Commit | Line | Data |
68dc0745 |
1 | =head1 NAME |
2 | |
c195e131 |
3 | perlfaq7 - General Perl Language Issues ($Revision: 10100 $) |
68dc0745 |
4 | |
5 | =head1 DESCRIPTION |
6 | |
7 | This section deals with general Perl language issues that don't |
8 | clearly fit into any of the other sections. |
9 | |
10 | =head2 Can I get a BNF/yacc/RE for the Perl language? |
11 | |
c8db1d39 |
12 | There is no BNF, but you can paw your way through the yacc grammar in |
13 | perly.y in the source distribution if you're particularly brave. The |
14 | grammar relies on very smart tokenizing code, so be prepared to |
15 | venture into toke.c as well. |
16 | |
17 | In the words of Chaim Frenkel: "Perl's grammar can not be reduced to BNF. |
18 | The work of parsing perl is distributed between yacc, the lexer, smoke |
19 | and mirrors." |
68dc0745 |
20 | |
d92eb7b0 |
21 | =head2 What are all these $@%&* punctuation signs, and how do I know when to use them? |
68dc0745 |
22 | |
23 | They are type specifiers, as detailed in L<perldata>: |
24 | |
ac003c96 |
25 | $ for scalar values (number, string or reference) |
26 | @ for arrays |
27 | % for hashes (associative arrays) |
28 | & for subroutines (aka functions, procedures, methods) |
29 | * for all types of that symbol name. In version 4 you used them like |
30 | pointers, but in modern perls you can just use references. |
68dc0745 |
31 | |
a6dd486b |
32 | There are couple of other symbols that you're likely to encounter that aren't |
33 | really type specifiers: |
68dc0745 |
34 | |
ac003c96 |
35 | <> are used for inputting a record from a filehandle. |
36 | \ takes a reference to something. |
68dc0745 |
37 | |
c47ff5f1 |
38 | Note that <FILE> is I<neither> the type specifier for files |
39 | nor the name of the handle. It is the C<< <> >> operator applied |
a6dd486b |
40 | to the handle FILE. It reads one line (well, record--see |
197aec24 |
41 | L<perlvar/$E<sol>>) from the handle FILE in scalar context, or I<all> lines |
68dc0745 |
42 | in list context. When performing open, close, or any other operation |
a6dd486b |
43 | besides C<< <> >> on files, or even when talking about the handle, do |
68dc0745 |
44 | I<not> use the brackets. These are correct: C<eof(FH)>, C<seek(FH, 0, |
45 | 2)> and "copying from STDIN to FILE". |
46 | |
47 | =head2 Do I always/never have to quote my strings or use semicolons and commas? |
48 | |
49 | Normally, a bareword doesn't need to be quoted, but in most cases |
50 | probably should be (and must be under C<use strict>). But a hash key |
51 | consisting of a simple word (that isn't the name of a defined |
c47ff5f1 |
52 | subroutine) and the left-hand operand to the C<< => >> operator both |
68dc0745 |
53 | count as though they were quoted: |
54 | |
ac003c96 |
55 | This is like this |
56 | ------------ --------------- |
57 | $foo{line} $foo{'line'} |
58 | bar => stuff 'bar' => stuff |
68dc0745 |
59 | |
60 | The final semicolon in a block is optional, as is the final comma in a |
61 | list. Good style (see L<perlstyle>) says to put them in except for |
62 | one-liners: |
63 | |
ac003c96 |
64 | if ($whoops) { exit 1 } |
65 | @nums = (1, 2, 3); |
66 | |
67 | if ($whoops) { |
68 | exit 1; |
69 | } |
68dc0745 |
70 | |
ac003c96 |
71 | @lines = ( |
68dc0745 |
72 | "There Beren came from mountains cold", |
73 | "And lost he wandered under leaves", |
ac003c96 |
74 | ); |
68dc0745 |
75 | |
76 | =head2 How do I skip some return values? |
77 | |
78 | One way is to treat the return values as a list and index into it: |
79 | |
ac003c96 |
80 | $dir = (getpwnam($user))[7]; |
68dc0745 |
81 | |
82 | Another way is to use undef as an element on the left-hand-side: |
83 | |
ac003c96 |
84 | ($dev, $ino, undef, undef, $uid, $gid) = stat($file); |
197aec24 |
85 | |
49d635f9 |
86 | You can also use a list slice to select only the elements that |
87 | you need: |
88 | |
89 | ($dev, $ino, $uid, $gid) = ( stat($file) )[0,1,4,5]; |
68dc0745 |
90 | |
91 | =head2 How do I temporarily block warnings? |
92 | |
9f1b1f2d |
93 | If you are running Perl 5.6.0 or better, the C<use warnings> pragma |
94 | allows fine control of what warning are produced. |
95 | See L<perllexwarn> for more details. |
96 | |
ac003c96 |
97 | { |
9f1b1f2d |
98 | no warnings; # temporarily turn off warnings |
99 | $a = $b + $c; # I know these might be undef |
ac003c96 |
100 | } |
6670e5e7 |
101 | |
28b41a80 |
102 | Additionally, you can enable and disable categories of warnings. |
103 | You turn off the categories you want to ignore and you can still |
104 | get other categories of warnings. See L<perllexwarn> for the |
105 | complete details, including the category names and hierarchy. |
106 | |
107 | { |
108 | no warnings 'uninitialized'; |
109 | $a = $b + $c; |
110 | } |
9f1b1f2d |
111 | |
112 | If you have an older version of Perl, the C<$^W> variable (documented |
113 | in L<perlvar>) controls runtime warnings for a block: |
68dc0745 |
114 | |
ac003c96 |
115 | { |
68dc0745 |
116 | local $^W = 0; # temporarily turn off warnings |
117 | $a = $b + $c; # I know these might be undef |
ac003c96 |
118 | } |
68dc0745 |
119 | |
120 | Note that like all the punctuation variables, you cannot currently |
121 | use my() on C<$^W>, only local(). |
122 | |
68dc0745 |
123 | =head2 What's an extension? |
124 | |
a6dd486b |
125 | An extension is a way of calling compiled C code from Perl. Reading |
126 | L<perlxstut> is a good place to learn more about extensions. |
68dc0745 |
127 | |
128 | =head2 Why do Perl operators have different precedence than C operators? |
129 | |
130 | Actually, they don't. All C operators that Perl copies have the same |
131 | precedence in Perl as they do in C. The problem is with operators that C |
