Creating a Data Load Configuration

Each data load in RDMP has a single [LoadMetadata] object which acts as the root for the configuration. A given configuration will load one or more tables (which must have primary keys). The tables loaded includes all the tables that underly the Catalogues which are ­associated with the [LoadMetadata]. For example if a Catalogue 'Biochemistry' includes a join on two tables 'Header' and 'Results' then a [LoadMetadata] which includes 'Biochemistry' would include both the 'Header' and 'Results' tables.

Create a new Biochemistry CSV file with BadMedicine and add a primary key to the 'Biochemistry' table in your example data. The key should be a composite primary key of:

• chi

• SampleDate

• TestCode

Command Line: This can be done from the CLI using:

./rdmp AlterTableCreatePrimaryKey TableInfo:*Biochemistry* ColumnInfo:*Biochemistry*chi* ColumnInfo:*SampleDate* ColumnInfo:*TestCode*

If your dataset already has a primary key then drop it and Sync the RDMP TableInfo. The SQL to do this will depend on your DBMS. For example in SQL Server it is:

Create a new [LoadMetadata] by clicking New under Data Load on the home screen or the ''Load Configurations' tree collection.

Command Line: This can be done from the CLI using:

./rdmp CreateNewLoadMetadata Catalogue:Biochemistry

All loads require a file system to load data from. This is the case even if data comes from another source (e.g. remote database table). Set the load directory to a location on disk. If you intend to run the load from a seperate server (e.g. automation) then this should be a network drive so that data analysts and automation server can both access it.

Command Line: This can be done from the CLI using:

./rdmp CreateNewDataLoadDirectory LoadMetadata:*Biochemistry c:/temp/biochem