo = '150' | '105'; # yields string '155' (under ASCII) |
1644 | |
1645 | $baz = 0+$foo & 0+$bar; # both ops explicitly numeric |
1646 | $biz = "$foo" ^ "$bar"; # both ops explicitly stringy |
a0d0e21e |
1647 | |
55497cff |
1648 | =head2 Integer Arithmetic |
a0d0e21e |
1649 | |
1650 | By default Perl assumes that it must do most of its arithmetic in |
1651 | floating point. But by saying |
1652 | |
1653 | use integer; |
1654 | |
1655 | you may tell the compiler that it's okay to use integer operations |
1656 | from here to the end of the enclosing BLOCK. An inner BLOCK may |
54310121 |
1657 | countermand this by saying |
a0d0e21e |
1658 | |
1659 | no integer; |
1660 | |
1661 | which lasts until the end of that BLOCK. |
1662 | |
55497cff |
1663 | The bitwise operators ("&", "|", "^", "~", "<<", and ">>") always |
2c268ad5 |
1664 | produce integral results. (But see also L<Bitwise String Operators>.) |
1665 | However, C<use integer> still has meaning |
55497cff |
1666 | for them. By default, their results are interpreted as unsigned |
1667 | integers. However, if C<use integer> is in effect, their results are |
5f05dabc |
1668 | interpreted as signed integers. For example, C<~0> usually evaluates |
5a964f20 |
1669 | to a large integral value. However, C<use integer; ~0> is -1 on twos-complement machines. |
68dc0745 |
1670 | |
1671 | =head2 Floating-point Arithmetic |
1672 | |
1673 | While C<use integer> provides integer-only arithmetic, there is no |
1674 | similar ways to provide rounding or truncation at a certain number of |
1675 | decimal places. For rounding to a certain number of digits, sprintf() |
1676 | or printf() is usually the easiest route. |
1677 | |
5a964f20 |
1678 | Floating-point numbers are only approximations to what a mathematician |
1679 | would call real numbers. There are infinitely more reals than floats, |
1680 | so some corners must be cut. For example: |
1681 | |
1682 | printf "%.20g\n", 123456789123456789; |
1683 | # produces 123456789123456784 |
1684 | |
1685 | Testing for exact equality of floating-point equality or inequality is |
1686 | not a good idea. Here's a (relatively expensive) work-around to compare |
1687 | whether two floating-point numbers are equal to a particular number of |
1688 | decimal places. See Knuth, volume II, for a more robust treatment of |
1689 | this topic. |
1690 | |
1691 | sub fp_equal { |
1692 | my ($X, $Y, $POINTS) = @_; |
1693 | my ($tX, $tY); |
1694 | $tX = sprintf("%.${POINTS}g", $X); |
1695 | $tY = sprintf("%.${POINTS}g", $Y); |
1696 | return $tX eq $tY; |
1697 | } |
1698 | |
68dc0745 |
1699 | The POSIX module (part of the standard perl distribution) implements |
1700 | ceil(), floor(), and a number of other mathematical and trigonometric |
1701 | functions. The Math::Complex module (part of the standard perl |
1702 | distribution) defines a number of mathematical functions that can also |
1703 | work on real numbers. Math::Complex not as efficient as POSIX, but |
1704 | POSIX can't work with complex numbers. |
1705 | |
1706 | Rounding in financial applications can have serious implications, and |
1707 | the rounding method used should be specified precisely. In these |
1708 | cases, it probably pays not to trust whichever system rounding is |
1709 | being used by Perl, but to instead implement the rounding function you |
1710 | need yourself. |
5a964f20 |
1711 | |
1712 | =head2 Bigger Numbers |
1713 | |
1714 | The standard Math::BigInt and Math::BigFloat modules provide |
1715 | variable precision arithmetic and overloaded operators. |
1716 | At the cost of some space and considerable speed, they |
1717 | avoid the normal pitfalls associated with limited-precision |
1718 | representations. |
1719 | |
1720 | use Math::BigInt; |
1721 | $x = Math::BigInt->new('123456789123456789'); |
1722 | print $x * $x; |
1723 | |
1724 | # prints +15241578780673678515622620750190521 |