3 perlvar - Perl predefined variables
7 =head2 Predefined Names
9 The following names have special meaning to Perl. Most
10 punctuation names have reasonable mnemonics, or analogs in the
11 shells. Nevertheless, if you wish to use long variable names,
16 at the top of your program. This aliases all the short names to the long
17 names in the current package. Some even have medium names, generally
18 borrowed from B<awk>. In general, it's best to use the
20 use English '-no_match_vars';
22 invocation if you don't need $PREMATCH, $MATCH, or $POSTMATCH, as it avoids
23 a certain performance hit with the use of regular expressions. See
26 Variables that depend on the currently selected filehandle may be set by
27 calling an appropriate object method on the IO::Handle object, although
28 this is less efficient than using the regular built-in variables. (Summary
29 lines below for this contain the word HANDLE.) First you must say
33 after which you may use either
41 Each method returns the old value of the IO::Handle attribute.
42 The methods each take an optional EXPR, which, if supplied, specifies the
43 new value for the IO::Handle attribute in question. If not supplied,
44 most methods do nothing to the current value--except for
45 autoflush(), which will assume a 1 for you, just to be different.
47 Because loading in the IO::Handle class is an expensive operation, you should
48 learn how to use the regular built-in variables.
50 A few of these variables are considered "read-only". This means that if
51 you try to assign to this variable, either directly or indirectly through
52 a reference, you'll raise a run-time exception.
54 You should be very careful when modifying the default values of most
55 special variables described in this document. In most cases you want
56 to localize these variables before changing them, since if you don't,
57 the change may affect other modules which rely on the default values
58 of the special variables that you have changed. This is one of the
59 correct ways to read the whole file at once:
61 open my $fh, "<", "foo" or die $!;
62 local $/; # enable localized slurp mode
66 But the following code is quite bad:
68 open my $fh, "<", "foo" or die $!;
69 undef $/; # enable slurp mode
73 since some other module, may want to read data from some file in the
74 default "line mode", so if the code we have just presented has been
75 executed, the global value of C<$/> is now changed for any other code
76 running inside the same Perl interpreter.
78 Usually when a variable is localized you want to make sure that this
79 change affects the shortest scope possible. So unless you are already
80 inside some short C<{}> block, you should create one yourself. For
84 open my $fh, "<", "foo" or die $!;
91 Here is an example of how your own code can go broken:
99 # do something with $_
102 You probably expect this code to print:
110 Why? Because nasty_break() modifies C<$_> without localizing it
111 first. The fix is to add local():
115 It's easy to notice the problem in such a short example, but in more
116 complicated code you are looking for trouble if you don't localize
117 changes to the special variables.
119 The following list is ordered by scalar variables first, then the
120 arrays, then the hashes.
129 The default input and pattern-searching space. The following pairs are
132 while (<>) {...} # equivalent only in while!
133 while (defined($_ = <>)) {...}
144 Here are the places where Perl will assume $_ even if you
151 The following functions:
153 abs, alarm, chomp, chop, chr, chroot, cos, defined, eval, exp, glob,
154 hex, int, lc, lcfirst, length, log, lstat, mkdir, oct, ord, pos, print,
155 quotemeta, readlink, readpipe, ref, require, reverse (in scalar context only),
156 rmdir, sin, split (on its second argument), sqrt, stat, study, uc, ucfirst,
161 All file tests (C<-f>, C<-d>) except for C<-t>, which defaults to STDIN.
167 The pattern matching operations C<m//>, C<s///> and C<tr///> (aka C<y///>)
168 when used without an C<=~> operator.
172 The default iterator variable in a C<foreach> loop if no other
173 variable is supplied.
177 The implicit iterator variable in the grep() and map() functions.
181 The implicit variable of given().
185 The default place to put an input record when a C<< <FH> >>
186 operation's result is tested by itself as the sole criterion of a C<while>
187 test. Outside a C<while> test, this will not happen.
191 As C<$_> is a global variable, this may lead in some cases to unwanted
192 side-effects. As of perl 5.9.1, you can now use a lexical version of
193 C<$_> by declaring it in a file or in a block with C<my>. Moreover,
194 declaring C<our $_> restores the global C<$_> in the current scope.
196 (Mnemonic: underline is understood in certain operations.)
207 Special package variables when using sort(), see L<perlfunc/sort>.
208 Because of this specialness $a and $b don't need to be declared
209 (using use vars, or our()) even when using the C<strict 'vars'> pragma.
210 Don't lexicalize them with C<my $a> or C<my $b> if you want to be
211 able to use them in the sort() comparison block or function.
220 Contains the subpattern from the corresponding set of capturing
221 parentheses from the last pattern match, not counting patterns
222 matched in nested blocks that have been exited already. (Mnemonic:
223 like \digits.) These variables are all read-only and dynamically
224 scoped to the current BLOCK.
231 The string matched by the last successful pattern match (not counting
232 any matches hidden within a BLOCK or eval() enclosed by the current
233 BLOCK). (Mnemonic: like & in some editors.) This variable is read-only
234 and dynamically scoped to the current BLOCK.
236 The use of this variable anywhere in a program imposes a considerable
237 performance penalty on all regular expression matches. See L</BUGS>.
239 See L</@-> for a replacement.
244 This is similar to C<$&> (C<$POSTMATCH>) except that it does not incur the
245 performance penalty associated with that variable, and is only guaranteed
246 to return a defined value when the pattern was compiled or executed with
254 The string preceding whatever was matched by the last successful
255 pattern match (not counting any matches hidden within a BLOCK or eval
256 enclosed by the current BLOCK). (Mnemonic: C<`> often precedes a quoted
257 string.) This variable is read-only.
259 The use of this variable anywhere in a program imposes a considerable
260 performance penalty on all regular expression matches. See L</BUGS>.
262 See L</@-> for a replacement.
267 This is similar to C<$`> ($PREMATCH) except that it does not incur the
268 performance penalty associated with that variable, and is only guaranteed
269 to return a defined value when the pattern was compiled or executed with
277 The string following whatever was matched by the last successful
278 pattern match (not counting any matches hidden within a BLOCK or eval()
279 enclosed by the current BLOCK). (Mnemonic: C<'> often follows a quoted
282 local $_ = 'abcdefghi';
284 print "$`:$&:$'\n"; # prints abc:def:ghi
286 This variable is read-only and dynamically scoped to the current BLOCK.
288 The use of this variable anywhere in a program imposes a considerable
289 performance penalty on all regular expression matches. See L</BUGS>.
291 See L</@-> for a replacement.
296 This is similar to C<$'> (C<$POSTMATCH>) except that it does not incur the
297 performance penalty associated with that variable, and is only guaranteed
298 to return a defined value when the pattern was compiled or executed with
301 =item $LAST_PAREN_MATCH
304 X<$+> X<$LAST_PAREN_MATCH>
306 The text matched by the last bracket of the last successful search pattern.
307 This is useful if you don't know which one of a set of alternative patterns
308 matched. For example:
310 /Version: (.*)|Revision: (.*)/ && ($rev = $+);
312 (Mnemonic: be positive and forward looking.)
