Data Mart Multilingual Configuration

Enable a Mart Schema for Local Languages

You can enable local language support for Data Mart.

To enable any of your Mart schemas to support local-language text:

1. From the Manage Mart window, click the Manage Data Marts link and then the Create or Edit link.
   You see the Define Data Mart Instance dialog box.

2. Select the Enable Internationalization check box.
   The following figure shows a mart schema that has been enabled to support local language text.

   

    

When you internationalize a mart, all the tables in that schema immediately receive a new key field, DMART_LANGUAGE. This includes tables like Trade Details that do not currently support local-language fields. A new tab, Languages, appears in every model setup wizard in that schema.

Language Selection in Models

You can add fields that are appropriate to a specific model type. If you add a wide-character text entity or security field to a model, you can populate local-language data to it. The languages populated are determined what you choose in the Languages tab of each model wizard. Fund Summary Model with French Selected as Second Language shows a Fund Summary model with French as the second language for building text fields.

A row of the Fund Summary table, with the French language designation and localized text values is displayed in Fund Summary Table Row.

When you build a model with one or more languages selected, you build a version of the table for each of the selected languages, plus English by default. Each version of your data consists of rows with the appropriate language code in the DMART_LANGUAGE column of the mart table you build. Note:

• If you have also selected one or more alternate base currencies in the Currencies tab of a model wizard, you build (1+L) x (1+C) complete versions of your data for that model's table, where L is the number of non-English languages you selected and C is the number of alternate base currencies you selected. This can increase build time and table rows depending upon the number of languages and currencies selected. If you also build more than one snapshot of the same model, this effect is further increased.

• Only text fields require translation, and only entity and security fields are translated. This means some models are not eligible for any local-language text. However, you can build all models with a given language designation to simplify your report queries for reporting in a particular language.

Convert an Existing Data Mart

Best practice is to use a new Data Mart schema to support multilingual text. This allows you to create new wide-character text fields in the models of that mart. However, you can convert an existing mart that includes non-wide-character text fields to an internationalized mart.

Complete the following procedure to convert and existing mart to support local-language text:

To convert existing mart to support local-language text:

1. From the Manage Mart window, click the Manage Data Marts link and then the Create or Edit link.
   You see the Define Data Mart Instance dialog box.

2. Select the Enable Internationalization check box. The selected mart can now support multilingual text.

3. Click OK to save your changes.

4. Click Close to close the Manage Data Mart Instances dialog box.
   If there are pre-existing text fields which you have changed to Wide Character data type to permit them to report non-English text, complete the following step.

5. In the Manage Mart window, click the Administration link and select Generate Scripts to Internationalize the Mart.

A script is created which you can give to your database administrator. This script can be used to convert table columns in that mart with a Wide Character underlying field attribute to the nvarchar data type as required for non-English reporting. You only need to select this option once when internationalizing a pre-existing mart containing text fields converted to wide character. Post-internationalization, all new wide character fields are set up in their tables as nvarchar.