Browse documentation

#Constraints

Data constraints apply business rules and data integrity rules during processing, so that the data quality stays high. A violated constraint throws a ConstraintViolationException with details about the row.

#Unique Constraints

Make sure that the values in specified columns are unique across the full dataset.

<?php

use function Flow\ETL\DSL\{data_frame, from_array, constraint_unique, to_output};

$dataFrame = data_frame()
    ->read(from_array([
        ['email' => 'user1@example.com', 'username' => 'user1'],
        ['email' => 'user2@example.com', 'username' => 'user2'],
        ['email' => 'user1@example.com', 'username' => 'user3'], // This will cause constraint violation
    ]))
    ->constrain(constraint_unique('email'))
    ->write(to_output())
    ->run();

#Multiple Column Constraints

Make sure that combinations across multiple columns are unique:

<?php

use function Flow\ETL\DSL\{constraint_unique};

$dataFrame = data_frame()
    ->read($extractor)
    ->constrain(constraint_unique('username', 'tenant_id'))
    ->write($loader)
    ->run();

#Custom Constraints

For a custom constraint, create a class that implements the Constraint interface:

<?php

use Flow\ETL\Operators\Constraint;
use Flow\ETL\Values\Rows;

final class AgeRangeConstraint implements Constraint
{
    public function __construct(private int $minAge, private int $maxAge) {}

    public function firstViolation(Rows $rows): ?int
    {
        foreach ($rows->column('age')->values() as $i => $age) {
            if ($age < $this->minAge || $age > $this->maxAge) {
                return $i;
            }
        }

        return null;
    }

    public function toString(): string
    {
        return "Age must be between {$this->minAge} and {$this->maxAge}";
    }

    public function violation(Rows $rows, int $index): string
    {
        return "Age {$rows->column('age')->value($index)} is outside allowed range";
    }
}

Found a typo or an outdated section? Edit this page on GitHub ↗