313 This variable is read-only and dynamically scoped to the current BLOCK.
315 =item $LAST_SUBMATCH_RESULT
320 The text matched by the used group most-recently closed (i.e. the group
321 with the rightmost closing parenthesis) of the last successful search
322 pattern. (Mnemonic: the (possibly) Nested parenthesis that most
325 This is primarily used inside C<(?{...})> blocks for examining text
326 recently matched. For example, to effectively capture text to a variable
327 (in addition to C<$1>, C<$2>, etc.), replace C<(...)> with
329 (?:(...)(?{ $var = $^N }))
331 By setting and then using C<$var> in this way relieves you from having to
332 worry about exactly which numbered set of parentheses they are.
334 This variable is dynamically scoped to the current BLOCK.
336 =item @LAST_MATCH_END
339 X<@+> X<@LAST_MATCH_END>
341 This array holds the offsets of the ends of the last successful
342 submatches in the currently active dynamic scope. C<$+[0]> is
343 the offset into the string of the end of the entire match. This
344 is the same value as what the C<pos> function returns when called
345 on the variable that was matched against. The I<n>th element
346 of this array holds the offset of the I<n>th submatch, so
347 C<$+[1]> is the offset past where $1 ends, C<$+[2]> the offset
348 past where $2 ends, and so on. You can use C<$#+> to determine
349 how many subgroups were in the last successful match. See the
350 examples given for the C<@-> variable.
355 Similar to C<@+>, the C<%+> hash allows access to the named capture
356 buffers, should they exist, in the last successful match in the
357 currently active dynamic scope.
359 For example, C<$+{foo}> is equivalent to C<$1> after the following match:
361 'foo' =~ /(?<foo>foo)/;
363 The keys of the C<%+> hash list only the names of buffers that have
364 captured (and that are thus associated to defined values).
366 The underlying behaviour of C<%+> is provided by the
367 L<Tie::Hash::NamedCapture> module.
369 B<Note:> C<%-> and C<%+> are tied views into a common internal hash
370 associated with the last successful regular expression. Therefore mixing
371 iterative access to them via C<each> may have unpredictable results.
372 Likewise, if the last successful match changes, then the results may be
375 =item HANDLE->input_line_number(EXPR)
377 =item $INPUT_LINE_NUMBER
382 X<$.> X<$NR> X<$INPUT_LINE_NUMBER> X<line number>
384 Current line number for the last filehandle accessed.
386 Each filehandle in Perl counts the number of lines that have been read
387 from it. (Depending on the value of C<$/>, Perl's idea of what
388 constitutes a line may not match yours.) When a line is read from a
389 filehandle (via readline() or C<< <> >>), or when tell() or seek() is
390 called on it, C<$.> becomes an alias to the line counter for that
393 You can adjust the counter by assigning to C<$.>, but this will not
394 actually move the seek pointer. I<Localizing C<$.> will not localize
395 the filehandle's line count>. Instead, it will localize perl's notion
396 of which filehandle C<$.> is currently aliased to.
398 C<$.> is reset when the filehandle is closed, but B<not> when an open
399 filehandle is reopened without an intervening close(). For more
400 details, see L<perlop/"IE<sol>O Operators">. Because C<< <> >> never does
401 an explicit close, line numbers increase across ARGV files (but see
402 examples in L<perlfunc/eof>).
404 You can also use C<< HANDLE->input_line_number(EXPR) >> to access the
405 line counter for a given filehandle without having to worry about
406 which handle you last accessed.
408 (Mnemonic: many programs use "." to mean the current line number.)
410 =item IO::Handle->input_record_separator(EXPR)
412 =item $INPUT_RECORD_SEPARATOR
417 X<$/> X<$RS> X<$INPUT_RECORD_SEPARATOR>
419 The input record separator, newline by default. This
420 influences Perl's idea of what a "line" is. Works like B<awk>'s RS
421 variable, including treating empty lines as a terminator if set to
422 the null string. (An empty line cannot contain any spaces
423 or tabs.) You may set it to a multi-character string to match a
424 multi-character terminator, or to C<undef> to read through the end
425 of file. Setting it to C<"\n\n"> means something slightly
426 different than setting to C<"">, if the file contains consecutive
427 empty lines. Setting to C<""> will treat two or more consecutive
428 empty lines as a single empty line. Setting to C<"\n\n"> will
429 blindly assume that the next input character belongs to the next
430 paragraph, even if it's a newline. (Mnemonic: / delimits
431 line boundaries when quoting poetry.)
433 local $/; # enable "slurp" mode
434 local $_ = <FH>; # whole file now here
437 Remember: the value of C<$/> is a string, not a regex. B<awk> has to be
438 better for something. :-)
440 Setting C<$/> to a reference to an integer, scalar containing an integer, or
441 scalar that's convertible to an integer will attempt to read records
442 instead of lines, with the maximum record size being the referenced
445 local $/ = \32768; # or \"32768", or \$var_containing_32768
446 open my $fh, "<", $myfile or die $!;
449 will read a record of no more than 32768 bytes from FILE. If you're
450 not reading from a record-oriented file (or your OS doesn't have
451 record-oriented files), then you'll likely get a full chunk of data
452 with every read. If a record is larger than the record size you've
453 set, you'll get the record back in pieces. Trying to set the record
454 size to zero or less will cause reading in the (rest of the) whole file.
456 On VMS, record reads are done with the equivalent of C<sysread>,
457 so it's best not to mix record and non-record reads on the same
458 file. (This is unlikely to be a problem, because any file you'd
459 want to read in record mode is probably unusable in line mode.)
460 Non-VMS systems do normal I/O, so it's safe to mix record and
461 non-record reads of a file.
463 See also L<perlport/"Newlines">. Also see C<$.>.
465 =item HANDLE->autoflush(EXPR)
467 =item $OUTPUT_AUTOFLUSH
470 X<$|> X<autoflush> X<flush> X<$OUTPUT_AUTOFLUSH>
472 If set to nonzero, forces a flush right away and after every write
473 or print on the currently selected output channel. Default is 0
474 (regardless of whether the channel is really buffered by the
475 system or not; C<$|> tells you only whether you've asked Perl
476 explicitly to flush after each write). STDOUT will
477 typically be line buffered if output is to the terminal and block
478 buffered otherwise. Setting this variable is useful primarily when
479 you are outputting to a pipe or socket, such as when you are running
480 a Perl program under B<rsh> and want to see the output as it's
481 happening. This has no effect on input buffering. See L<perlfunc/getc>
482 for that. See L<perldoc/select> on how to select the output channel.
483 See also L<IO::Handle>. (Mnemonic: when you want your pipes to be piping hot.)
485 =item IO::Handle->output_field_separator EXPR
487 =item $OUTPUT_FIELD_SEPARATOR
492 X<$,> X<$OFS> X<$OUTPUT_FIELD_SEPARATOR>
494 The output field separator for the print operator. If defined, this
495 value is printed between each of print's arguments. Default is C<undef>.
