3 perldebug - Perl debugging
7 First of all, have you tried using the B<-w> switch?
9 =head1 The Perl Debugger
11 If you invoke Perl with the B<-d> switch, your script runs under the
12 Perl source debugger. This works like an interactive Perl
13 environment, prompting for debugger commands that let you examine
14 source code, set breakpoints, get stack backtraces, change the values of
15 variables, etc. This is so convenient that you often fire up
16 the debugger all by itself just to test out Perl constructs
17 interactively to see what they do. For example:
21 In Perl, the debugger is not a separate program as it usually is in the
22 typical compiled environment. Instead, the B<-d> flag tells the compiler
23 to insert source information into the parse trees it's about to hand off
24 to the interpreter. That means your code must first compile correctly
25 for the debugger to work on it. Then when the interpreter starts up, it
26 preloads a Perl library file containing the debugger itself.
28 The program will halt I<right before> the first run-time executable
29 statement (but see below regarding compile-time statements) and ask you
30 to enter a debugger command. Contrary to popular expectations, whenever
31 the debugger halts and shows you a line of code, it always displays the
32 line it's I<about> to execute, rather than the one it has just executed.
34 Any command not recognized by the debugger is directly executed
35 (C<eval>'d) as Perl code in the current package. (The debugger uses the
36 DB package for its own state information.)
38 Leading white space before a command would cause the debugger to think
39 it's I<NOT> a debugger command but for Perl, so be careful not to do
42 =head2 Debugger Commands
44 The debugger understands the following commands:
50 Prints out a help message.
52 If you supply another debugger command as an argument to the C<h> command,
53 it prints out the description for just that command. The special
54 argument of C<h h> produces a more compact help listing, designed to fit
55 together on one screen.
57 If the output the C<h> command (or any command, for that matter) scrolls
58 past your screen, either precede the command with a leading pipe symbol so
59 it's run through your pager, as in
63 You may change the pager which is used via C<O pager=...> command.
67 Same as C<print {$DB::OUT} expr> in the current package. In particular,
68 because this is just Perl's own B<print> function, this means that nested
69 data structures and objects are not dumped, unlike with the C<x> command.
71 The C<DB::OUT> filehandle is opened to F</dev/tty>, regardless of
72 where STDOUT may be redirected to.
76 Evaluates its expression in list context and dumps out the result
77 in a pretty-printed fashion. Nested data structures are printed out
78 recursively, unlike the C<print> function.
80 The details of printout are governed by multiple C<O>ptions.
84 Display all (or some) variables in package (defaulting to the C<main>
85 package) using a data pretty-printer (hashes show their keys and values so
86 you see what's what, control characters are made printable, etc.). Make
87 sure you don't put the type specifier (like C<$>) there, just the symbol
92 Use C<~pattern> and C<!pattern> for positive and negative regexps.
94 Nested data structures are printed out in a legible fashion, unlike
95 the C<print> function.
97 The details of printout are governed by multiple C<O>ptions.
101 Same as C<V currentpackage [vars]>.
105 Produce a stack backtrace. See below for details on its output.
109 Single step. Executes until it reaches the beginning of another
110 statement, descending into subroutine calls. If an expression is
111 supplied that includes function calls, it too will be single-stepped.
115 Next. Executes over subroutine calls, until it reaches the beginning
116 of the next statement. If an expression is supplied that includes
117 function calls, those functions will be executed with stops before
122 Repeat last C<n> or C<s> command.
126 Continue, optionally inserting a one-time-only breakpoint
127 at the specified line or subroutine.
