› Workflow › Data models › Create a local data model using JSON

Create a local data model using JSON

You can define a data model local to a process, case, rule, or custom service using a JSON file.

Use the following best practices when you create a data model:

  • Keep only the fields you need to keep your data model lightweight.
  • Population of an object in the data model is non-strict by default:
    • It ignores extra fields in the incoming JSON that are not defined in the model.
    • It matches and populates only the fields that exist in both the JSON and the model.
    • It ignores unmatched fields.

When a data model is in use, removing a referenced field can cause a runtime error.

Procedure
  1. Open the process map for which to create a data model.
  2. On the Process modeling bar, select Data models.

    The Data models page appears.

    The local data model inherits the process category.

  3. Select New.
    The Add Data model dialog box is displayed.
  4. Enter a Name for the data model.
  5. Optional. Provide a Description of the data model.
  6. To add a JSON file, select Load from external source.

    The Load from external dialog box is displayed.

  7. On the Type list select JSON expression. (Default: JSON expression)
  8. Browse the file or drag the file into the File box.

    You can also paste the sample JSON format into a File content.

    Once you add the files, the file preview is displayed in the File content box. You can edit the file if required.

  9. Optional. To populate default values from the sample JSON data, select Populate defaults from sample JSON. (Default: Clear)
  10. Select Add.

    • If you choose to populate defaults from the sample JSON, a confirmation message appears. Select Yes to continue or No to cancel.

    • The JSON file is parsed, and the fields are displayed in the tree structure in the canvas.

    • You can change or rename the field properties or delete them from the tree.

    • The field names support these special characters (!@#$%^&*()_+-={}|:"<>?,./;'[]\).

    • If the JSON file cannot be parsed, an error message is displayed.

    • When you define a tree structure for the data model, but provide a different JSON sample, an error message is displayed that the existing model will be overwritten. You must confirm if you want to continue; this action cannot be undone.

  11. You can rearrange the fields using drag and drop. The updated field order is automatically reflected even in the Data object variable.

    • You can reorder the entire object, or fields that are at same level or within a list.

    • You can move the entire object or list up and down, or move fields from one object to another object; however, you cannot move the fields out of a list.

  12. Select Add.

    The data model is saved as part of the process definition.

    You can similarly create a data model local to a business rule, case, or custom service using a JSON file.