Cory G Watson > DBIx-Class-QueryLog > DBIx::Class::QueryLog::Analyzer

Download:
DBIx-Class-QueryLog-1.3.3.tar.gz

Dependencies

Annotate this POD

CPAN RT

Open  0
View/Report Bugs
Source  

NAME ^

DBIx::Class::QueryLog::Analyzer - Query Analysis

SYNOPSIS ^

Analyzes the results of a QueryLog. Create an Analyzer and pass it the QueryLog:

    my $schema = ... # Get your schema!
    my $ql = DBIx::Class::QueryLog->new;
    $schema->storage->debugobj($ql);
    $schema->storage->debug(1);
    ... # do some stuff!
    my $ana = DBIx::Class::QueryLog::Analyzer->new({ querylog => $ql });
    my @queries = $ana->get_sorted_queries;
    # or...
    my $totaled = $ana->get_totaled_queries;

METHODS ^

new

Create a new DBIx::Class::QueryLog::Analyzer

get_sorted_queries

Returns an arrayref of all Query objects, sorted by elapsed time (descending).

get_fastest_query_executions($sql_statement)

Returns an arrayref of Query objects representing in order of the fastest executions of a given statement. Accepts either SQL or a DBIx::Class::QueryLog::Query object. If given SQL, it must match the executed SQL, including placeholders.

  $ana->get_slowest_query_executions("SELECT foo FROM bar WHERE gorch = ?");

get_slowest_query_executions($sql_statement)

Opposite of get_fastest_query_executions. Same arguments.

get_totaled_queries

Returns hashref of the queries executed, with same-SQL combined and totaled. So if the same query is executed multiple times, it will be combined into a single entry. The structure is:

    $var = {
        'SQL that was EXECUTED' => {
            count           => 2,
            time_elapsed    => 1931,
            queries         => [
                DBIx::Class::QueryLog...,
                DBIx::Class::QueryLog...
            ]
        }
    }

This is useful for when you've fine-tuned individually slow queries and need to isolate which queries are executed a lot, so that you can determine which to focus on next.

To sort it you'll want to use something like this (sorry for the long line, blame perl...):

    my $analyzed = $ana->get_totaled_queries;
    my @keys = reverse sort {
            $analyzed->{$a}->{'time_elapsed'} <=> $analyzed->{$b}->{'time_elapsed'}
        } keys(%{ $analyzed });

So one could sort by count or time_elapsed.

get_totaled_queries_by_bucket

Same as get_totaled_queries, but breaks the totaled queries up by bucket:

$var = { 'bucket1' => { 'SQL that was EXECUTED' => { count => 2, time_elapsed => 1931, queries => [ DBIx::Class::QueryLog..., DBIx::Class::QueryLog... ] } } 'bucket2' => { ... } }

It is otherwise identical to get_totaled_queries

AUTHOR ^

Cory G Watson <gphat at cpan.org>

ACKNOWLEDGEMENTS ^

COPYRIGHT & LICENSE ^

Copyright 2009 Cory G Watson, all rights reserved.

This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

syntax highlighting: