Create mappings

A Mapping defines the transformation rules for integrating data from one or more source datastores into one or several target datastores.

These transformation rules are used by the templates attached to this mapping to generate a complete and optimized data integration process.

Create a mapping

To create a mapping:

  1. In the Project Explorer toolbar, click the mapping New Mapping button. The New Mapping wizard opens.

  2. Select the parent folder or project for your new resource.

  3. Enter a File Name and click Finish. The mapping file is created and the editor opens.

Add the source and target datastores

To add the source and target datastores:

  1. In the Project Explorer, expand the metadata containing a source or target datastore (table, flat file, XML file…​) that you want to use in your mapping.

  2. Drag and drop the datastore from the Project Explorer into the mapping editor.

  3. Select this datastore in the editor. In the properties view, set the following properties:

    Property Description


    Alias used in the expressions to refer to this datastore. By default, it is the name of the datastore.

    Use CDC

    Check this box to enable consumption of changed data from a source datastore, captured via the CDC feature. Refer to Change Data Capture (CDC) for more information.

    Enable Rejects Detection

    Check this box to enable rejects management on a target datastore. Refer to Rejects management for more information.

    Update Key Provider

    Defines the key used to identify existing records in the context of updates:

    • Manual Definition: Select this value to manually define the columns to use for the key. Select each column on the datasource and click the key button to add it to the definition of the key.


    • Select an existing constraint from the metadata.

    Staging Area

    Select the datastore to use as the staging area. This option is useful to benefit from the transformation capabilities of another datastore that are not available in the current datastore.

    For example, using SQL functions in a mapping that involves a JSON or an XML file.

    When defining a staging area, a block representing the load process template is automatically added to the current datastore.


    Add a tag to the table (used in process templates).


    Free form text.

    Integration Sequence

    Specifies the order in which tables without any mutual dependencies must be loaded.


    Defines the order of a source datastore in the FROM clause generated when loading a target.

  4. Repeat the same operations for all datastores from which data is to be extracted (the sources) and into which data is to be loaded (the targets).

  5. Press Ctrl+S to save the mapping.

To create a link between two datastores:

  1. Select a column from a source datastore in the mapping diagram.

  2. Drag this column onto another source column in the mapping diagram.

  3. Release the mouse button. You are prompted to select the type of link to create:

    link types

    • Join: Creates a new join between the two datastores. Refer to Join sources for more information.

    • Map - current field: Creates a mapping (source-target link) between the two columns.

    • Map - all matching fields: Creates a mapping (source-target link) between the two columns and between each other group of source and target columns having the same name.

    • Exists / Not Exists: Adds an Exists/Not Exists operation to filter data from the source datastore depending on data from the target datastore (see Exists/Not Exists filters).

You are prompted to select the type of link only if Semarchy xDI Designer detects that several types of links can be created. If only one type of link can be created, the accurate type will be selected automatically.
When creating a mapping link between datastores, blocks representing the Load templateLoad and Integration templateIntegration process templates are automatically added to the upper part of the target datastore. See Configure templates.


  1. It is not possible to use a Staging Area as a source for another Staging Area if both use different connections.

  2. A datastore cannot be linked to another datastore with both mapping and join link types.

Configure mappings

Once a mapping link has been created between datastores, it can be configured from the Properties view of the mapped column in the target datastore.

The following properties are available:

  • Enable: Enables or disables the mapping.

  • Execution Location: Defines where the mapping is executed:

    • Source: The mapping will be executed by the source datastore.

    • Staging Area: The mapping will be executed in the staging area.

    • Target: The mapping will be executed by the target datastore.

  • Use as Key: Select this option to use the selected column as part of the unique key for this mapping in order to identify records for reject management and target records update purposes. Several columns can participate in the unique key in a mapping. These columns can be the same as the columns involved in primary or alternate key constraints for this datastore, or be distinct.

  • Enable Insert: Enables inserting data with this mapping.

  • Enable Update: Enables updating data with this mapping.

  • Aggregate: Select this option if the selected column’s code contains an aggregate expression. Other (non-aggregated) columns will be added to the GROUP BY clause of the generated queries.

  • Tag: Add a tag to the mapping. Tags are used in process templates.

  • Description: Free form text.

Mapping expressions

A mapping expression describes how columns in a source datastore are used to populate columns in a target datastore, and how this data is processed.


A mapping, filter, or join expression must use the language and functions from the engine that will process it, as defined by the Execution Location property. In a typical situation where the datastore or staging area is on a relational database, mapping expressions are SQL expressions for that database engine. You can use literal values, column names, database functions, and operators in these expressions.

Edit mapping expressions

To edit a mapping expression:

  • Either use the Expression Editor:

    1. Open the Expression Editor view and make sure the Expression Editor lock button lock option is not enabled. If it is enabled, press the Expression Editor icon button to unlock the view.