132 | doesn't have, especially functions that give a list context to everything |
a6dd486b |
133 | on their right, eg. print, chmod, exec, and so on. Such functions are |
68dc0745 |
134 | called "list operators" and appear as such in the precedence table in |
135 | L<perlop>. |
136 | |
137 | A common mistake is to write: |
138 | |
ac003c96 |
139 | unlink $file || die "snafu"; |
68dc0745 |
140 | |
141 | This gets interpreted as: |
142 | |
ac003c96 |
143 | unlink ($file || die "snafu"); |
68dc0745 |
144 | |
145 | To avoid this problem, either put in extra parentheses or use the |
146 | super low precedence C<or> operator: |
147 | |
ac003c96 |
148 | (unlink $file) || die "snafu"; |
149 | unlink $file or die "snafu"; |
68dc0745 |
150 | |
151 | The "English" operators (C<and>, C<or>, C<xor>, and C<not>) |
152 | deliberately have precedence lower than that of list operators for |
153 | just such situations as the one above. |
154 | |
155 | Another operator with surprising precedence is exponentiation. It |
156 | binds more tightly even than unary minus, making C<-2**2> product a |
157 | negative not a positive four. It is also right-associating, meaning |
158 | that C<2**3**2> is two raised to the ninth power, not eight squared. |
159 | |
c8db1d39 |
160 | Although it has the same precedence as in C, Perl's C<?:> operator |
161 | produces an lvalue. This assigns $x to either $a or $b, depending |
162 | on the trueness of $maybe: |
163 | |
ac003c96 |
164 | ($maybe ? $a : $b) = $x; |
c8db1d39 |
165 | |
68dc0745 |
166 | =head2 How do I declare/create a structure? |
167 | |
168 | In general, you don't "declare" a structure. Just use a (probably |
169 | anonymous) hash reference. See L<perlref> and L<perldsc> for details. |
170 | Here's an example: |
171 | |
ac003c96 |
172 | $person = {}; # new anonymous hash |
173 | $person->{AGE} = 24; # set field AGE to 24 |
174 | $person->{NAME} = "Nat"; # set field NAME to "Nat" |
175 | |
68dc0745 |
176 | If you're looking for something a bit more rigorous, try L<perltoot>. |
177 | |
178 | =head2 How do I create a module? |
179 | |
7678cced |
180 | (contributed by brian d foy) |
68dc0745 |
181 | |
7678cced |
182 | L<perlmod>, L<perlmodlib>, L<perlmodstyle> explain modules |
cf525c36 |
183 | in all the gory details. L<perlnewmod> gives a brief |
7678cced |
184 | overview of the process along with a couple of suggestions |
185 | about style. |
65acb1b1 |
186 | |
7678cced |
187 | If you need to include C code or C library interfaces in |
188 | your module, you'll need h2xs. h2xs will create the module |
189 | distribution structure and the initial interface files |
190 | you'll need. L<perlxs> and L<perlxstut> explain the details. |
7207e29d |
191 | |
7678cced |
192 | If you don't need to use C code, other tools such as |
193 | ExtUtils::ModuleMaker and Module::Starter, can help you |
194 | create a skeleton module distribution. |
195 | |
196 | You may also want to see Sam Tregar's "Writing Perl Modules |
197 | for CPAN" ( http://apress.com/book/bookDisplay.html?bID=14 ) |
198 | which is the best hands-on guide to creating module |
199 | distributions. |
65acb1b1 |
200 | |
c195e131 |
201 | =head2 How do I adopt or take over a module already on CPAN? |
202 | |
203 | (contributed by brian d foy) |
204 | |
205 | The easiest way to take over a module is to have the current |
206 | module maintainer either make you a co-maintainer or transfer |
207 | the module to you. |
208 | |
209 | If you can't reach the author for some reason (e.g. email bounces), |
210 | the PAUSE admins at modules@perl.org can help. The PAUSE admins |
211 | treat each case individually. |
212 | |
213 | =over 4 |
214 | |
215 | =item |
216 | |
217 | Get a login for the Perl Authors Upload Server (PAUSE) if you don't |
218 | already have one: http://pause.perl.org |
219 | |
220 | =item |
221 | |
222 | Write to modules@perl.org explaining what you did to contact the |
223 | current maintainer. The PAUSE admins will also try to reach the |
224 | maintainer. |
225 | |
226 | =item |
227 | |
228 | Post a public message in a heavily trafficked site announcing your |
229 | intention to take over the module. |
230 | |
231 | =item |
232 | |
233 | Wait a bit. The PAUSE admins don't want to act too quickly in case |
234 | the current maintainer is on holiday. If there's no response to |
235 | private communication or the public post, a PAUSE admin can transfer |
236 | it to you. |
237 | |
238 | =back |
239 | |
68dc0745 |
240 | =head2 How do I create a class? |
241 | |
242 | See L<perltoot> for an introduction to classes and objects, as well as |
243 | L<perlobj> and L<perlbot>. |
244 | |
245 | =head2 How can I tell if a variable is tainted? |
246 | |
213329dd |
247 | You can use the tainted() function of the Scalar::Util module, available |
248 | from CPAN (or included with Perl since release 5.8.0). |
249 | See also L<perlsec/"Laundering and Detecting Tainted Data">. |
68dc0745 |
250 | |
251 | =head2 What's a closure? |
252 | |
253 | Closures are documented in L<perlref>. |
254 | |
255 | I<Closure> is a computer science term with a precise but |
322be77c |
256 | hard-to-explain meaning. Usually, closures are implemented in Perl as |
257 | anonymous subroutines with lasting references to lexical variables |
258 | outside their own scopes. These lexicals magically refer to the |
259 | variables that were around when the subroutine was defined (deep |
260 | binding). |
261 | |
262 | Closures are most often used in programming languages where you can |
263 | have the return value of a function be itself a function, as you can |
264 | in Perl. Note that some languages provide anonymous functions but are |
265 | not capable of providing proper closures: the Python language, for |
68dc0745 |