496 (Mnemonic: what is printed when there is a "," in your print statement.)
498 =item IO::Handle->output_record_separator EXPR
500 =item $OUTPUT_RECORD_SEPARATOR
505 X<$\> X<$ORS> X<$OUTPUT_RECORD_SEPARATOR>
507 The output record separator for the print operator. If defined, this
508 value is printed after the last of print's arguments. Default is C<undef>.
509 (Mnemonic: you set C<$\> instead of adding "\n" at the end of the print.
510 Also, it's just like C<$/>, but it's what you get "back" from Perl.)
512 =item $LIST_SEPARATOR
515 X<$"> X<$LIST_SEPARATOR>
517 This is like C<$,> except that it applies to array and slice values
518 interpolated into a double-quoted string (or similar interpreted
519 string). Default is a space. (Mnemonic: obvious, I think.)
521 =item $SUBSCRIPT_SEPARATOR
526 X<$;> X<$SUBSEP> X<SUBSCRIPT_SEPARATOR>
528 The subscript separator for multidimensional array emulation. If you
529 refer to a hash element as
535 $foo{join($;, $a, $b, $c)}
539 @foo{$a,$b,$c} # a slice--note the @
543 ($foo{$a},$foo{$b},$foo{$c})
545 Default is "\034", the same as SUBSEP in B<awk>. If your
546 keys contain binary data there might not be any safe value for C<$;>.
547 (Mnemonic: comma (the syntactic subscript separator) is a
548 semi-semicolon. Yeah, I know, it's pretty lame, but C<$,> is already
549 taken for something more important.)
551 Consider using "real" multidimensional arrays as described
554 =item HANDLE->format_page_number(EXPR)
556 =item $FORMAT_PAGE_NUMBER
559 X<$%> X<$FORMAT_PAGE_NUMBER>
561 The current page number of the currently selected output channel.
563 (Mnemonic: % is page number in B<nroff>.)
565 =item HANDLE->format_lines_per_page(EXPR)
567 =item $FORMAT_LINES_PER_PAGE
570 X<$=> X<$FORMAT_LINES_PER_PAGE>
572 The current page length (printable lines) of the currently selected
573 output channel. Default is 60.
575 (Mnemonic: = has horizontal lines.)
577 =item HANDLE->format_lines_left(EXPR)
579 =item $FORMAT_LINES_LEFT
582 X<$-> X<$FORMAT_LINES_LEFT>
584 The number of lines left on the page of the currently selected output
587 (Mnemonic: lines_on_page - lines_printed.)
589 =item @LAST_MATCH_START
592 X<@-> X<@LAST_MATCH_START>
594 $-[0] is the offset of the start of the last successful match.
595 C<$-[>I<n>C<]> is the offset of the start of the substring matched by
596 I<n>-th subpattern, or undef if the subpattern did not match.
598 Thus after a match against $_, $& coincides with C<substr $_, $-[0],
599 $+[0] - $-[0]>. Similarly, $I<n> coincides with C<substr $_, $-[n],
600 $+[n] - $-[n]> if C<$-[n]> is defined, and $+ coincides with
601 C<substr $_, $-[$#-], $+[$#-] - $-[$#-]>. One can use C<$#-> to find the last
602 matched subgroup in the last successful match. Contrast with
603 C<$#+>, the number of subgroups in the regular expression. Compare
606 This array holds the offsets of the beginnings of the last
607 successful submatches in the currently active dynamic scope.
608 C<$-[0]> is the offset into the string of the beginning of the
609 entire match. The I<n>th element of this array holds the offset
610 of the I<n>th submatch, so C<$-[1]> is the offset where $1
611 begins, C<$-[2]> the offset where $2 begins, and so on.
613 After a match against some variable $var:
617 =item C<$`> is the same as C<substr($var, 0, $-[0])>
619 =item C<$&> is the same as C<substr($var, $-[0], $+[0] - $-[0])>
621 =item C<$'> is the same as C<substr($var, $+[0])>
623 =item C<$1> is the same as C<substr($var, $-[1], $+[1] - $-[1])>
625 =item C<$2> is the same as C<substr($var, $-[2], $+[2] - $-[2])>
627 =item C<$3> is the same as C<substr($var, $-[3], $+[3] - $-[3])>
634 Similar to C<%+>, this variable allows access to the named capture buffers
635 in the last successful match in the currently active dynamic scope. To
636 each capture buffer name found in the regular expression, it associates a
637 reference to an array containing the list of values captured by all
638 buffers with that name (should there be several of them), in the order
643 if ('1234' =~ /(?<A>1)(?<B>2)(?<A>3)(?<B>4)/) {
644 foreach my $bufname (sort keys %-) {
645 my $ary = $-{$bufname};
646 foreach my $idx (0..$#$ary) {
647 print "\$-{$bufname}[$idx] : ",
648 (defined($ary->[$idx]) ? "'$ary->[$idx]'" : "undef"),
661 The keys of the C<%-> hash correspond to all buffer names found in
662 the regular expression.
664 The behaviour of C<%-> is implemented via the
665 L<Tie::Hash::NamedCapture> module.
667 B<Note:> C<%-> and C<%+> are tied views into a common internal hash
668 associated with the last successful regular expression. Therefore mixing
669 iterative access to them via C<each> may have unpredictable results.
670 Likewise, if the last successful match changes, then the results may be
673 =item HANDLE->format_name(EXPR)
678 X<$~> X<$FORMAT_NAME>
680 The name of the current report format for the currently selected output
681 channel. Default is the name of the filehandle. (Mnemonic: brother to
684 =item HANDLE->format_top_name(EXPR)
686 =item $FORMAT_TOP_NAME
689 X<$^> X<$FORMAT_TOP_NAME>
691 The name of the current top-of-page format for the currently selected
692 output channel. Default is the name of the filehandle with _TOP
693 appended. (Mnemonic: points to top of page.)
695 =item IO::Handle->format_line_break_characters EXPR
697 =item $FORMAT_LINE_BREAK_CHARACTERS
700 X<$:> X<FORMAT_LINE_BREAK_CHARACTERS>
702 The current set of characters after which a string may be broken to
703 fill continuation fields (starting with ^) in a format. Default is
704 S<" \n-">, to break on whitespace or hyphens. (Mnemonic: a "colon" in
705 poetry is a part of a line.)
707 =item IO::Handle->format_formfeed EXPR
709 =item $FORMAT_FORMFEED
712 X<$^L> X<$FORMAT_FORMFEED>
714 What formats output as a form feed. Default is \f.
719 X<$^A> X<$ACCUMULATOR>
721 The current value of the write() accumulator for format() lines. A format
722 contains formline() calls that put their result into C<$^A>. After
723 calling its format, write() prints out the contents of C<$^A> and empties.