131 List next window of lines.
135 List C<incr+1> lines starting at C<min>.
139 List lines C<min> through C<max>. C<l -> is synonymous to C<->.
147 List first window of lines from subroutine.
151 List previous window of lines.
155 List window (a few lines) around the current line.
159 Return debugger pointer to the last-executed line and
164 Switch to viewing a different file or eval statement. If C<filename>
165 is not a full filename as found in values of %INC, it is considered as
170 Search forwards for pattern; final / is optional.
174 Search backwards for pattern; final ? is optional.
178 List all breakpoints and actions.
182 List subroutine names [not] matching pattern.
186 Toggle trace mode (see also C<AutoTrace> C<O>ption).
190 Trace through execution of expr. For example:
193 Stack dump during die enabled outside of evals.
195 Loading DB routines from perl5db.pl patch level 0.94
196 Emacs support available.
198 Enter h or `h h' for help.
205 DB<3> t print foo() * bar()
206 main::((eval 172):3): print foo() + bar();
207 main::foo((eval 168):2):
208 main::bar((eval 170):2):
211 or, with the C<O>ption C<frame=2> set,
215 DB<5> t print foo() * bar()
225 =item b [line] [condition]
227 Set a breakpoint. If line is omitted, sets a breakpoint on the line
228 that is about to be executed. If a condition is specified, it's
229 evaluated each time the statement is reached and a breakpoint is taken
230 only if the condition is true. Breakpoints may be set on only lines
231 that begin an executable statement. Conditions don't use B<if>:
234 b 237 ++$count237 < 11
237 =item b subname [condition]
239 Set a breakpoint at the first line of the named subroutine.
241 =item b postpone subname [condition]
243 Set breakpoint at first line of subroutine after it is compiled.
245 =item b load filename
247 Set breakpoint at the first executed line of the file. Filename should
248 be a full name as found in values of %INC.
250 =item b compile subname
252 Sets breakpoint at the first statement executed after the subroutine
257 Delete a breakpoint at the specified line. If line is omitted, deletes
258 the breakpoint on the line that is about to be executed.
262 Delete all installed breakpoints.
264 =item a [line] command
266 Set an action to be done before the line is executed.
267 The sequence of steps taken by the debugger is
269 1. check for a breakpoint at this line
270 2. print the line if necessary (tracing)
271 3. do any actions associated with that line
272 4. prompt user if at a breakpoint or in single-step
275 For example, this will print out C<$foo> every time line
278 a 53 print "DB FOUND $foo\n"
282 Delete all installed actions.
284 =item O [opt[=val]] [opt"val"] [opt?]...
286 Set or query values of options. val defaults to 1. opt can
287 be abbreviated. Several options can be listed.
291 =item C<recallCommand>, C<ShellBang>
293 The characters used to recall command or spawn shell. By
294 default, these are both set to C<!>.
298 Program to use for output of pager-piped commands (those
299 beginning with a C<|> character.) By default,
300 C<$ENV{PAGER}> will be used.
304 Run Tk while prompting (with ReadLine).
306 =item C<signalLevel>, C<warnLevel>, C<dieLevel>
308 Level of verbosity. By default the debugger is in a sane verbose mode,
309 thus it will print backtraces on all the warnings and die-messages
310 which are going to be printed out, and will print a message when
311 interesting uncaught signals arrive.
313 To disable this behaviour, set these values to 0. If C<dieLevel> is 2,
314 then the messages which will be caught by surrounding C<eval> are also
319 Trace mode (similar to C<t> command, but can be put into
324 File or pipe to print line number info to. If it is a pipe (say,
325 C<|visual_perl_db>), then a short, "emacs like" message is used.
327 =item C<inhibit_exit>
329 If 0, allows I<stepping off> the end of the script.
333 affects printing of return value after C<r> command.
337 affects printing messages on entry and exit from subroutines. If
338 C<frame & 2> is false, messages are printed on entry only. (Printing
339 on exit may be useful if inter(di)spersed with other messages.)
341 If C<frame & 4>, arguments to functions are printed as well as the
342 context and caller info. If C<frame & 8>, overloaded C<stringify> and
343 C<tie>d C<FETCH> are enabled on the printed arguments. The length at
344 which the argument list is truncated is governed by the next option:
348 length at which the argument list is truncated when C<frame> option's
353 The following options affect what happens with C<V>, C<X>, and C<x>
358 =item C<arrayDepth>, C<hashDepth>
360 Print only first N elements ('' for all).
362 =item C<compactDump>, C<veryCompact>
364 Change style of array and hash dump. If C<compactDump>, short array
365 may be printed on one line.