266 | example. For more information on closures, check out any textbook on |
267 | functional programming. Scheme is a language that not only supports |
268 | but encourages closures. |
269 | |
322be77c |
270 | Here's a classic non-closure function-generating function: |
68dc0745 |
271 | |
ac003c96 |
272 | sub add_function_generator { |
273 | return sub { shift() + shift() }; |
274 | } |
68dc0745 |
275 | |
ac003c96 |
276 | $add_sub = add_function_generator(); |
277 | $sum = $add_sub->(4,5); # $sum is 9 now. |
68dc0745 |
278 | |
322be77c |
279 | The anonymous subroutine returned by add_function_generator() isn't |
280 | technically a closure because it refers to no lexicals outside its own |
281 | scope. Using a closure gives you a I<function template> with some |
282 | customization slots left out to be filled later. |
68dc0745 |
283 | |
284 | Contrast this with the following make_adder() function, in which the |
285 | returned anonymous function contains a reference to a lexical variable |
286 | outside the scope of that function itself. Such a reference requires |
287 | that Perl return a proper closure, thus locking in for all time the |
288 | value that the lexical had when the function was created. |
289 | |
ac003c96 |
290 | sub make_adder { |
291 | my $addpiece = shift; |
292 | return sub { shift() + $addpiece }; |
293 | } |
294 | |
295 | $f1 = make_adder(20); |
296 | $f2 = make_adder(555); |
68dc0745 |
297 | |
298 | Now C<&$f1($n)> is always 20 plus whatever $n you pass in, whereas |
299 | C<&$f2($n)> is always 555 plus whatever $n you pass in. The $addpiece |
300 | in the closure sticks around. |
301 | |
302 | Closures are often used for less esoteric purposes. For example, when |
303 | you want to pass in a bit of code into a function: |
304 | |
ac003c96 |
305 | my $line; |
306 | timeout( 30, sub { $line = <STDIN> } ); |
68dc0745 |
307 | |
c47ff5f1 |
308 | If the code to execute had been passed in as a string, |
309 | C<< '$line = <STDIN>' >>, there would have been no way for the |
310 | hypothetical timeout() function to access the lexical variable |
311 | $line back in its caller's scope. |
68dc0745 |
312 | |
322be77c |
313 | Another use for a closure is to make a variable I<private> to a |
314 | named subroutine, e.g. a counter that gets initialized at creation |
315 | time of the sub and can only be modified from within the sub. |
316 | This is sometimes used with a BEGIN block in package files to make |
317 | sure a variable doesn't get meddled with during the lifetime of the |
318 | package: |
319 | |
ac003c96 |
320 | BEGIN { |
321 | my $id = 0; |
322 | sub next_id { ++$id } |
323 | } |
322be77c |
324 | |
325 | This is discussed in more detail in L<perlsub>, see the entry on |
326 | I<Persistent Private Variables>. |
327 | |
46fc3d4c |
328 | =head2 What is variable suicide and how can I prevent it? |
329 | |
9e72e4c6 |
330 | This problem was fixed in perl 5.004_05, so preventing it means upgrading |
331 | your version of perl. ;) |
46fc3d4c |
332 | |
9e72e4c6 |
333 | Variable suicide is when you (temporarily or permanently) lose the value |
334 | of a variable. It is caused by scoping through my() and local() |
335 | interacting with either closures or aliased foreach() iterator variables |
336 | and subroutine arguments. It used to be easy to inadvertently lose a |
337 | variable's value this way, but now it's much harder. Take this code: |
338 | |
ac003c96 |
339 | my $f = 'foo'; |
340 | sub T { |
341 | while ($i++ < 3) { my $f = $f; $f .= "bar"; print $f, "\n" } |
342 | } |
343 | |
344 | T; |
345 | print "Finally $f\n"; |
46fc3d4c |
346 | |
9e72e4c6 |
347 | If you are experiencing variable suicide, that C<my $f> in the subroutine |
348 | doesn't pick up a fresh copy of the C<$f> whose value is <foo>. The output |
349 | shows that inside the subroutine the value of C<$f> leaks through when it |
350 | shouldn't, as in this output: |
351 | |
352 | foobar |
353 | foobarbar |
354 | foobarbarbar |
355 | Finally foo |
356 | |
46fc3d4c |
357 | The $f that has "bar" added to it three times should be a new C<$f> |
9e72e4c6 |
358 | C<my $f> should create a new lexical variable each time through the loop. |
359 | The expected output is: |
360 | |
361 | foobar |
362 | foobar |
363 | foobar |
364 | Finally foo |
46fc3d4c |
365 | |
d92eb7b0 |
366 | =head2 How can I pass/return a {Function, FileHandle, Array, Hash, Method, Regex}? |
68dc0745 |
367 | |
d92eb7b0 |
368 | With the exception of regexes, you need to pass references to these |
68dc0745 |
369 | objects. See L<perlsub/"Pass by Reference"> for this particular |
370 | question, and L<perlref> for information on references. |
371 | |
ac9dac7f |
372 | See "Passing Regexes", later in L<perlfaq7>, for information on |
373 | passing regular expressions. |
a6dd486b |
374 | |
68dc0745 |
375 | =over 4 |
376 | |
377 | =item Passing Variables and Functions |
378 | |
a6dd486b |
379 | Regular variables and functions are quite easy to pass: just pass in a |
68dc0745 |
380 | reference to an existing or anonymous variable or function: |
381 | |
ac003c96 |
382 | func( \$some_scalar ); |
68dc0745 |
383 | |
ac003c96 |
384 | func( \@some_array ); |
385 | func( [ 1 .. 10 ] ); |
68dc0745 |
386 | |
ac003c96 |
387 | func( \%some_hash ); |
388 | func( { this => 10, that => 20 } ); |
68dc0745 |
389 | |
ac003c96 |
390 | func( \&some_func ); |
391 | func( sub { $_[0] ** $_[1] } ); |
68dc0745 |
392 | |
393 | =item Passing Filehandles |
394 | |
49d635f9 |
395 | As of Perl 5.6, you can represent filehandles with scalar variables |
396 | which you treat as any other scalar. |
397 | |
398 | open my $fh, $filename or die "Cannot open $filename! $!"; |
399 | func( $fh ); |
197aec24 |
400 | |
49d635f9 |
401 | sub func { |