724 So you never really see the contents of C<$^A> unless you call
725 formline() yourself and then look at it. See L<perlform> and
726 L<perlfunc/formline()>.
731 X<$?> X<$CHILD_ERROR>
733 The status returned by the last pipe close, backtick (C<``>) command,
734 successful call to wait() or waitpid(), or from the system()
735 operator. This is just the 16-bit status word returned by the
736 traditional Unix wait() system call (or else is made up to look like it). Thus, the
737 exit value of the subprocess is really (C<<< $? >> 8 >>>), and
738 C<$? & 127> gives which signal, if any, the process died from, and
739 C<$? & 128> reports whether there was a core dump. (Mnemonic:
740 similar to B<sh> and B<ksh>.)
742 Additionally, if the C<h_errno> variable is supported in C, its value
743 is returned via $? if any C<gethost*()> function fails.
745 If you have installed a signal handler for C<SIGCHLD>, the
746 value of C<$?> will usually be wrong outside that handler.
748 Inside an C<END> subroutine C<$?> contains the value that is going to be
749 given to C<exit()>. You can modify C<$?> in an C<END> subroutine to
750 change the exit status of your program. For example:
753 $? = 1 if $? == 255; # die would make it 255
756 Under VMS, the pragma C<use vmsish 'status'> makes C<$?> reflect the
757 actual VMS exit status, instead of the default emulation of POSIX
758 status; see L<perlvms/$?> for details.
760 Also see L<Error Indicators>.
762 =item ${^CHILD_ERROR_NATIVE}
763 X<$^CHILD_ERROR_NATIVE>
765 The native status returned by the last pipe close, backtick (C<``>)
766 command, successful call to wait() or waitpid(), or from the system()
767 operator. On POSIX-like systems this value can be decoded with the
768 WIFEXITED, WEXITSTATUS, WIFSIGNALED, WTERMSIG, WIFSTOPPED, WSTOPSIG
769 and WIFCONTINUED functions provided by the L<POSIX> module.
771 Under VMS this reflects the actual VMS exit status; i.e. it is the same
772 as $? when the pragma C<use vmsish 'status'> is in effect.
777 The I<object reference> to the Encode object that is used to convert
778 the source code to Unicode. Thanks to this variable your perl script
779 does not have to be written in UTF-8. Default is I<undef>. The direct
780 manipulation of this variable is highly discouraged.
787 X<$!> X<$ERRNO> X<$OS_ERROR>
789 If used numerically, yields the current value of the C C<errno>
790 variable, or in other words, if a system or library call fails, it
791 sets this variable. This means that the value of C<$!> is meaningful
792 only I<immediately> after a B<failure>:
794 if (open my $fh, "<", $filename) {
795 # Here $! is meaningless.
798 # ONLY here is $! meaningful.
800 # Already here $! might be meaningless.
802 # Since here we might have either success or failure,
803 # here $! is meaningless.
805 In the above I<meaningless> stands for anything: zero, non-zero,
806 C<undef>. A successful system or library call does B<not> set
807 the variable to zero.
809 If used as a string, yields the corresponding system error string.
810 You can assign a number to C<$!> to set I<errno> if, for instance,
811 you want C<"$!"> to return the string for error I<n>, or you want
812 to set the exit value for the die() operator. (Mnemonic: What just
815 Also see L<Error Indicators>.
824 Each element of C<%!> has a true value only if C<$!> is set to that
825 value. For example, C<$!{ENOENT}> is true if and only if the current
826 value of C<$!> is C<ENOENT>; that is, if the most recent error was
827 "No such file or directory" (or its moral equivalent: not all operating
828 systems give that exact error, and certainly not all languages).
829 To check if a particular key is meaningful on your system, use
830 C<exists $!{the_key}>; for a list of legal keys, use C<keys %!>.
831 See L<Errno> for more information, and also see above for the
834 =item $EXTENDED_OS_ERROR
837 X<$^E> X<$EXTENDED_OS_ERROR>
839 Error information specific to the current operating system. At
840 the moment, this differs from C<$!> under only VMS, OS/2, and Win32
841 (and for MacPerl). On all other platforms, C<$^E> is always just
844 Under VMS, C<$^E> provides the VMS status value from the last
845 system error. This is more specific information about the last
846 system error than that provided by C<$!>. This is particularly
847 important when C<$!> is set to B<EVMSERR>.
849 Under OS/2, C<$^E> is set to the error code of the last call to
850 OS/2 API either via CRT, or directly from perl.
852 Under Win32, C<$^E> always returns the last error information
853 reported by the Win32 call C<GetLastError()> which describes
854 the last error from within the Win32 API. Most Win32-specific
855 code will report errors via C<$^E>. ANSI C and Unix-like calls
856 set C<errno> and so most portable Perl code will report errors
859 Caveats mentioned in the description of C<$!> generally apply to
860 C<$^E>, also. (Mnemonic: Extra error explanation.)
862 Also see L<Error Indicators>.
869 The Perl syntax error message from the last eval() operator.
870 If $@ is the null string, the last eval() parsed and executed
871 correctly (although the operations you invoked may have failed in the
872 normal fashion). (Mnemonic: Where was the syntax error "at"?)
874 Warning messages are not collected in this variable. You can,
875 however, set up a routine to process warnings by setting C<$SIG{__WARN__}>
878 Also see L<Error Indicators>.
885 X<$$> X<$PID> X<$PROCESS_ID>
887 The process number of the Perl running this script. You should
888 consider this variable read-only, although it will be altered
889 across fork() calls. (Mnemonic: same as shells.)
891 Note for Linux users: on Linux, the C functions C<getpid()> and
892 C<getppid()> return different values from different threads. In order to
893 be portable, this behavior is not reflected by C<$$>, whose value remains
894 consistent across threads. If you want to call the underlying C<getpid()>,
895 you may use the CPAN module C<Linux::Pid>.
902 X<< $< >> X<$UID> X<$REAL_USER_ID>
904 The real uid of this process. (Mnemonic: it's the uid you came I<from>,
905 if you're running setuid.) You can change both the real uid and
906 the effective uid at the same time by using POSIX::setuid(). Since
907 changes to $< require a system call, check $! after a change attempt to
908 detect any possible errors.
910 =item $EFFECTIVE_USER_ID
915 X<< $> >> X<$EUID> X<$EFFECTIVE_USER_ID>
917 The effective uid of this process. Example:
919 $< = $>; # set real to effective uid
920 ($<,$>) = ($>,$<); # swap real and effective uid
922 You can change both the effective uid and the real uid at the same
923 time by using POSIX::setuid(). Changes to $> require a check to $!
924 to detect any possible errors after an attempted change.
926 (Mnemonic: it's the uid you went I<to>, if you're running setuid.)
927 C<< $< >> and C<< $> >> can be swapped only on machines
928 supporting setreuid().
935 X<$(> X<$GID> X<$REAL_GROUP_ID>
937 The real gid of this process. If you are on a machine that supports
938 membership in multiple groups simultaneously, gives a space separated
939 list of groups you are in. The first number is the one returned by
940 getgid(), and the subsequent ones by getgroups(), one of which may be
941 the same as the first number.