369 Whether to print contents of globs.
373 Dump arrays holding debugged files.
375 =item C<DumpPackages>
377 Dump symbol tables of packages.
379 =item C<quote>, C<HighBit>, C<undefPrint>
381 Change style of string dump. Default value of C<quote> is C<auto>, one
382 can enable either double-quotish dump, or single-quotish by setting it
383 to C<"> or C<'>. By default, characters with high bit set are printed
388 I<very> rudimentally per-package memory usage dump. Calculates total
389 size of strings in variables in the package.
393 During startup options are initialized from C<$ENV{PERLDB_OPTS}>.
394 You can put additional initialization options C<TTY>, C<noTTY>,
395 C<ReadLine>, and C<NonStop> there.
399 &parse_options("NonStop=1 LineInfo=db.out AutoTrace");
401 The script will run without human intervention, putting trace information
402 into the file I<db.out>. (If you interrupt it, you would better reset
403 C<LineInfo> to something "interactive"!)
409 The TTY to use for debugging I/O.
413 If set, goes in C<NonStop> mode, and would not connect to a TTY. If
414 interrupt (or if control goes to debugger via explicit setting of
415 $DB::signal or $DB::single from the Perl script), connects to a TTY
416 specified by the C<TTY> option at startup, or to a TTY found at
417 runtime using C<Term::Rendezvous> module of your choice.
419 This module should implement a method C<new> which returns an object
420 with two methods: C<IN> and C<OUT>, returning two filehandles to use
421 for debugging input and output correspondingly. Method C<new> may
422 inspect an argument which is a value of C<$ENV{PERLDB_NOTTY}> at
423 startup, or is C<"/tmp/perldbtty$$"> otherwise.
427 If false, readline support in debugger is disabled, so you can debug
428 ReadLine applications.
432 If set, debugger goes into noninteractive mode until interrupted, or
433 programmatically by setting $DB::signal or $DB::single.
437 Here's an example of using the C<$ENV{PERLDB_OPTS}> variable:
439 $ PERLDB_OPTS="N f=2" perl -d myprogram
441 will run the script C<myprogram> without human intervention, printing
442 out the call tree with entry and exit points. Note that C<N f=2> is
443 equivalent to C<NonStop=1 frame=2>. Note also that at the moment when
444 this documentation was written all the options to the debugger could
445 be uniquely abbreviated by the first letter (with exception of
448 Other examples may include
450 $ PERLDB_OPTS="N f A L=listing" perl -d myprogram
452 - runs script noninteractively, printing info on each entry into a
453 subroutine and each executed line into the file F<listing>. (If you
454 interrupt it, you would better reset C<LineInfo> to something
458 $ env "PERLDB_OPTS=R=0 TTY=/dev/ttyc" perl -d myprogram
460 may be useful for debugging a program which uses C<Term::ReadLine>
461 itself. Do not forget detach shell from the TTY in the window which
462 corresponds to F</dev/ttyc>, say, by issuing a command like
466 See L<"Debugger Internals"> below for more details.
468 =item E<lt> [ command ]
470 Set an action (Perl command) to happen before every debugger prompt.
471 A multi-line command may be entered by backslashing the newlines. If
472 C<command> is missing, resets the list of actions.
474 =item E<lt>E<lt> command
476 Add an action (Perl command) to happen before every debugger prompt.
477 A multi-line command may be entered by backslashing the newlines.
481 Set an action (Perl command) to happen after the prompt when you've
482 just given a command to return to executing the script. A multi-line
483 command may be entered by backslashing the newlines. If C<command> is
484 missing, resets the list of actions.