402 | my $passed_fh = shift; |
197aec24 |
403 | |
ac003c96 |
404 | my $line = <$passed_fh>; |
49d635f9 |
405 | } |
197aec24 |
406 | |
49d635f9 |
407 | Before Perl 5.6, you had to use the C<*FH> or C<\*FH> notations. |
a6dd486b |
408 | These are "typeglobs"--see L<perldata/"Typeglobs and Filehandles"> |
c8db1d39 |
409 | and especially L<perlsub/"Pass by Reference"> for more information. |
410 | |
d92eb7b0 |
411 | =item Passing Regexes |
412 | |
413 | To pass regexes around, you'll need to be using a release of Perl |
414 | sufficiently recent as to support the C<qr//> construct, pass around |
415 | strings and use an exception-trapping eval, or else be very, very clever. |
68dc0745 |
416 | |
d92eb7b0 |
417 | Here's an example of how to pass in a string to be regex compared |
418 | using C<qr//>: |
68dc0745 |
419 | |
ac003c96 |
420 | sub compare($$) { |
421 | my ($val1, $regex) = @_; |
422 | my $retval = $val1 =~ /$regex/; |
d92eb7b0 |
423 | return $retval; |
ac003c96 |
424 | } |
425 | $match = compare("old McDonald", qr/d.*D/i); |
d92eb7b0 |
426 | |
427 | Notice how C<qr//> allows flags at the end. That pattern was compiled |
428 | at compile time, although it was executed later. The nifty C<qr//> |
429 | notation wasn't introduced until the 5.005 release. Before that, you |
430 | had to approach this problem much less intuitively. For example, here |
431 | it is again if you don't have C<qr//>: |
432 | |
ac003c96 |
433 | sub compare($$) { |
434 | my ($val1, $regex) = @_; |
435 | my $retval = eval { $val1 =~ /$regex/ }; |
68dc0745 |
436 | die if $@; |
437 | return $retval; |
ac003c96 |
438 | } |
68dc0745 |
439 | |
ac003c96 |
440 | $match = compare("old McDonald", q/($?i)d.*D/); |
68dc0745 |
441 | |
442 | Make sure you never say something like this: |
443 | |
ac003c96 |
444 | return eval "\$val =~ /$regex/"; # WRONG |
68dc0745 |
445 | |
d92eb7b0 |
446 | or someone can sneak shell escapes into the regex due to the double |
68dc0745 |
447 | interpolation of the eval and the double-quoted string. For example: |
448 | |
ac003c96 |
449 | $pattern_of_evil = 'danger ${ system("rm -rf * &") } danger'; |
68dc0745 |
450 | |
ac003c96 |
451 | eval "\$string =~ /$pattern_of_evil/"; |
68dc0745 |
452 | |
453 | Those preferring to be very, very clever might see the O'Reilly book, |
454 | I<Mastering Regular Expressions>, by Jeffrey Friedl. Page 273's |
455 | Build_MatchMany_Function() is particularly interesting. A complete |
456 | citation of this book is given in L<perlfaq2>. |
457 | |
458 | =item Passing Methods |
459 | |
460 | To pass an object method into a subroutine, you can do this: |
461 | |
ac003c96 |
462 | call_a_lot(10, $some_obj, "methname") |
463 | sub call_a_lot { |
464 | my ($count, $widget, $trick) = @_; |
465 | for (my $i = 0; $i < $count; $i++) { |
466 | $widget->$trick(); |
467 | } |
468 | } |
68dc0745 |
469 | |
a6dd486b |
470 | Or, you can use a closure to bundle up the object, its |
471 | method call, and arguments: |
68dc0745 |
472 | |
ac003c96 |
473 | my $whatnot = sub { $some_obj->obfuscate(@args) }; |
474 | func($whatnot); |
475 | sub func { |
476 | my $code = shift; |
477 | &$code(); |
478 | } |
68dc0745 |
479 | |
480 | You could also investigate the can() method in the UNIVERSAL class |
481 | (part of the standard perl distribution). |
482 | |
483 | =back |
484 | |
485 | =head2 How do I create a static variable? |
486 | |
6670e5e7 |
487 | (contributed by brian d foy) |
68dc0745 |
488 | |
6670e5e7 |
489 | Perl doesn't have "static" variables, which can only be accessed from |
490 | the function in which they are declared. You can get the same effect |
491 | with lexical variables, though. |
492 | |
493 | You can fake a static variable by using a lexical variable which goes |
a05e4845 |
494 | out of scope. In this example, you define the subroutine C<counter>, and |
6670e5e7 |
495 | it uses the lexical variable C<$count>. Since you wrap this in a BEGIN |
496 | block, C<$count> is defined at compile-time, but also goes out of |
497 | scope at the end of the BEGIN block. The BEGIN block also ensures that |
498 | the subroutine and the value it uses is defined at compile-time so the |
499 | subroutine is ready to use just like any other subroutine, and you can |
500 | put this code in the same place as other subroutines in the program |
501 | text (i.e. at the end of the code, typically). The subroutine |
502 | C<counter> still has a reference to the data, and is the only way you |
503 | can access the value (and each time you do, you increment the value). |
504 | The data in chunk of memory defined by C<$count> is private to |
505 | C<counter>. |
506 | |
ac003c96 |
507 | BEGIN { |
508 | my $count = 1; |
509 | sub counter { $count++ } |
510 | } |
511 | |
512 | my $start = counter(); |
513 | |
514 | .... # code that calls counter(); |
515 | |
516 | my $end = counter(); |
68dc0745 |
517 | |
6670e5e7 |
518 | In the previous example, you created a function-private variable |
519 | because only one function remembered its reference. You could define |
520 | multiple functions while the variable is in scope, and each function |
521 | can share the "private" variable. It's not really "static" because you |
522 | can access it outside the function while the lexical variable is in |
523 | scope, and even create references to it. In this example, |
524 | C<increment_count> and C<return_count> share the variable. One |
525 | function adds to the value and the other simply returns the value. |
526 | They can both access C<$count>, and since it has gone out of scope, |
527 | there is no other way to access it. |
68dc0745 |
528 | |
ac003c96 |
529 | BEGIN { |
530 | my $count = 1; |
531 | sub increment_count { $count++ } |
532 | sub return_count { $count } |
