CSV file format to import mappings in extension mapping documents

Purpose

Use a CSV file to define and import extract, transform, and load (ETL) transactions that occur outside of IBM® InfoSphere® Information Server.

Syntax

The following syntax rules govern how to correctly define applications, stored procedure definitions, or files in the extension mapping document in CSV format.

Column Headings

The following column headings are required and must be in the first row of the extension mapping document. The headings are:

• Source Columns
• Target Columns
• Rule
• Function
• Specification Description

Do not change the names of the column headings. You can change the column order. This syntax makes the import file compatible with other InfoSphere Information Server products.

If you want to assign custom attributes to a mapping, put the name of each custom attribute in a column heading after the column Specification Description. The custom attribute must exist in the catalog.

Source Columns, Target Columns

If the context of an asset is not in the catalog, that mapping is not imported into the catalog.

The asset name and its context are not case-sensitive. For example, MyApp.SrcObjType.SrcMthd.Param1 and MYAPP.SRCOBJTYPE.SRCMTHD.PARAM1 refer to the same asset and context.

If the asset name contains a comma (,) it must be preceded by a backslash (\). Otherwise, the asset name must not contain an ellipsis (...), quotation marks ("), single quotation marks ('), or other special characters.

Rule

Text that describes the purpose or business reason of the mapping. For example, a rule might be Removes duplicate subscribers from billing list.

Function

Text that describes what the mapping does. For example, a function might be Compares subscriber names in 2 tables.

Specification Description

Text that describes the movement of data, the process in which this movement takes place, or additional information.

Custom attribute columns

The name of the custom attribute is the column heading. The name must be at least one alphanumeric character in length and must be unique. The name must not contain quotation marks (") or single quotation marks (').

A custom attribute can have one or multiple values. Custom attribute values are in the cell of the mapping row that the custom attribute is assigned to. Multiple values must be separated by a semicolon (;).

The custom attribute name and its values are not case-sensitive.

The custom attribute must exist in the catalog. If the custom attribute does not exist in the catalog, the row is imported but the column is ignored.

Special characters and language support

Commas

A comma (,) is the only accepted delimiter.

Quotation marks

The use and treatment of quotation marks (") vary according to the column.

Table 1. Use or treatment of quotation marks according to column

Column	Use or treatment of quotation marks
Source Columns Target Columns	Quotation marks are removed during import.
Rule Function	If quotation marks occur in the text, they must be enclosed within quotation marks (" " ").
Specification Description	Quotation marks are required around the entire text of the field if the text includes non-alphanumeric characters, such as mathematical symbols or commas.

Language support

The CSV file must be in UTF-8 or in ANSI encoding to be compatible with all languages.

You can type the values in any language, but do not change the names of the column headings.