486 =item E<gt>E<gt> command
488 Adds an action (Perl command) to happen after the prompt when you've
489 just given a command to return to executing the script. A multi-line
490 command may be entered by backslashing the newlines.
494 Set an action (debugger command) to happen before every debugger prompt.
495 A multi-line command may be entered by backslashing the newlines. If
496 C<command> is missing, resets the list of actions.
500 Add an action (debugger command) to happen before every debugger prompt.
501 A multi-line command may be entered by backslashing the newlines.
505 Redo a previous command (default previous command).
509 Redo number'th-to-last command.
513 Redo last command that started with pattern.
514 See C<O recallCommand>, too.
518 Run cmd in a subprocess (reads from DB::IN, writes to DB::OUT)
519 See C<O shellBang> too.
523 Display last n commands. Only commands longer than one character are
524 listed. If number is omitted, lists them all.
528 Quit. ("quit" doesn't work for this.) This is the only supported way
529 to exit the debugger, though typing C<exit> twice may do it too.
531 Set an C<O>ption C<inhibit_exit> to 0 if you want to be able to I<step
532 off> the end the script. You may also need to set C<$finished> to 0 at
533 some moment if you want to step through global destruction.
537 Restart the debugger by B<exec>ing a new session. It tries to maintain
538 your history across this, but internal settings and command line options
541 Currently the following setting are preserved: history, breakpoints,
542 actions, debugger C<O>ptions, and the following command line
543 options: B<-w>, B<-I>, and B<-e>.
547 Run debugger command, piping DB::OUT to current pager.
551 Same as C<|dbcmd> but DB::OUT is temporarily B<select>ed as well.
552 Often used with commands that would otherwise produce long
557 =item = [alias value]
559 Define a command alias, like
563 or list current aliases.
567 Execute command as a Perl statement. A missing semicolon will be
572 The expression is evaluated, and the methods which may be applied to
573 the result are listed.
577 The methods which may be applied to objects in the C<package> are listed.
581 =head2 Debugger input/output
587 The debugger prompt is something like
595 where that number is the command number, which you'd use to access with
596 the builtin B<csh>-like history mechanism, e.g., C<!17> would repeat
597 command number 17. The number of angle brackets indicates the depth of
598 the debugger. You could get more than one set of brackets, for example, if
599 you'd already at a breakpoint and then printed out the result of a
600 function call that itself also has a breakpoint, or you step into an
601 expression via C<s/n/t expression> command.
603 =item Multiline commands
605 If you want to enter a multi-line command, such as a subroutine
606 definition with several statements, or a format, you may escape the
607 newline that would normally end the debugger command with a backslash.
611 cont: print "ok\n"; \
618 Note that this business of escaping a newline is specific to interactive
619 commands typed into the debugger.
621 =item Stack backtrace
623 Here's an example of what a stack backtrace via C<T> command might
626 $ = main::infested called from file `Ambulation.pm' line 10
627 @ = Ambulation::legs(1, 2, 3, 4) called from file `camel_flea' line 7
628 $ = main::pests('bactrian', 4) called from file `camel_flea' line 4
630 The left-hand character up there tells whether the function was called
631 in a scalar or list context (we bet you can tell which is which). What
632 that says is that you were in the function C<main::infested> when you ran
633 the stack dump, and that it was called in a scalar context from line 10
634 of the file I<Ambulation.pm>, but without any arguments at all, meaning
635 it was called as C<&infested>. The next stack frame shows that the
636 function C<Ambulation::legs> was called in a list context from the
637 I<camel_flea> file with four arguments. The last stack frame shows that
638 C<main::pests> was called in a scalar context, also from I<camel_flea>,
641 Note that if you execute C<T> command from inside an active C<use>
642 statement, the backtrace will contain both C<L<perlfunc/require>>
643 frame and an C<L<perlfunc/eval EXPR>>) frame.
