Skip to main content

Dimensions

A summary table can have one or more dimensions. Dimensions are properties of the source entity that are used to group the data. They have distinct, descriptive values.

In a summary table, a dimension is defined using the #[Analytics\Dimension] attribute.

Source Definition​

The most important property of a dimension is the source argument. This argument specifies how we get the value from the source entity.

use Doctrine\DBAL\Types\Types;
use Doctrine\ORM\Mapping as ORM;
use Rekalogika\Analytics\Attribute as Analytics;
use Rekalogika\Analytics\ValueResolver\PropertyValueResolver;

class YourSummary extends Summary
{
    #[ORM\Column(type: Types::STRING)]
    #[Analytics\Dimension(
        source: new PropertyValueResolver('status'),
    )]
    private ?string $status = null;
}

The source can be of these types:

  • string: The name of the property in the source entity; can also be a path that will traverse the entity tree. The framework will try to autodetect how to get the value from the source entity.
  • A ValueResolver instance: Defines a DQL expression on how to get the value from the source entity.
  • An array of string or ValueResolver instances: Explained TBD.

Hierarchical Dimensions​

A dimension can be hierarchical. A common example is the time dimension. The time dimension can be further divided into year, month, day, and more. The caller will then be able to decide whether to group the time by year, month, or day, etc.

use Doctrine\ORM\Mapping as ORM;
use Rekalogika\Analytics\Attribute as Analytics;
use Rekalogika\Analytics\Model\Hierarchy\TimeDimensionHierarchy;
use Rekalogika\Analytics\ValueResolver\PropertyValueResolver;

class YourSummary extends Summary
{
    #[ORM\Embedded()]
    #[Analytics\Dimension(
        source: new PropertyValueResolver('time'),
        sourceTimeZone: new \DateTimeZone('UTC'),
        summaryTimeZone: new \DateTimeZone('Asia/Jakarta'),
    )]
    private TimeDimensionHierarchy $time;
}

The framework provides these predefined time hierarchies:

  • TimeDimensionHierarchy
  • DateDimensionHierarchy
  • SimpleTimeDimensionHierarchy
  • SimpleDateDimensionHierarchy