533 | } |
68dc0745 |
534 | |
6670e5e7 |
535 | To declare a file-private variable, you still use a lexical variable. |
536 | A file is also a scope, so a lexical variable defined in the file |
537 | cannot be seen from any other file. |
68dc0745 |
538 | |
6670e5e7 |
539 | See L<perlsub/"Persistent Private Variables"> for more information. |
540 | The discussion of closures in L<perlref> may help you even though we |
541 | did not use anonymous subroutines in this answer. See |
542 | L<perlsub/"Persistent Private Variables"> for details. |
c8db1d39 |
543 | |
68dc0745 |
544 | =head2 What's the difference between dynamic and lexical (static) scoping? Between local() and my()? |
545 | |
a6dd486b |
546 | C<local($x)> saves away the old value of the global variable C<$x> |
547 | and assigns a new value for the duration of the subroutine I<which is |
68dc0745 |
548 | visible in other functions called from that subroutine>. This is done |
549 | at run-time, so is called dynamic scoping. local() always affects global |
550 | variables, also called package variables or dynamic variables. |
551 | |
552 | C<my($x)> creates a new variable that is only visible in the current |
a6dd486b |
553 | subroutine. This is done at compile-time, so it is called lexical or |
68dc0745 |
554 | static scoping. my() always affects private variables, also called |
555 | lexical variables or (improperly) static(ly scoped) variables. |
556 | |
557 | For instance: |
558 | |
ac003c96 |
559 | sub visible { |
560 | print "var has value $var\n"; |
561 | } |
562 | |
563 | sub dynamic { |
564 | local $var = 'local'; # new temporary value for the still-global |
565 | visible(); # variable called $var |
566 | } |
567 | |
568 | sub lexical { |
569 | my $var = 'private'; # new private variable, $var |
570 | visible(); # (invisible outside of sub scope) |
571 | } |
572 | |
573 | $var = 'global'; |
574 | |
575 | visible(); # prints global |
576 | dynamic(); # prints local |
577 | lexical(); # prints global |
68dc0745 |
578 | |
579 | Notice how at no point does the value "private" get printed. That's |
580 | because $var only has that value within the block of the lexical() |
581 | function, and it is hidden from called subroutine. |
582 | |
583 | In summary, local() doesn't make what you think of as private, local |
584 | variables. It gives a global variable a temporary value. my() is |
585 | what you're looking for if you want private variables. |
586 | |
197aec24 |
587 | See L<perlsub/"Private Variables via my()"> and |
13a2d996 |
588 | L<perlsub/"Temporary Values via local()"> for excruciating details. |
68dc0745 |
589 | |
590 | =head2 How can I access a dynamic variable while a similarly named lexical is in scope? |
591 | |
49d635f9 |
592 | If you know your package, you can just mention it explicitly, as in |
593 | $Some_Pack::var. Note that the notation $::var is B<not> the dynamic $var |
594 | in the current package, but rather the one in the "main" package, as |
595 | though you had written $main::var. |
596 | |
597 | use vars '$var'; |
598 | local $var = "global"; |
599 | my $var = "lexical"; |
68dc0745 |
600 | |
49d635f9 |
601 | print "lexical is $var\n"; |
602 | print "global is $main::var\n"; |
68dc0745 |
603 | |
49d635f9 |
604 | Alternatively you can use the compiler directive our() to bring a |
605 | dynamic variable into the current lexical scope. |
68dc0745 |
606 | |
49d635f9 |
607 | require 5.006; # our() did not exist before 5.6 |
608 | use vars '$var'; |
68dc0745 |
609 | |
49d635f9 |
610 | local $var = "global"; |
611 | my $var = "lexical"; |
612 | |
613 | print "lexical is $var\n"; |
614 | |
615 | { |
ac003c96 |
616 | our $var; |
617 | print "global is $var\n"; |
49d635f9 |
618 | } |
68dc0745 |
619 | |
620 | =head2 What's the difference between deep and shallow binding? |
621 | |
622 | In deep binding, lexical variables mentioned in anonymous subroutines |
623 | are the same ones that were in scope when the subroutine was created. |
624 | In shallow binding, they are whichever variables with the same names |
625 | happen to be in scope when the subroutine is called. Perl always uses |
626 | deep binding of lexical variables (i.e., those created with my()). |
627 | However, dynamic variables (aka global, local, or package variables) |
628 | are effectively shallowly bound. Consider this just one more reason |
629 | not to use them. See the answer to L<"What's a closure?">. |
630 | |
04d666b1 |
631 | =head2 Why doesn't "my($foo) = E<lt>FILEE<gt>;" work right? |
68dc0745 |
632 | |
c8db1d39 |
633 | C<my()> and C<local()> give list context to the right hand side |
c47ff5f1 |
634 | of C<=>. The <FH> read operation, like so many of Perl's |
c8db1d39 |
635 | functions and operators, can tell which context it was called in and |
636 | behaves appropriately. In general, the scalar() function can help. |
637 | This function does nothing to the data itself (contrary to popular myth) |
638 | but rather tells its argument to behave in whatever its scalar fashion is. |
639 | If that function doesn't have a defined scalar behavior, this of course |
640 | doesn't help you (such as with sort()). |
68dc0745 |
641 | |
642 | To enforce scalar context in this particular case, however, you need |
643 | merely omit the parentheses: |
644 | |
ac003c96 |
645 | local($foo) = <FILE>; # WRONG |
646 | local($foo) = scalar(<FILE>); # ok |
647 | local $foo = <FILE>; # right |
68dc0745 |
648 | |
649 | You should probably be using lexical variables anyway, although the |
650 | issue is the same here: |
651 | |
ac003c96 |
652 | my($foo) = <FILE>; # WRONG |
653 | my $foo = <FILE>; # right |
68dc0745 |
654 | |
54310121 |
655 | =head2 How do I redefine a builtin function, operator, or method? |