647 Listing given via different flavors of C<l> command looks like this:
651 102:b @isa{@i,$pack} = ()
652 103 if(exists $i{$prevpack} || exists $isa{$pack});
656 107==> if(exists $isa{$pack});
658 109:a if ($extra-- > 0) {
659 110: %isa = ($pack,1);
661 Note that the breakable lines are marked with C<:>, lines with
662 breakpoints are marked by C<b>, with actions by C<a>, and the
663 next executed line is marked by C<==E<gt>>.
667 When C<frame> option is set, debugger would print entered (and
668 optionally exited) subroutines in different styles.
670 What follows is the start of the listing of
672 env "PERLDB_OPTS=f=1 N" perl -d -V
679 entering Config::BEGIN
680 Package lib/Exporter.pm.
682 Package lib/Config.pm.
683 entering Config::TIEHASH
684 entering Exporter::import
685 entering Exporter::export
686 entering Config::myconfig
687 entering Config::FETCH
688 entering Config::FETCH
689 entering Config::FETCH
690 entering Config::FETCH
695 entering Config::BEGIN
696 Package lib/Exporter.pm.
699 Package lib/Config.pm.
700 entering Config::TIEHASH
701 exited Config::TIEHASH
702 entering Exporter::import
703 entering Exporter::export
704 exited Exporter::export
705 exited Exporter::import
707 entering Config::myconfig
708 entering Config::FETCH
710 entering Config::FETCH
712 entering Config::FETCH
716 in $=main::BEGIN() from /dev/nul:0
717 in $=Config::BEGIN() from lib/Config.pm:2
718 Package lib/Exporter.pm.
720 Package lib/Config.pm.
721 in $=Config::TIEHASH('Config') from lib/Config.pm:644
722 in $=Exporter::import('Config', 'myconfig', 'config_vars') from /dev/nul:0
723 in $=Exporter::export('Config', 'main', 'myconfig', 'config_vars') from li
724 in @=Config::myconfig() from /dev/nul:0
725 in $=Config::FETCH(ref(Config), 'package') from lib/Config.pm:574
726 in $=Config::FETCH(ref(Config), 'baserev') from lib/Config.pm:574
727 in $=Config::FETCH(ref(Config), 'PATCHLEVEL') from lib/Config.pm:574
728 in $=Config::FETCH(ref(Config), 'SUBVERSION') from lib/Config.pm:574
729 in $=Config::FETCH(ref(Config), 'osname') from lib/Config.pm:574
730 in $=Config::FETCH(ref(Config), 'osvers') from lib/Config.pm:574
734 in $=main::BEGIN() from /dev/nul:0
735 in $=Config::BEGIN() from lib/Config.pm:2
736 Package lib/Exporter.pm.
738 out $=Config::BEGIN() from lib/Config.pm:0
739 Package lib/Config.pm.
740 in $=Config::TIEHASH('Config') from lib/Config.pm:644
741 out $=Config::TIEHASH('Config') from lib/Config.pm:644
742 in $=Exporter::import('Config', 'myconfig', 'config_vars') from /dev/nul:0
743 in $=Exporter::export('Config', 'main', 'myconfig', 'config_vars') from lib/
744 out $=Exporter::export('Config', 'main', 'myconfig', 'config_vars') from lib/