943 However, a value assigned to C<$(> must be a single number used to
944 set the real gid. So the value given by C<$(> should I<not> be assigned
945 back to C<$(> without being forced numeric, such as by adding zero. Note
946 that this is different to the effective gid (C<$)>) which does take a
949 You can change both the real gid and the effective gid at the same
950 time by using POSIX::setgid(). Changes to $( require a check to $!
951 to detect any possible errors after an attempted change.
953 (Mnemonic: parentheses are used to I<group> things. The real gid is the
954 group you I<left>, if you're running setgid.)
956 =item $EFFECTIVE_GROUP_ID
961 X<$)> X<$EGID> X<$EFFECTIVE_GROUP_ID>
963 The effective gid of this process. If you are on a machine that
964 supports membership in multiple groups simultaneously, gives a space
965 separated list of groups you are in. The first number is the one
966 returned by getegid(), and the subsequent ones by getgroups(), one of
967 which may be the same as the first number.
969 Similarly, a value assigned to C<$)> must also be a space-separated
970 list of numbers. The first number sets the effective gid, and
971 the rest (if any) are passed to setgroups(). To get the effect of an
972 empty list for setgroups(), just repeat the new effective gid; that is,
973 to force an effective gid of 5 and an effectively empty setgroups()
974 list, say C< $) = "5 5" >.
976 You can change both the effective gid and the real gid at the same
977 time by using POSIX::setgid() (use only a single numeric argument).
978 Changes to $) require a check to $! to detect any possible errors
979 after an attempted change.
981 (Mnemonic: parentheses are used to I<group> things. The effective gid
982 is the group that's I<right> for you, if you're running setgid.)
984 C<< $< >>, C<< $> >>, C<$(> and C<$)> can be set only on
985 machines that support the corresponding I<set[re][ug]id()> routine. C<$(>
986 and C<$)> can be swapped only on machines supporting setregid().
991 X<$0> X<$PROGRAM_NAME>
993 Contains the name of the program being executed.
995 On some (read: not all) operating systems assigning to C<$0> modifies
996 the argument area that the C<ps> program sees. On some platforms you
997 may have to use special C<ps> options or a different C<ps> to see the
998 changes. Modifying the $0 is more useful as a way of indicating the
999 current program state than it is for hiding the program you're
1000 running. (Mnemonic: same as B<sh> and B<ksh>.)
1002 Note that there are platform specific limitations on the maximum
1003 length of C<$0>. In the most extreme case it may be limited to the
1004 space occupied by the original C<$0>.
1006 In some platforms there may be arbitrary amount of padding, for
1007 example space characters, after the modified name as shown by C<ps>.
1008 In some platforms this padding may extend all the way to the original
1009 length of the argument area, no matter what you do (this is the case
1010 for example with Linux 2.2).
1012 Note for BSD users: setting C<$0> does not completely remove "perl"
1013 from the ps(1) output. For example, setting C<$0> to C<"foobar"> may
1014 result in C<"perl: foobar (perl)"> (whether both the C<"perl: "> prefix
1015 and the " (perl)" suffix are shown depends on your exact BSD variant
1016 and version). This is an operating system feature, Perl cannot help it.
1018 In multithreaded scripts Perl coordinates the threads so that any
1019 thread may modify its copy of the C<$0> and the change becomes visible
1020 to ps(1) (assuming the operating system plays along). Note that
1021 the view of C<$0> the other threads have will not change since they
1022 have their own copies of it.
1027 The index of the first element in an array, and of the first character
1028 in a substring. Default is 0, but you could theoretically set it
1029 to 1 to make Perl behave more like B<awk> (or Fortran) when
1030 subscripting and when evaluating the index() and substr() functions.
1031 (Mnemonic: [ begins subscripts.)
1033 As of release 5 of Perl, assignment to C<$[> is treated as a compiler
1034 directive, and cannot influence the behavior of any other file.
1035 (That's why you can only assign compile-time constants to it.)
1036 Its use is highly discouraged.
1038 Note that, unlike other compile-time directives (such as L<strict>),
1039 assignment to C<$[> can be seen from outer lexical scopes in the same file.
1040 However, you can use local() on it to strictly bind its value to a
1046 The version + patchlevel / 1000 of the Perl interpreter. This variable
1047 can be used to determine whether the Perl interpreter executing a
1048 script is in the right range of versions. (Mnemonic: Is this version
1049 of perl in the right bracket?) Example:
1051 warn "No checksumming!\n" if $] < 3.019;
1053 See also the documentation of C<use VERSION> and C<require VERSION>
1054 for a convenient way to fail if the running Perl interpreter is too old.
1056 The floating point representation can sometimes lead to inaccurate
1057 numeric comparisons. See C<$^V> for a more modern representation of
1058 the Perl version that allows accurate string comparisons.
1063 X<$^C> X<$COMPILING>
1065 The current value of the flag associated with the B<-c> switch.
1066 Mainly of use with B<-MO=...> to allow code to alter its behavior
1067 when being compiled, such as for example to AUTOLOAD at compile
1068 time rather than normal, deferred loading. Setting
1069 C<$^C = 1> is similar to calling C<B::minus_c>.
1074 X<$^D> X<$DEBUGGING>
1076 The current value of the debugging flags. (Mnemonic: value of B<-D>
1077 switch.) May be read or set. Like its command-line equivalent, you can use
1078 numeric or symbolic values, eg C<$^D = 10> or C<$^D = "st">.
1080 =item ${^RE_DEBUG_FLAGS}
1082 The current value of the regex debugging flags. Set to 0 for no debug output
1083 even when the re 'debug' module is loaded. See L<re> for details.
1085 =item ${^RE_TRIE_MAXBUF}
1087 Controls how certain regex optimisations are applied and how much memory they
1088 utilize. This value by default is 65536 which corresponds to a 512kB temporary
1089 cache. Set this to a higher value to trade memory for speed when matching
1090 large alternations. Set it to a lower value if you want the optimisations to
1091 be as conservative of memory as possible but still occur, and set it to a
1092 negative value to prevent the optimisation and conserve the most memory.
1093 Under normal situations this variable should be of no interest to you.
1095 =item $SYSTEM_FD_MAX
1098 X<$^F> X<$SYSTEM_FD_MAX>
1100 The maximum system file descriptor, ordinarily 2. System file
1101 descriptors are passed to exec()ed processes, while higher file
1102 descriptors are not. Also, during an open(), system file descriptors are
1103 preserved even if the open() fails. (Ordinary file descriptors are
1104 closed before the open() is attempted.) The close-on-exec
1105 status of a file descriptor will be decided according to the value of
1106 C<$^F> when the corresponding file, pipe, or socket was opened, not the
1111 WARNING: This variable is strictly for internal use only. Its availability,
1112 behavior, and contents are subject to change without notice.