68dc0745 |
656 | |
657 | Why do you want to do that? :-) |
658 | |
659 | If you want to override a predefined function, such as open(), |
660 | then you'll have to import the new definition from a different |
4a4eefd0 |
661 | module. See L<perlsub/"Overriding Built-in Functions">. There's |
65acb1b1 |
662 | also an example in L<perltoot/"Class::Template">. |
68dc0745 |
663 | |
664 | If you want to overload a Perl operator, such as C<+> or C<**>, |
665 | then you'll want to use the C<use overload> pragma, documented |
666 | in L<overload>. |
667 | |
668 | If you're talking about obscuring method calls in parent classes, |
669 | see L<perltoot/"Overridden Methods">. |
670 | |
671 | =head2 What's the difference between calling a function as &foo and foo()? |
672 | |
673 | When you call a function as C<&foo>, you allow that function access to |
a6dd486b |
674 | your current @_ values, and you bypass prototypes. |
675 | The function doesn't get an empty @_--it gets yours! While not |
68dc0745 |
676 | strictly speaking a bug (it's documented that way in L<perlsub>), it |
677 | would be hard to consider this a feature in most cases. |
678 | |
c8db1d39 |
679 | When you call your function as C<&foo()>, then you I<do> get a new @_, |
68dc0745 |
680 | but prototyping is still circumvented. |
681 | |
682 | Normally, you want to call a function using C<foo()>. You may only |
683 | omit the parentheses if the function is already known to the compiler |
684 | because it already saw the definition (C<use> but not C<require>), |
685 | or via a forward reference or C<use subs> declaration. Even in this |
686 | case, you get a clean @_ without any of the old values leaking through |
687 | where they don't belong. |
688 | |
689 | =head2 How do I create a switch or case statement? |
690 | |
f449fe8a |
691 | If one wants to use pure Perl and to be compatible with Perl versions |
692 | prior to 5.10, the general answer is to write a construct like this: |
c8db1d39 |
693 | |
ac003c96 |
694 | for ($variable_to_test) { |
695 | if (/pat1/) { } # do something |
696 | elsif (/pat2/) { } # do something else |
697 | elsif (/pat3/) { } # do something else |
698 | else { } # default |
699 | } |
68dc0745 |
700 | |
f449fe8a |
701 | Here's a simple example of a switch based on pattern matching, |
702 | lined up in a way to make it look more like a switch statement. |
8305e449 |
703 | We'll do a multiway conditional based on the type of reference stored |
c8db1d39 |
704 | in $whatchamacallit: |
705 | |
706 | SWITCH: for (ref $whatchamacallit) { |
68dc0745 |
707 | |
708 | /^$/ && die "not a reference"; |
709 | |
710 | /SCALAR/ && do { |
711 | print_scalar($$ref); |
712 | last SWITCH; |
713 | }; |
714 | |
715 | /ARRAY/ && do { |
716 | print_array(@$ref); |
717 | last SWITCH; |
718 | }; |
719 | |
720 | /HASH/ && do { |
721 | print_hash(%$ref); |
722 | last SWITCH; |
723 | }; |
724 | |
725 | /CODE/ && do { |
726 | warn "can't print function ref"; |
727 | last SWITCH; |
728 | }; |
729 | |
730 | # DEFAULT |
731 | |
732 | warn "User defined type skipped"; |
733 | |
734 | } |
735 | |
f449fe8a |
736 | See L<perlsyn> for other examples in this style. |
c8db1d39 |
737 | |
738 | Sometimes you should change the positions of the constant and the variable. |
739 | For example, let's say you wanted to test which of many answers you were |
740 | given, but in a case-insensitive way that also allows abbreviations. |
741 | You can use the following technique if the strings all start with |
a6dd486b |
742 | different characters or if you want to arrange the matches so that |
c8db1d39 |
743 | one takes precedence over another, as C<"SEND"> has precedence over |
744 | C<"STOP"> here: |
745 | |
ac003c96 |
746 | chomp($answer = <>); |
747 | if ("SEND" =~ /^\Q$answer/i) { print "Action is send\n" } |
748 | elsif ("STOP" =~ /^\Q$answer/i) { print "Action is stop\n" } |
749 | elsif ("ABORT" =~ /^\Q$answer/i) { print "Action is abort\n" } |
750 | elsif ("LIST" =~ /^\Q$answer/i) { print "Action is list\n" } |
751 | elsif ("EDIT" =~ /^\Q$answer/i) { print "Action is edit\n" } |
c8db1d39 |
752 | |
197aec24 |
753 | A totally different approach is to create a hash of function references. |
c8db1d39 |
754 | |
ac003c96 |
755 | my %commands = ( |
756 | "happy" => \&joy, |
757 | "sad", => \&sullen, |
758 | "done" => sub { die "See ya!" }, |
759 | "mad" => \&angry, |
760 | ); |
761 | |
762 | print "How are you? "; |
763 | chomp($string = <STDIN>); |
764 | if ($commands{$string}) { |
765 | $commands{$string}->(); |
766 | } else { |
767 | print "No such command: $string\n"; |
768 | } |
c8db1d39 |
769 | |
f449fe8a |
770 | Note that starting from version 5.10, Perl has now a native switch |
771 | statement. See L<perlsyn>. |
772 | |
773 | Starting from Perl 5.8, a source filter module, C<Switch>, can also be |
774 | used to get switch and case. Its use is now discouraged, because it's |
775 | not fully compatible with the native switch of Perl 5.10, and because, |
776 | as it's implemented as a source filter, it doesn't always work as intended |
777 | when complex syntax is involved. |
778 | |
49d635f9 |
779 | =head2 How can I catch accesses to undefined variables, functions, or methods? |
68dc0745 |
780 | |
781 | The AUTOLOAD method, discussed in L<perlsub/"Autoloading"> and |
782 | L<perltoot/"AUTOLOAD: Proxy Methods">, lets you capture calls to |
783 | undefined functions and methods. |
784 | |
785 | When it comes to undefined variables that would trigger a warning |
49d635f9 |
786 | under C<use warnings>, you can promote the warning to an error. |
68dc0745 |
787 | |
49d635f9 |
788 | use warnings FATAL => qw(uninitialized); |
68dc0745 |
789 | |
790 | =head2 Why can't a method included in this same file be found? |
