2 X<warning, lexical> X<warnings> X<warning>
4 perllexwarn - Perl Lexical Warnings
8 The C<use warnings> pragma enables to control precisely what warnings are
9 to be enabled in which parts of a Perl program. It's a more flexible
10 alternative for both the command line flag B<-w> and the equivalent Perl
13 This pragma works just like the C<strict> pragma.
14 This means that the scope of the warning pragma is limited to the
15 enclosing block. It also means that the pragma setting will not
16 leak across files (via C<use>, C<require> or C<do>). This allows
17 authors to independently define the degree of warning checks that will
18 be applied to their module.
20 By default, optional warnings are disabled, so any legacy code that
21 doesn't attempt to control the warnings will work unchanged.
23 All warnings are enabled in a block by either of these:
28 Similarly all warnings are disabled in a block by either of these:
33 For example, consider the code below:
43 The code in the enclosing block has warnings enabled, but the inner
44 block has them disabled. In this case that means the assignment to the
45 scalar C<$c> will trip the C<"Scalar value @a[0] better written as $a[0]">
46 warning, but the assignment to the scalar C<$b> will not.
48 =head2 Default Warnings and Optional Warnings
50 Before the introduction of lexical warnings, Perl had two classes of
51 warnings: mandatory and optional.
53 As its name suggests, if your code tripped a mandatory warning, you
54 would get a warning whether you wanted it or not.
55 For example, the code below would always produce an C<"isn't numeric">
56 warning about the "2:".
60 With the introduction of lexical warnings, mandatory warnings now become
61 I<default> warnings. The difference is that although the previously
62 mandatory warnings are still enabled by default, they can then be
63 subsequently enabled or disabled with the lexical warning pragma. For
64 example, in the code below, an C<"isn't numeric"> warning will only
65 be reported for the C<$a> variable.
71 Note that neither the B<-w> flag or the C<$^W> can be used to
72 disable/enable default warnings. They are still mandatory in this case.
74 =head2 What's wrong with B<-w> and C<$^W>
76 Although very useful, the big problem with using B<-w> on the command
77 line to enable warnings is that it is all or nothing. Take the typical
78 scenario when you are writing a Perl program. Parts of the code you
79 will write yourself, but it's very likely that you will make use of
80 pre-written Perl modules. If you use the B<-w> flag in this case, you
81 end up enabling warnings in pieces of code that you haven't written.
83 Similarly, using C<$^W> to either disable or enable blocks of code is
84 fundamentally flawed. For a start, say you want to disable warnings in
85 a block of code. You might expect this to be enough to do the trick:
93 When this code is run with the B<-w> flag, a warning will be produced
94 for the C<$a> line: C<"Reversed += operator">.
96 The problem is that Perl has both compile-time and run-time warnings. To
97 disable compile-time warnings you need to rewrite the code like this:
105 The other big problem with C<$^W> is the way you can inadvertently
106 change the warning setting in unexpected places in your code. For example,
107 when the code below is run (without the B<-w> flag), the second call
108 to C<doit> will trip a C<"Use of uninitialized value"> warning, whereas
123 This is a side-effect of C<$^W> being dynamically scoped.
125 Lexical warnings get around these limitations by allowing finer control
126 over where warnings can or can't be tripped.
128 =head2 Controlling Warnings from the Command Line
130 There are three Command Line flags that can be used to control when
131 warnings are (or aren't) produced:
138 This is the existing flag. If the lexical warnings pragma is B<not>
139 used in any of you code, or any of the modules that you use, this flag
140 will enable warnings everywhere. See L<Backward Compatibility> for
141 details of how this flag interacts with lexical warnings.
146 If the B<-W> flag is used on the command line, it will enable all warnings
147 throughout the program regardless of whether warnings were disabled
148 locally using C<no warnings> or C<$^W =0>. This includes all files that get
149 included via C<use>, C<require> or C<do>.
150 Think of it as the Perl equivalent of the "lint" command.
155 Does the exact opposite to the B<-W> flag, i.e. it disables all warnings.
159 =head2 Backward Compatibility
161 If you are used to working with a version of Perl prior to the
162 introduction of lexically scoped warnings, or have code that uses both
163 lexical warnings and C<$^W>, this section will describe how they interact.
165 How Lexical Warnings interact with B<-w>/C<$^W>:
171 If none of the three command line flags (B<-w>, B<-W> or B<-X>) that
172 control warnings is used and neither C<$^W> nor the C<warnings> pragma
173 are used, then default warnings will be enabled and optional warnings
175 This means that legacy code that doesn't attempt to control the warnings
180 The B<-w> flag just sets the global C<$^W> variable as in 5.005. This
181 means that any legacy code that currently relies on manipulating C<$^W>
182 to control warning behavior will still work as is.
186 Apart from now being a boolean, the C<$^W> variable operates in exactly
187 the same horrible uncontrolled global way, except that it cannot
188 disable/enable default warnings.
192 If a piece of code is under the control of the C<warnings> pragma,
193 both the C<$^W> variable and the B<-w> flag will be ignored for the
194 scope of the lexical warning.
198 The only way to override a lexical warnings setting is with the B<-W>
199 or B<-X> command line flags.
