CSV columns and column values

The following documentation describes the CSV columns and the column values.

Important: The first line in the data file must be a header line (JobID, RecordID, Name, and Value). The header line is required for better readability.

Column and column values have the following meanings:

JobID
  • ID of the DataStage® job that created the data file. You can use the JobID to identify which records in the sample CSV output file table correspond to a specific running of the IBM® InfoSphere® DataStage job.
  • The JobID must be unique for each job instance.
RecordID
  • RecordID identifies an item in the data file.
  • Rows within a job that have the same record ID correspond to a single item.
  • Rows for a single item must be placed contiguously in the file and not separated from one another.
  • IDs must be unique in a job.
Name
  • Identifies a row as corresponding to a category, catalog, attribute, or action mode. When you import data, specifies the action to take.
  • If the name refers to an attribute, then you must specify the full path for the attribute.
  • If the attribute is localized, then you must ensure that the path is appropriate for the locale. The path of the localized attribute in the data file must also have the locale information like mySpec/name/en_US.
  • With relationship attributes, the related catalog and the primary key of the related item must be specified separated by a pipe delimiter.
  • The order of occurrence is important for the data to be imported correctly.
Value
  • When the "Name" is "Catalog" then "Value" gives the name of the catalog. This overrides the default catalog that is mentioned in the property file.
  • When the "Name" is "Category" then "Value" gives the entire path of the category that includes the hierarchy information. Hence you will be able to support multiple hierarchies.
  • When the "Name" is "Action Mode" then "Value" can be any one of these "Create", "Update", or "Delete". "Action Mode" value is case-sensitive.
    Note: "Create" mode is for creating a new item, which does not exist; "Delete" mode is for deleting an existing item, which is not checked out; and "Update" mode is for updating an existing item, which is not currently checked out. In other cases, an error occurs.
  • For an attribute, "Value" holds the value of the attribute.