Mapping use cases

This page covers common and advanced mapping scenarios : type conversions, complex conditions, and multi-field logic.

Map a TEXT field to a NUMBER field

When numbers are stored as text, formats can vary (decimal separators, thousands separators, units, placeholders), making conversion ambiguous and error-prone. To be filterable and sortable, numbers must be stored as numbers, if they are stored as text, sorting and range filters will not behave reliably.

Examples of ambiguous formats:

• 1,000.90 (US) is equivalent to 1 000,90 (many European formats).
• 1,000 (US) is not equivalent to 1,000 (Europe, where , can be a decimal separator).
• Some legacy systems use placeholders like NA or - to mean "no value".

Mixing number formats across data sources increases the risk of conversion errors and inconsistent data.

WARNING

Parsing text into numbers is inherently ambiguous and error-prone. Keep inputs as consistent as possible (same locale/format) and add validation checks.

Use a template (Twig) to normalize the text value before writing to a NUMBER field.

A basic example to handle spaces:

Replace PRICE_TEXT with your source field key.

{% set v = targetItem.fields.PRICE_TEXT.value.data | default('') | trim %}
{% set v = v | replace({' ':''}) %}
{{ v }}

A more complete example that handles thousands separators, decimal separators, and common placeholders (NA, N/A, -):

Replace PRICE_TEXT with your source field key.

{# Read and normalize common number-as-text formats #}
{% set raw = targetItem.fields.PRICE_TEXT.value.data | default('') | trim %}
{% set v = raw | lower %}

{# Common "no value" placeholders #}
{% if v in ['', 'na', 'n/a', '-', '—'] %}
  {{ '' }}
{% else %}
  {# Remove spaces (including non-breaking), apostrophes, and some currency symbols #}
  {% set v = v | replace({' ':'', ' ':'', "'":'', '€':'', '$':'', '£':''}) %}

  {% set hasComma = (v | split(',') | length) > 1 %}
  {% set hasDot = (v | split('.') | length) > 1 %}

  {% if hasComma and hasDot %}
    {% set afterComma = v | split(',') | last %}
    {% set afterDot = v | split('.') | last %}

    {# Heuristic: 1–2 digits → decimal separator #}
    {% if afterDot|length in [1,2] %}
      {% set v = v | replace({',':''}) %}
    {% elseif afterComma|length in [1,2] %}
      {% set v = (v | replace({'.':''})) | replace({',':'.'}) %}
    {% else %}
      {% set v = (v | replace({',':''})) | replace({'.':''}) %}
    {% endif %}

  {% elseif hasComma %}
    {% set parts = v | split(',') %}
    {% if parts|length == 2 and (parts[1]|length in [1,2]) %}
      {% set v = v | replace({',':'.'}) %}
    {% else %}
      {% set v = v | replace({',':''}) %}
    {% endif %}

  {% elseif hasDot %}
    {% set parts = v | split('.') %}
    {% if parts|length != 2 or (parts[1]|length not in [1,2]) %}
      {# Most likely a thousands separator #}
      {% set v = v | replace({'.':''}) %}
    {% endif %}
  {% endif %}

  {{ v }}
{% endif %}

Map a TEXT or NUMBER field to a SINGLE-SELECT or CLASSIFICATION field

Text or Number values cannot always be predicted or normalized in a reliable way.

For example, the color "green" might appear as:

• Green
• green
• GREEN
• 104 (a color code)

Since these values cannot be normalized automatically, the approach depends on the number of predictable input values.

Few predictable values - one rule per input

Create one rule per input value using conditions : each rule handles one specific case explicitly.

1. In Action, select Use a default value and choose the target option or category to apply.
2. In Only if, set the conditions describing when this value must be used (for example when source text equals "Green").
3. Before saving, click on Add new rule.
4. Give a meaningful rule name, for example: Green for "Green".
5. Click on Save and add a new rule and repeat the process for each predictable input value.

You will end up with multiple rules, each one handling a specific input value.

Many predictable values - use a template

For a larger number of values, a template is more efficient:

{% if targetItem.fields["SOURCE"].value.data|default('')|trim|upper == 'GREEN' %}
  GREEN
{% elseif targetItem.fields["SOURCE"].value.data|default('')|trim == '104' %}
  GREEN
{% endif %}

Replace SOURCE with your source field key and add as many conditions as needed.

Map a CLASSIFICATION field depending on multiple source fields

Sometimes a classification depends on the combination of several source fields rather than a single one.

In this example:

• The retailer target table contains:
  • EAN — the product EAN 13
  • Classification — the website classification, with possible values:
    • Vacuum cleaner wireless
    • Vacuum cleaner with wire
• The supplier source table contains:
  • EAN 13 — the product EAN 13
  • Typology — product typology, with values:
    • Vacuum cleaner
    • Coffee machine
  • Type — specific to Vacuum cleaner, with values:
    • Wireless
    • With wire

The business rule is:

• If Typology is Vacuum cleaner and Type is Wireless Then Classification = Vacuum cleaner wireless
• If Typology is Vacuum cleaner and Type is With wire Then Classification = Vacuum cleaner with wire

Create one rule per case on the same target field.

First rule : Vacuum cleaner wireless:

1. In Action, select Use a default value and choose Vacuum cleaner wireless.
2. In Only if, add:
   • A condition on Typology equals Vacuum cleaner
   • A condition on Type equals Wireless
3. Click + Add new rule and name it Classification – vacuum cleaner wireless.

Second rule : Vacuum cleaner with wire:

1. In Action, select Use a default value and choose Vacuum cleaner with wire.
2. In Only if, add:
   • A condition on Typology equals Vacuum cleaner
   • A condition on Type equals With wire
3. Click Save and add a new rule.

Order the rules as needed — usually the more specific rules first.

Rules are evaluated in order. Make sure the most specific rules come first.