1114 This variable contains compile-time hints for the Perl interpreter. At the
1115 end of compilation of a BLOCK the value of this variable is restored to the
1116 value when the interpreter started to compile the BLOCK.
1118 When perl begins to parse any block construct that provides a lexical scope
1119 (e.g., eval body, required file, subroutine body, loop body, or conditional
1120 block), the existing value of $^H is saved, but its value is left unchanged.
1121 When the compilation of the block is completed, it regains the saved value.
1122 Between the points where its value is saved and restored, code that
1123 executes within BEGIN blocks is free to change the value of $^H.
1125 This behavior provides the semantic of lexical scoping, and is used in,
1126 for instance, the C<use strict> pragma.
1128 The contents should be an integer; different bits of it are used for
1129 different pragmatic flags. Here's an example:
1131 sub add_100 { $^H |= 0x100 }
1138 Consider what happens during execution of the BEGIN block. At this point
1139 the BEGIN block has already been compiled, but the body of foo() is still
1140 being compiled. The new value of $^H will therefore be visible only while
1141 the body of foo() is being compiled.
1143 Substitution of the above BEGIN block with:
1145 BEGIN { require strict; strict->import('vars') }
1147 demonstrates how C<use strict 'vars'> is implemented. Here's a conditional
1148 version of the same lexical pragma:
1150 BEGIN { require strict; strict->import('vars') if $condition }
1154 The %^H hash provides the same scoping semantic as $^H. This makes it
1155 useful for implementation of lexically scoped pragmas. See L<perlpragma>.
1160 X<$^I> X<$INPLACE_EDIT>
1162 The current value of the inplace-edit extension. Use C<undef> to disable
1163 inplace editing. (Mnemonic: value of B<-i> switch.)
1168 By default, running out of memory is an untrappable, fatal error.
1169 However, if suitably built, Perl can use the contents of C<$^M>
1170 as an emergency memory pool after die()ing. Suppose that your Perl
1171 were compiled with C<-DPERL_EMERGENCY_SBRK> and used Perl's malloc.
1174 $^M = 'a' x (1 << 16);
1176 would allocate a 64K buffer for use in an emergency. See the
1177 F<INSTALL> file in the Perl distribution for information on how to
1178 add custom C compilation flags when compiling perl. To discourage casual
1179 use of this advanced feature, there is no L<English|English> long name for
1187 The name of the operating system under which this copy of Perl was
1188 built, as determined during the configuration process. The value
1189 is identical to C<$Config{'osname'}>. See also L<Config> and the
1190 B<-V> command-line switch documented in L<perlrun>.
1192 In Windows platforms, $^O is not very helpful: since it is always
1193 C<MSWin32>, it doesn't tell the difference between
1194 95/98/ME/NT/2000/XP/CE/.NET. Use Win32::GetOSName() or
1195 Win32::GetOSVersion() (see L<Win32> and L<perlport>) to distinguish
1196 between the variants.
1200 An internal variable used by PerlIO. A string in two parts, separated
1201 by a C<\0> byte, the first part describes the input layers, the second
1202 part describes the output layers.
1209 The internal variable for debugging support. The meanings of the
1210 various bits are subject to change, but currently indicate:
1216 Debug subroutine enter/exit.
1220 Line-by-line debugging.
1224 Switch off optimizations.
1228 Preserve more data for future interactive inspections.
1232 Keep info about source lines on which a subroutine is defined.
1236 Start with single-step on.
1240 Use subroutine address instead of name when reporting.
1244 Report C<goto &subroutine> as well.
1248 Provide informative "file" names for evals based on the place they were compiled.
1252 Provide informative names to anonymous subroutines based on the place they
1257 Debug assertion subroutines enter/exit.
1261 Some bits may be relevant at compile-time only, some at
1262 run-time only. This is a new mechanism and the details may change.
1264 =item $LAST_REGEXP_CODE_RESULT
1267 X<$^R> X<$LAST_REGEXP_CODE_RESULT>
1269 The result of evaluation of the last successful C<(?{ code })>
1270 regular expression assertion (see L<perlre>). May be written to.
1272 =item $EXCEPTIONS_BEING_CAUGHT
1275 X<$^S> X<$EXCEPTIONS_BEING_CAUGHT>
1277 Current state of the interpreter.
1280 --------- -------------------
1281 undef Parsing module/eval
1282 true (1) Executing an eval
1285 The first state may happen in $SIG{__DIE__} and $SIG{__WARN__} handlers.
1292 The time at which the program began running, in seconds since the
1293 epoch (beginning of 1970). The values returned by the B<-M>, B<-A>,
1294 and B<-C> filetests are based on this value.
1298 Reflects if taint mode is on or off. 1 for on (the program was run with
1299 B<-T>), 0 for off, -1 when only taint warnings are enabled (i.e. with
1300 B<-t> or B<-TU>). This variable is read-only.
1304 Reflects certain Unicode settings of Perl. See L<perlrun>
1305 documentation for the C<-C> switch for more information about
1306 the possible values. This variable is set during Perl startup
1307 and is thereafter read-only.
1311 This variable controls the state of the internal UTF-8 offset caching code.
1312 1 for on (the default), 0 for off, -1 to debug the caching code by checking
1313 all its results against linear scans, and panicking on any discrepancy.
1315 =item ${^UTF8LOCALE}
1317 This variable indicates whether an UTF-8 locale was detected by perl at
1318 startup. This information is used by perl when it's in
1319 adjust-utf8ness-to-locale mode (as when run with the C<-CL> command-line
1320 switch); see L<perlrun> for more info on this.
1325 X<$^V> X<$PERL_VERSION>
1327 The revision, version, and subversion of the Perl interpreter, represented
1328 as a C<version> object.
1330 This variable first appeared in perl 5.6.0; earlier versions of perl will
1331 see an undefined value. Before perl 5.10.0 $^V was represented as a v-string.
1333 $^V can be used to determine whether the Perl interpreter executing a
1334 script is in the right range of versions. (Mnemonic: use ^V for Version
1337 warn "Hashes not randomized!\n" if !$^V or $^V lt v5.8.1
1339 To convert C<$^V> into its string representation use sprintf()'s
1340 C<"%vd"> conversion:
1342 printf "version is v%vd\n", $^V; # Perl's version
1344 See the documentation of C<use VERSION> and C<require VERSION>
1345 for a convenient way to fail if the running Perl interpreter is too old.
1347 See also C<$]> for an older representation of the Perl version.
1354 The current value of the warning switch, initially true if B<-w>
1355 was used, false otherwise, but directly modifiable. (Mnemonic:
1356 related to the B<-w> switch.) See also L<warnings>.
1358 =item ${^WARNING_BITS}
1360 The current set of warning checks enabled by the C<use warnings> pragma.
1361 See the documentation of C<warnings> for more details.
1363 =item ${^WIN32_SLOPPY_STAT}
1365 If this variable is set to a true value, then stat() on Windows will
1366 not try to open the file. This means that the link count cannot be
1367 determined and file attributes may be out of date if additional
1368 hardlinks to the file exist. On the other hand, not opening the file
1369 is considerably faster, especially for files on network drives.