    2. Select the column from the target datastore for which you want to update the mapping expression.

    3. The Expression Editor view is updated and shows the selected column’s mapping expression.

    4. Edit the expression and save the editor.

      Drag and drop columns from datastores onto the Expression Editor to add them to the current expression.
      Press the Expression Editor lock button button to lock the content of the Expression Editor to the currently selected element from the mapping diagram.
  • Or use the Mapper Tool:

    1. Right-click the target datastore in a mapping and select Mapper Tool. The Mapper Tool opens.

    2. Select the target column for which you want to edit the mapping expression. The expression is displayed in the edition zone.

      Mapper Tool

    3. Edit the mapping expression.

    4. Close the Mapper Tool window.

    5. Save the editor.

Think of a mapping expression as an element of the column list in a SELECT statement.

Examples of valid mapping expressions using SQL syntax:

  • SRC.FIRST_NAME || ' ' || SRC.LAST_NAME (concatenates values from different columns)

  • SUM(ORDERLINE.AMOUNT) (aggregate field)

  • 'ADMIN' (constant value)

Target columns in expressions

Mapping expressions are typically used to select and manipulate data before it reaches the target datastore. In certain cases, you can also manipulate the data after it reaches the target, by referencing a target column in a field expression. The following conditions have to be met:

  • The Execution Location must be set to Target.

  • In the mapping, your datastore is using a compatible template.

  • You cannot have both a UDF and a target column in an expression.

At this time, INTEGRATION HSQL and INTEGRATION Rdbms templates are supported.

Column icons

Source and target columns are displayed in the mapping editor with an icon that contains the first letter of their datatype. This icon appears in grey when the source column is not used in the mapping or when the target column is not mapped.

In addition, target columns are tagged with icons to identify their key properties. The various icons and their meaning are listed in the following table.

Icon Meaning

col key

The yellow key indicates that the column is part of the key. The white star in the upper right corner indicates that the column is not nullable. If reject management is activated, rows with null values for this column are rejected. The letter represents the column’s data type (I: Integer, V: Varchar, etc.)

col not nullable and checked

The star in the upper right corner means that the column is not nullable and Semarchy xDI checks the null values for this column.

col not nullable and not checked

The cross in the upper right corner means that the column is not nullable but Semarchy xDI does not check the null values for this column.

col nullable and not checked

No sign in the upper right corner means that the column is nullable and that Semarchy xDI does not check the null values for this column.

col nullable and checked

The plus sign in the upper right corner means that the column is nullable but Semarchy xDI checks the null values for this column.

exec source

This expression runs in the source.

exec staging

This expression runs in the staging area.

exec target

This expression runs in the target.

col four letters

These four letters have the following meaning:

  • I: Enable Insert is selected for this mapping.

  • U: Enable Update is selected for this mapping.

  • A: Aggregate is selected for this mapping.

  • T: One or more tags are set for this mapping.

Configure templates

Templates in Semarchy xDI are generic and configurable elements used to generate process steps. Templates are automatically added to Stages and target datastores depending on mapping definition:

  • Load templates are used to generate process steps loading data from a source. A Load template iconLoad template is automatically added to Stages and to target datastores for each source datastore mapped with them and using a different technology.

  • Integration templates are used to generate process steps in charge of integrating data into the target. An Integration template iconIntegration template is automatically added to target datastores mapped with at least one source or Stage.

  • Stage templates are used to generate process steps that load data in a Stage. A Stage template iconStage template is automatically added to Stages mapped with a source.

  • Reject templates are used to generate process steps that handle rejects occurring while loading and integrating data. A Reject template iconReject template is automatically added to target datastores for which reject management is enabled.

To configure a template added to a mapping:

  1. Select the template’s icon.

  2. The Properties view shows the template’s configuration. This configuration is composed of:

    • The template file that will be used to generate the process (Template LOAD Rdbms to in the example below).

    • The parameters that apply to this template.

      Template properties

  3. If required, use the drop-down list to select the expected template file to use for this mapping.

    To managed the installed templates, refer to Import templates
  4. Review and edit template parameters. To modify a parameter:

    1. Click the parameter’s label to make it editable.

    2. Edit the parameter’s value.

      Click the label of a populated parameter to revert to its default value.
      The most frequent template parameters are listed in Template parameters

Computed fields

A Computed Field defined on a given datastore is calculated on the fly during mapping execution but is not stored.

To create a Computed Field on a mapping:

  1. Right-click a column from the datastore and select Create Computed Field.

  2. If the Computed Field is the first one for this datastore: enter a Container’s Computed Field Alias, which defines the label of the container that will contain all Computed Fields for this datastore.

  3. A new Computed Field is added to the container named after Container’s Computed Field Alias. By default, this field has the following properties:

    • Name: newField_<sequence number>

    • Expression: <datastore name>.<selected column name>

  4. Click the name of the Computed Field to edit it.

  5. Double-click the Computed Field’s value to edit it in the Mapper Tool.

  6. Exit the Mapper Tool and save the editor.

Computed Fields can be created only on objects having transformation capabilities (RDBMS mostly, with SQL syntax).
It is possible to create a Computed Field from another Computed Field, which is useful to chain transformations or operations.

Copy-paste operations in mappings

You can copy and paste different mappings elements within mappings as well as to other mappings.

Use the keyboard controls Ctrl+C and Ctrl+V or +C and +V on Mac.