203 The combined effect of 3 & 4 is that it will allow code which uses
204 the C<warnings> pragma to control the warning behavior of $^W-type
205 code (using a C<local $^W=0>) if it really wants to, but not vice-versa.
207 =head2 Category Hierarchy
208 X<warning, categories>
210 A hierarchy of "categories" have been defined to allow groups of warnings
211 to be enabled/disabled in isolation.
213 The current hierarchy is:
225 | +- experimental::lexical_subs
321 Just like the "strict" pragma any of these categories can be combined
323 use warnings qw(void redefine);
324 no warnings qw(io syntax untie);
326 Also like the "strict" pragma, if there is more than one instance of the
327 C<warnings> pragma in a given scope the cumulative effect is additive.
329 use warnings qw(void); # only "void" warnings enabled
331 use warnings qw(io); # only "void" & "io" warnings enabled
333 no warnings qw(void); # only "io" warnings enabled
335 To determine which category a specific warning has been assigned to see
338 Note: In Perl 5.6.1, the lexical warnings category "deprecated" was a
339 sub-category of the "syntax" category. It is now a top-level category
342 =head2 Fatal Warnings
345 The presence of the word "FATAL" in the category list will escalate any
346 warnings detected from the categories specified in the lexical scope
347 into fatal errors. In the code below, the use of C<time>, C<length>
348 and C<join> can all produce a C<"Useless use of xxx in void context">
356 use warnings FATAL => qw(void);
364 When run it produces this output
366 Useless use of time in void context at fatal line 3.
367 Useless use of length in void context at fatal line 7.
369 The scope where C<length> is used has escalated the C<void> warnings
370 category into a fatal error, so the program terminates immediately it
371 encounters the warning.
373 To explicitly turn off a "FATAL" warning you just disable the warning
374 it is associated with. So, for example, to disable the "void" warning
375 in the example above, either of these will do the trick:
377 no warnings qw(void);
378 no warnings FATAL => qw(void);
380 If you want to downgrade a warning that has been escalated into a fatal
381 error back to a normal warning, you can use the "NONFATAL" keyword. For
382 example, the code below will promote all warnings into fatal errors,
383 except for those in the "syntax" category.
385 use warnings FATAL => 'all', NONFATAL => 'syntax';
387 =head2 Reporting Warnings from a Module
388 X<warning, reporting> X<warning, registering>
390 The C<warnings> pragma provides a number of functions that are useful for
391 module authors. These are used when you want to report a module-specific
392 warning to a calling module has enabled warnings via the C<warnings>
395 Consider the module C<MyMod::Abc> below.
399 use warnings::register;
403 if ($path !~ m#^/#) {
404 warnings::warn("changing relative path to /var/abc")
405 if warnings::enabled();
406 $path = "/var/abc/$path";
412 The call to C<warnings::register> will create a new warnings category
413 called "MyMod::Abc", i.e. the new category name matches the current
414 package name. The C<open> function in the module will display a warning
415 message if it gets given a relative path as a parameter. This warnings
416 will only be displayed if the code that uses C<MyMod::Abc> has actually
417 enabled them with the C<warnings> pragma like below.
420 use warnings 'MyMod::Abc';
422 abc::open("../fred.txt");
424 It is also possible to test whether the pre-defined warnings categories are
425 set in the calling module with the C<warnings::enabled> function. Consider
426 this snippet of code:
431 warnings::warnif("deprecated",
432 "open is deprecated, use new instead");
440 The function C<open> has been deprecated, so code has been included to
441 display a warning message whenever the calling module has (at least) the
442 "deprecated" warnings category enabled. Something like this, say.
444 use warnings 'deprecated';
447 MyMod::Abc::open($filename);
449 Either the C<warnings::warn> or C<warnings::warnif> function should be
450 used to actually display the warnings message. This is because they can
451 make use of the feature that allows warnings to be escalated into fatal
452 errors. So in this case
455 use warnings FATAL => 'MyMod::Abc';
457 MyMod::Abc::open('../fred.txt');
459 the C<warnings::warnif> function will detect this and die after
460 displaying the warning message.
462 The three warnings functions, C<warnings::warn>, C<warnings::warnif>
463 and C<warnings::enabled> can optionally take an object reference in place
464 of a category name. In this case the functions will use the class name
465 of the object as the warnings category.
467 Consider this example:
472 use warnings::register;
485 if ($value % 2 && warnings::enabled($self))
486 { warnings::warn($self, "Odd numbers are unsafe") }
493 $self->check($value);
501 use warnings::register;
503 our @ISA = qw( Original );
513 The code below makes use of both modules, but it only enables warnings from
518 use warnings 'Derived';
519 my $a = Original->new();
521 my $b = Derived->new();
524 When this code is run only the C<Derived> object, C<$b>, will generate
527 Odd numbers are unsafe at main.pl line 7
529 Notice also that the warning is reported at the line where the object is first
532 When registering new categories of warning, you can supply more names to
533 warnings::register like this:
536 use warnings::register qw(format precision);
540 warnings::warnif('MyModule::format', '...');
544 L<warnings>, L<perldiag>.