1371 This variable could be set in the F<sitecustomize.pl> file to
1372 configure the local Perl installation to use "sloppy" stat() by
1373 default. See L<perlrun> for more information about site
1376 =item $EXECUTABLE_NAME
1379 X<$^X> X<$EXECUTABLE_NAME>
1381 The name used to execute the current copy of Perl, from C's
1382 C<argv[0]> or (where supported) F</proc/self/exe>.
1384 Depending on the host operating system, the value of $^X may be
1385 a relative or absolute pathname of the perl program file, or may
1386 be the string used to invoke perl but not the pathname of the
1387 perl program file. Also, most operating systems permit invoking
1388 programs that are not in the PATH environment variable, so there
1389 is no guarantee that the value of $^X is in PATH. For VMS, the
1390 value may or may not include a version number.
1392 You usually can use the value of $^X to re-invoke an independent
1393 copy of the same perl that is currently running, e.g.,
1395 @first_run = `$^X -le "print int rand 100 for 1..100"`;
1397 But recall that not all operating systems support forking or
1398 capturing of the output of commands, so this complex statement
1399 may not be portable.
1401 It is not safe to use the value of $^X as a path name of a file,
1402 as some operating systems that have a mandatory suffix on
1403 executable files do not require use of the suffix when invoking
1404 a command. To convert the value of $^X to a path name, use the
1405 following statements:
1407 # Build up a set of file names (not command names).
1411 {$this_perl .= $Config{_exe}
1412 unless $this_perl =~ m/$Config{_exe}$/i;}
1414 Because many operating systems permit anyone with read access to
1415 the Perl program file to make a copy of it, patch the copy, and
1416 then execute the copy, the security-conscious Perl programmer
1417 should take care to invoke the installed copy of perl, not the
1418 copy referenced by $^X. The following statements accomplish
1419 this goal, and produce a pathname that can be invoked as a
1420 command or referenced as a file.
1423 $secure_perl_path = $Config{perlpath};
1425 {$secure_perl_path .= $Config{_exe}
1426 unless $secure_perl_path =~ m/$Config{_exe}$/i;}
1431 The special filehandle that iterates over command-line filenames in
1432 C<@ARGV>. Usually written as the null filehandle in the angle operator
1433 C<< <> >>. Note that currently C<ARGV> only has its magical effect
1434 within the C<< <> >> operator; elsewhere it is just a plain filehandle
1435 corresponding to the last file opened by C<< <> >>. In particular,
1436 passing C<\*ARGV> as a parameter to a function that expects a filehandle
1437 may not cause your function to automatically read the contents of all the
1443 contains the name of the current file when reading from <>.
1448 The array @ARGV contains the command-line arguments intended for
1449 the script. C<$#ARGV> is generally the number of arguments minus
1450 one, because C<$ARGV[0]> is the first argument, I<not> the program's
1451 command name itself. See C<$0> for the command name.
1456 The special filehandle that points to the currently open output file
1457 when doing edit-in-place processing with B<-i>. Useful when you have
1458 to do a lot of inserting and don't want to keep modifying $_. See
1459 L<perlrun> for the B<-i> switch.
1464 The array @F contains the fields of each line read in when autosplit
1465 mode is turned on. See L<perlrun> for the B<-a> switch. This array
1466 is package-specific, and must be declared or given a full package name
1467 if not in package main when running under C<strict 'vars'>.
1472 The array @INC contains the list of places that the C<do EXPR>,
1473 C<require>, or C<use> constructs look for their library files. It
1474 initially consists of the arguments to any B<-I> command-line
1475 switches, followed by the default Perl library, probably
1476 F</usr/local/lib/perl>, followed by ".", to represent the current
1477 directory. ("." will not be appended if taint checks are enabled, either by
1478 C<-T> or by C<-t>.) If you need to modify this at runtime, you should use
1479 the C<use lib> pragma to get the machine-dependent library properly
1482 use lib '/mypath/libdir/';
1485 You can also insert hooks into the file inclusion system by putting Perl
1486 code directly into @INC. Those hooks may be subroutine references, array
1487 references or blessed objects. See L<perlfunc/require> for details.
1494 Within a subroutine the array @_ contains the parameters passed to that
1495 subroutine. See L<perlsub>.
1500 The hash %INC contains entries for each filename included via the
1501 C<do>, C<require>, or C<use> operators. The key is the filename
1502 you specified (with module names converted to pathnames), and the
1503 value is the location of the file found. The C<require>
1504 operator uses this hash to determine whether a particular file has
1505 already been included.
1507 If the file was loaded via a hook (e.g. a subroutine reference, see
1508 L<perlfunc/require> for a description of these hooks), this hook is
1509 by default inserted into %INC in place of a filename. Note, however,
1510 that the hook may have set the %INC entry by itself to provide some more
1518 The hash %ENV contains your current environment. Setting a
1519 value in C<ENV> changes the environment for any child processes
1520 you subsequently fork() off.
1527 The hash C<%SIG> contains signal handlers for signals. For example:
1529 sub handler { # 1st argument is signal name
1531 print "Caught a SIG$sig--shutting down\n";
1536 $SIG{'INT'} = \&handler;
1537 $SIG{'QUIT'} = \&handler;
1539 $SIG{'INT'} = 'DEFAULT'; # restore default action
1540 $SIG{'QUIT'} = 'IGNORE'; # ignore SIGQUIT
1542 Using a value of C<'IGNORE'> usually has the effect of ignoring the
1543 signal, except for the C<CHLD> signal. See L<perlipc> for more about
1546 Here are some other examples:
1548 $SIG{"PIPE"} = "Plumber"; # assumes main::Plumber (not recommended)
1549 $SIG{"PIPE"} = \&Plumber; # just fine; assume current Plumber
1550 $SIG{"PIPE"} = *Plumber; # somewhat esoteric
1551 $SIG{"PIPE"} = Plumber(); # oops, what did Plumber() return??
1553 Be sure not to use a bareword as the name of a signal handler,
1554 lest you inadvertently call it.
1556 If your system has the sigaction() function then signal handlers are
1557 installed using it. This means you get reliable signal handling.
1559 The default delivery policy of signals changed in Perl 5.8.0 from
1560 immediate (also known as "unsafe") to deferred, also known as
1561 "safe signals". See L<perlipc> for more information.
1563 Certain internal hooks can be also set using the %SIG hash. The
1564 routine indicated by C<$SIG{__WARN__}> is called when a warning message is
1565 about to be printed. The warning message is passed as the first
1566 argument. The presence of a C<__WARN__> hook causes the ordinary printing
1567 of warnings to C<STDERR> to be suppressed. You can use this to save warnings
1568 in a variable, or turn warnings into fatal errors, like this:
1570 local $SIG{__WARN__} = sub { die $_[0] };
1573 As the C<'IGNORE'> hook is not supported by C<__WARN__>, you can
1574 disable warnings using the empty subroutine:
1576 local $SIG{__WARN__} = sub {};
1578 The routine indicated by C<$SIG{__DIE__}> is called when a fatal exception
1579 is about to be thrown. The error message is passed as the first
1580 argument. When a C<__DIE__> hook routine returns, the exception
1581 processing continues as it would have in the absence of the hook,
1582 unless the hook routine itself exits via a C<goto>, a loop exit, or a C<die()>.