745 out $=Exporter::import('Config', 'myconfig', 'config_vars') from /dev/nul:0
746 out $=main::BEGIN() from /dev/nul:0
747 in @=Config::myconfig() from /dev/nul:0
748 in $=Config::FETCH(ref(Config), 'package') from lib/Config.pm:574
749 out $=Config::FETCH(ref(Config), 'package') from lib/Config.pm:574
750 in $=Config::FETCH(ref(Config), 'baserev') from lib/Config.pm:574
751 out $=Config::FETCH(ref(Config), 'baserev') from lib/Config.pm:574
752 in $=Config::FETCH(ref(Config), 'PATCHLEVEL') from lib/Config.pm:574
753 out $=Config::FETCH(ref(Config), 'PATCHLEVEL') from lib/Config.pm:574
754 in $=Config::FETCH(ref(Config), 'SUBVERSION') from lib/Config.pm:574
758 in $=main::BEGIN() from /dev/nul:0
759 in $=Config::BEGIN() from lib/Config.pm:2
760 Package lib/Exporter.pm.
762 out $=Config::BEGIN() from lib/Config.pm:0
763 Package lib/Config.pm.
764 in $=Config::TIEHASH('Config') from lib/Config.pm:644
765 out $=Config::TIEHASH('Config') from lib/Config.pm:644
766 in $=Exporter::import('Config', 'myconfig', 'config_vars') from /dev/nul:0
767 in $=Exporter::export('Config', 'main', 'myconfig', 'config_vars') from lib/E
768 out $=Exporter::export('Config', 'main', 'myconfig', 'config_vars') from lib/E
769 out $=Exporter::import('Config', 'myconfig', 'config_vars') from /dev/nul:0
770 out $=main::BEGIN() from /dev/nul:0
771 in @=Config::myconfig() from /dev/nul:0
772 in $=Config::FETCH('Config=HASH(0x1aa444)', 'package') from lib/Config.pm:574
773 out $=Config::FETCH('Config=HASH(0x1aa444)', 'package') from lib/Config.pm:574
774 in $=Config::FETCH('Config=HASH(0x1aa444)', 'baserev') from lib/Config.pm:574
775 out $=Config::FETCH('Config=HASH(0x1aa444)', 'baserev') from lib/Config.pm:574
779 In all the cases indentation of lines shows the call tree, if bit 2 of
780 C<frame> is set, then a line is printed on exit from a subroutine as
781 well, if bit 4 is set, then the arguments are printed as well as the
782 caller info, if bit 8 is set, the arguments are printed even if they
783 are tied or references.
785 When a package is compiled, a line like this
789 is printed with proper indentation.
793 =head2 Debugging compile-time statements
795 If you have any compile-time executable statements (code within a BEGIN
796 block or a C<use> statement), these will C<NOT> be stopped by debugger,
797 although C<require>s will (and compile-time statements can be traced
798 with C<AutoTrace> option set in C<PERLDB_OPTS>). From your own Perl
799 code, however, you can
800 transfer control back to the debugger using the following statement,
801 which is harmless if the debugger is not running:
805 If you set C<$DB::single> to the value 2, it's equivalent to having
806 just typed the C<n> command, whereas a value of 1 means the C<s>
807 command. The C<$DB::trace> variable should be set to 1 to simulate
808 having typed the C<t> command.
810 Another way to debug compile-time code is to start debugger, set a
811 breakpoint on I<load> of some module thusly
813 DB<7> b load f:/perllib/lib/Carp.pm
814 Will stop on load of `f:/perllib/lib/Carp.pm'.
816 and restart debugger by C<R> command (if possible). One can use C<b
817 compile subname> for the same purpose.
819 =head2 Debugger Customization
821 Most probably you not want to modify the debugger, it contains enough
822 hooks to satisfy most needs. You may change the behaviour of debugger
823 from the debugger itself, using C<O>ptions, from the command line via
824 C<PERLDB_OPTS> environment variable, and from I<customization files>.
826 You can do some customization by setting up a F<.perldb> file which
827 contains initialization code. For instance, you could make aliases
828 like these (the last one is one people expect to be there):
830 $DB::alias{'len'} = 's/^len(.*)/p length($1)/';
831 $DB::alias{'stop'} = 's/^stop (at|in)/b/';
832 $DB::alias{'ps'} = 's/^ps\b/p scalar /';
833 $DB::alias{'quit'} = 's/^quit(\s*)/exit\$/';
835 One changes options from F<.perldb> file via calls like this one;
837 parse_options("NonStop=1 LineInfo=db.out AutoTrace=1 frame=2");
839 (the code is executed in the package C<DB>). Note that F<.perldb> is
840 processed before processing C<PERLDB_OPTS>. If F<.perldb> defines the
841 subroutine C<afterinit>, it is called after all the debugger
842 initialization ends. F<.perldb> may be contained in the current
843 directory, or in the C<LOGDIR>/C<HOME> directory.
