CSV file format for extended data sources

Purpose

Use a CSV file to import the following main types of extended data sources into the catalog: application, stored procedure definitions, and files. The CSV file uses specific syntax for each of these assets to create them in the catalog.

Templates and samples

From the Import Extended Data Sources pane (Catalog > Information Assets > Import Extended Data Sources), download the compressed file that contains CSV files with samples of application, stored procedure definition, and file extended data sources. The samples have all sections and columns in the correct syntax for each asset. Modify these samples to suit your needs.

Syntax

Application, stored procedure definitions, and files have unique sections in the CSV file. Therefore, create separate CSV files for the general types that you want to import. To define an application, stored procedure definition, and file in the CSV file, follow these syntax rules.

Sections

The sections for application, stored procedure definition, and file define the context of each asset to import into the catalog.

You must list the sections in the CSV file in the specified sequence of the extended data source, starting with the top level. The section sequence is listed in the next table.

For example, the CSV file for application has a section for application assets, followed by sections for object type assets, method assets, input parameter assets, and output value assets. If, instead of this sequence, you put the object type section before the application section, the context of an object type asset refers to an application asset that has not yet been created in the catalog. As a result, that row in the Object Type section is not imported.

For each row, enter the section name in the following format:

+++ section_name - Begin +++

+++ section_name - End +++

where section_name corresponds to the extended data source to be imported.

The following table lists the section names and their sequence in the CSV file.

Table 1. Section names and their sequence in the CSV file

Main type of extended data source to import	Value of section_name and the sequence of sections in the CSV file
Application	Application Object type Method The following sections can be in any sequence: Input parameter Output value
Stored procedure definition	Stored procedure definition The following sections can be in any sequence: In parameter Out parameter Inout parameter Result column
File	File

Be sure to put a space before and after the minus sign (-), between the plus signs (+++) and section_name, and between Begin and the plus signs.

Do not put empty rows within a section.

There is no limit to the number of rows in each section.

The name of the extended data source is not case-sensitive. For example, the names myApp1 and MYAPP1 refer to the same asset.

Columns and column headings

The columns of all sections must be in this sequence:

1. Name of the extended data source

   A value in this field is required. If no name is listed, the row is ignored.

   Do not use the following special characters in the name: , " [ ] '.

2. Parents of the extended data source, if any exist

   The sections Application, Stored Procedure Definition, and File do not have this column.

   All other sections contain a column for each of its parent sections, according to its context. For example, the section for an input parameter asset has three columns for parents: application, object type, and method. The section for an in parameter asset has one column for its parent, stored procedure definition.

   If no value is listed in a parent column or if the value does not match the extended data source type, the row is ignored. For example, in a section for a method, MyApp1 is in the Name column, the Object Type column is empty, and NewMethodTest is in the Method column. This row is ignored during import because a method must have a parent of object type.

3. Description

   A value in this field is optional. It is the last column of every section.

Column headings are required and must be in the row immediately after the Begin row for a section.

Do not change the name of the column headings and do not delete columns.

Comments

• To include a comment in a row, insert an asterisk (*). The entire row is ignored during import.
• Put an asterisk only at the beginning of the row. Do not put comments within a row.
• You can include empty rows between sections to make the CSV file easier to read. You do not need an asterisk if the entire row is empty.

Special characters and language support

Commas

If you create a CSV file instead of modifying a template, a comma (,) is the only accepted delimiter.

Quotation marks

Quotation marks (") are required if the value in a field includes non-alphanumeric characters such as mathematical symbols or commas. For example, the following value in the Description field must be enclosed within quotation marks because of the embedded commas: “This is the first, middle, and last description of the application.”

If quotation marks occur in the text, they must be enclosed within quotation marks (" " ").

If the value in a field does not include non-alphanumeric characters, the quotation marks are removed during import.

The quotation marks (') are not valid and the row is ignored during import.

Language support

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

You can type the value in a field in any language, but do not change the column headings and the section headings. The headings must remain in English.