1583 The C<__DIE__> handler is explicitly disabled during the call, so that you
1584 can die from a C<__DIE__> handler. Similarly for C<__WARN__>.
1586 Due to an implementation glitch, the C<$SIG{__DIE__}> hook is called
1587 even inside an eval(). Do not use this to rewrite a pending exception
1588 in C<$@>, or as a bizarre substitute for overriding C<CORE::GLOBAL::die()>.
1589 This strange action at a distance may be fixed in a future release
1590 so that C<$SIG{__DIE__}> is only called if your program is about
1591 to exit, as was the original intent. Any other use is deprecated.
1593 C<__DIE__>/C<__WARN__> handlers are very special in one respect:
1594 they may be called to report (probable) errors found by the parser.
1595 In such a case the parser may be in inconsistent state, so any
1596 attempt to evaluate Perl code from such a handler will probably
1597 result in a segfault. This means that warnings or errors that
1598 result from parsing Perl should be used with extreme caution, like
1601 require Carp if defined $^S;
1602 Carp::confess("Something wrong") if defined &Carp::confess;
1603 die "Something wrong, but could not load Carp to give backtrace...
1604 To see backtrace try starting Perl with -MCarp switch";
1606 Here the first line will load Carp I<unless> it is the parser who
1607 called the handler. The second line will print backtrace and die if
1608 Carp was available. The third line will be executed only if Carp was
1611 See L<perlfunc/die>, L<perlfunc/warn>, L<perlfunc/eval>, and
1612 L<warnings> for additional information.
1616 =head2 Error Indicators
1617 X<error> X<exception>
1619 The variables C<$@>, C<$!>, C<$^E>, and C<$?> contain information
1620 about different types of error conditions that may appear during
1621 execution of a Perl program. The variables are shown ordered by
1622 the "distance" between the subsystem which reported the error and
1623 the Perl process. They correspond to errors detected by the Perl
1624 interpreter, C library, operating system, or an external program,
1627 To illustrate the differences between these variables, consider the
1628 following Perl expression, which uses a single-quoted string:
1631 open my $pipe, "/cdrom/install |" or die $!;
1633 close $pipe or die "bad pipe: $?, $!";
1636 After execution of this statement all 4 variables may have been set.
1638 C<$@> is set if the string to be C<eval>-ed did not compile (this
1639 may happen if C<open> or C<close> were imported with bad prototypes),
1640 or if Perl code executed during evaluation die()d . In these cases
1641 the value of $@ is the compile error, or the argument to C<die>
1642 (which will interpolate C<$!> and C<$?>). (See also L<Fatal>,
1645 When the eval() expression above is executed, open(), C<< <PIPE> >>,
1646 and C<close> are translated to calls in the C run-time library and
1647 thence to the operating system kernel. C<$!> is set to the C library's
1648 C<errno> if one of these calls fails.
1650 Under a few operating systems, C<$^E> may contain a more verbose
1651 error indicator, such as in this case, "CDROM tray not closed."
1652 Systems that do not support extended error messages leave C<$^E>
1655 Finally, C<$?> may be set to non-0 value if the external program
1656 F</cdrom/install> fails. The upper eight bits reflect specific
1657 error conditions encountered by the program (the program's exit()
1658 value). The lower eight bits reflect mode of failure, like signal
1659 death and core dump information See wait(2) for details. In
1660 contrast to C<$!> and C<$^E>, which are set only if error condition
1661 is detected, the variable C<$?> is set on each C<wait> or pipe
1662 C<close>, overwriting the old value. This is more like C<$@>, which
1663 on every eval() is always set on failure and cleared on success.
1665 For more details, see the individual descriptions at C<$@>, C<$!>, C<$^E>,
1668 =head2 Technical Note on the Syntax of Variable Names
1670 Variable names in Perl can have several formats. Usually, they
1671 must begin with a letter or underscore, in which case they can be
1672 arbitrarily long (up to an internal limit of 251 characters) and
1673 may contain letters, digits, underscores, or the special sequence
1674 C<::> or C<'>. In this case, the part before the last C<::> or
1675 C<'> is taken to be a I<package qualifier>; see L<perlmod>.
1677 Perl variable names may also be a sequence of digits or a single
1678 punctuation or control character. These names are all reserved for
1679 special uses by Perl; for example, the all-digits names are used
1680 to hold data captured by backreferences after a regular expression
1681 match. Perl has a special syntax for the single-control-character
1682 names: It understands C<^X> (caret C<X>) to mean the control-C<X>
1683 character. For example, the notation C<$^W> (dollar-sign caret
1684 C<W>) is the scalar variable whose name is the single character
1685 control-C<W>. This is better than typing a literal control-C<W>
1688 Finally, new in Perl 5.6, Perl variable names may be alphanumeric
1689 strings that begin with control characters (or better yet, a caret).
1690 These variables must be written in the form C<${^Foo}>; the braces
1691 are not optional. C<${^Foo}> denotes the scalar variable whose
1692 name is a control-C<F> followed by two C<o>'s. These variables are
1693 reserved for future special uses by Perl, except for the ones that
1694 begin with C<^_> (control-underscore or caret-underscore). No
1695 control-character name that begins with C<^_> will acquire a special
1696 meaning in any future version of Perl; such names may therefore be
1697 used safely in programs. C<$^_> itself, however, I<is> reserved.
1699 Perl identifiers that begin with digits, control characters, or
1700 punctuation characters are exempt from the effects of the C<package>
1701 declaration and are always forced to be in package C<main>; they are
1702 also exempt from C<strict 'vars'> errors. A few other names are also
1703 exempt in these ways:
1711 In particular, the new special C<${^_XYZ}> variables are always taken
1712 to be in package C<main>, regardless of any C<package> declarations
1717 Due to an unfortunate accident of Perl's implementation, C<use
1718 English> imposes a considerable performance penalty on all regular
1719 expression matches in a program, regardless of whether they occur
1720 in the scope of C<use English>. For that reason, saying C<use
1721 English> in libraries is strongly discouraged. See the
1722 Devel::SawAmpersand module documentation from CPAN
1723 ( http://www.cpan.org/modules/by-module/Devel/ )
1724 for more information. Writing C<use English '-no_match_vars';>
1725 avoids the performance penalty.
1727 Having to even think about the C<$^S> variable in your exception
1728 handlers is simply wrong. C<$SIG{__DIE__}> as currently implemented
1729 invites grievous and difficult to track down errors. Avoid it
1730 and use an C<END{}> or CORE::GLOBAL::die override instead.