column_aggregator_task.md

ColumnAggregatorTask

For each configured column, collects the input rows that contain this column (and match the optional condition), and outputs all the groups once all previous tasks are resolved.

Task reference

• Service: CleverAge\ProcessBundle\Task\ColumnAggregatorTask
• Blocking task

Accepted inputs

array: an associative array that should contain the configured columns (a column is present as soon as its key exists, even if its value is null)

Possible outputs

array: associative array indexed by column name, each entry being:

• <reference_key>: the column name
• <aggregation_key>: list of the whole input rows that contain this column and matched the condition

Columns that never matched are absent; if nothing matched, the output is an empty array.

Options

Code	Type	Required	Default	Description
columns	array	X		List of column keys to aggregate on
reference_key	string		column	Key holding the column name in each output group
aggregation_key	string		values	Key holding the aggregated rows in each output group
condition	array		[]	Conditions (match, not_match, empty, not_empty, match_regexp, not_match_regexp), see ConditionTrait
ignore_missing	bool		false	If true, missing columns only log a warning instead of throwing an \UnexpectedValueException

The condition is checked, for each column, against an array with two keys: input_column_value (the value of the column) and input (the whole row). Use property paths such as [input_column_value] or [input][status].

Examples

• Aggregate rows whose col1 equals A

# Task configuration level
iterate:
  service: '@CleverAge\ProcessBundle\Task\InputIteratorTask'
  outputs: [aggregate_a]
aggregate_a:
  service: '@CleverAge\ProcessBundle\Task\ColumnAggregatorTask'
  options:
    columns: [col1]
    condition:
      match:
        '[input_column_value]': A

With inputs {col1: A, col2: 1}, {col1: B, col2: 2} and {col1: A, col2: 3}, the output is {col1: {column: col1, values: [{col1: A, col2: 1}, {col1: A, col2: 3}]}}.