Condensed output of Data::Dumper used on DateTime

andreas1234567 has asked for the wisdom of the Perl Monks concerning the following question:

Honorable Monks,

I'm looking for a way to increase the readability of log files.

If a reference to a DateTime object gets dumped, then the output can span multiple lines of relatively uninteresting and confusing output. Is there a way to override the 'dump' call so that what gets logged is a single line (e.g. $dt->dmy . q{ } .$dt->hms or equivalent)? This would be analogous to the TO_JSON support in JSON::XS.

$ perl -wle 'use Data::Dumper; use DateTime; my $dt = DateTime::->now(
+); print Dumper($dt); print $dt->dmy . q{ } .$dt->hms'
$VAR1 = bless( {
                 'local_rd_secs' => 44783,
                 'local_rd_days' => 734534,
                 'rd_nanosecs' => 0,
                 'locale' => bless( {
                                      'default_time_format_length' => 
+'medium',
                                      'native_territory' => 'United St
+ates',
                                      'native_language' => 'English',
                                      'native_complete_name' => 'Engli
+sh United States',
                                      'en_language' => 'English',
                                      'id' => 'en_US',
                                      'default_date_format_length' => 
+'medium',
                                      'en_complete_name' => 'English U
+nited States',
                                      'en_territory' => 'United States
+'
                                    }, 'DateTime::Locale::en_US' ),
                 'local_c' => {
                                'hour' => 12,
                                'second' => 23,
                                'month' => 2,
                                'quarter' => 1,
                                'day_of_year' => 32,
                                'day_of_quarter' => 32,
                                'minute' => 26,
                                'day' => 1,
                                'day_of_week' => 3,
                                'year' => 2012
                              },
                 'utc_rd_secs' => 44783,
                 'formatter' => undef,
                 'tz' => bless( {
                                  'name' => 'UTC'
                                }, 'DateTime::TimeZone::UTC' ),
                 'utc_year' => 2013,
                 'utc_rd_days' => 734534,
                 'offset_modifier' => 0
               }, 'DateTime' );

01-02-2012 12:26:23
[download]

--
No matter how great and destructive your problems may seem now, remember, you've probably only seen the tip of them. [1]

Comment on Condensed output of Data::Dumper used on DateTime Select or Download Code

Replies are listed 'Best First'.
Re: Condensed output of Data::Dumper used on DateTime by jmcnamara (Monsignor) on Feb 01, 2012 at 13:11 UTC
Here is an article by brian d foy in the Effective Perler about using Data::Dump filters for nicer pretty-printing in relation to DateTime objects. I don't know if similar overloading is possible with Data::Dumper. -- John.	[reply]
Re: Condensed output of Data::Dumper used on DateTime by educated_foo (Vicar) on Feb 01, 2012 at 16:17 UTC
Yes. Set `$Data::Dumper::Freezer` to the name of the method you want called, then implement that method in DateTime, e.g. `$Data::Dumper::Freezer = 'my_dumper'; sub DateTime::my_dumper { ... }` [download] Even better, you can localize the change to `$Data::Dumper::Freezer` so pretty-printing only happens when you want it to. EDIT: Corrected to `DateTime::my_dumper`...	[reply] [d/l] [select]
Re: Condensed output of Data::Dumper used on DateTime by tobyink (Canon) on Feb 01, 2012 at 14:03 UTC
Dump Data::Dumper. Use Data::Printer instead. Data::Dumper generates executable Perl code as its output. That's great. But when debugging, do you actually need executable Perl to be dumped? Probably not. You can probably get more readable information if you sacrifice your dump's executability. One of the examples given in the Data::Printer pod is dealing with DateTime.	[reply]
Re: Condensed output of Data::Dumper used on DateTime by thargas (Deacon) on Feb 01, 2012 at 13:26 UTC
Data::Dump's `dump_filtered()` will allow you to apply whatever filters seem right to you. I don't see any such hook in Data::Dumper	[reply] [d/l]
Re: Condensed output of Data::Dumper used on DateTime by andreas1234567 (Vicar) on Feb 28, 2012 at 22:26 UTC
A complete working sample using Data::Dumper where an object having a DateTime reference can be printed in short and long versions depending on `$Data::Dumper::Freezer`: `# Foo.pm package Foo; use strict; use warnings; use DateTime; sub new { my $class = shift; my $self = { _timevalue => DateTime->now, }; bless( $self, $class ); return $self; } # Modify _timevalue and return blessed reference sub _dumper_hook { $_[0] = bless { %{ $_[0] }, _timevalue => $_[0]->{_timevalue}->ymd . q{ } . $_[0]->{_timevalue}->hms, }, ref( $_[0] ); } 1; __END__` [download] Test: `# Foo.t: use strict; use warnings; use Data::Dumper; use Test::More; plan tests => 2; use Foo; my $f = Foo::->new; { local $Data::Dumper::Freezer = '_dumper_hook'; like( Dumper($f), qr/\d{4}-\d{2}-\d{2}\s\d{2}:\d{2}:\d{2}/, q{prints short version} ); } unlike( Dumper($f), qr/\d{4}-\d{2}-\d{2}\s\d{2}:\d{2}:\d{2}/, q{prints long version} ); __END__` [download] Run: $ perl Foo.t 1..2 ok 1 - prints short version ok 2 - prints long version $ [download] -- No matter how great and destructive your problems may seem now, remember, you've probably only seen the tip of them. [1]	[reply] [d/l] [select]


Your skill will accomplish what the force of many cannot
	PerlMonks