791 | |
792 | Some possible reasons: your inheritance is getting confused, you've |
793 | misspelled the method name, or the object is of the wrong type. Check |
a6dd486b |
794 | out L<perltoot> for details about any of the above cases. You may |
795 | also use C<print ref($object)> to find out the class C<$object> was |
796 | blessed into. |
68dc0745 |
797 | |
798 | Another possible reason for problems is because you've used the |
799 | indirect object syntax (eg, C<find Guru "Samy">) on a class name |
800 | before Perl has seen that such a package exists. It's wisest to make |
801 | sure your packages are all defined before you start using them, which |
802 | will be taken care of if you use the C<use> statement instead of |
a6dd486b |
803 | C<require>. If not, make sure to use arrow notation (eg., |
c47ff5f1 |
804 | C<< Guru->find("Samy") >>) instead. Object notation is explained in |
68dc0745 |
805 | L<perlobj>. |
806 | |
c8db1d39 |
807 | Make sure to read about creating modules in L<perlmod> and |
ae93639c |
808 | the perils of indirect objects in L<perlobj/"Method Invocation">. |
c8db1d39 |
809 | |
68dc0745 |
810 | =head2 How can I find out my current package? |
811 | |
812 | If you're just a random program, you can do this to find |
813 | out what the currently compiled package is: |
814 | |
ac003c96 |
815 | my $packname = __PACKAGE__; |
68dc0745 |
816 | |
a6dd486b |
817 | But, if you're a method and you want to print an error message |
68dc0745 |
818 | that includes the kind of object you were called on (which is |
819 | not necessarily the same as the one in which you were compiled): |
820 | |
ac003c96 |
821 | sub amethod { |
822 | my $self = shift; |
823 | my $class = ref($self) || $self; |
824 | warn "called me from a $class object"; |
825 | } |
68dc0745 |
826 | |
46fc3d4c |
827 | =head2 How can I comment out a large block of perl code? |
828 | |
659cfd94 |
829 | You can use embedded POD to discard it. Enclose the blocks you want |
7678cced |
830 | to comment out in POD markers. The <=begin> directive marks a section |
831 | for a specific formatter. Use the C<comment> format, which no formatter |
832 | should claim to understand (by policy). Mark the end of the block |
833 | with <=end>. |
46fc3d4c |
834 | |
ac003c96 |
835 | # program is here |
836 | |
837 | =begin comment |
838 | |
839 | all of this stuff |
840 | |
841 | here will be ignored |
842 | by everyone |
843 | |
7678cced |
844 | =end comment |
ac003c96 |
845 | |
846 | =cut |
847 | |
848 | # program continues |
46fc3d4c |
849 | |
f05bbc40 |
850 | The pod directives cannot go just anywhere. You must put a |
851 | pod directive where the parser is expecting a new statement, |
852 | not just in the middle of an expression or some other |
659cfd94 |
853 | arbitrary grammar production. |
fc36a67e |
854 | |
f05bbc40 |
855 | See L<perlpod> for more details. |
c8db1d39 |
856 | |
65acb1b1 |
857 | =head2 How do I clear a package? |
858 | |
859 | Use this code, provided by Mark-Jason Dominus: |
860 | |
ac003c96 |
861 | sub scrub_package { |
862 | no strict 'refs'; |
863 | my $pack = shift; |
864 | die "Shouldn't delete main package" |
865 | if $pack eq "" || $pack eq "main"; |
866 | my $stash = *{$pack . '::'}{HASH}; |
867 | my $name; |
868 | foreach $name (keys %$stash) { |
869 | my $fullname = $pack . '::' . $name; |
870 | # Get rid of everything with that name. |
871 | undef $$fullname; |
872 | undef @$fullname; |
873 | undef %$fullname; |
874 | undef &$fullname; |
875 | undef *$fullname; |
876 | } |
65acb1b1 |
877 | } |
65acb1b1 |
878 | |
197aec24 |
879 | Or, if you're using a recent release of Perl, you can |
65acb1b1 |
880 | just use the Symbol::delete_package() function instead. |
881 | |
d92eb7b0 |
882 | =head2 How can I use a variable as a variable name? |
883 | |
884 | Beginners often think they want to have a variable contain the name |
885 | of a variable. |
886 | |
ac003c96 |
887 | $fred = 23; |
888 | $varname = "fred"; |
889 | ++$$varname; # $fred now 24 |
d92eb7b0 |
890 | |
891 | This works I<sometimes>, but it is a very bad idea for two reasons. |
892 | |
a6dd486b |
893 | The first reason is that this technique I<only works on global |
894 | variables>. That means that if $fred is a lexical variable created |
895 | with my() in the above example, the code wouldn't work at all: you'd |
896 | accidentally access the global and skip right over the private lexical |
897 | altogether. Global variables are bad because they can easily collide |
898 | accidentally and in general make for non-scalable and confusing code. |
d92eb7b0 |
899 | |
900 | Symbolic references are forbidden under the C<use strict> pragma. |
901 | They are not true references and consequently are not reference counted |
902 | or garbage collected. |
903 | |
904 | The other reason why using a variable to hold the name of another |
a6dd486b |
905 | variable is a bad idea is that the question often stems from a lack of |
d92eb7b0 |
906 | understanding of Perl data structures, particularly hashes. By using |
907 | symbolic references, you are just using the package's symbol-table hash |
908 | (like C<%main::>) instead of a user-defined hash. The solution is to |
909 | use your own hash or a real reference instead. |
910 | |
ac003c96 |
911 | $USER_VARS{"fred"} = 23; |
912 | $varname = "fred"; |
913 | $USER_VARS{$varname}++; # not $$varname++ |
d92eb7b0 |
914 | |
915 | There we're using the %USER_VARS hash instead of symbolic references. |
916 | Sometimes this comes up in reading strings from the user with variable |
917 | references and wanting to expand them to the values of your perl |
918 | program's variables. This is also a bad idea because it conflates the |
919 | program-addressable namespace and the user-addressable one. Instead of |
