1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243
|
# You may distribute under the terms of either the GNU General Public License
# or the Artistic License (the same terms as Perl itself)
#
# (C) Paul Evans, 2020-2022 -- leonerd@leonerd.org.uk
package Metrics::Any::Adapter::Test 0.10;
use v5.14;
use warnings;
use base qw( Metrics::Any::AdapterBase::Stored );
use Carp;
use List::Util 1.29 qw( pairs );
use Scalar::Util qw( reftype );
=head1 NAME
C<Metrics::Any::Adapter::Test> - a metrics reporting adapter for unit testing
=head1 SYNOPSIS
use Test2::V0;
use Metrics::Any::Adapter 'Test';
{
Metrics::Any::Adapter::Test->clear;
# perform some work in the code under test
is( Metrics::Any::Adapter::Test->metrics,
"an_expected_metric = 1\n",
'Metrics were reported while doing something'
);
}
=head1 DESCRIPTION
This L<Metrics::Any> adapter type stores reported metrics locally, allowing
access to them by the L</metrics> method. This is useful to use in a unit test
to check that the code under test reports the correct metrics.
This adapter supports timer metrics by storing as distributions. By default,
distributions store only a summary, giving the count and total duration. If
required, the full values can be stored by setting L</use_full_distributions>.
For predictable output of timer metrics in unit tests, a unit test may wish to
use the L</override_timer_duration> method.
This adapter type supports batch mode reporting. Callbacks are invoked at the
beginning of the L</metrics> method.
=cut
my $singleton;
sub new
{
my $class = shift;
return $singleton //= $class->SUPER::new( @_ );
}
# We're a singleton instance, so we can just store state in file-scoped
# lexicals
my $timer_duration;
my $use_full_distributions;
=head1 METHODS
=cut
=head2 metrics
$result = Metrics::Any::Adapter::Test->metrics
This class method returns a string describing all of the stored metric values.
Each is reported on a line formatted as
name = value
Each line, including the final one, is terminated by a linefeed. The metrics
are sorted alphabetically. Any multi-part metric names will be joined with
underscores (C<_>).
Metrics that have additional labels are formatted with additional label names
and label values in declared order after the name and before the C<=> symbol:
name l1:v1 l2:v2 = value
=cut
=head2 use_full_distributions
Metrics::Any::Adapter::Test->use_full_distributions; # enables the option
Metrics::Any::Adapter::Test->use_full_distributions( $enable );
I<Since version 0.08.>
If enabled, this option stores the full value of every reported observation
into distributions, rathr than just the count-and-total summary.
Full value distributions will be formatted as a sequence of lines containing
the count of observations at that particular value, in square brackets,
followed by the summary count.
name[v1] = c1
name[v2] = c2
...
name_count = c
In order not to be too sensitive to numerical rounding errors, values are
stored to only 3 decimal places.
=cut
sub use_full_distributions
{
shift;
( $use_full_distributions ) = @_ ? @_ : ( 1 );
}
use constant {
DIST_COUNT => 0,
DIST_TOTAL => 1,
};
sub store_distribution
{
shift;
my ( $storage, $amount ) = @_;
if( $use_full_distributions ) {
$storage //= {};
$storage->{ sprintf "%.3f", $amount }++;
}
else {
$storage //= [ 0, 0 ];
$storage->[DIST_COUNT] += 1;
$storage->[DIST_TOTAL] += $amount;
}
return $storage;
}
*store_timer = \&store_distribution;
sub metrics
{
my $self = shift;
ref $self or $self = Metrics::Any::Adapter::Test->new;
my @ret;
$self->walk( sub {
my ( $type, $name, $labels, $value ) = @_;
$name .= sprintf " %s:%s", $_->key, $_->value for pairs @$labels;
if( $type eq "counter" or $type eq "gauge" ) {
push @ret, "$name = $value";
}
elsif( $type eq "distribution" or $type eq "timer" ) {
if( reftype $value eq "HASH" ) {
my $total = 0;
foreach my $k ( sort { $a <=> $b } keys %$value ) {
my $v = $value->{$k};
# Trim trailing zeroes for neatness
$k =~ s/\.0+$/./; $k =~ s/\.$//;
push @ret, "$name\[$k] = $v";
$total += $v;
}
push @ret, "${name}_count = $total";
}
else {
push @ret, "${name}_count = " . $value->[DIST_COUNT];
push @ret, "${name}_total = " . $value->[DIST_TOTAL];
}
}
else {
warn "Unsure how to handle metric of type $type\n";
}
} );
return join "", map { "$_\n" } @ret;
}
=head2 clear
Metrics::Any::Adapter::Test->clear
This class method removes all of the stored values of reported metrics.
=cut
sub clear
{
my $self = shift;
ref $self or $self = Metrics::Any::Adapter::Test->new;
$self->clear_values;
undef $timer_duration;
}
=head2 override_timer_duration
Metrics::Any::Adapter::Test->override_timer_duration( $duration )
This class method sets a duration value, that any subsequent call to
C<inc_timer> will use instead of the value the caller actually passed in. This
will ensure reliably predictable output in unit tests.
Any value set here will be cleared by L</clear>.
=cut
sub override_timer_duration
{
shift;
( $timer_duration ) = @_;
}
sub report_timer
{
my $self = shift;
my ( $handle, $duration, @labelvalues ) = @_;
$duration = $timer_duration if defined $timer_duration;
$self->SUPER::report_timer( $handle, $duration, @labelvalues );
}
=head1 AUTHOR
Paul Evans <leonerd@leonerd.org.uk>
=cut
0x55AA;
|