845 If you want to modify the debugger, copy F<perl5db.pl> from the Perl
846 library to another name and modify it as necessary. You'll also want
847 to set your C<PERL5DB> environment variable to say something like this:
849 BEGIN { require "myperl5db.pl" }
851 As the last resort, one can use C<PERL5DB> to customize debugger by
852 directly setting internal variables or calling debugger functions.
854 =head2 Readline Support
856 As shipped, the only command line history supplied is a simplistic one
857 that checks for leading exclamation points. However, if you install
858 the Term::ReadKey and Term::ReadLine modules from CPAN, you will
859 have full editing capabilities much like GNU I<readline>(3) provides.
860 Look for these in the F<modules/by-module/Term> directory on CPAN.
862 A rudimentary command line completion is also available.
863 Unfortunately, the names of lexical variables are not available for
866 =head2 Editor Support for Debugging
868 If you have GNU B<emacs> installed on your system, it can interact with
869 the Perl debugger to provide an integrated software development
870 environment reminiscent of its interactions with C debuggers.
872 Perl is also delivered with a start file for making B<emacs> act like a
873 syntax-directed editor that understands (some of) Perl's syntax. Look in
874 the I<emacs> directory of the Perl source distribution.
876 (Historically, a similar setup for interacting with B<vi> and the
877 X11 window system had also been available, but at the time of this
878 writing, no debugger support for B<vi> currently exists.)
880 =head2 The Perl Profiler
882 If you wish to supply an alternative debugger for Perl to run, just
883 invoke your script with a colon and a package argument given to the B<-d>
884 flag. One of the most popular alternative debuggers for Perl is
885 B<DProf>, the Perl profiler. As of this writing, B<DProf> is not
886 included with the standard Perl distribution, but it is expected to
887 be included soon, for certain values of "soon".
889 Meanwhile, you can fetch the Devel::Dprof module from CPAN. Assuming
890 it's properly installed on your system, to profile your Perl program in
891 the file F<mycode.pl>, just type:
893 perl -d:DProf mycode.pl
895 When the script terminates the profiler will dump the profile information
896 to a file called F<tmon.out>. A tool like B<dprofpp> (also supplied with
897 the Devel::DProf package) can be used to interpret the information which is
900 =head2 Debugger support in perl
902 When you call the B<caller> function (see L<perlfunc/caller>) from the
903 package DB, Perl sets the array @DB::args to contain the arguments the
904 corresponding stack frame was called with.
906 If perl is run with B<-d> option, the following additional features
913 Perl inserts the contents of C<$ENV{PERL5DB}> (or C<BEGIN {require
914 'perl5db.pl'}> if not present) before the first line of the
919 The array C<@{"_<$filename"}> is the line-by-line contents of
920 $filename for all the compiled files. Same for C<eval>ed strings which
921 contain subroutines, or which are currently executed. The C<$filename>
922 for C<eval>ed strings looks like C<(eval 34)>.
926 The hash C<%{"_<$filename"}> contains breakpoints and action (it is
927 keyed by line number), and individual entries are settable (as opposed
928 to the whole hash). Only true/false is important to Perl, though the
929 values used by F<perl5db.pl> have the form
930 C<"$break_condition\0$action">. Values are magical in numeric context:
931 they are zeros if the line is not breakable.
933 Same for evaluated strings which contain subroutines, or which are
934 currently executed. The C<$filename> for C<eval>ed strings looks like
939 The scalar C<${"_<$filename"}> contains C<"_<$filename">. Same for
940 evaluated strings which contain subroutines, or which are currently
941 executed. The C<$filename> for C<eval>ed strings looks like C<(eval
946 After each C<require>d file is compiled, but before it is executed,
947 C<DB::postponed(*{"_<$filename"})> is called (if subroutine
948 C<DB::postponed> exists). Here the $filename is the expanded name of
949 the C<require>d file (as found in values of C<%INC>).
