Overview

The mapping step ensures that incoming data, whether from suppliers or other systems, is accurately organized according to the expected structure. It allows the user to map the source columns of the input to the attribute list of the destination.

Configuration

Below are some key parameters specific to the Mapping configuration:

• fields - override of project fields. The properties below are specific to the Field object in the mapping config:
  • mode what to do when mapping multiple columns into a single field
    • join - join them with a separator
    • first - use the first non empty value
  • protected bool(False) - prevent the user from modifying the field in the UI. Useful when assigning a default in the config. For example, the client has a rule that says the product name is the brand - product name. We create the field by concatenating the two fields. Protected true will ensure that the user cannot do anything.
  • default: can assign default values to the columns if there is no value
    • type:
      • constant a hard coded value will be populated in the output column
      • column: if we want to read values from a column
      • metadata rarely used, can be used to input metadata such as job creator name or portal name
    • column: the value of a field from the job (e.g. ean, sku… ) when type=column
    • constant: the value of the hard coded content when type=constant
    • metadata: the source of the metadata when type=metadata
  • editable_default: only used when default.type=constant. Adds an input field to the mapping UI that allows dynamically modifying the default value.
• include_fields - Use only these fields from the project - See Normalization#include_exclude
• exclude_fields - Use all fields in the project except the ones in this group - See Normalization#include_exclude
• groups: same as in the Normalization step
• max_requirement_level: see the detailed explanation
• Specific mapping options
  • default_join_separator: If we select multiple source columns, how will we concatenate them.
  • block_learning: Prevents saving the mappings for the demo because otherwise, we would have problems for the next demo (rename to block_learning_demo?). False by default.

Op Status Mapping

• stats
• original mapping: what SDM mapped automatically
• user mapping: what the user mapped afterwards

This information will be particularly useful to retrace the events in case the customer has issues/complains.

Mapping results

• old: field as it is entered in the source file.
• group: of the new field.
• counter: how many times this mapping was done.
• new: destination field.

Use Cases

• Structuring Supplier Catalogs: Mapping helps format supplier catalogs to align with your internal system’s expectations.
• Concatenating Fields: Combine values from multiple fields (e.g., joining product names and brands).
• Transforming Data: Modify data by converting it to uppercase or applying other transformations.

Interface

The attributes are grouped by Display Groups, such as "Main Attributes" and "Other Attributes." 

You can map multiple source columns to a single destination attribute by applying either:

• Concatenation Logic: Combine columns with a separator.
• Selection Logic: Choose the first non-empty column.

(More information on Display Groups: Fields, Groups, Display Groups & Available Locales)

The “Unmapped fields only” option is enabled by default, allowing users to focus on mapping new fields. SDM’s memory tracks how often specific source columns are mapped to a destination attribute, automatically favoring the most frequently mapped columns.

Example: If “product name” has been mapped to “name” 100 times and “nom” mapped to “name” 90 times, SDM will prioritize mapping “product name” in future files that contain both columns.

Limitations and Known Issues

• Display Groups: Fields are grouped by "main_attributes" and "other_attributes," a legacy structure that limits flexibility.
  • Mapping per Field x Group: Mapping is performed for each field x group combination rather than per field alone. (See Groups)
  • No Advanced Customization: Custom transformations (e.g., concatenation or uppercasing) cannot be applied directly within the Mapping step config.
    • Example activite_animal: If we create main_attributes or other_attributes groups in the config, we need to think about excluding the groups at the Normalization and Extraction steps.
• Each workflow begins automatically with a Mapping step.
• Cannot customize the mapped value, for example by concatenating or uppercasing it