920 | reading a string and expanding it to the actual contents of your program's |
921 | own variables: |
922 | |
ac003c96 |
923 | $str = 'this has a $fred and $barney in it'; |
924 | $str =~ s/(\$\w+)/$1/eeg; # need double eval |
d92eb7b0 |
925 | |
a6dd486b |
926 | it would be better to keep a hash around like %USER_VARS and have |
d92eb7b0 |
927 | variable references actually refer to entries in that hash: |
928 | |
ac003c96 |
929 | $str =~ s/\$(\w+)/$USER_VARS{$1}/g; # no /e here at all |
d92eb7b0 |
930 | |
931 | That's faster, cleaner, and safer than the previous approach. Of course, |
932 | you don't need to use a dollar sign. You could use your own scheme to |
933 | make it less confusing, like bracketed percent symbols, etc. |
934 | |
ac003c96 |
935 | $str = 'this has a %fred% and %barney% in it'; |
936 | $str =~ s/%(\w+)%/$USER_VARS{$1}/g; # no /e here at all |
d92eb7b0 |
937 | |
a6dd486b |
938 | Another reason that folks sometimes think they want a variable to |
939 | contain the name of a variable is because they don't know how to build |
940 | proper data structures using hashes. For example, let's say they |
941 | wanted two hashes in their program: %fred and %barney, and that they |
942 | wanted to use another scalar variable to refer to those by name. |
d92eb7b0 |
943 | |
ac003c96 |
944 | $name = "fred"; |
945 | $$name{WIFE} = "wilma"; # set %fred |
d92eb7b0 |
946 | |
ac003c96 |
947 | $name = "barney"; |
948 | $$name{WIFE} = "betty"; # set %barney |
d92eb7b0 |
949 | |
950 | This is still a symbolic reference, and is still saddled with the |
951 | problems enumerated above. It would be far better to write: |
952 | |
ac003c96 |
953 | $folks{"fred"}{WIFE} = "wilma"; |
954 | $folks{"barney"}{WIFE} = "betty"; |
d92eb7b0 |
955 | |
956 | And just use a multilevel hash to start with. |
957 | |
958 | The only times that you absolutely I<must> use symbolic references are |
959 | when you really must refer to the symbol table. This may be because it's |
960 | something that can't take a real reference to, such as a format name. |
961 | Doing so may also be important for method calls, since these always go |
962 | through the symbol table for resolution. |
963 | |
964 | In those cases, you would turn off C<strict 'refs'> temporarily so you |
965 | can play around with the symbol table. For example: |
966 | |
ac003c96 |
967 | @colors = qw(red blue green yellow orange purple violet); |
968 | for my $name (@colors) { |
969 | no strict 'refs'; # renege for the block |
970 | *$name = sub { "<FONT COLOR='$name'>@_</FONT>" }; |
971 | } |
d92eb7b0 |
972 | |
973 | All those functions (red(), blue(), green(), etc.) appear to be separate, |
974 | but the real code in the closure actually was compiled only once. |
975 | |
976 | So, sometimes you might want to use symbolic references to directly |
977 | manipulate the symbol table. This doesn't matter for formats, handles, and |
a6dd486b |
978 | subroutines, because they are always global--you can't use my() on them. |
979 | For scalars, arrays, and hashes, though--and usually for subroutines-- |
980 | you probably only want to use hard references. |
d92eb7b0 |
981 | |
5cd0b561 |
982 | =head2 What does "bad interpreter" mean? |
983 | |
571e049f |
984 | (contributed by brian d foy) |
985 | |
5cd0b561 |
986 | The "bad interpreter" message comes from the shell, not perl. The |
987 | actual message may vary depending on your platform, shell, and locale |
988 | settings. |
989 | |
990 | If you see "bad interpreter - no such file or directory", the first |
991 | line in your perl script (the "shebang" line) does not contain the |
6670e5e7 |
992 | right path to perl (or any other program capable of running scripts). |
5cd0b561 |
993 | Sometimes this happens when you move the script from one machine to |
ac9dac7f |
994 | another and each machine has a different path to perl--/usr/bin/perl |
571e049f |
995 | versus /usr/local/bin/perl for instance. It may also indicate |
6670e5e7 |
996 | that the source machine has CRLF line terminators and the |
997 | destination machine has LF only: the shell tries to find |
571e049f |
998 | /usr/bin/perl<CR>, but can't. |
5cd0b561 |
999 | |
1000 | If you see "bad interpreter: Permission denied", you need to make your |
1001 | script executable. |
1002 | |
1003 | In either case, you should still be able to run the scripts with perl |
1004 | explicitly: |
1005 | |
1006 | % perl script.pl |
1007 | |
1008 | If you get a message like "perl: command not found", perl is not in |
1009 | your PATH, which might also mean that the location of perl is not |
1010 | where you expect it so you need to adjust your shebang line. |
1011 | |
500071f4 |
1012 | =head1 REVISION |
1013 | |
c195e131 |
1014 | Revision: $Revision: 10100 $ |
500071f4 |
1015 | |
c195e131 |
1016 | Date: $Date: 2007-10-21 20:59:30 +0200 (Sun, 21 Oct 2007) $ |
500071f4 |
1017 | |
1018 | See L<perlfaq> for source control details and availability. |
1019 | |
68dc0745 |
1020 | =head1 AUTHOR AND COPYRIGHT |
1021 | |
ee891a00 |
1022 | Copyright (c) 1997-2007 Tom Christiansen, Nathan Torkington, and |
7678cced |
1023 | other authors as noted. All rights reserved. |
5a964f20 |
1024 | |
5a7beb56 |
1025 | This documentation is free; you can redistribute it and/or modify it |
1026 | under the same terms as Perl itself. |
5a964f20 |
1027 | |
1028 | Irrespective of its distribution, all code examples in this file |
1029 | are hereby placed into the public domain. You are permitted and |
1030 | encouraged to use this code in your own programs for fun |
1031 | or for profit as you see fit. A simple comment in the code giving |
1032 | credit would be courteous but is not required. |
a6dd486b |
1033 | |