953 After each subroutine C<subname> is compiled existence of
954 C<$DB::postponed{subname}> is checked. If this key exists,
955 C<DB::postponed(subname)> is called (if subroutine C<DB::postponed>
960 A hash C<%DB::sub> is maintained, with keys being subroutine names,
961 values having the form C<filename:startline-endline>. C<filename> has
962 the form C<(eval 31)> for subroutines defined inside C<eval>s.
966 When execution of the application reaches a place that can have
967 a breakpoint, a call to C<DB::DB()> is performed if any one of
968 variables $DB::trace, $DB::single, or $DB::signal is true. (Note that
969 these variables are not C<local>izable.) This feature is disabled when
970 the control is inside C<DB::DB()> or functions called from it (unless
971 C<$^D & (1E<lt>E<lt>30)>).
975 When execution of the application reaches a subroutine call, a call
976 to C<&DB::sub>(I<args>) is performed instead, with C<$DB::sub> being
977 the name of the called subroutine. (Unless the subroutine is compiled
978 in the package C<DB>.)
982 Note that no subroutine call is possible until C<&DB::sub> is defined
983 (for subroutines outside of package C<DB>). (This restriction is
986 (In fact, for the standard debugger the same is true if C<$DB::deep>
987 (how many levels of recursion deep into the debugger you can go before
988 a mandatory break) is not defined.)
990 With the recent updates the minimal possible debugger consists of one
995 which is quite handy as contents of C<PERL5DB> environment
998 env "PERL5DB=sub DB::DB {}" perl -d your-script
1000 Another (a little bit more useful) minimal debugger can be created
1001 with the only line being
1003 sub DB::DB {print ++$i; scalar <STDIN>}
1005 This debugger would print the sequential number of encountered
1006 statement, and would wait for your C<CR> to continue.
1008 The following debugger is quite functional:
1013 sub sub {print ++$i, " $sub\n"; &$sub}
1016 It prints the sequential number of subroutine call and the name of the
1017 called subroutine. Note that C<&DB::sub> should be compiled into the
1020 =head2 Debugger Internals
1022 At the start, the debugger reads your rc file (F<./.perldb> or
1023 F<~/.perldb> under Unix), which can set important options. This file may
1024 define a subroutine C<&afterinit> to be executed after the debugger is
1027 After the rc file is read, the debugger reads environment variable
1028 PERLDB_OPTS and parses it as a rest of C<O ...> line in debugger prompt.
1030 It also maintains magical internal variables, such as C<@DB::dbline>,
1031 C<%DB::dbline>, which are aliases for C<@{"::_<current_file"}>
1032 C<%{"::_<current_file"}>. Here C<current_file> is the currently
1033 selected (with the debugger's C<f> command, or by flow of execution)
1036 Some functions are provided to simplify customization. See L<"Debugger
1037 Customization"> for description of C<DB::parse_options(string)>. The
1038 function C<DB::dump_trace(skip[, count])> skips the specified number
1039 of frames, and returns an array containing info about the caller
1040 frames (all if C<count> is missing). Each entry is a hash with keys
1041 C<context> (C<$> or C<@>), C<sub> (subroutine name, or info about
1042 eval), C<args> (C<undef> or a reference to an array), C<file>, and
1045 The function C<DB::print_trace(FH, skip[, count[, short]])> prints
1046 formatted info about caller frames. The last two functions may be
1047 convenient as arguments to C<E<lt>>, C<E<lt>E<lt>> commands.
1049 =head2 Other resources
1051 You did try the B<-w> switch, didn't you?
1055 You cannot get the stack frame information or otherwise debug functions
1056 that were not compiled by Perl, such as C or C++ extensions.
1058 If you alter your @_ arguments in a subroutine (such as with B<shift>
1059 or B<pop>, the stack backtrace will not show the original values.