SYNOPSIS

Input

A record stream.

Output

Records with an extra field of your choice to hold an id for each unique value of a field of your choice.

DESCRIPTION

By selecting a field from a list of available fields on the input, it can map each unique value in the selected field to a corresponding unique id that is invented at the time a value is first seen, and kept for a later discoveries of the same value. This value/id pair is stored in a database table, if you do not chose read-only mode of operation. Similar, if the database table already contains value/id pairs, these are looked up and used as ids for values already in the table.

Note

This tool, together with other ODBC tools, has a simple database editor/viewer built in that can be utilized for setting up the table needed for operation.

As a prerequisite you will need to have the IP of your server, and have created a database on the server.

At the time of writing, the following databases are supported: MSSQL, MySQL, PostgreSQL, Sybase.

OPTIONS

Various options

The options are decided by the ODBC driver. Most are intuitive, some are not, but can often just be left at their default.

Each driver may vary in how they name the fields that need to be filled in. One may call the username UID, and another just Username, etc.

Table

The table you wish to access.

Key field

The field to uniquely identify. The tool must be connected to a record stream before any fields are shown.

Number field

The name of the column where ids will be stored.

Cache size

The size of the memory cache. Increase this if the performance is gettings poor. This ensures that more value/id pairs can be stored in the memory and decreases the need for consulting the database.

Read only

The dimension table will not be updated, only read. The ids not in the table will receive the id -1.